reactions on polystyrene beads.7 Unfortunately, the linker
was degraded significantly under acidic conditions (e.g.,
TsOH), resulting in premature cleavage of the substrates from
the solid support. We report herein our solution to this
problem, entailing the synthesis and evaluation of a new,
robust tert-butyldiarylsilyl (TBDAS) linker 3.8
We then evaluated various methods of activating the silyl
hydride resin 6b for alcohol loading (Table 1) including
Table 1. Activation of the Silyl Hydride Resin 6b and Alcohol
Loading
By analogy to the tert-butyldiphenylsilyl (TBDPS) pro-
tecting group for alcohols,9 we envisioned that the TBDAS
linker 3 would be significantly more stable to acids than 1
or 2. Moreover, since the solution-phase analogues of 1 and
2 are uncommon, we recognized that the direct cor-
respondence between 3 and the well-established TBDPS
protecting group would offer a further advantage with respect
to predicting the stability of the linker and the reactivity of
TBDAS-linked substrates.
activating reagent
NBSa
NBSa
NBSa
NBSb
TfOHc
trichloroisocyanuric acida imidazole
1,3-dichloro-5,5-dimethyl- imidazole
hydantoind
basee
Et3N, DMAP CH2Cl2 69
Et3N, DMAP DMF 21
solvent yield (%)f
Our initial approach to 3 involved lithiation of bromopoly-
styrene resin 42,7 (Scheme 1), followed by treatment with
imidazole
imidazole
2,6-lutidine
CH2Cl2 78
CH2Cl2 73
CH2Cl2
0
CH2Cl2 23g
Scheme 1. Synthesis of the TBDAS Linker.
CH2Cl2 93-100
a Performed with 2 equiv. b Performed with 10 equiv. c Performed with
6 equiv. d Performed with 12 equiv. e Base (8-12 equiv) was added to the
washed resin in the solvent indicated, followed by 3 equiv of N-Fmoc-â-
alaninol. f Yields determined by Fmoc quantitation, relative to Si loading
level determined by elemental analysis. g Not optimized.
N-bromosuccinimide,16 triflic acid,17 and trichloroisocyanuric
acid.4 The best results were achieved via chlorination using
1,3-dichloro-5,5-dimethylhydantoin,5b which afforded highly
efficient loading of N-Fmoc-â-alaninol (Fmoc quantitation)
relative to the silyl hydride loading level.18
We also investigated loading of the TBDAS linker onto
brominated polystyrene SynPhase L-Series Lanterns,19 which
were lithiated, silylated, chlorinated, and loaded with N-Fmoc-
â-alaninol as described above. Elemental analysis indicated
16-19 µequiv of Si per Lantern (45-55% yield based on
the initial Br loading level). Residual bromine was detected
at 5.7-6.7 µequiv of Br per Lantern (16-19%).20 Fmoc
quantitation indicated 13-18 µequiv of alcohol loaded per
Lantern (37-52% overall yield). Although these yields are
lower than those achieved using the polystyrene beads above,
tert-butyldichlorophenylsilane 5a, which is commercially
available, albeit at relatively high cost.10 A large molar excess
of 5a was used, both to drive the silylation reaction to
completion and to reduce the amount of anticipated intrabead
cross-linking. The support-bound silyl chloride 6a was then
treated with N-Fmoc-â-alaninol and imidazole in CH2Cl2.
We were encouraged to find that standard Fmoc quantita-
tion11 indicated 38% alcohol loading. However, in view of
the expensive dichlorosilane 5a required, we elected to
pursue a second-generation approach to provide more practi-
cal access to the TBDAS linker.
Thus, we synthesized the known monochlorosilane 5b12
by addition of t-BuLi to dichlorophenylsilane. The freshly
prepared and distilled monochlorosilane was then used to
silylate lithiated bromopolystyrene to afford the shelf-stable
silyl hydride resin 6b. After significant optimization, linker
loading levels of 1.03-1.39 mequiv/g were achieved (Si
elemental analysis), corresponding to a 60-81% yield
relative to the initial bromine loading level (see Supporting
Information). Residual bromine was also detected at 0-10%
of the initial loading level (Br elemental analysis).13 Additives
such as HMPA, CuCN,14 or TMSCN15 did not increase the
loading level.
(13) An alternative metalation with i-Pr(n-Bu)2MgLi has been reported
recently: Thomas, G. L.; Ladlow, M.; Spring, D. R. Org. Biomol. Chem.
2004, 2, 1679-1681. This procedure also provided efficient metalation
(0.6% residual Br, elemental analysis) but did not improve net linker loading
(49% yield, Si elemental analysis).
(14) Shirahata, A. Tetrahedron Lett. 1989, 30, 6393-6394.
(15) Lennon, P. J.; Mack, D. P.; Thompson, Q. E. Organometallics 1989,
8, 1121-1122.
(16) Petit, M.; Chouraqui, G.; Aubert, C.; Malacria, M. Org. Lett. 2003,
5, 2037-2040.
(17) (a) Hu, Y.; Porco, J. A. Tetrahedron Lett. 1999, 40, 3289-3292.
(b) Smith, E. M. Tetrahedron Lett. 1999, 40, 3285-3288.
(18) Overall loading level with 3 is comparable to the levels we achieved
using 1 (62-65%, Fmoc quantitation, relative to initial Br loading level)
and 2 (40-59%).
(7) Polymer Laboratories: PL-PBS Resin, 150-300 µm, 25%
p-bromostyrene, 1% divinylbenzene, 2.0 mequiv Br/g.
(8) For a report of an amide-linked anilino-tert-butylphenylsilyl linker,
see: Mullen, D. G.; Barany, G. J. Org. Chem. 1988, 53, 5240-5248.
(9) Hanessian, S.; Lavallee, P. Can. J. Chem. 1975, 53, 2975-2977.
(10) Lancaster Synthesis, $94.80/5 g.
(19) Mimotopes: SynPhase PS L-Series brominated Lanterns, 35 µequiv
Br/Lantern.
(11) Meienhofer, J.; Waki, M.; Heimer, E. P.; Lambros, T. J.; Makofske,
R. C.; Chang, C. D. Int. J. Pept. Protein Res. 1979, 13, 35-42.
(12) Maier, G.; Kratt, A.; Schick, A.; Reisenauer, H. P.; Barbosa, F.;
Gescheidt, G. Eur. J. Org. Chem. 2000, 1107-1112.
(20) Use of the alternative i-Pr(n-Bu)2MgLi metalation procedure (ref
13) yielded Lanterns with 16.1 µequiv of residual Br (46% of initial loading,
Br elemental analysis) and 12.7 µequiv of linker loading (36% yield, Si
elemental analysis).
1778
Org. Lett., Vol. 7, No. 9, 2005