Polysiloxane-bound ligand accelerated catalysis: A modular approach to heterogeneous and homogeneous macromolecular asymmetric dihydroxylation ligands

Article abstract of DOI:10.1039/b406341d

Polysiloxane acts as a modular scaffold for macromolecular reagent development. Two separate components were covalently integrated into one material, one constituent provided reagent functionality, the other modulated solubility. In particular cinchona alkaloid based ligands used in the osmium tetroxide catalyzed asymmetric dihydroxylation (AD) reaction were covalently attached to commercially available polysiloxane. To enhance the reactivity of these polymeric ligands, multifunctional reagents were designed to include both the cinchona alkaloid and an alkoxyethylester solubilizing moiety providing random co-polymers. While the mono-functional materials led to heterogeneous conditions, the bifunctional polymers resulted in homogeneous reaction mixtures. Although both reagent types provided diol products with excellent yield and selectivity (>99% ee in nearly quantitative yield) the homogeneous analog has nearly twice the reactivity. Every repeat unit in the heterogeneous material was functionalized along the polysiloxane backbone while approximately half of these sites contained ligand in the homogeneous version. This approach led to macromolecular catalysts with high loadings of ligand and therefore materials with very low equivalent weights. Although these polymers are highly loaded they do maintain reactivity on a par with their free ligand counterpart. Using straightforward purification techniques (i.e. precipitation, simple filtration, or ultrafiltration) these polymeric ligands were easily separated from diol product and reused multiple times. Polysiloxane is a viable support for the catalysis of AD reactions and may provide a generally useful backbone for other catalytic systems.

Full text of DOI:10.1039/b406341d

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Polysiloxane-bound ligand accelerated catalysis: a modular
approach to heterogeneous and homogeneous macromolecular
asymmetric dihydroxylation ligands
Michael S. DeClue†a and Jay S. Siegel*b
a
Department of Chemistry, University of California-San Diego, La Jolla, California 92093-0358
Organisch-Chemisches Institut, Universität Zürich, Winterthurerstrasse 190, CH-8057 Zürich,
b
Switzerland
Received 28th April 2004, Accepted 26th May 2004
First published as an Advance Article on the web 15th July 2004
Polysiloxane acts as a modular scaffold for macromolecular reagent development. Two separate components were
covalently integrated into one material, one constituent provided reagent functionality, the other modulated solubility.
In particular cinchona alkaloid based ligands used in the osmium tetroxide catalyzed asymmetric dihydroxylation (AD)
reaction were covalently attached to commercially available polysiloxane. To enhance the reactivity of these polymeric
ligands, multifunctional reagents were designed to include both the cinchona alkaloid and an alkoxyethylester solubilizing
moiety providing random co-polymers. While the mono-functional materials led to heterogeneous conditions, the bi-
functional polymers resulted in homogeneous reaction mixtures. Although both reagent types provided diol products
with excellent yield and selectivity (>99% ee in nearly quantitative yield) the homogeneous analog has nearly twice the
reactivity. Every repeat unit in the heterogeneous material was functionalized along the polysiloxane backbone while
approximately half of these sites contained ligand in the homogeneous version. This approach led to macromolecular
catalysts with high loadings of ligand and therefore materials with very low equivalent weights. Although these polymers
are highly loaded they do maintain reactivity on a par with their free ligand counterpart. Using straightforward purification
techniques (i.e. precipitation, simple filtration, or ultrafiltration) these polymeric ligands were easily separated from diol
product and reused multiple times. Polysiloxane is a viable support for the catalysis of AD reactions and may provide a
generally useful backbone for other catalytic systems.
Despite the high solubility of polysiloxane in a wide variety of sol-
Introduction
vents28 cinchona alkaloid derivatized polysiloxanes were only spar-
ingly soluble under AD reaction conditions. However, bi-functional
polymers, with a second functionality that grants solubility, allowed
for the rapid development of soluble polymeric cinchona alkaloid
based reagents for the AD reaction. The PMHS support satisfied the
criteria for an inexpensive material while maintaining high catalyst
loadings giving reagents with relatively low equivalent weights.
The Sharpless asymmetric dihydroxylation (AD) reaction makes use
of catalytic osmium tetroxide in the presence of cinchona alkaloid
derivatives to oxidize olefins enantioselectively and form a wide
variety of vicinal diols.1–4 Immobilization of cinchona derivatives
can enhance the separation of AD reaction products while facilitat-
ing the recovery and reuse of the chiral ligand, as well as the toxic
metal.5–8 The isolation of product can in general be simplified by
converting the reagent into a macromolecule.9,10 AD ligands have
been engineered onto a variety of macromolecular supports provid-
ing heterogeneous5,11 and homogeneous6,12–16 reaction conditions.
Although the recyclable nature of heterogeneous AD ligands is an
advantageous trait, these reagents can suffer from reduced reactiv-
ity leading to extended reaction times and lower selectivity.5,17,18
Desired properties with reactivity comparable to the equivalent free
ligand are possible when AD ligands are tethered to supports that
provide homogeneous reaction conditions.6,13,19–21 However, a major
drawback to homogeneous macromolecular reagents is the excess of
inert material needed to enhance the solubility of the reagent. Cur-
rent homogeneous macromolecular reagents are thus plagued by low
catalyst loadings leading to materials with large equivalent weights
making them relatively unattractive, especially in the industrial set-
tings for which they were designed. Eliminating or reducing the inert
material in scaffolds for polymeric reagents while maintaining high
reactivity and selectivity is the next crucial step in this field.
Results and discussion
Initial siloxane ligand
To understand the feasibility of polysiloxane as a support for
AD ligands a mono-functional cinchona polymer was prepared
(Scheme 1). Although, it was conceivable that quinidine itself
could be directly hydrosilated to polysiloxane without chemical
modification this approach was not successful. Simple modifica-
tion of hydroquinidine hydrochloride provided a compound that
was directly grafted onto polysiloxane. The reaction of hydroquini-
dine hydrochloride with 10-undecenoyl chloride in the presence
of triethylamine provided 1. Hydrosilation of 1 to PMHS was
accomplished using a catalytic amount of dichlorodi(cyclopenta-
dienyl)platinum(II) (Cl2Ptdcp) with toluene as solvent.26,27 The
hydrosilation reaction could be considered >99% complete when
the Si–H bond vanished from a convenient window in the IR spec-
trum (2160 cm−1).29 Polymer purification was achieved by an initial
toluene–hexanes precipitation, followed by a repetitive reprecipita-
tion procedure using tetrahydrofuran as solvent and hexanes as a
precipitant. Disappearance of olefin signals from 1H NMR verified
that the polymeric ligand was clean of starting material. Evapora-
tion of residual solvents from the precipitate yielded 2 as a tan
solid. Polymer 2 was marginally soluble in acetone and essentially
insoluble in hexanes, tert-butyl alcohol and methanol; however, 2
was highly soluble in tetrahydrofuran, toluene, CHCl3 and CH2Cl2,
which allowed for its full characterization.
We thus sought a solution to the drawbacks associated with
macromolecular reagent development by applying knowledge
gained from our previous work with photorefractive polymers.22,23
Specifically, we previously developed a modular approach to dis-
cover and optimize holographic materials using polysiloxane as
a modifiable scaffold. Functionalized olefins were grafted onto
commercial poly(methylhydrosiloxane)24 (PMHS) in mono- and
bi-functional formats, by platinum catalyzed hydrosilation.22,25–28
Ligand 2 was initially screened for AD activity by assaying its
ability to convert trans-stilbene into hydrobenzoin using stan-
† Present Address: Swiss Federal Institute of Technology, Laboratory of
Organic Chemistry, CH-8093, Zurich, Switzerland.
T h i s j o u r n a l i s
©
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 8 7

View Article Online
Scheme 1 Cinchona alkaloid attached to polysiloxane.
dard reaction conditions.5,6 These conditions converted the olefin
acrylates in general cannot be attached to PMHS by this method.
(1.0 equiv.) to diol product using a mixture of 4-methylmorpholine
N-oxide (NMO, 1.5 equiv.), tetraethylammonium acetate tetra-
hydrate (1.0 equiv.), a catalytic amount of OsO4 (0.01 equiv.) and
the polymeric ligand (0.25 equiv.) in an acetone–water (10:1, v/v)
solvent mixture at 0 °C. Upon complete transformation a simple
filtration procedure was used to isolate cleanly the desired product
from the insoluble ligand. Analysis of the diol product using optical
rotation provided a quick assay for the enantiomeric excess (ee).30
Under these conditions the preliminary results for 2 were encourag-
ing since the diol product was obtained in 60% yield with 70% ee
under 24 h. By comparison, the hydrobenzoin product was obtained
in 85% yield and 82% ee using the analogous free ligand under
similar AD conditions.1
Allyl acetate could be grafted to PMHS, but the resulting polymer
was not soluble in acetone.
Ethylene glycol mono-allyl ether was treated with acetyl chloride
to provide 3 followed by attachment to polysiloxane (Scheme 2).
The resulting polymer 4 was soluble in acetone and tert-butyl al-
cohol. The solublizing group 3 was grafted to polysiloxane in the
presence of the cinchona alkaloid derivative 1 in a single synthetic
step to provide the bi-functional polymer 5. The relative ratio of
grafted groups in bi-functional material 5 can be determined by
comparing the integration of unique signals stemming from each
graft type in the 1H NMR spectrum. Furthermore, this ratio can be
regulated empirically by modifying the initial olefin concentrations
prior to grafting.22
The bi-functional polymer 5 was completely soluble in acetone
and tert-butyl alcohol providing homogeneous conditions for typi-
cal AD reactions. However, the material precipitated when a high
concentration of water was added; such a solubility difference al-
lows for simple separation of product from reagent. Ligand 5 gave
diol product from trans-stilbene with similar yield and selectivity as
that provided by 2; however, this reaction was complete in less than
5 h compared to 24 h as found with compound 2 under identicalAD
conditions. Once a component capable of increasing the derivatized
polysiloxane’s solubility and reactivity in typical AD conditions
was identified, the issue of selectivity could be addressed.
Soluble siloxane ligand
With the precedent established for polysiloxane’s ability to func-
tion as a scaffold for solid phase reagents, it became a priority to
improve and refine the process while determining its scope. The
preliminary results for 2 indicated that the reactivity and selectivity
were lower than desired for AD reactions. Increasing the solubility
of this reagent in acetone should increase reactivity, and varying the
steric bulk at the O9 hydroxyl on the cinchona alkaloid is known to
improve the stereoselectivity of ligands used in the AD reaction.31,32
Integrating these aspects into the design of future materials was easy
to implement.
Compound 2 was already slightly soluble in acetone; a slight
modification of this material made the reagent completely soluble
and hence attained the desired increase in reactivity. A search for
bi-functional materials with a component that enhanced solubil-
ity of the polymer material in acetone–water as well as tert-butyl
alcohol/water solvent mixtures was carried out.
Array of ligands
To improve the stereoselectivity of these materials a limited search
for attachable cinchona alkaloid derivatives with steric linkers cova-
lently bound at the O9 hydroxyl was initiated (Scheme 3). Addition
of aromatic substituents to the O9 position of the cinchona alkaloid
has led to increased enantioselectivity in AD reactions.31,32 Initially,
two isomeric phthalic acid moieties were integrated into the O9
position to provide 6 and 7 containing terminal olefins. Changing
the regio-chemistry from a 1,4-terephthalic acid to the 1,2-phthalic
acid derivatized linker probed the polymer chain’s influence on the
selectivity. The larger phthalazine linker was also incorporated into
The search for polysiloxane solubilizing moieties began by
looking at simple commercially available polyether and ester ma-
terials with terminal olefins. Attempts to graft di(ethylene glycol)
ethyl ether acrylate to PMHS did not produce the desired polymer
product. This occurrence, as well as other cases,33 indicated that
Scheme 2 Modular approach towards soluble siloxane ligand.
2 2 8 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8

View Article Online
Scheme 3 Modified cinchona alkaloid ligands with increased steric bulk on the O9 hydroxyl.
the O9 position in a two-step chemical process to provide 9. The
to convert all starting material to product compared to its soluble
steric phthalazine linker may have an enhanced effect on the stereo-
chemical outcome of diols produced when these solid phase ligands
are used to dihydroxylate olefins.
counterpart 11 under identical conditions.
Limited concentration of OsO4
Once compounds 6, 7 and 9 were prepared they were individually
attached to PMHS, using the standard conditions already described,
to provide mono-functional materials 10, 12 and 14 accordingly
(Fig. 1). Furthermore, the ligands 6, 7 and 9 were separately grafted
together with 3 to provide bi-functional materials 11, 13, and 15, re-
spectively. As before, the bi-functional materials 11, 13 and 15 were
soluble under common AD conditions, whereas mono-functional
polymers 10, 12 and 14 were not.
The new class of ligands 10–14 (0.25 equiv.) was individually
tested for their ability to dihydroxylate trans-stilbene (1.0 equiv.)
using OsO4 (0.01 equiv.), NMO (1.5 equiv.) and tetraethyl-
ammonium acetate tetrahydrate (1.0 equiv.) in an acetone/water
(10:1, v/v) solvent mixture at 0 °C (Table 1). The modified materi-
als have improved reactivity and selectivity compared to the initial
ligands 2 and 5. The second generation of polysiloxane materials
were capable of converting trans-stilbene to hydrobenzoin in as
short as 2 h with >80% isolated yield and >80% ee. However, after
an initial increase in selectivity and yield (entry 1) each subsequent
modification of the steric linker (i.e. ligands 12 and 14) reflected
essentially no benefit. Furthermore, the reactivity difference de-
creased between heterogeneous and homogeneous materials. For
example, insoluble ligand 10 required only 30 additional minutes
To understand the performance of polysiloxane bound AD ligands
better, the dependence of OsO4 concentration relative to hetero-
geneous ligand 10 and homogeneous ligand 11 was investigated
(Table 2). The soluble ligands showed dramatically improved re-
activity over the insoluble ligands when a limited concentration of
OsO4 was used. For example, complete conversion to diol product
using 0.1 mol% OsO4 required 46 h with the heterogeneous ligand
10 (entry 1) compared to 29 h for the homogeneous ligand 11 (entry
4) under matching conditions.
Bis-cinchona alkaloid ligands
A slight modification of the synthesis Sharpless used to prepare
(DHQD)2PHAL3,34 was applied to make a dimeric cinchona li-
gand with a terminal olefin capable of attachment to polysiloxane
(Scheme 4). Exchange of one chlorine atom for quinidine on 1,4-di-
chlorophthalazine to provide 16 occurred via nucleophilic aromatic
substitution at slightly elevated temperature. Hydroquinidine was
then covalently linked to 16 by displacement of the second chloride
on the phthalazine heterocycle at a higher temperature yielding 17.
Compound 17 can be attached to polysiloxane using the standard
conditions to provide both mono- and bi-functional polymers. The
Fig. 1 Modified cinchona alkaloid ligands attached to polysiloxane.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 8 9

View Article Online
Table 1 Dihydroxylation reactions using polysiloxane ligandsa
Entry
Ligand
Olefin
OsO4 (mol%)
Amount of ligandb (mol%)
Reaction time/h
Yieldc (%)
eed (%)
1
2
3
4
5
10
11
12
13
14
1
1
1
1
1
25
25
25
25
25
2.5
2.0
2.0
2.0
3.0
85
77
72
83
86
83
79
79
82
79
a Reaction conditions: NMO (1.5 equiv.) as secondary oxidizer, Et4N+CH3CO2− (1.0 equiv.), acetone–water (10:1, v/v), 0–4 °C. bAmount of ligand used was
based on alkaloid incorporation. c Isolated yield. d Enantiomeric excess was determined by comparing optical rotation with the literature values.3,30
a
Table 2 Dihydroxylation reactions with polysiloxane ligands using limited OsO4
Entry
Ligand
Olefin
OsO4 (mol%)
Amount of ligandb (mol%)
Reaction time/h
Yieldc (%)
eed (%)
1
2
3
4
5
6
10
10
10
11
11
11
0.1
0.5
1.0
0.1
0.5
1.0
25
25
25
25
25
25
46
79
85
85
72
73
77
69
82
83
76
85
79
3.5
2.5
29
3.0
2.0
a Reaction conditions: NMO (1.5 equiv.) as secondary oxidizer, Et4N+CH3CO2− (1.0 equiv.), acetone–water (10:1, v/v), 0–4 °C. bAmount of ligand used was
based on alkaloid incorporation. c Isolated yield. d Enantiomeric excess was determined by comparing optical rotation with the literature values.3,30
Scheme 4 (DHQD)2PHAL attached to polysiloxane.
bi-functional polymer 19 led to homogeneous conditions when
subjected to common AD solvents while the mono-functional
polymer 18 gave heterogeneous conditions. Reagent 19 has a
much lower equivalent weight compared to similar homogeneous
macromolecular reagents designed for the AD reaction13 and only
about 30% more than the simple free ligand. The siloxane based re-
agent 19 has an equivalent weight of around 1000 g mol−1 compared
to >5000 g mol−1 for the analogous material designed by Janda and
800 g mol−1 for (DHQD)2PHAL. This low equivalent weight is a
simple result of the high loading factor possible in PMHS.
The mono-functional heterogeneous ligand 18 was evaluated for
its ability to convert trans-stilbene to hydrobenzoin (Table 3). This
polysiloxane bound dimeric ligand was superior to all previous si-
loxane materials tried.At first, the material was tested using standard
conditions established by Sharpless5 for two different secondary oxi-
dants.35,36 Based on superior yield and ee it was rapidly established
that the K3Fe(CN)6 conditions were optimal for this new ligand.
Enantioselectivity equivalent to the free ligand (DHQD)2PHAL was
achieved using polymeric ligand 18 and K3Fe(CN)6 as secondary ox-
idant.3 Next, experiments to establish the minimal amount of ligand
needed to achieve optimal results were conducted. It was shown
that ligand concentrations as low as 5 mol% (based on the number
of active sites) produced a desired outcome with 1 mol% OsO4. By
comparison, the commonly usedAD-mix  contains 10 mol% of the
ligand (DHQD)2PHAL and 0.2 mol% osmium.3
To further evaluate the scope and utility of these materials an
array of olefins was dihydroxylated with polysiloxane bound di-
meric cinchona alkaloid ligands 18 and 19. Initially, a variety of
olefin substrates based on a styrene core were used to determine the
scope of the heterogeneous polysiloxane ligand 18 (Table 4). The
majority of substrates were screened using K3Fe(CN)6 as the opti-
mal secondary oxidant with 5 mol% of ligand based on the alkaloid
incorporation to the polymer support. These conditions proved to
be very good for oxidizing olefins having multiple substitutions.
However, the ligand’s performance dropped as the degree of substi-
tution on the alkene was decreased. The data showed that ligand 18
worked best with 1,2-trans- and 1,1-disubstituted alkenes followed
by styrene and last 1,2-cis-disubstituted olefins.
Homogeneous bi-functional ligand 19 was screened under several
different conditions for its ability to convert olefin to diol product
(Table 5). Low yields of hydrobenzoin product were observed when
25 mol% of ligand was used (entry 1). When a lower concentration
of ligand was employed (entries 2–3) the hydrobenzoin yield went up
and the ee remained high. The soluble siloxane-bound ligands were
quickly separated from diol product by precipitating the crude reac-
tion mixture into an excess of water. The low yields could arise from
2 2 9 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8

View Article Online
Table 3 Dihydroxylation reactions using the heterogeneous ligand 18
Entry
Olefin
OsO4 (mol%)
Amount of liganda (mol%)
Secondary oxidantb
Reaction time/h
Yieldc (%)
eed (%)
1
2
3
4
5
1
1
1
1
1
25
25
10
5
NMO
3
24
24
24
3
87
94
92
98
85
93
99
95
93
91
K3Fe(CN)6
K3Fe(CN)6
K3Fe(CN)6
NMO
5
aAmount of ligand used was based on alkaloid incorporation. b With NMO (1.5 equiv.) acetone–water (10:1, v/v) solvent was used at 0–4 °C and with
K3Fe(CN)6 (3.0 equiv.) tert-butyl alcohol–water (1:1, v/v) solvent was used at 0–4 °C. c Isolated yield. d Enantiomeric excess was determined by comparing
optical rotation with the literature values.3,30
Table 4 Scope of heterogeneous ligand 18
Entry
Olefin
OsO4 (mol%)
Amount of liganda (mol%)
Secondary oxidantb
Reaction time/h
Yieldc (%)
eed (%)
1
2
1
1
5
5
NMO
K3Fe(CN)6
2.3
18
92
92
84
92
3
4
1
1
5
5
NMO
K3Fe(CN)6
2
2
89
91
64
92
5
1
5
K3Fe(CN)6
24
91
96
6
7
1
1
5
5
K3Fe(CN)6
K3Fe(CN)6
18
8
73
70
85
36
aAmount of ligand used was based on alkaloid incorporation. b With NMO (1.5 equiv.) acetone–water (10:1, v/v) solvent was used at 0–4 °C and with
K3Fe(CN)6 (3.0 equiv.) tert-butyl alcohol–water (1:1, v/v) solvent was used at 0–4 °C. c Isolated yield. d Enantiomeric excess was determined by comparing
optical rotation with the literature values.3,30,37,38
Table 5 Dihydroxylation reactions using the homogeneous ligand 19
Entry
Olefin
OsO4 (mol%)
Amount of liganda (mol%)
Secondary oxidantb
Reaction time/h
Yieldc (%)
eed (%)
1
2
3
4
1
1
1
1
25
10
5
NMO
NMO
NMO
K3Fe(CN)6
2.5
3.0
2.0
21
65
71
84
88
96
91
92
83
5
5
6
1
1
5
5
NMO
K3Fe(CN)6
2.5
14
91
93
89
92
7
8
1
1
5
5
NMO
K3Fe(CN)6
1
2
86
93
76
90
9
1
5
K3Fe(CN)6
12.5
90
91
aAmount of ligand used was based on alkaloid incorporation. b With NMO (1.5 equiv.) acetone–water (10:1, v/v) solvent was used at 0–4 °C and with
K3Fe(CN)6 (3.0 equiv.) tert-butyl alcohol–water (1:1, v/v) solvent was used at 0–4 °C. c Isolated yield. d Enantiomeric excess was determined by comparing
optical rotation with the literature values.3,30,37
inclusion of the diol by the precipitated polymer ligand 19 during the
workup.39 Higher yields of diols were observed as product polarity
increased, consistent with the assumption of co-precipitation.
While the selectivity and yield remained essentially the same for
the two ligands, formation of diol product using the heterogeneous
catalyst 18 (Table 4) was in general slower compared to the reaction
with homogeneous ligand 19 (Table 5) under identical conditions;
in some specific cases the reaction times for 19 were half of what
was needed for 18. Polysiloxane based catalysts 18 and 19 have a
broad scope, in the conversion of many substrates nearly quantita-
tively to a single enantiomer using about 5 mol% catalyst.
recovery of the heterogeneous ligand with partial recovery of OsO4.
When 25 mol% of 18 was used to catalyze the AD of trans-stilbene
under NMO conditions (Table 3, entry 1) the recovered ligand can
be reused to catalyze the identical transformation. Recycled ligand
18 can provide hydrobenzoin product after reacting for 2 days with
88% yield and 94% ee without the addition of more OsO4. Although
2 days is significantly longer than the initial reaction time of 3 h the
product was obtained with comparable quantity and purity. A wide
variety of cinchona alkaloid supports also suffer from the general
problem of OsO4 leaching11,13,40,41 with only limited examples suc-
cessfully retaining the metal.42,43 In general, successive trials of
recovered macromolecular bound AD ligands require additional
OsO4 to retain high reactivity and selectivity.
Recycling ligands
To establish the extent these materials can be recycled ligand
18 (0.1 equiv.) was recovered and reused multiple times for
dihydroxylation of 1.0 equiv. of trans-stilbene (Fig. 2). For each
consecutive iteration of the recycled ligand a fresh aliquot of OsO4
(0.01 equiv.) was added. Using K3Fe(CN)6 (3.0 equiv.) and tert-
An important aspect of solid phase reagents is that they can be
reused multiple times to decrease waste and reduce the cost of a
particular synthetic transformation per mole of reagent. Simple
filtration of reactions employing 18 allowed essentially complete
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 9 1

View Article Online
butyl alcohol–water (1:1, v/v) as the solvent, four iterations of the
ligand were made to prepare hydrobenzoin product without any loss
of reactivity or selectivity. For example, in the fourth trial only 26 h
was needed to reach completion compared to the initial reaction
requiring 24 h.
Conclusions
Polysiloxane is a viable support for solid phase synthesis, as well
as a useful material for mediating AD reactions. It is inexpensive
and simple to modify, allowing for rapid innovations. Unlike typi-
cal polymer supports this material allows for high catalyst loadings
while maintaining excellent reactivity. Macromolecular reagents
with comparatively low equivalent weights (500–1000 g mol−1) are
possible with polysiloxane. The performance of the siloxane bound
ligands was equivalent to the analogous free reagent, and also
maintained the desired macromolecular properties. Heterogeneous
and homogeneous ligands were easily separated from product and
were reused multiple times. While heterogeneous siloxane ligands
have an enhanced affinity towards OsO4, the homogeneous analogs
displayed greater reactivity due to their superior solubility. Expan-
sion of siloxane bound reagents beyond the AD reaction should be
particularly interesting for industrial chemical processes.
Experimental
General methods
NMR spectroscopy (1H and 13C) was recorded on Varian (Mercury
300 MHz, Mercury 400 MHz, or Unity 500 MHz) spectrometers.
1H NMR spectra were referenced to tetramethylsilane (TMS) at
0.00 ppm. 13C NMR spectra were referenced to 77.0 ppm in chloro-
form-d (CDCl3) and 53.5 ppm in dichloromethane-d2 (CD2Cl2).
High-resolution mass spectra (HRMS) were obtained from the
University of California at Riverside’s mass spectrometry facility
in the fast atom bombardment (FAB) or electron impact (EI) mode.
Ultraviolet spectroscopy (UV-vis) was recorded on a Perkin-Elmer
Lambda 19 UV/vis/NIR spectrophotometer. Infrared (IR) spectra
were recorded on a Perkin-Elmer 1420 IR spectrophotometer.
Melting points (mp) were obtained on a Mel-Temp melting point
apparatus and were recorded uncorrected.
All experiments were carried out under argon in freshly distilled
solvents under anhydrous conditions unless otherwise noted. Com-
mercial chemicals were acquired from Sigma–Aldrich with the
exception of dichlorodi(cyclopentadienyl)platinum(II) that was
obtained from Strem Chemicals, Inc. Commercial chemicals were
used as supplied unless otherwise stated. Osmium tetroxide (OsO4)
was stored at −5 °C under an atmosphere of argon. Tetrahydrofuran
(THF) was distilled from sodium/benzophenone; toluene and
dichloromethane (CH2Cl2) were distilled from calcium hydride;
triethylamine was distilled from NaOH, and pyridine was distilled
from KOH. Preparative column chromatography was performed
with silica gel (230–425 mesh) from Fisher Scientific Company.
Thin-layer chromatography (TLC) was performed on aluminium
backed silica gel 250 m F254 plates from Whatman. Centrifugal
filter (2 mL capacity) with a nominal molecular weight limit cel-
lulose membrane was acquired from Millipore (catalogue number
UFC4 LTK 25).
Fig. 2 Recycling heterogeneous ligand 18 using a fresh aliquot of OsO4
with each iteration and K3Fe(CN)6 as the secondary oxidant.
When the homogeneous ligand 19 was used for the AD of olefins
its recovery and reuse was possible by adding an excess of H2O
(≥4/1, v/v) to the completed reaction. The polymer ligand was
insoluble when subjected to a high concentration of H2O and was
selectively precipitated. Filtration of the precipitate allowed for the
recovery of the ligand while extraction of the mother-liquor with
CH2Cl2 provided the diol product. Although this method recovered
a significant amount of the ligand (>90%) it unfortunately retained
only a fractional amount of OsO4. When 25 mol% of 19 was used
to dihydroxylate trans-stilbene under the NMO conditions (Table
5, entry 1) the recovered ligand will not catalyze a second itera-
tion unless additional OsO4 was added. Using the recovered ligand
with fresh substrate and a secondary oxidizer resulted in detection
of very little hydrobenzoin product even after a reaction time of 2
weeks. However, when an additional aliquot of OsO4 (0.01 equiv.)
was added the reaction was complete in only 3 h with a nearly iden-
tical outcome as the initial trial.
Ultrafiltration
Separation methods ranging from selective precipitation to
multi-phasic techniques have been employed to isolate various
homogeneous macromolecule reagents from their correspond-
ing products.44–47 We investigated ultrafiltration as an alternative
method to help separate the diol product from the homogeneous
macromolecular ligand 19. A 30 kD centrifugal concentrator com-
patible with organic solvents was successfully used to isolate diol
product from the soluble siloxane catalyst. The apparatus used had
a filtering volume capacity of 2 mL that allowed the reactions to
be run inside the filtering unit. trans-Stilbene was dihydroxylated
inside the ultrafiltration device. Initially, the polymer ligand 19
(0.011 mmol) was pre-filtered from an acetone solution to remove
the low molecular weight oligomers. Then the dihydroxylation re-
action proceeded by adding all necessary components directly into
the filtration device. Using the standard conditions for NMO as a
secondary oxidant the dihydroxylation of trans-stilbene yielded
hydrobenzoin in 86% yield and 91% ee in 3 h. The use of the cen-
trifugal concentrator allowed the trivial separation of diol product
from ligand in a reasonable amount of time. In a second iteration
the apparatus containing filtered ligand was reused to dihydroxylate
trans-stilbene to provide hydrobenzoin with nearly identical results
(87% yield, 94% ee, 3 h). In principle, a larger version of this ap-
paratus could be used to prepare diol product cleanly and efficiently
on an industrial scale with reduced waste.
Optical rotation measurements were performed on a Perkin-
Elmer model 241 polarimeter. The specific rotation was determined
using the following formula:
o
α
lc
α t
=
[ ]λ
where [] = specific rotation, t = temperature in degrees Celsius,
 = wavelength of incident light (for the sodium D lamp, indicated
simply by “D”,  = 589 nm, the yellow emission line of hot so-
dium vapor),  = observed optical rotation in degrees, l = length
of sample container in decimeters (its value was 1, i.e. 10 cm),
c = concentration (grams per milliliter of solution). The optical
purity (op) was determined by the following formula:










α
[ ]observed
α
op = 100


[ ]


pure enantiomer 
The rough equality of op and ee for the reaction was confirmed by
Mosher’s analysis48,49 for a few representative diol products.
2 2 9 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8

View Article Online
General hydrosilation procedure
alkaloid incorporation) and acetone (2 mL). This mixture was
stirred at room temperature until the polymeric ligand was com-
pletely dissolved (ca. 5 min). Then the apparatus was centrifuged to
a concentrated volume of ca. 30 L (ca.1.0 h). The mother-liquor
was discarded since it contained only low molecular weight oligo-
mers. Then the filtration cup was charged with 4-methylmorpho-
line N-oxide (20 mg, 0.165 mmol), tetraethylammonium acetate
tetrahydrate (29 mg, 0.11 mmol), acetone (1.8 mL), water (0.2 mL)
and OsO4 in tert-butyl alcohol (14 L of OsO4 2.5 wt% solution,
0.0011 mmol). After stirring the solution for 5 min at 0–4 °C, trans-
stilbene (21 mg, 0.11 mmol) was added in one portion. The reaction
mixture was stirred vigorously at 0–4 °C until the olefin disappeared
as judged by TLC. Then the apparatus was centrifuged to a concen-
trated volume of ca. 30 L (ca.1.0 h). The filter cup was washed
with acetone (3 × 2 mL) and centrifuged to a concentrated volume
each time. The combined mother-liquors were worked up similar to
the heterogeneous description below considering that the polymer
ligand was thoroughly removed.
The procedure was adapted from Strohriegl.26,27 To a two-neck
round-bottom flask equipped with condenser, gas inlet, stir bar, and
glass stopper was added toluene and poly(methylhydrosiloxane)24
Mn
(PMHS, 0.2–0.3 M, 1 equiv.,
= 9500) under a flow of argon.
The olefin(s) (1.1–1.3 equiv.) was then added to the mixture
followed by dichlorodi(cyclopentadienyl)platinum(II) (1.0 mg,
0.0025 mmol). The mixture was stirred at elevated temperature
(60–65 °C). The reaction progress was monitored using IR spec-
troscopy. After an initial reaction time of 20 h, an aliquot of the neat
reaction solution was evaporated on NaCl plates, and the IR spec-
trum was recorded to follow the disappearance of the Si–H stretch
at 2150 cm1. Additional dichlorodi(cyclopentadienyl)platinum(II)
(ca. 1 mg) was added at regular intervals until the IR spectrum of
an aliquot showed no residual Si–H stretch. If the reaction was in-
complete, then additional dichlorodi(cyclopentadienyl)platinum(II)
(ca. 1 mg) was added. This cycle was continued at regular time in-
tervals (ca. 1 h) until the silicon-hydrogen signal was no longer
present in the IR spectrum. In most cases the reactions required
little or no additional dichlorodi(cyclopentadienyl)platinum(II) to
reach completion. Upon completion, the reaction was cooled to
room temperature. Then the reaction solution was added dropwise
into an excess of hexanes. Unless otherwise stated this caused the
polymeric product to precipitate. The precipitate was collected by
centrifuge and decanted. Then the precipitated polymer residue was
dissolved in a minimal amount of THF (amount varies on substrate
and scale) needed to completely dissolve the crude polymer. This
THF solution was precipitated again into an excess of hexanes. The
precipitation process was continued until the polymer residue is free
of monomers as determined by 1H NMR. Unless otherwise stated,
a total of three precipitations were all that was needed for clean
product. The residual solvents were removed from the polymer
residue under reduced pressure. In most cases the products become
a solid foam after solvents were evacuated by vacuum. In the case
of bi-functional polymers the ratio was determined by integration
of unique signals from 1H NMR.
Asymmetric dihydroxylation work-up procedure
Heterogeneous ligands. Once the reaction was complete the vial
was centrifuged and the reaction solution decanted leaving behind
the heterogeneous ligand ready for reuse without further purifica-
tion. The mother-liquor was treated with solid sodium metabisulfite
(500 mg) for 5 min followed by Na2SO4 to remove water. All solids
were removed by filtration and washed with CH2Cl2 (3 × 10 mL).
The combined filtrates were evaporated to dryness giving nearly
pure dihydroxylated product. Further purification using silica gel
column chromatography was performed when needed.
Homogeneous ligands. To the completed reaction was added
H2O (15 mL) to selectively precipitate the siloxane ligand. This
cloudy solution was centrifuged then decanted leaving behind the
recycled ligand. No further purification of the ligand was performed.
The mother-liquor was extracted using CH2Cl2 (3 × 10 mL). The or-
ganic layers were combined and stirred over sodium metabisulfite
(500 mg) for 15 min followed by Na2SO4. All solids were removed
by filtration and washed with CH2Cl2 (3 × 10 mL). The combined
filtrates were evaporated to dryness giving nearly pure dihydroxyl-
ated product. Further purification of the dihydroxylated product
using silica gel column chromatography was performed when
needed.
General asymmetric dihydroxylation procedure
NMO secondary oxidant. The procedure was adapted from
Janda.6 A 10 mL screw-top vial equipped with magnetic stir bar was
charged with the polymer catalyst (0.011–0.55 mmol, 0.05–0.25
equiv. based on alkaloid incorporation), 4-methylmorpholine N-
oxide (39 mg, 0.33 mmol), tetraethylammonium acetate tetrahydrate
(57 mg, 0.22 mmol), acetone (3.6 mL), water (0.4 mL) and OsO4 in
tert-butyl alcohol (28 L of OsO4 2.5 wt% solution, 0.0022 mmol).
After stirring the solution for 5 min at 0–4 °C, the olefin (0.22 mmol)
was added in one portion. The reaction mixture was stirred vigor-
ously at 0–4 °C until the olefin completely disappeared as judged by
TLC. The work-up procedure varied depending if heterogeneous or
homogeneous ligands were employed (see work-up).
Undec-10-enoic
acid
((2R,5R)-(+)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl
ester (1). To a 250 mL two-neck round-bottom flask equipped with
stir bar, 100 mL dropping addition funnel, condenser and gas inlet
was charged hydroquinidine hydrochloride (3.34 g, 9.2 mmol),
CH2Cl2 (50 mL) and triethylamine (6.4 mL, 46 mmol) under a
flow of argon. A solution of 10-undecenoyl chloride (2.96 mL,
13.8 mmol) in CH2Cl2 (10 mL) was prepared in the addition funnel
and added dropwise into the stirred reaction solution at 0 °C. The re-
action was then brought to room temperature and stirred overnight.
Then the reaction mixture was carefully poured into a separatory
funnel containing water (100 mL). The organic layer was separated
and washed with saturated aqueous NaHCO3 (1 × 50 mL) solution
and dried over MgSO4. The solvent was removed using a rotary
evaporator to yield a crude yellow oil. Purification via flash chro-
matography using silica as absorbent (acetone–EtOAc, 4:1) yielded
a light yellow oil (3.8 g, 84%), []2D2 = 34.4 (c 0.53, EtOH). 1H NMR
(400 MHz, CDCl3):  0.92 (t, J = 7.2 Hz, 3H), 1.22–1.67 (m, 18H),
1.71–1.81 (m, 2H), 2.02 (q, J = 6.8 Hz, 2H), 2.38 (t, J = 6.4 Hz,
2H), 2.62–2.82 (m, 3H), 2.85–2.94 (m, 1H), 2.28 (q, J = 8.4 Hz,
1H), 3.91 (s, 3H), 4.92 (dd, J = 10.0, 0.8 Hz, 1H), 4.98 (dd, J = 17.2,
2.0 Hz, 1H), 5.74–5.85 (m, 1H), 6.53 (d, J = 7.6 Hz, 1H), 7.32–7.36
(m, 1H), 7.37 (d, J = 2.4 Hz, 1H), 7.42 (d, J = 2.4 Hz, 1H), 8.01
(d, J = 9.2 Hz, 1H), 8.73 (d, J = 4.8 Hz, 1H). 13C NMR (100 MHz,
CDCl3):  12.05, 23.43, 24.90, 25.47, 26.07, 27.20, 28.84, 29.00,
29.08, 29.15, 29.24, 33.73, 34.46, 37.31, 49.87, 50.71, 55.48, 59.01,
73.22, 101.20, 113.95, 118.39, 121.59, 126.84, 131.51, 138.81,
K3Fe(CN)6 secondary oxidant. The procedure was adapted from
Sharpless.5 To a 10 mL screw-top vial equipped with magnetic stir
bar and wrapped in aluminium foil was charged with the polymer
catalyst (0.011–0.55 mmol, 0.05–0.25 equiv. based on alkaloid
incorporation), potassium ferricyanide (217 mg, 0.66 mmol), po-
tassium carbonate (91 mg, 0.66 mmol), tert-butyl alcohol (2 mL),
water (2 mL), and OsO4 in tert-butyl alcohol (28 L of OsO4
2.5 wt% solution, 0.0022 mmol). After stirring the solution for
10 min at 0–4 °C, the olefin (0.22 mmol) was added in one por-
tion. The reaction mixture was stirred vigorously at 0–4 °C until
the olefin disappeared as judged by TLC. The work-up procedure
varied depending if heterogeneous or homogeneous ligands were
employed (see work-up).
Centrifugal filter apparatus. To the filtration cup section of a
2 mL capacity centrifugal filter (30,000 nominal molecular weight
limit regenerated cellulose membrane, Millipore catalogue number
UFC4 LTK 25) equipped with magnetic stir bar was charged with
the polymer catalyst 19 (12 mg, 0.011 mmol, 0.1 equiv. based on
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 9 3

View Article Online
143.75, 144.45, 147.14, 157.50, 172.42. HRMS-EI+ m/z: found
s, 4H). 13C NMR (100 MHz, CDCl3):  −2.65, −0.32, 12.16, 13.57,
492.3349; calc. (C31H44N2O3) 492.3352.
18.03, 21.05, 23.15, 23.92, 25.45, 25.92, 26.68, 27.35, 27.49, 29.75,
29.81, 29.95, 30.10, 30.19, 34.03, 35.21, 36.70, 50.19, 50.95, 55.35,
60.02, 63.97, 68.92, 73.07, 74.23, 102.38, 118.34, 122.44, 128.35,
132.54, 145.01, 145.37, 147.82, 159.73, 170.65, 173.12. IR (neat):
Poly(methylundecanoic
acid
((2R,5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl ester
siloxane) (2). Refer to general hydrosilation procedure. The reaction
was run using 1 (2.6 g, 5.2 mmol), PMHS (280 L, 4.7 mmol), tolu-
ene (30 mL) and dichlorodi(cyclopentadienyl)platinum(II) (20 mg,
0.05 mmol) to provide a tan solid (1.1 g, 48%), equiv. wt. 552. 1H
NMR (500 MHz, CD2Cl2):  −0.02–0.30 (br s, 3H), 0.44–0.64 (br
s, 2H), 0.82–1.01 (br s, 3H), 1.08–1.90 (br m, 24H), 2.25–2.50 (br
s, 2H), 2.57–2.81 (br s, 3H), 2.83–3.02 (br s, 1H), 3.16–3.35 (br s,
1H), 3.82–4.13 (br s, 3H), 6.32–6.77 (br s, 1H), 7.18–7.41 (br m,
2H), 7.43–7.62 (br s, 1H), 7.86–8.14 (br s, 1H), 8.54–8.82 (br s, 1H).
13C NMR (125 MHz, CD2Cl2):  −2.40, 12.25, 18.13, 23.59, 25.45,
25.93, 26.64, 27.32, 27.46, 29.70, 29. 81, 29.95, 30.10, 30.15, 34.03,
34.95, 37.70, 50.19, 51.15, 56.33, 60.22, 73.45, 102.28, 119.14,
122.44, 127.73, 132.34, 145.10, 145.30, 148.02, 158.65, 173.21. IR
(neat): Si–O 1010, CO 1725, C–H 2895 cm−1.

Si–O 1040, CO1735, C–H 2910 cm−1.
Terephthalic acid 1-allyl ester 4-[((2R,5R)-(−)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester (6). A 250 mL two-neck round-bottom flask equipped with
stir bar, rubber septum, condenser and gas inlet was charged with
terephthaloyl chloride (2.03 g, 10 mmol), CH2Cl2 (50 mL) and tri-
ethylamine (5.6 mL, 40 mmol) under a flow of argon. Allyl alcohol
(680 L, 10 mmol) was added via syringe to the reaction solution
at 0 °C in a dropwise fashion. After the addition was complete the
reaction was stirred for 10 min at room temperature. Then hydro-
quinidine hydrochloride (3.63 g, 10 mmol) was added in one por-
tion. The rubber septum was exchanged with a glass stopper and
the reaction mixture was heated to 35 °C with vigorous stirring.
The consumption of hydroquinide hydrochloride was monitored
using TLC (SiO2, acetone–EtOAc, 3:1). After 1 h the reaction
was cooled to room temperature followed by the slow addition of a
saturated aqueous solution of NaHCO3 (100 mL). This mixture was
then extracted using CH2Cl2 (3 × 50 mL). All organic layers were
combined and dried over MgSO4. The solvent was removed using
a rotary evaporator to yield a crude white product. Purification via
flash chromatography with silica as absorbent (acetone–EtOAc,
3:1) yielded a white solid (3.4 g, 66%), mp 52–54 °C, []2D2 = −76.1
(c 1.08, EtOH). 1H NMR (500 MHz, CDCl3):  0.90 (t, J = 6.5 Hz,
3H), 1.40–1.53 (m, 4H), 1.55–1.64 (m, 2H), 1.75 (s, 1H), 1.91 (t,
J = 11.5 Hz, 1H), 2.73 (t, J = 10.0 Hz, 2H), 2.82 (t, J = 12.0 Hz,
1H), 2.93 (t, J = 13.0 Hz, 1H), 3.46 (q, J = 8.0 Hz, 1H), 3.99 (s,
3H), 4.85 (d, J = 4.5 Hz, 2H), 5.30 (d, J = 11.0 Hz, 1H), 5.42 (d,
J = 17.5 Hz, 1H), 5.99–6.09 (m, 1H), 6.79 (d, J = 7.0 Hz, 1H), 7.39
(d, J = 8.5 Hz, 1H), 7.46 (d, J = 4.5 Hz, 1H), 7.54 (s, 1H), 8.04
(d, J = 9.0 Hz, 1H), 8.14–8.23 (m, 4H), 8.76 (d, J = 4.0 Hz, 1H).
13C NMR (125 MHz, CDCl3):  11.62, 23.37, 25.16, 25.84, 26.89,
37.09, 49.69, 50.58, 55.31, 59.25, 65.77, 74.52, 101.23, 118.49,
118.55, 121.75, 126.94, 129.49, 129.64, 131.71, 131.72, 133.44,
134.25, 143.55, 144.67, 147.34, 157.86, 164.79, 165.10. HRMS-EI+
[M + H]+: found 515.2554; calc. (C31H35N2O5) 515.2546.
Acetic acid 2-allyloxyethyl ester (3). A 250 mL two-neck
round-bottom flask equipped with stir bar, rubber septum, con-
denser and gas inlet was charged with ethylene glycol monoallyl
ether (5.3 mL, 50 mmol), CH2Cl2 (85 mL), and triethylamine
(13.9 mL, 100 mmol) under a flow of argon. Acetyl chloride
(4.3 mL, 60 mmol) was added via syringe to the reaction solution
at 0 °C in a dropwise fashion. The reaction mixture was warmed to
room temperature and stirred overnight. Then the reaction solution
was poured slowly into a saturated aqueous solution of NaHCO3
(50 mL). This mixture was extracted with Et2O (2 × 75 mL). The
ether layers were combined, dried over MgSO4, filtered and evapo-
rated under reduced pressure to yield a yellow liquid. Purification
via simple vacuum distillation provided a colorless liquid (5.2 g,
1
72%), bp 30–32 °C (1 mm Hg). H NMR (400 MHz, CDCl3): 
2.09 (s, 3H), 3.62 (t, J = 4.8 Hz, 2 H), 4.04 (dt, J = 5.6, 1.2 Hz,
2H), 4.23 (t, J = 4.8 Hz, 2 H), 5.21 (dd, J = 10.0, 1.6 HZ, 1H), 5.30
(dd, J = 17.2, 2.0 Hz, 1H), 5.86–5.97 (m, 1H); lit.50  2.02 (s, 3H),
3.55 (t, 2H), 3.96 (d, 2H), 4.15 (t, 2H), 5.1–6.1 (m, 3H). 13C NMR
(100 MHz, CDCl3):  21.0, 63.6, 67.8, 72.1, 117.3, 134.2, 170.8.
Poly(methylacetic acid 2-propoxyethyl ester siloxane) (4).
Refer to general hydrosilation procedure. The reaction was run using
3 (1.0 g, 6.9 mmol), PMHS (340 L, 5.7 mmol), toluene (12 mL)
and dichlorodi(cyclopentadienyl)platinum(II) (1 mg, 0.0025 mmol)
to provide a colorless oil (860 mg, 74%), equiv. wt. 204. 1H NMR
(300 MHz, CDCl3):  −0.10–0.20 (br s, 3H), 0.44–0.58 (br m,
2H), 1.54–1.68 (br m, 2H), 1.98–2.18 (br s, 3H), 3.36–3.46 (br t,
J = 6.6 Hz, 2H), 3.56–3.68 (br t, J = 4.8 Hz, 2H), 4.16–4.26 (br t,
J = 4.5 Hz, 2H). 13C NMR (100 MHz, CDCl3):  −0.3, 13.5, 21.0,
23.1, 63.5, 68.4, 73.7, 170.6. IR (neat): Si–O 1120, CO 1730, C–H
2930 cm−1.
Phthalic acid 1-allyl ester 2-[((2R,5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester (7). The synthesis and purification of this compound was
analogous to 6. Combining phthaloyl dichloride (2.03 g, 10 mmol),
CH2Cl2 (50 mL), triethylamine (5.6 mL, 40 mmol), allyl alcohol
(680 L, 10 mmol), and hydroquinidine hydrochloride (3.63 g,
10 mmol) in the manner indicated above yielded a white solid
(3.0 g, 57%), mp 22–24 °C, []2D2 = +61.4 (c 1.3, EtOH). 1H NMR
(400 MHz, CDCl3):  0.89 (t, J = 7.2 Hz, 3H), 1.40–1.65 (m, 6H),
1.77–1.89 (m, 2H), 2.64–2.79 (m, 3H), 2.87–2.94 (m, 1H), 3.38
(q, J = 8.8 Hz, 1H), 3.95 (s, 3H), 4.43 (dd, J = 12.8, 6.0 Hz, 1H),
4.55 (dd, J = 13.2, 5.6 Hz, 1H), 5.10 (d, J = 10.4 Hz, 1H), 5.18 (dd,
J = 17.2, 1.2 Hz, 1H), 5.64–5.75 (m, 1H), 6.69 (d, J = 8.0 Hz, 1H),
7.38 (dd, J = 4.6, 2.8 Hz, 1H), 7.44 (d, J = 4.8 Hz, 1H), 7.49 (d,
J = 2.4 Hz, 1H), 7.53–7.60 (m, 2H), 7.69–7.78 (m, 2H), 8.02 (d,
J = 9.2 Hz, 1H), 8.76 (d, J = 4.4 Hz, 1H). 13C NMR (100 MHz,
CDCl3):  12.12, 24.19, 25.56, 26.23, 27.33, 37.47, 49.90, 50.77,
55.58, 59.76, 66.20, 74.75, 101.52, 118.34, 119.06, 121.75, 127.14,
128.44, 129.07, 130.96, 131.20, 131.23, 131.57, 131.61, 131.93,
143.55, 144.61, 147.30, 157.58, 166.29, 166.62. HRMS-FAB+
[M + H]+: found 515.2571; calc. (C31H35N2O5) 515.2546.
Poly(methylundecanoic
acid
((2R,5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl
ester siloxane-co-methylacetic acid 2-propoxyethyl ester silox-
ane) (5). Refer to general hydrosilation procedure. The reaction was
run using 1 (245 mg, 0.5 mmol), 3 (72 mg, 0.5 mmol), PMHS (50 L,
0.83 mmol), toluene (7.5 mL) and dichlorodi(cyclopentadienyl)-
platinum(II) (5 mg, 0.013 mmol) to provide a yellow solid (160 mg,
51%). This procedure gave a material with a ratio of 4:7 for the
cinchona alkaloid to the soluble linker, respectively. The ratio was
1
determined by comparing the integration of a unique H NMR
signal from the cinchona alkaloid ( 8.7, 1H) to a unique signal
from the soluble linker ( 2.0, 3H). The H NMR integration data
1
is reported relative to four cinchona units, equiv. wt. 909. 1H NMR
(300 MHz, CDCl3):  −0.16–0.21 (br s, 33H), 0.33–0.59 (br s,
22H), 0.76–0.95 (br s, 12H), 1.00–1.77 (br m, 110H), 1.91–2.11 (br
s, 21H), 2.24–2.45 (br m, 8H) 2.57–2.81 (br s, 12 H), 2.83–3.02 (br
s, 4H), 3.12–3.26 (br s, 4H), 3.28–3.43 (br s, 14H), 3.47–3.64 (br s,
14H), 3.83–4.00 (br s, 12H), 4.04–4.27 (br s, 14H), 6.45–6.76 (br
s, 4H), 7.17–7.55 (br m, 12H), 7.88–8.08 (br m, 4H), 8.59–8.80 (br
1-Allyloxy-4-chlorophthalazine (8). The procedure was
adapted from Sharpless.3,34 A 250 mL two-neck round-bottom flask
equipped with stir bar, rubber septum, condenser and gas inlet was
charged with sodium hydride (390 mg, 9.8 mmol, 60% in mineral
oil), and THF (30 mL) under a flow of argon. Allyl alcohol (313 L,
4.6 mmol) was then added in a dropwise fashion to the stirred re-
action solution at room temperature. This mixture was stirred for
2 2 9 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8

View Article Online
30 min, then 1,4-dichlorophthalazine (1.0 g, 5.0 mmol) was added
in one portion. The rubber septum was replaced with a glass stop-
per and the reaction was stirred at reflux. The reaction mixture
gradually changed colors from a cloudy white, to a yellow then
finally brown suspension after addition of 1,4-dichlorophthalazine.
The consumption of 1,4-dichlorophthalazine was monitored using
TLC (SiO2, hexanes–EtOAc, 1:4). After stirring the reaction at
reflux overnight it was cooled to room temperature. Excess NaH
was quenched by the slow addition of a water–THF (75 mL, 1:1
v/v) mixture to the vigorously stirred reaction at 0 °C. This solution
was extracted with CH2Cl2 (2 × 75 mL). All organic layers were
combined and dried over MgSO4. The solvent was removed using
a rotary evaporator to yield a crude yellow solid. Purification via
flash chromatography with silica as absorbent (EtOAc–hexanes,
3:7) yielded a white solid (680 mg, 67%), mp 90–91 °C, 1H NMR
(400 MHz, CDCl3):  5.17 (d, J = 5.6 Hz, 2H), 5.36 (dd, J = 10.4,
1.6 Hz, 1H), 5.53 (dd, J = 17.2, 1.6 Hz, 1H), 6.16–6.28 (m, 1H),
7.88–7.97 (m, 2H), 8.18 (dd, J = 7.2, 1.6 Hz, 1H), 8.25 (dd, J = 6.4,
2.4 Hz, 1H). 13C NMR (100 MHz, CDCl3):  68.35, 118.46, 121.37,
123.47, 124.95, 127.48, 132.29, 132.67, 132.96, 149.91, 159.69.
HRMS-EI+ m/z: found 220.0415; calc. (C11H9N2OCl) 220.0403.
UV-vis (CHCl3) max: 274, 309 nm.
(br s, 1H), 2.28–2.44 (br s, 2H), 2.86–3.08 (br s, 2H), 3.10–3.20 (br
s, 1H), 3.22–3.38 (br s, 1H), 3.44–3.60 (br s, 1H), 3.86–4.02 (br s,
3H), 4.16–4.34 (br s, 2H), 7.28–7.46 (br m, 2H), 7.54–7.72 (br m,
2H), 7.94–8.04 (br m, 1H), 8.06–8.26 (br m, 4H), 8.62–8.74 (br s,
1H). 13C NMR (100 MHz, CDCl3):  −2.82, 11.67, 13.01, 20.98,
22.25, 25.17, 25.83, 26.09, 36.03, 49.06, 49.64, 56.14, 58.29, 67.69,
73.01, 100.69, 117.41, 122.58, 126.02, 129.00, 129.62, 130.73,
131.60, 133.92, 141.78, 144.35, 146.86, 158.46, 164.14, 171.33. IR
(neat): Si–O 1100, CO 1720, C–H 2945 cm−1.
Poly(methylterephthalic acid 1-[((2R, 5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester 4-propyl ester siloxane-co-methylacetic acid 2-propoxy-
ethyl ester siloxane) (11). Refer to general hydrosilation proce-
dure. The reaction was run using 6 (875 mg, 1.7 mmol), 3 (187 mg,
1.3 mmol), PMHS (150 L, 2.5 mmol), toluene (10 mL) and
dichlorodi(cyclopentadienyl)platinum(II) (2 mg, 0.005 mmol). The
purification was modified after the first precipitation of the toluene
reaction mixture into hexanes. Not all of the precipitated material
was soluble in THF. Thus the polymer residue was titurated with
THF (3 × 10 mL). The insoluble residue was discarded. All THF
fractions were combined and evaporated under reduced pressure.
The tan oily residue that remained after evaporation was dissolved
in a minimal amount of THF (4 mL) and precipitated into an excess
of hexanes (8 mL). No further precipitations were needed to reach
a pure material. The residual solvents were removed from the poly-
mer residue under reduced pressure to provide a tan solid (370 mg,
38%). This procedure gave a material with a ratio of 5:7 for the cin-
chona alkaloid to the soluble linker, respectively. The ratio was de-
termined by comparing the integration of a unique 1H NMR signal
from the cinchona alkaloid ( 3.9, 3H) to a unique signal from the
soluble linker ( 2.1, 3H). The 1H NMR integration data is reported
relative to five cinchona units, equiv. wt. 861. 1H NMR (400 MHz,
CDCl3):  −0.6–0.32 (br s, 36H), 0.42–0.72 (br m, 24H), 0.81–0.98
(br s, 15H), 1.43–1.70 (br m, 49H), 1.72–1.91 (br m, 15H), 1.93–
2.12 (br s, 21H), 2.54–3.13 (br m, 20H), 3.31–3.50 (br m, 19H),
3.52–3.66 (br s, 14H), 3.88–4.01 (br s, 15H), 4.08–4.22 (br s, 14H),
4.24–4.36 (br s, 10H), 6.85–7.00 (br s, 5H), 7.32–7.45 (br m, 10H),
7.49–7.62 (br m, 5H), 7.96–8.21 (br m, 25H), 8.64–8.75 (br m, 5H).
13C NMR (100 MHz, CDCl3):  −2.91, −0.63, 11.79, 13.02, 13.25,
20.86, 22.24, 22.88, 23.37, 25.27, 26.01, 26.35, 36.74, 49.48, 50.27,
55.65, 58.85, 63.42, 67.57, 68.28, 73.02, 73.56, 100.88, 117.01,
122.01, 126.45, 129.27, 129.40, 131.43, 133.42, 140.30, 142.74,
144.22, 146.84, 157.92, 164.09, 165.07, 170.59. IR (neat): Si–O
1180, CO 1730, C–H 2940 cm−1.
1-Allyloxy-4-[((2R,5R)-(−)-5-ethyl-1-azabicyclo[2.2.2]oct-2-
yl)(6-methoxyquinolin-4-yl)-(S)-methoxy]phthalazine (9). The
procedure was adapted from Sharpless.3,34 A 100 mL two-neck
round-bottom flask equipped with stir bar, glass stopper, condenser
and gas inlet was charged with sodium hydride (330 mg, 8.3 mmol,
60% in mineral oil) and N,N-dimethylacetamide (20 mL) under a
flow of argon. Hydroquinidine hydrochloride (1.4 g, 3.8 mmol)
was then added in three equal portions to the stirred reaction so-
lution at room temperature. This mixture was stirred for 30 min
at 100 °C over which time the reaction changed in color from a
cloudy white to an orange and finally a green suspension. The reac-
tion was cooled to room temperature and 8 (800 mg, 3.6 mmol) was
added all at once. The reaction mixture was again stirred at 100 °C.
The consumption of hydroquinidine hydrochloride was monitored
using TLC (SiO2, EtOAc–MeOH, 4:1). After stirring the reaction
at 100 °C overnight it was cooled to room temperature. Excess NaH
was quenched by the slow addition of a water–THF (10 mL, 1:1 v/v)
mixture to the vigorously stirred reaction at 0 °C. Then the solution
was diluted with water (50 mL). This solution was extracted with
CH2Cl2 (2 × 50 mL). The organic layers were combined and washed
with water (5 × 75 mL), dried over MgSO4, filtered and evaporated
to yield a crude white solid. Purification via flash chromatography
with silica as absorbent (EtOAc–MeOH, 4:1) yielded a white solid
(1.5 g, 82%), mp 65–68 °C, []2D2 = −189.5 (c 1.2, EtOH). 1H NMR
(400 MHz, CDCl3):  0.90 (t, J = 6.8 Hz, 3H), 1.42–1.64 (m, 6H),
1.76 (s, 1H), 2.12 (t, J = 12.0 Hz, 1H), 2.70–2.80 (m, 1H), 2.85 (d,
J = 9.6 Hz, 1H), 2.92 (d, J = 8.0 Hz, 2H), 3.49 (q, J = 8.4 Hz, 1H),
3.99 (s, 3H), 4.93–5.07 (m, 2H), 5.27 (d, J = 10.4 Hz, 1H), 5.41 (dd,
J = 17.6, 1.2 Hz, 1H), 6.08–6.20 (m, 1H), 7.15 (d, J = 5.6 Hz, 1H),
7.35 (dd, J = 8.8, 2.4 Hz, 1H), 7.47 (d, J = 4.4 Hz, 1H), 7.64 (d,
J = 2.4 Hz, 1H), 7.82–7.94 (m, 2H), 7.99 (d, J = 8.8 Hz, 1H), 8.18
(d, J = 8.0 Hz, 1H), 8.35 (d, J = 8.0 Hz, 1H), 8.65 (d, J = 4.4 Hz,
1H). 13C NMR (100 MHz, CDCl3):  12.05, 23.13, 25.50, 26.52,
27.39, 37.53, 50.10, 50.93, 55.62, 59.88, 67.61, 76.78, 102.14,
117.58, 118.51, 121.47, 122.21, 122.25, 122.35, 123.32, 126.96,
131.42, 131.82, 131.87, 132.77, 144.39, 144.54, 147.19, 156.19,
157.14, 157.30. HRMS-EI+ m/z: found 510.2634; calc. (C31H34N4O3)
510.2631. UV-vis (CHCl3) max: 282, 310, 334 nm.
Poly(methylphthalic
acid
1-[((2R,5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester 2-propyl ester siloxane) (12). Refer to general hydrosila-
tion procedure. The reaction was run using 7 (1.0 g, 1.9 mmol),
PMHS (94 L, 0.44 mmol), toluene (8 mL) and dichloro-
di(cyclopentadienyl)platinum(II) (6 mg, 0.015 mmol) to provide
1
a tan solid (500 mg, 55%), equiv. wt. 575. H NMR (400 MHz,
CDCl3):  −0.10–0.20 (br s, 3H), 0.32–0.56 (br s, 2H), 0.70–0.86
(br m, 3H), 1.08–1.18 (br m, 2H), 1.24–1.56 (br m, 6H), 1.58–1.78
(br m, 2H), 2.52–2.80 (br s, 3H), 2.82–2.96 (br s, 1H), 3.28–3.40
(br m, 1H), 3.62–3.74 (br m, 2H), 3.78–4.04 (br s, 3H), 6.65–6.85
(br s, 1H), 7.08–7.13 (br d, J = 9.2 Hz, 1H), 7.28–2.34 (br d, J =
4.4 Hz, 1H), 7.42–7.58 (br m, 2H), 7.61–7.68 (br d, J = 7.2 Hz, 1H),
7.76–7.82 (br d, J = 9.2 Hz, 1H), 7.88–7.94 (br d, J = 7.6 Hz, 1H),
7.95–8.03 (br s, 1H), 8.62–8.67 (br d, J = 4.4 Hz, 1H). 13C NMR
(100 MHz, CDCl3):  −2.72, 11.69, 19.88, 24.37, 24.59, 25.32,
25.47, 26.11, 35.53, 49.39, 49.79, 55.23, 57.30, 71.23, 74.50,
100.44, 116.43, 122.57, 125.49, 126.21, 128.75, 128.87, 130.59,
131.06, 131.21, 131.50, 141.18, 144.16, 146.03, 158.09, 165.86,
166.27. IR (neat): Si–O 1120, CO 1730, C–H 2945 cm−1.
Poly(methylterephthalic acid 1-[((2R,5R)-5-ethyl-1-aza-
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester 4-propyl ester siloxane) (10). Refer to general hydrosilation
procedure. The reaction was run using 6 (272 mg, 0.53 mmol),
PMHS (26 L, 0.44 mmol), toluene (5 mL) and dichloro-
di(cyclopentadienyl)platinum(II) (2 mg, 0.005 mmol) to provide
Poly(methylphthalic
acid
1-[((2R,5R)-5-ethyl-1-aza-
1
a tan solid (190 mg, 75%), equiv. wt. 575. H NMR (400 MHz,
bicyclo[2.2.2]oct-2-yl)(6-methoxyquinolin-4-yl)-(S)-methyl]
ester 2-propyl ester siloxane-co-methylacetic acid 2-propoxy-
ethyl ester siloxane) (13). Refer to general hydrosilation proce-
CDCl3):  −0.05–0.30 (br s, 3H), 0.52–0.70 (br s, 2H), 0.78–1.00
(br s, 3H), 1.30–1.84 (br m, 6H), 1.88–1.98 (br s, 1H), 2.04–2.26
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 9 5

View Article Online
dure. The reaction was run using 7 (1.75 mg, 3.4 mmol), 3 (353 mg,
2.45 mmol), PMHS (280 L, 4.68 mmol), toluene (20 mL) and
dichlorodi(cyclopentadienyl)platinum(II) (6 mg, 0.015 mmol). The
purification was modified after the first precipitation of the toluene
reaction mixture into hexanes. Not all of the precipitated material
was soluble in THF. Thus the polymer residue was titurated with
chloroform (2 × 5 mL). The insoluble residue was discarded. All
chloroform fractions were combined and precipitated into an ex-
cess of hexanes (15 mL). No further precipitations were needed to
reach a pure material. The residual solvents were removed from
the polymer residue under reduced pressure to provide a tan solid
(940 mg, 52%). This procedure gave a material with a ratio of 3:2
for the cinchona alkaloid to the soluble linker, respectively. The
ratio was determined by comparing the integration of a unique 1H
NMR signal from the cinchona alkaloid ( 4.0, 3H) to a unique
26.53, 26.75, 37.28, 49.95, 50.73, 55.65, 58.97, 59.82, 63.52,
68.31, 69.29, 73.66, 76.47, 102.20, 118.55, 121.68, 122.21, 122.37,
122.89, 125.29, 127.05, 131.93, 132.22, 133.15, 144.22, 144.75,
147.34, 156.90, 157.22, 157.89, 170.96. IR (neat): Si–O 1040, CO
1745, C–H 2940 cm−1.
1-Chloro-4-[(6-methoxyquinolin-4-yl)-((2R, 5R)-(−)-5-vinyl-
1-azabicyclo[2.2.2]oct-2-yl)-(S)-methoxy]phthalazine (16). The
procedure was adapted from Sharpless.3,34 A 500 mL two-neck
round-bottom flask equipped with stir bar, glass stopper, condenser
and gas inlet was charged with sodium hydride (2.0 g, 50 mmol,
60% in mineral oil), and THF (125 mL) under a flow of argon.
Quinidine (7.45 g, 23 mmol) was then added in one portion to the
stirred reaction solution at room temperature. This mixture was
stirred at reflux for 30 min. Then 1,4-dichlorophthalazine (5.0 g,
25 mmol) was added all at once to the reaction cooled to room
temperature. This mixture was then stirred at reflux. The reaction
mixture gradually changed colors from a cloudy white to a brown
suspension after addition of 1,4-dichlorophthalazine. The consump-
tion of quinidine was monitored using TLC (SiO2, EtOAc–MeOH,
4:1). After stirring the reaction at reflux overnight it was cooled to
room temperature. Excess NaH was quenched by the slow addition
of a water–THF (75 mL, 1:1 v/v) mixture to the vigorously stirred
reaction at 0 °C. This solution was diluted with water (100 mL)
then extracted with CH2Cl2 (3 × 100 mL). All organic layers were
combined and dried over MgSO4. The solvent was removed using
rotary evaporator to yield a crude red oil. Purification via flash
chromatography with silica as absorbent (EtOAc–MeOH, 9:1)
yielded a white solid (7.4 g, 66%), mp 94–96 °C, []2D2 = −248.7
1
signal from the soluble linker ( 2.1, 3H). The H NMR integra-
tion data is reported relative to three cinchona units, equiv. wt. 712.
1H NMR (400 MHz, CDCl3):  −0.60–0.32 (br s, 15H), 0.39–0.66
(br s, 10H), 0.75–0.98 (br t, J = 7.2 Hz, 9H), 1.12–1.28 (br s, 6H),
1.33–1.94 (br m, 28H), 1.99–2.16 (br s, 6H), 2.67–2.88 (br s, 9H),
2.91–3.06 (br s, 3H), 3.17–3.48 (br m, 7H), 3.51–3.68 (br s, 4H),
3.70–3.82 (br m, 6H), 3.85–4.06 (br s, 9H), 4.11–4.29 (br s, 4H),
6.76–6.95 (br s, 3H), 7.11–7.89 (br m, 21H), 7.93–8.12 (br s, 3H),
8.66–8.82 (br s, 3H). 13C NMR (100 MHz, CDCl3):  −2.72, −0.60,
11.73, 12.99, 20.91, 21.97, 22.93, 24.35, 24.55, 25.29, 25.44, 26.09,
35.51, 49.74, 50.34, 55.20, 57.32, 63.52, 67.61, 68.32, 73.63, 74.47,
100.46, 116.44, 122.56, 125.51, 126.17, 128.16, 128.89, 130.50,
130.95, 131.38, 131.83, 141.20, 144.02, 146.01, 158.11, 165.88,
170.64, 172.50. IR (neat): Si–O 1060, CO 1730, C–H 2940 cm−1.
1
(c 1.06, EtOH). H NMR (500 MHz, CDCl3):  1.56–1.70 (m,
Poly(methyl-1-[((2R,5R)-5-ethyl-1-azabicyclo[2.2.2]oct-2-
yl)(6-methoxyquinolin-4-yl)-(S)-methoxy]-4-propoxyphthala-
zine siloxane) (14). Refer to general hydrosilation procedure.
The reaction was run using 9 (1.35 g, 2.6 mmol), PMHS (130 L,
2.16 mmol), toluene (10 mL) and dichlorodi(cyclopentadienyl)-
platinum(II) (4 mg, 0.01 mmol) to provide a tan solid (560 mg,
2H), 1.89 (s, 1H), 1.97 (s, 1H), 2.17 (t, J = 11.0 Hz, 1H), 2.32 (q,
J = 8.0 Hz, 1H), 2.74–2.83 (m, 1H), 2.87–2.94 (m, 1H), 2.95–3.02
(m, 1H), 3.07–3.15 (m, 1H), 3.52 (q, J = 8.5 Hz, 1H), 4.00 (s, 3H),
5.10 (d, J = 17.5 Hz, 1H), 5.13 (d, J = 10.5 Hz, 1H), 6.02–6.12 (m,
1H), 7.31 (d, J = 6.0 Hz, 1H), 7.37 (dd, J = 9.5, 2.5 Hz, 1H), 7.45
(d, J = 4.5 Hz, 1H), 7.63 (d, J = 2.5 Hz, 1H), 7.96–8.01 (m, 3H),
8.17–8.22 (m, 1H), 8.40–8.45 (m, 1H), 8.66 (d, J = 4.0 Hz, 1H).
13C NMR (125 MHz, CDCl3):  23.44, 26.41, 27.65, 39.53, 49.49,
49.88, 55.66, 59.84, 76.17, 101.72, 114.88, 118.35, 121.53, 121.96,
123.26, 125.50, 127.29, 127.87, 131.73, 133.09, 133.41, 140.40,
144.06, 144.78, 147.36, 150.43, 157.95, 159.31. HRMS-FAB+
[M + H]+: found 487.1897; calc. (C28H28ClN4O2) 487.1901. UV-vis
(CHCl3) max: 281, 335 nm.
1
45%), equiv. wt. 571. H NMR (400 MHz, CDCl3):  −0.20–0.20
(br s, 3H), 0.50–0.70 (br s, 2H), 0.75–0.95 (br s, 3H), 1.30–1.65
(br s, 6H), 1.70–1.95 (br m, 3H), 2.05–2.30 (br s, 1H), 2.55–3.05
(br m, 4H), 3.30–3.55 (br s, 1H), 3.85–4.05 (br s, 3H), 4.20–4.50
(br s, 2H), 7.10–7.35 (br m, 2H), 7.40–7.50 (br s, 1H), 7.55–7.70 (br
s, 1H), 7.80–8.00 (br m, 3H), 8.05–8.15 (br s, 1H), 8.20–8.35 (br s,
1H), 8.50–8.65 (br m, 1H). 13C NMR (100 MHz, CDCl3):  −2.88,
11.98, 13.21, 22.52, 25.44, 26.50, 26.92, 37.37, 49.99, 50.76,
55.80, 59.07, 59.86, 69.28, 76.48, 102.12, 118.36, 121.95, 122.19,
122.35, 122.74, 123.34, 126.90, 131.73, 132.01, 132.95, 144.38,
144.62 146.99, 155.70, 156.92, 157.60. IR (neat): Si–O 1090, C–H
2950 cm−1. UV-vis (CHCl3) max: 290, 311, 334 nm.
1-[((2R,5R)-(−)-5-Ethyl-1-azabicyclo[2.2.2]oct-2-yl)(6-
methoxyquinolin-4-yl)(S)-methoxy]-4-[(6-methoxyquinolin-4-
yl)((2R, 5R)-5-vinyl-1-azabicyclo[2.2.2]oct-2-yl)-(S)-methoxy]
phthalazine (17). The procedure was adapted from Sharpless.3,34
A
250 mL two-neck round-bottom flask equipped with stir bar, glass
stopper, condenser and gas inlet was charged with sodium hydride
(1.3 g, 32 mmol, 60% in mineral oil), and N,N-dimethylacetamide
(70 mL) under a flow of argon. Hydroquinidine hydrochloride
(5.3 g, 14.5 mmol) was then added in one portion to the stirred
reaction solution at room temperature. This mixture was stirred for
15 min at 100 °C. The reaction was cooled to room temperature and
16 (6.7 g, 13.8 mmol) was added all at once. The reaction was again
stirred at 100 °C. The consumption of 16 was monitored using TLC
(SiO2, EtOAc–MeOH, 3:2). After stirring the reaction at 100 °C
overnight it was cooled to room temperature. Excess NaH was
quenched by the slow addition of water (100 mL) to the vigorously
stirred reaction at 0 °C. A white precipitate occurred upon the addi-
tion of water that was collected using a Buchner funnel. The solid
was washed with water (30 mL) and dried under reduced pressure
to yield a white solid (9.0 g). This white solid was recrystallized
from methanol (75 mL) to yield white crystals (6.4 g, 60%), mp
Poly(methyl-1-[((2R,5R)-5-ethyl-1-azabicyclo[2.2.2]oct-2-
yl)(6-methoxyquinolin-4-yl)-(S)-methoxy]-4-propoxyphthalazi-
nesiloxane-co-methylacetic acid 2-propoxyethyl ester siloxane)
(15). Refer to general hydrosilation procedure. The reaction was run
using 9 (820 mg, 1.6 mmol), 3 (184 mg, 1.3 mmol), PMHS (140 L,
2.4 mmol), toluene (8 mL) and dichlorodi(cyclopentadienyl)-
platinum(II) (3 mg, 0.01 mmol) to provide a tan solid (390 mg,
42%). This procedure gave a material with a ratio of 13:10 for
the cinchona alkaloid to the soluble linker, respectively. The ratio
was determined by comparing the integration of a unique 1H NMR
signal from the cinchona alkaloid ( 3.9, 3H) to a unique signal
1
from the soluble linker ( 2.0, 3H). The H NMR integration data
1
is reported relative to 13 cinchona units, equiv. wt. 728. H NMR
(500 MHz, CDCl3):  −0.33–0.22 (br s, 69H), 0.35–0.50 (br s, 20H),
0.53–0.65 (br s, 26H), 0.72–0.93 (br m, 39H), 1.27–1.62 (br s, 98H),
1.67–1.85 (br m, 39H), 1.89–2.05 (br s, 30H), 2.09–2.21 (br s, 13H),
2.61–2.99 (br m, 52H), 3.08–3.59 (br m, 53H), 3.82–4.00 (br s,
39H), 4.02–4.20 (br s, 20H), 4.22–4.44 (br s, 26H), 7.07–7.21 (br
s, 13H), 7.23–7.34 (br m, 13H), 7.36–7.47 (br s, 13H), 7.52–8.34
(br m, 78H), 8.52–8.68 (br s, 13H). 13C NMR (125 MHz, CDCl3):
 −3.12, −0.90, 11.57, 11.79, 12.96, 22.32, 22.82, 25.27, 26.32,
22
1
136–138 °C, []D = −257.9 (c 1.7, EtOH). H NMR (400 MHz,
CDCl3):  0.80 (t, J = 6.8 Hz, 3H), 1.26 (s, 1H), 1.34–1.46 (m, 4H),
1.48–1.60 (m, 4H), 1.69 (s, 1H), 1.81 (s, 1H), 1.96 (t, J = 12.8 Hz,
1H), 2.08 (t, J = 12.8 Hz, 1H), 2.21 (q, J = 8.4 Hz, 1H), 2.52–2.88
(m, 7H), 2.92–3.00 (m, 1H), 3.41 (q, J = 8.8 Hz, 2H), 3.90 (s, 6H),
2 2 9 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8

View Article Online
4.98 (d, J = 13.6 Hz, 2H), 5.88–6.00 (m, 1H), 6.97 (d, J = 6.4 Hz,
1H), 7.03 (d, J = 5.6 Hz, 1H), 7.32–7.40 (m, 2H), 7.43 (d, J =
4.8 Hz, 1H), 7.45 (d, J = 4.8 Hz, 1H), 7.52–7.59 (m, 2H), 7.89–7.96
(m, 2H), 7.99 (dd, J = 9.0, 2.0 Hz, 2H), 8.30–8.38 (m, 2H), 8.65
(t, J = 4.0 Hz, 2H). 13C NMR (100 MHz, CDCl3):  11.99, 23.36,
23.60, 25.36, 26.34, 26.55, 27.36, 27.80, 37.46 39.71, 49.49, 49.85,
49.96, 50.87, 55.58, 55.61, 60.07, 60.22, 76.05, 76.28, 101.83,
101.93, 114.47, 118.11, 118.37, 121.59, 121.67, 122.22, 122.24,
122.60, 122.85, 127.06, 127.09, 127.16, 127.18, 131.33, 131.90,
131.97, 140.11, 144.43, 144.69, 144.75, 147.14, 156.06, 156.12,
156.18, 157.26, 157.33. HRMS-FAB+ [M + H]+: found 777.4141;
calc. (C48H53N6O4) 777.4128. UV-vis (CHCl3) max: 281, 334 nm.
Acknowledgements
We would like to thank Ms Denise Scofield, Dr KenWoycechowsky,
Prof. Franco Cozzi and Prof.Amir H. Hoveyda for their helpful sug-
gestions. This work was financially supported by US-NSF, (Grant
CHE-0213323) AFSOR, and Swiss SNF.
References
1 S. G. Hentges and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102,
4263–4265.
2 E. N. Jacobsen, I. Marko, W. S. Mungall, G. Schroeder and K. B.
Sharpless, J. Am. Chem. Soc., 1988, 110, 1968–1970.
3 K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Hartung,
K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, D. Xu and X.-L.
Zhang, J. Org. Chem., 1992, 57, 2768–2771.
Poly(methylDHQD2PHAL siloxane) (18). Refer to general
hydrosilation procedure. The reaction was run using 17 (2.75 g,
3.5 mmol),PMHS(170L,2.9mmol),toluene(20mL)anddichloro-
di(cyclopentadienyl)platinum(II) (1 mg, 0.0025 mmol). The purifi-
cation was difficult and required special conditions to separate the
polymer product from 17. After the initial precipitation subsequent
precipitations were done by dissolving the crude material in toluene
(5 mL) and dripping a drop at a time into hexanes (10 mL). A total
of six precipitations from toluene to hexanes were needed for clean
material. The residual solvents were removed from the polymer
residue under reduced pressure to provide a tan solid (520 mg,
22%), equiv. wt. 837. 1H NMR (500 MHz, CDCl3):  −0.10–0.20 (br
s, 3H), 0.35–0.60 (br s, 2H), 0.72–0.84 (br s, 3H), 1.24–1.58 (br m,
10H), 1.62–1.72 (br s, 2H), 1.90–2.02 (br s, 2H), 2.50–2.86 (br m,
6H), 3.28–3.58 (br s, 8H), 3.76–3.96 (br s, 6H), 6.86–7.04 (br s, 2H),
7.30–7.46 (br m, 4H), 7.48–7.60 (br s, 2H), 7.84–8.06 (br s, 4H),
8.24–8.40 (br s, 2H), 8.54–8.70 (br s, 2H). 13C NMR (125 MHz,
CDCl3):  −1.00, 11.79, 14.56, 22.91, 23.32, 24.84, 25.21, 26.24,
26.43, 27.22, 27.71, 37.38, 39.63, 49.65, 49.94, 50.05, 50.84, 55.55,
55.69, 60.04, 60.25, 76.02, 76.32, 101.73, 102.04, 118.19, 118.56,
121.60, 121.81, 122.22, 122.43, 122.73, 122.88, 126.89, 127.01,
127.22, 127.32, 132.17, 132.30, 140.37, 144.70, 144.87, 144.94,
147.35, 156.29, 156.39, 156.49, 157.61, 157.73. IR (neat): Si–O
1050, C–H 2930 cm−1. UV-vis (CHCl3) max: 281, 336 nm.
4 H. C. Kolb, M. S. VanNieuwenhze and K. B. Sharpless, Chem. Rev.,
1994, 94, 2483–2547.
5 (a) B. M. Kim and K. B. Sharpless, Tetrahedron Lett., 1990, 31, 3003–
3006; (b) I. Motorina and C. M. Credden, Org. Lett., 2001, 3, 2325.
6 H. Han and K. D. Janda, J. Am. Chem. Soc., 1996, 118, 7632–7633.
7 P. Salvadori, D. Pini, A. Petri and A. Mandoli, Catalytic Heterogeneous
Enantioselective Dihydroxylation and Epoxidation, Wiley-VCH Verlag,
Weinheim, Germany, 2000.
8 T. Frenzel, W. Solodenko and A. Kirschning, Polymeric Materials in
Organic Synthesis and Catalysis, Wiley-VCH: Weinheim, 2003.
9 N. E. Leadbeater and M. Marco, Chem. Rev., 2002, 102, 3217–3273.
10 M. Benaglia, A. Puglisi and F. Cozzi, Chem. Rev., 2003, 103, 3401–
3429.
11 C. Bolm and A. Gerlach, Eur. J. Org. Chem., 1998, 21–27.
12 C. Bolm and A. Gerlach, Angew. Chem., Int. Ed. Engl., 1997, 36,
741–743.
13 H. Han and K. D. Janda, Tetrahedron Lett., 1997, 38, 1527–1530.
14 D. J. Gravert and K. D. Janda, Chem. Rev., 1997, 97, 489–509.
15 P. H. Toy and K. D. Janda, Acc. Chem. Res., 2000, 33, 546–554.
16 T. J. Dickerson, N. N. Reed, K. D. Janda, Polymeric Materials in Or-
ganic Synthesis and Catalysis, Wiley-VCH, Weinheim, 2003.
17 B. B. Lohray, E. Nandanan and V. Bhushan, Tetrahedron Lett., 1994, 35,
6559–6562.
18 A. Petri, D. Pini and P. Salvadori, Tetrahedron Lett., 1995, 36, 1549–
1552.
19 F. Sieber, P. Wentworth Jr. and K. D. Janda, J. Comb. Chem., 1999, 1,
540–546.
20 F. Sieber, P. Wentworth Jr., J. D. Toker, A. D. Wentworth, W. A. Metz,
N. N. Reed and K. D. Janda, J. Org. Chem., 1999, 64, 5188–5192.
21 T. S. Reger and K. D. Janda, J. Am. Chem. Soc., 2000, 122, 6929–6934.
Poly(methylDHQD2PHAL siloxane-co-methylacetic acid
2-propoxyethyl ester siloxane) (19). Refer to general hydrosila-
tion procedure. The reaction was run using 17 (1.2 g, 1.5 mmol), 3
(165 mg, 1.15 mmol), PMHS (130 L, 2.2 mmol), toluene (15 mL)
and dichlorodi(cyclopentadienyl)platinum(II) (1 mg, 0.005 mmol).
The purification was difficult and required special conditions to
separate the polymer product from 17. After the initial precipitation
subsequent precipitations were done by dissolving the crude mate-
rial in toluene (5 mL) and dripping a drop at a time into hexanes
(10 mL). A total of six precipitations from toluene to hexanes were
needed for clean material. The residual solvents were removed
from the polymer residue under reduced pressure to provide a tan
solid (250 mg, 22%). This procedure gave a material with a ratio
of 5:7 for the cinchona alkaloid to the soluble linker, respectively.
The ratio was determined by comparing the integration of a unique
1H NMR signal from the cinchona alkaloid ( 8.6, 2H) to a unique
signal from the soluble linker ( 4.2, 2H). The 1H NMR integration
data is reported relative to five cinchona units, equiv. wt. 1123. 1H
NMR (500 MHz, CDCl3):  −0.07–0.26 (br s, 36H), 0.37–0.61 (br
s, 24H), 0.71–0.82 (br s, 15H), 1.30–1.81 (br m, 74H), 1.88–2.12
(br m, 31H), 2.55–2.84 (br m, 30H), 3.27–3.52 (br m, 54H),
3.54–3.65 (br s, 14H), 3.78–3.95 (br m, 30H), 4.07–4.23 (br s,
14H), 6.84–7.05 (br s, 10H), 7.26–7.44 (br m, 20H), 7.46–7.61 (br
s, 10H), 7.82–8.07 (br m, 20H), 8.23–8.42 (br s, 10H), 8.53–8.70 (br
s, 10H). 13C NMR (125 MHz, CDCl3):  −1.19, −0.94, 11.76, 12.85,
14.53, 20.78, 22.61, 22.74, 23.20, 24.76, 25.18, 26.21, 26.35, 27.16,
27.63, 37.33, 39.62, 49.63, 49.90, 50.01, 50.80, 55.33, 55.54, 59.98,
60.19, 63.56, 68.37, 73.69, 76.25, 76.36, 101.89, 101.97, 118.32,
118.53, 121.82, 121.97, 122.17, 122.42, 122.71, 122.86, 126.98,
127.11, 127.19, 127.28, 132.16, 132.30, 140.35, 144.56, 144.69,
144.90, 147.31, 156.26, 156.38, 156.44, 157.42, 157.65, 170.96. IR
22 M. S. Bratcher,
M. S. DeClue, A. Grunnet-Jepsen,
D. Wright,
B. R. Smith, W. E. Moerner and J. S. Siegel, J. Am. Chem. Soc., 1998,
120, 9680–9681.
23 D. Wright, U. Gubler, W. E. Moerner, M. S. DeClue and J. S. Siegel, J.
Phys. Chem. B, 2003, 107, 4732–4737.
24 Commercial poly(methylhydrosiloxane) available through Aldrich with
Mn = 9500.
25 R. N. Meals, Pure Appl. Chem., 1966, 13, 141–157.
26 P. Strohriegl, Makromol. Chem., Rapid Commun., 1986, 7, 771–775.
27 M. Lux, P. Strohriegl and H. Höcker, Makromol. Chem., 1987, 188,
811–820.
28 S. J. Clarson and J. A. Semlyen, Siloxane Polymers, PTR Prentice Hall,
NJ, 1993.
29 G. Nestor, M. S. White, G. W. Gray, D. Lacey and K. J. Toyne, Makro-
mol. Chem., 1987, 188, 2759–2767.
30 The rough equality of optical purity and ee was confirmed by Mosher’s
analysis.
31 Y. Ogino, H. Chen, E. Manoury, T. Shibata, M. Beller, D. Lubben and
K. B. Sharpless, Tetrahedron Lett., 1991, 32, 5761–5764.
32 W. Amberg, Y. L. Bennani, R. K. Chadha, G. A. Crispino, W. D. Davis,
J. Hartung, K. S. Jeong, Y. Ogino, T. Shibata and K. B. Sharpless, J.
Org. Chem., 1993, 58, 844–849.
33 M. S. DeClue, PhD Thesis, University of Califorina – San Diego, CA,
2003.
34 H. C. Kolb, P. G. Andersson,Y. L. Bennani, G. A. Crispino, K. S. Jeong,
H. L. Kwong and K. B. Sharpless, J. Am. Chem. Soc., 1993, 115, 12226–
12227.
35 V. Van Rheenen, R. C. Kelly and D. Y. Cha, Tetrahedron Lett., 1976,
1973–1976.
36 M. Minato, K. Yamamoto and J. Tsuji, J. Org. Chem., 1990, 55,
766–768.
37 P.-O. Norrby, H. Becker and K. B. Sharpless, J. Am. Chem. Soc., 1996,
118, 35–42.
38 D. R. Boyd,
N. D. Sharma,
N. I. Bowers,
P. A. Goodrich,
(neat): Si–O 1050, CO 1740, C–H 2940 cm−1. UV-vis (CHCl3) max
:
M. R. Groocock, A. J. Blacker, D. A. Clarke, T. Howard and H. Dalton,
Tetrahedron: Asymmetry, 1996, 7, 1559–1562.
281, 336 nm.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8
2 2 9 7

View Article Online
39 3,3′,5,5′-Tetramethylstilbene was synthesized and used to confirm the
inclusion of apolar diol products using ligand 19. Tetramethylstilbene has
a distinctly different 1H NMR spectrum than the dihydroxylated product
1,2-bis(3,5-dimethylphenyl)ethane-1,2-diol.The 1HNMRspectrumofthe
crude precipitate from the AD reaction using 19 and tetramethylstilbene
revealed nearly complete co-precipitation of product and ligand.
40 P. Salvadori, D. Pini and A. Petri, J. Am. Chem. Soc., 1997, 119, 6929–
6930.
41 C. E. Song, J. W. Yang and H.-J. Ha, Tetrahedron: Asymmetry, 1997, 8,
841–844.
42 S. Nagayama, M. Endo and S. Kobayashi, J. Org. Chem., 1998, 63,
6094–6095.
43 S. Kobayashi, T. Ishida and R. Akiyama, Org. Lett., 2001, 3, 2649–
2652.
44 D. E. Bergbreiter and C. Li, Org. Lett., 2003, 5, 2445–2447.
45 D. E. Bergreiter, P. L. Osburn, T. Smith, C. Li and J. D. Frels, J. Am.
Chem. Soc., 2003, 125, 6254–6260.
46 D. E. Berbreiter and J. Li, Chem. Commun., 2004, 42–43.
47 D. E. Bergbreiter and J. Li, Top. Curr. Chem., 2004, in press.
48 J. A. Dale, D. L. Dull and H. S. Mosher, J. Org. Chem., 1969, 34,
2543–2549.
49 J. A. Dale and H. S. Mosher, J. Am. Chem. Soc., 1973, 95, 512–519.
50 M. P. Bertrand, J. M. Surzur, M. Boyer and M. L. Mihailovic, Tetra-
hedron, 1979, 35, 1365–1372.
2 2 9 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 2 8 7 – 2 2 9 8