Functional Polymers from (Vinyl)polystyrene. Recyclable Polymer-Supported Organosilicon Protecting Groups for Solid-Phase Synthesis

Article abstract of DOI:10.1021/jo960603m

Dicobalt octacarbonyl catalyzed the hydrosilylation of residual vinyl groups on insoluble porous beads of divinylbenzene-rich copolymer, exclusively with β orientation, by either dialkylhalosilanes or alkyldihalosilanes. The resulting silyl halide functionalities, now bound to a cross-linked polystyrene matrix through stable dimethylene spacers, could then be used to protect and immobilize hydroxyl-containing free molecules for their further modification in the course of solid-phase synthesis. The reaction proved selective for primary hydroxyls in the presence of secondary, and secondary over tertiary. Together with product free alcohols, later cleavage of these silyl ethers with aqueous acid or base also released the corresponding silanol groups, while HF, under particularly mild and selective conditions, left silyl fluorides that themselves could then be very conveniently regenerated by BCl₃/CH₂Cl₂ back to the polymer-supported silyl chlorides. For a variety of possible applications, these solid-phase protecting groups offer control, yield, recovery, and reusability, as well as other possible advantages involving site-site isolation of functional groups within the cross-linked polystyrene matrix.

Full text of DOI:10.1021/jo960603m

J . Org. Chem. 1997, 62, 6183-6186
6183
F u n ction a l P olym er s fr om (Vin yl)p olystyr en e. Recycla ble
P olym er -Su p p or ted Or ga n osilicon P r otectin g Gr ou p s for
Solid -P h a se Syn th esis
Brent R. Stranix and Hua Qin Liu
Dept. of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal, Quebec H3A 2K6, Canada
Graham D. Darling*
Active Materials Inc., P.O. Box 47, Russell, Ontario K4R 1C7, Canada
Received April 1, 1996 (Revised Manuscript Received April 16, 1997X)
Dicobalt octacarbonyl catalyzed the hydrosilylation of residual vinyl groups on insoluble porous
beads of divinylbenzene-rich copolymer, exclusively with â orientation, by either dialkylhalosilanes
or alkyldihalosilanes. The resulting silyl halide functionalities, now bound to a cross-linked
polystyrene matrix through stable dimethylene spacers, could then be used to protect and immobilize
hydroxyl-containing free molecules for their further modification in the course of solid-phase
synthesis. The reaction proved selective for primary hydroxyls in the presence of secondary, and
secondary over tertiary. Together with product free alcohols, later cleavage of these silyl ethers
with aqueous acid or base also released the corresponding silanol groups, while HF, under
particularly mild and selective conditions, left silyl fluorides that themselves could then be very
conveniently regenerated by BCl3/CH2Cl2 back to the polymer-supported silyl chlorides. For a
variety of possible applications, these solid-phase protecting groups offer control, yield, recovery,
and reusability, as well as other possible advantages involving site-site isolation of functional groups
within the cross-linked polystyrene matrix.
For many conceivable organic applications involving
protecting groups, the use of insoluble supports bearing
organosilyl halide moieties would, by analogy with other
types of solid-phase synthesis, greatly ease product
separation, improve yields, allow automation, and pos-
sibly show other benefits from site separation and a
polymer microenvironment. However, to become widely
adopted as tools for organic synthesis, such materials
would have to be easy enough to prepare, or regenerate
for reuse, or both. Thus, though halosilyl-functionalized
cross-linked polystyrene Ps-SiRR′-X (R ) R′ ) Me or
Ph; X ) Cl or other leaving group) has already proven
useful,1 the difficulty and expense of its preparation using
organolithium reagents, and the inefficiency of its regen-
eration due to the lability of its aryl-silicon linkage
toward the Lewis acid reagents needed for its regenera-
tion,2 have limited its application. In common with many
other functional polymers from (chloromethyl)polysty-
rene, the alternative Ps-CH2-SiRR′-X would have a
benzylic bond that is still fragile toward other species (i.e.,
fluoride).3,4 Functional polymers with longer spacers
would be chemically much more robust: but, though Ps-
CH2CH2CH2-SiRR′-X has been prepared and regener-
ated from the silanol,5,6 the preparation was involved, and
the regeneration conditions did not apply to the silyl
fluoride of greater practical interest.
Controlled polymerization of technical-grade divinyl-
benzene:ethylstyrene results in macroporous beads of
cross-linked polystyrene that bear many residual vinyl
groups (typical Xf ) degree of functionalization ) 0.20
to over 0.40 mol of functional groups per mole of polymer
repeating units). Anti-Markovnikov addition of various
small molecules HZ to such (vinyl)polystyrene Ps-
CH)CH2 (also commercially available as Amberlite
XAD-2 and XAD-4) gives many useful spacer-containing
functional polymers Ps-CH2CH2-Z, in which functional
groups are accessible, stably-linked, and mobile.7-9 Hy-
drosilylation of alkenes is a well-known means for
preparing organosilicon compounds,10 and by this reac-
tion, as catalyzed by dicyclopentadienylplatinum(II) chlo-
ride, poly(methylhydrosiloxane) has been grafted onto
commercial (vinyl)polystyrene.11 However, many such
organoplatinum catalysts cause mixed R and â addition
to vinylarenes or, from being colloids, cannot penetrate
a cross-linked polymer matrix.11,12 Procedures using an
alternative, free-radical mechanism, though able to add
HSiCl3 and HSiMe2Ph to Ps-CH)CH2, proved ineffective
(6) Stover, H.; Lu, D. H.; Zhi, P.; Fre´chet, J . M. J . Polymer Bull.
1991, 575-582.
* To whom correspondence should be addressed. E-mail: darlingg@
activematerials.ca.
(7) Gao, J . P.; Morin, F. G.; Darling, G. D. Macromolecules 1993,
26, 1196-98.
X Abstract published in Advance ACS Abstracts, J uly 15, 1997.
(1) Danishefsky, S. J .; McClure, K. F.; Randolph, J . T.; Ruggeri, R.
B. Science 1993, 260, 1307.
(8) Stranix, B. R.; Darling, G. D. Biotechnol. Tech. 1995, 9, 75-80.
(9) Stranix, B. R.; Gao, J . P.; Barghi, R.; Salha, J .; Darling, G. D. J .
Org. Chem., in press.
(2) Chan, T. H.; Huang, W. Q. J . Chem. Soc., Chem. Commun. 1985,
909-911.
(10) Ojima, I. In The Chemistry of Organic Silicon Compounds;
Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1989; Vol. 2, pp 1479-
1526.
(11) Zhengpu, Z.; Hodge, P.; Stratford, P. W. Reactive Polym. 1991,
15, 71-77.
(12) Lewis, L. N.; Lewis, N.; Uriarte, R. J . In Homogeneous Transi-
tion Metal Catalyzed Reactions; Moser, W. R., Slocum, D. W., Eds.;
Advances in Chemistry Series 230; American Chemical Society:
Washington, D.C., 1992; pp 541-549.
(3) Darling, G. D.; Fre´chet, J . M. J . J . Org. Chem. 1986, 51, 2270-
2276.
(4) Darling, G. D.; Fre´chet, J . M. J . In Chemical Reactions on
Polymers; Benham, J . L., Kinstle, J . F., Eds.; American Chemical
Society: Washington, DC, 1988; Vol. 364, p 24-36.
(5) Fre´chet, J . M. J .; Darling, G. D.; Itsuno, S.; Lu, P. Z.; Vivas de
Meftaji, M.; Rolls, W. A., J r. Pure Appl. Chem. 1988, 60, 353-364.
S0022-3263(96)00603-2 CCC: $14.00 © 1997 American Chemical Society

6184 J . Org. Chem., Vol. 62, No. 18, 1997
Stranix et al.
Ta ble 2. Selectivity of Resin 2-Cl for Alcoh ols
alcohols added alcohols protected
EtOH:i-PrOH 1:1
i-PrOH:t-BuOH 1:1
EtOH:i-PrOH:t-BuOH 1:1:1
Sch em e 1
EtOH only
i-PrOH only
EtOH only
Sch em e 2
Ta ble 1. Mod ifica tion of Resin 1 by Co2(CO)8-Ca ta lyzed
Hyd r osilyla tion
mass of
Xf of
%
1 (g)
Xf of 1
product
product
conversion
1
50
1
30
25
0.43
0.30
0.25
0.30
0.30
2-Cl
2-Cl
3-Cl
3-Cl
4-Cl
0.23
0.15
0.12
0.14
0.10
53
50
48
47
33
with such silanes as HSiMe2Cl, which cannot as easily
form radical intermediates.9
Dicobalt octacarbonyl Co2(CO)8 has been established
as an effective, homogeneous catalyst for the exclusive
â-hydrosilylation of various substituted styrenes.13-15
Following its application to our analogous polymer
modification, we can now report on a quite facile method
for preparing solid-phase organosilicon protecting groups,
along with equally straightforward procedures for their
application and regeneration.16
Resu lts a n d Discu ssion
a
Key: (a) Pyr, CH2Cl2; (b) BzCl, pyr, CH2Cl2; (c) HF, H2O,
MeCN; (d) BCl3, CH2Cl2.
The action of an excess of several HSiRR′Cl on (vi-
nyl)polystyrene in the presence of 2-4 mol % Co2(CO)8
gave, after a few days, exclusive â-hydrosilylation of
polymer vinyls, as determined by FT-IR, solid-phase 13C
and 29Si NMR, and elemental analysis (Scheme 1, Table
1). Supported on a hydrophobic polymer matrix, the
resulting silyl chloride groups were found to be relatively
stable in air, with only a little hydrolysis from atmo-
spheric moisture after several weeks in the open.
The resulting solid-phase trialkylchlorosilanes were
reacted with alcohols in the presence of pyridine or tri-
ethylamine, in a suitable solvent, to form the correspond-
ing trialkylsilyl ethers. Elemental analysis of the product
from a nitrogen-containing alcohol confirmed spectral
analyses showing complete reaction. Solid-phase 13C
NMR experiments on Ps-CH2CH2-Si(CH3)2-O(CH2)3CH3
(2-OBu ) showed, by retention of the peak corresponding
to Si-O-CH2 in the dipolar dephased (DD) spectrum,
that the protected and supported alcohol moiety was very
mobile (liquid-like), thus unhindered, thus probably
accessible.7 The use of alcohol mixtures showed selective
protection of primary hydroxyls in the presence of
secondary, and secondary over tertiary (Table 2).
by wash, excess benzoyl chloride + base, wash, then
cleavage of silyl ether, gave only the 3-monobenzoylated
product in high isolated yield, and none of the dibenzoy-
lated or the 1-monobenzoylated products (Scheme 2). The
small quantity of starting material that was also recov-
ered can be explained by diprotection of the starting diol,
through reaction with two insufficiently-separated17 Si-
Cl groups within the 2-Cl resin. This side reaction,
further cross-linking the polymer matrix as well as
decreasing final yields of desired product, could arise
from primary hydroxyls of diol molecules still diffusing
into the resin, having to compete with certain secondary
hydroxyls of diol already bound. This model is supported
by the observation that a greater excess of diol as starting
material (from 1.2- to 1.5- to 3.5-fold) actually results in
less diol appearing as side product at reaction’s end (from
15% to 12% to 10%, respectively). Use of more hindered
silyl halide groups for still better selectivity of primary
> secondary hydroxyl, and of lower-functionalized resin
for better site-isolation, should help even more.
Somewhat similar to certain small-molecule stud-
ies,13,14 except for using better-swelling solvents than
methanol or ethanol, conditions were explored for acid-
and base-catalyzed hydrolysis and fluoride-induced cleav-
age of sample polymeric menthyl silyl ethers 2-OMen
and 3-OMen (Table 3). Because of deviations from
pseudo-first-order kinetics (probably due to reaction
heterogeneity and site variability), and for other practical
reasons, the relative stability of each was here expressed
as the time needed for its apparent 100% deprotection
and release of menthol (t100).
1,3-Butanediol, bearing both primary and secondary
hydroxyls on the same molecule, normally reacts with
an excess of acylating agent to give mixtures of 1-monoa-
cylated and 1,3-diacylated products. In a demonstration
of solid-phase synthesis, its addition to 2-Cl, followed
(13) Sommer, L. H. Stereochemistry, Mechanism and Silicon;
McGraw-Hill: New York, 1965.
(14) Barton, T. J .; Tully, C. R. J . Org. Chem. 1978, 43, 3649-3653.
(15) Magomedov, G. K. I.; Andrianov, K. A.; Shkolnik, O. V.;
Izmailov, B. A.; Kalinin, V. N. J . Organomet. Chem. 1978, 149, 29-
36.
(16) Stranix, B. R.; Liu, H. Q.; Darling, G. D. Abstracts of Papers,
27th Organosilicon Symposium, Troy, NY, Mar 18-19, 1994; Rensse-
laer Polytechnic Institute: Troy, 1995; Abstract P-76.
(17) Leznoff, C. C.; Fyles, T. M.; Weatherston, J . Can. J . Chem. 1977,
55, 1143-1149.

Functional Polymers from (Vinyl)polystyrene
J . Org. Chem., Vol. 62, No. 18, 1997 6185
Ta ble 3. Tim e for Com p lete Relea se of Men th ol fr om
P olym er ic Men th yl Silyl Eth er s (t100) u n d er Va r iou s
Con d ition s
later be weighed into a reaction vessel and then “acti-
vated” in situ to the silyl chloride form by treatment with
a commercial dichloromethane solution of boron trichlo-
ride, which also dried the solid powder of any remaining
adsorbed atmospheric moisture. Subsequent application
of vacuum, or flushing with dry nitrogen gas, then
removed BCl3 reagent, CH2Cl2 solvent, and BF3 side
product, all of them volatile, leaving a dry powder ready
for use in organic synthesis.
We continue to investigate these silyl-bearing polymers
and their further derivatives for application as solid-
phase reagents and catalysts, as well as protecting groups
for other solid-phase syntheses.
condnsa
2-OMen
3-OMen
1.0 M HCl
<30 sb
30 minc
1.0 M Me4N+-OH
0.22 M HF
2 hc
no reaction after 48 hd
30 min
10 min
a
1.0 mmol of substrate, 5 mL of 9:1 dioxane:water, 22-24 °C.
Desilylation was so fast that t100 could not be determined
b
accurately. c Kinetics curves of triplicate experiments matched to
d
(10%. Subsequent addition of 1 equiv of HFaq released all
menthol within 30 min.
Ta ble 4. Effect of Recyclin g on An a lysis of
2/3-p-OCH2CH2P h NO2
from 2-Cl
from 3-Cl
Exp er im en ta l Section
no. of recycles
% Si
% N
% Si
% N
Where specified “dry” below, aromatic solvents were distilled
from calcium hydride and other solvents and reagents stored
over molecular sieves. Otherwise, chemicals were used as
received. Organohalosilanes came from Hu¨ls America and
dicobalt octacarbonyl from J ohnson Matthey. (Vinyl)polysty-
rene (1) was prepared as previously described.7 Transmittance
FT-IR spectra were recorded of samples ground and pressed
into dry KBr pellets or spread onto IR-transparent silicon
wafers. 13C and 29Si CP-MAS (cross polarization/magic angle
spinning) and 13C CP-MAS-DD (cross polarization/magic angle
spinning/dipolar-dephasing, τ ) 45 µs; in the lists following,
peaks labeled “DD” persist here) solid-phase NMR spectra
were obtained on a Chemagnetics Inc. M-100 spectrometer.
Elemental analyses were done by Robertson Microlit Labora-
tories (NJ ).
0 (“fresh”)
1
2
2.54
2.55
2.54
1.30
1.30
1.25
2.30
2.30
2.28
1.17
1.17
0.84
After hydrolysis and drying, the spent resin 2-OH was
found to also contain Si-O-Si groups, according to solid-
phase 29Si NMR, further demonstrating that site isolation
was not complete. Possibly because of greater steric
hindrance, such siloxane cross-linking appeared negli-
gible under all conditions for producing 3-OH.
Fluoride cleaved all polymer-bound silyl ethers and
other silyl derivatives to the corresponding 2-F , 3-F ,
or 4-F , showing slight or no disiloxane formation by 29Si
NMR. Such reaction generally proved slower with tet-
rabutylammonium fluoride (TBAF) than with hydrofluo-
ric acid (>18 h vs. 30 min, for 3-OMen ), perhaps owing
to the more bulky and ionic nature of the former reagent
that hindered its penetration into the hydrophobic poly-
mer matrix, and to greater assistance by protons with
the latter. In surprisingly convenient fashion, boron
trichloride was then able to convert the solid-phase silyl
fluoride resins back to the starting silyl chlorides, a
reaction presumably driven by the conversion of admit-
tedly strong Si-F bonds to even stronger B-F and
evolution of gaseous BF3 side product. Elemental and
solid-phase NMR analyses, and experiments with [(fluo-
rodimethylsilyl)ethyl]benzene as a small molecule ana-
logue, confirmed that transformation was clean and
complete. Alternative regeneration of the fluoride through
triethylamine-catalyzed halogen exchange with Me3SiCl18
proved much less successful. The same BCl3 also con-
verted SiOR and SiOH to SiCl, presumably with borate
ester or anhydride as side products, whereas similar
reactions with SOCl2 or MeSO2Cl did not proceed to
completion.
Elemental analysis showed 2-p-OCH2CH2P h NO2 to
be the same whether produced from “fresh” or once- or
twice-recycled 2-Cl (Table 4), and the polymer showed
no spectral changes after two further such cycles. 3-Cl,
however, after recycling twice, could not be completely
regenerated even after 3 days, thus showing some loss
of active functionality.
On the basis of these results, we eventually took to
treating these organosilicon functional polymers with
aqueous hydrogen fluoride immediately after their prepa-
ration by hydrosilylation, thus transforming them to the
inert silyl fluoride forms most suitable for long-term
storage, while also washing out any remaining traces of
dicobaltoctacarbonyl catalyst. As needed, portions could
â-Hyd r osilyla tion of (vin yl)p olystyr en e (gen er a l p r o-
ced u r e). (Vinyl)polystyrene (1), of Xf 0.20-0.45 (25-50 g, 55-
111 mmol CH)CH2), was oven-dried for 2 days and then
transferred to a flame-dried, two-neck round-bottom flask,
which was then purged with nitrogen. To this was added 100
mL of dry toluene, and the suspension was heated on an oil
bath with slow stirring at 60 °C. A dialkylchlorosilane (4-8-
fold excess) or alkyldichlorosilane (4-fold excess) was then
slowly injected. Dicobalt octacarbonyl (2-4 mol %) was finally
added portionwise, whereupon the mixture turned deep blue
to brown. Stirring continued for 5 days, with 0.2-0.3 g of
suspended solids being sampled after 2 and 4 days for FT-IR
assay. The suspension was then filtered under nitrogen and
the residue placed in a Soxhlet extractor, extracted with dry
benzene or toluene until the filtrate was colorless, and then
dried in a vacuum oven at 40 °C for 24 h to give 2-Cl, 3-Cl,
and 4-Cl as pale blue beads.
[(Ch lor od im eth ylsilyl)eth yl]p olystyr en e (2-Cl). From
1 (C10H12
)0.45(C10H10)0.25(C8H7CH)CH2)0.30. Product 2-Cl: FTIR
(KBr) 1262 (SiCH3), 1078 (SiO), 472 cm-1 (SiCl); solid-phase
29Si NMR (300 MHz) δ 29.9 (CH2SiMe2Cl), 6.3 ppm (w, SiOSi);
13C CP-MAS NMR (100 MHz) δ 146 (DD), 129, 43, 31, 16 (DD),
3 ppm (DD, SiCH3). Anal. Calcd for (C10H12
)0.45(C10H10
)
0.40
-
(C12H17SiCl)0.15 (50% conversion): Si, 2.90. Found: Si, 2.88.
[(Ch lor od iisop r op ylsilyl)et h yl]p olyst yr en e (3-Cl).
From
1
(C10H12
)
0.45(C10H10
)
0.25(C8H7CH)CH2)0.30
.
Product
3-Cl: FTIR (KBr) 1453, 1364, 1062, 997, 888, 572 cm-1; solid-
phase 29Si NMR (300 MHz) δ 33.9 ppm (SiCl), 5.1 ppm (w,
SiOSi). Anal. Calcd for (C10H12)0.45(C10H10)0.41(C16H25SiCl)0.14
(47% conversion): Si, 2.58. Found: Si, 2.56.
[(Dich lor om eth ylsilyl)eth yl]p olystyr en e (4-Cl). From
1 (C10H12)0.45(C10H10)0.25(C8H7CH)CH2)0.30. Product 4-Cl: FTIR
(Si wafer) 1264, 1078, 997, 888, 551 cm-1; solid-phase 29Si NMR
(300 MHz) δ 30.6 ppm (SiCl); CP-MAS 13C NMR (100 MHz) δ
146 (DD), 129, 43, 31, 16 (DD), 5 ppm (DD; SiCH3). Anal.
Calcd for (C10H12)0.45(C10H10)0.45(C15H22SiCl2)0.10 (33% conver-
sion): Si, 2.58. Found: Si, 2.56.
[[[(p -Nit r op h e n e t h yl)oxy]d im e t h ylsilyl]e t h yl]p oly-
st yr en e (2-p -OCH2CH 2P h NO2). Resin 2-Cl (C10H12
)
-
0.45
(C10H10)0.40(C12H17SiCl)0.15 (0.50 g, 0.51 mmol Cl) was placed
in a 25 mL round-bottom flask containing p-nitrophenethyl
alcohol (0.20 g, 1.2 mmol) and dry CH2Cl2:pyridine 5:1 (6 mL,
12 mmol pyr). The reaction was stirred for 8 h at room
(18) Kanner, B.; Baily, D. L. U.S. Patent 3 128 297, 1964.

6186 J . Org. Chem., Vol. 62, No. 18, 1997
Stranix et al.
temperature under N2 and filtered and the residue extracted
in a Soxhlet overnight with CHCl3 and then dried in vacuo 24
h to yield 0.51 g of 2-p-OCH2CH2PhNO2 as pale gray beads:
29Si solid-phase NMR (300 MHz) δ 14.3 ppm. Anal. Calcd
isolated yield): 1H NMR (CD3OD, TMS) δ 8.1 (dd 2H), 7.5 (m
3H), 5.4 (m 1H), 3.5 (q 2H), 1.9 (dd 2H), 1.7 ppm (d 3H); 13C
NMR (270 MHz, CD3OD, TMS) δ 167, 133, 130, 129, 128, 68,
57, 40, 20 ppm.
for (C10H12)0.45(C10H10)0.40(C20H24SiNO3)0.15 (100% conversion):
Kin et ics of Clea va ge of P olym er -Su p p or t ed Silyl
Eth er s 2-OMen a n d 3-OMen . Resins 2-OMen or 3-O-
Men (1 g, ca. 1 mmol of -OR) were added to 1 M HCl or 1 M
(Me)4N+-OH (5 mL, 5 mmol) in dioxane:water 9:1, with 1 mM
triglyme as internal standard, at 22-24 °C, and the release
of menthol into the liquid phase was monitored by gas
chromatography. The rates were obtained by comparison with
1 mmol of each substrate hydrolyzed with 52% hydrofluoric
acid (0.04 mL, 1.1 mmol HF) in dioxane:water 9:1 with 1 mM
triglyme as internal standard, where complete conversion after
0.5 h was confirmed by 29Si solid-state NMR on the 2-F and
3-F products.
Si, 2.55; N, 1.27. Found: Si, 2.54; N, 1.30.
[[[(p -N it r o p h e n e t h y l)o x y ]d iis o p r o p y ls ily l]e t h y l]-
polystyr en e (3-p-OCH2CH2P h NO2). Resin 3-Cl (C10H12
)
-
0.45
(C10H10)0.41(C16H25SiCl)0.14 (0.50 g, 0.46 mmol Cl) was treated
as for 2-p-OCH2CH2P h NO2 to yield 0.50 g of 3-p-OCH2-
CH2P h NO2 as light gray beads: FTIR (KBr) 1610, 1455, 1342,
1255, 1062, 918 cm-1 29Si NMR δ 13.8 (SiOCH2), 7.6 ppm (w,
;
SiOSi). Anal. Calcd for (C10H12)0.45 (C10H12)0.41(C24H32SiNO3)0.14
(100% conversion): Si, 2.30; N, 1.15. Found: Si, 2.30; N, 1.17.
[(Bu t oxyd im et h ylsilyl)et h yl]p olyst yr en e (2-OBu ).
Resin 2-Cl (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (0.50 g, 0.51
mmol Cl) was reacted as in 2-p-OCH2CH2P h NO2, with
n-butyl alcohol (0.20 g, 1.2 mmol): 13C NMR δ 146, 130 (DD),
61 (DD), 43, 31 (DD), 18 (DD), 15 (DD), 1 ppm (DD).
[(Men th oxydim eth ylsilyl)eth yl]polystyr en e (2-OMen ).
[(F lu or od im eth ylsilyl)eth yl]p olystyr en e (2-F ). Resin
2-Cl (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (12.21 g, 12.3 mmol
Cl) was placed in a 50 mL polypropylene tube and swollen with
25 mL of MeCN. Hydrofluoric acid 52% (5 mL, 145 mmol of
HF) was added and the tube shaken for 3 h. The suspension
was then filtered and the residue washed with 200 mL of
MeCN and then dried in vacuo at 60 °C for 24 h, yielding 12.17
g of 2-F as light gray beads: FTIR 1256 (SiCH3), 875 cm-1
(SiF); solid-phase 29Si NMR (300 MHz) δ 29.3 (d, SiF), 6.3 (w,
Resin 2-Cl (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (5.00 g, 5.1
mmol Cl) was added to (-)-menthol (2.00 g, 12.7 mmol) in
CH2Cl2:pyridine 5:1 (10 mL, 20 mmol pyr) and the mixture
stirred at room temperature 12 h. The beads were then
filtered and extracted in a Soxhlet for 12 h with CHCl3 and
then dried in vacuo overnight, yielding 5.16 g of 2-OMen as
light gray beads: FTIR (Si wafer) 1455, 1342, 1255, 1062, 918
cm-1; solid-phase 29Si NMR (300 MHz) δ 13.2 ppm (SiOCH<),
7.8 ppm (w, SiOSi).
SiOSi). Anal. Calcd for (C10H12)0.45(C10H10)0.40((C12H17Si)2O)0.015-
(C12H17SiF)0.12 (80% conversion, remainder siloxane): F, 1.60.
Found: F, 1.64.
[(Flu or odiisopr opylsilyl)eth yl]polystyr en e (3-F). Resin
3-Cl (C10H12)0.45(C10H10)0.41(C16H25SiCl)0.14 (1.02 g, 0.96 mmol
[(Men t h oxyd iisop r op ylsilyl)et h yl]p olyst yr en e (3-O-
Men ). Resin 3-Cl (C10H12)0.45(C10H10)0.41(C16H25SiCl)0.14 (4.60
g, 4.30 mmol Cl) was treated as in the preparation of 2-Cl,
giving 5.00 g of 3-OMen : solid-phase 29Si NMR (300 MHz) δ
12.1 ppm.
Selectivity in for m in g silyl eth er s fr om 2-Cl. Resin
2-Cl (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (0.50 g, 0.51 mmol
Cl) was reacted with mixed alcohols (0.1 mL, ca. 2.5 mmol) in
toluene:triethylamine 14:1 (6 mL, 3 mmol R3N) as for 2-p-
OCH2CH2P h NO2. The suspended solid products were then
recovered by filtration (0.35 g, 0.35 mmol 2-OR), washed with
toluene, treated with 1 M TBAF in THF (2 mL, 2 mmol)
overnight, and extracted with 2 mL of ether, and the filtrate
was analyzed for relative content of each alcohol by gas
chromatography.
Cl) was placed in a 50 mL polypropylene tube with 20 mL of
dioxane and hydrofluoric acid 52% (5 mL, 145 mmol of HF).
The suspension was then shaken 12 h and filtered and the
residue washed 5× with 20 mL acetone and dried in vacuo at
65 °C for 24 h, yielding 0.99 g 3-F as light gray beads: FTIR
(KBr) 1610, 1342, 998, 880 cm-1; solid-phase 29Si NMR (300
MHz) δ 26.7 (d, SiF), 2.5 ppm (w, SiOSi). Anal. Calcd for
(C10H12
Found: F, 1.40.
Regen er a tion of [(Ch lor od im eth ylsilyl)eth yl]p olysty-
r en e (2-Cl). Resin 3-F (C10H12)0.45(C10H10)0.40((C12H17Si)2O)0.015
)0.45(C10H12)0.41(C16H25SiF)0.14 (100% conversion): F, 1.47.
-
(C12H17SiF)0.12 (1.26 g, 1.29 mmol F) was added to a polypro-
pylene tube in an ice bath and purged with N2, 1 M BCl3 in
CH2Cl2 (5 mL, 5 mmol) was added slowly, and the mixture
was then stirred for 3 h. The ice bath was then removed, and
N2 was passed over the suspension until all liquids had
disappeared. The remaining beads were rinsed with CH2Cl2
under N2 and then vacuum dried for 12 h, yielding 1.21 g of
2-Cl as light brown beads: FTIR (KBr in air) 1258, 1079 cm-1
(w); solid-phase 29Si NMR (300 MHz) δ 30.1 ppm; 13C CP-MAS
NMR (100 MHz) δ 146 (DD), 129, 43, 31, 16 (DD), 3 ppm (DD).
Use of 2-Cl in th e Solid -P h a se Syn th esis of 3-Hy-
d r oxy-1-m eth ylp r op yl Ben zoa te (H-OBu OBz). Resin
2-Cl (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (25.01 g, 25 mmol
Cl) was added to 1,3-butanediol (6.00 g, 67 mmol) in CH2Cl2:
pyridine 5:1 (50 mL, 100 mmol of pyr), and the mixture was
rapidly stirred overnight. The suspension was then filtered
and the residue washed with warm CHCl3; FTIR of a dried
small portion showed a strong peak at 3550 cm-1
. The
Anal. Calcd for (C10H12)0.45(C10H10)0.40(C12H17SiCl)0.15 (100%
remaining beads were then resuspended in a solution contain-
ing benzoyl chloride (7.01 g, 50 mmol) in CH2Cl2:pyridine 5:1,
refluxed 24 h, and then recovered by filtration and washed
with CHCl3; another small dried portion showed a different
strong FTIR peak at 1746 cm-1. The remaining polymer was
suspended in a mixture of 52% hydrofluoric acid (10 mL, 270
mmol HF) and 50 mL of MeCN for 3 h, the suspension was
then filtered, and the filtrate was extracted with ethyl acetate
(2 × 100 mL), GC analysis of which showed 10% starting 1,3-
butanediol, besides a single other product peak. Evacuation
of the solvent and Kugelrohr distillation (12 mm Hg, 130-
140 °C) afforded 3.6 g of H-OBu OBz as a clear oil (90%
conversion): Cl, 3.51. Found: Cl, 3.48.
Ack n ow led gm en t. Thanks for financial support are
due to the Natural Sciences and Engineering Research
Council (NSERC) and Respiratory Health Network of
Centres of Excellence (RHNCE, Inspiraplex) of Canada
and the Fonds pour Chercheur et l’Aide a la Recherche
(FCAR) of Que´bec. We also thank F. G. Morin for the
solid-phase NMR analyses.
J O960603M