Organocatalytic Regioselective Chlorosilylation of Oxirane Derivatives: Mild and Effective Insertion of Bulky Silyl Chloride by Using 4-Methoxypyridine N-Oxide

Article abstract of DOI:10.1002/adsc.201600209

The highly regioselective organocatalytic chlorosilylation of oxirane derivatives using bulky silyl reagents such as (tert-butyl)diphenylsilyl chloride (TBDPSCl) or triphenylsilyl chloride (Ph₃SiCl) was developed. The reaction was effectively catalyzed with 4-methoxypyridine N-oxide in the presence of sodium sulfate (Na₂SO₄). Several silylated halohydrins were obtained with complete regioselectivity and the reaction was applied to the synthesis of carvedilol. (Figure presented.) .

Full text of DOI:10.1002/adsc.201600209

COMMUNICATIONS
DOI: 10.1002/adsc.201600209
Organocatalytic Regioselective Chlorosilylation of Oxirane
Derivatives: Mild and Effective Insertion of Bulky Silyl Chloride
by Using 4-Methoxypyridine N-Oxide
Keisuke Yoshida,a,* Hina Suzuki,a Hiroki Inoue,a Kohei Matsui,a Yuta Fujino,a
Yohei Kanoko,a Yukihiro Itatsu,a and Ken-ichi Takaoa,*
a
Department of Applied Chemistry, Keio University, Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
Fax: (+81)-45-566-1551; phone: (+81)-45-566-1570; e-mail: yoshida@applc.keio.ac.jp or takao@applc.keio.ac.jp
Received: February 22, 2016; Revised: April 5, 2016; Published online: June 2, 2016
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201600209.
Abstract: The highly regioselective organocatalytic
chlorosilylation of oxirane derivatives using bulky
silyl reagents such as (tert-butyl)diphenylsilyl chlo-
ride (TBDPSCl) or triphenylsilyl chloride (Ph3SiCl)
was developed. The reaction was effectively cata-
lyzed with 4-methoxypyridine N-oxide in the pres-
ence of sodium sulfate (Na2SO4). Several silylated
halohydrins were obtained with complete regiose-
lectivity and the reaction was applied to the synthe-
sis of carvedilol.
good regioselective insertion into 1,2-epoxy com-
pounds. These methods could provide an alternative
approach for the synthesis of halohydrins because of
the easy removal of the trimethylsilyl (TMS) group.
However, it is difficult to maintain the TMS group
during transformations under harsh conditions when
using trimethysilylated halohydrins as synthetic inter-
mediates and therefore a method for inserting a bulki-
er silyl reagent such as TBS or TBDPS is required.
We recently reported that the combination of
TBDPSCl, DIPEA, and pyrrolidinopyridine N-oxide
(PPYO) catalyst silylated various secondary alcohols
(Scheme 1a),[12] and that amine-free silyl etherification
of various primary and secondary alcohols was ach-
ieved by using 4-methylpyridine N-oxide and MS 4Š
powder (Scheme 1b).[13] Other organocatalytic reac-
tions using N-oxide derivatives have been also report-
Keywords: amine N-oxides; organocatalysis; regio-
selectivity; silicon
1,2-Epoxy compounds are important building blocks ed by some groups.[14] In our reactions, pyridine N-
in organic chemistry. Regioselective ring-opening re- oxide derivatives showed excellent catalytic activity
actions of epoxides with various nucleophiles are and compatibility with bulky silyl reagents.[15] The for-
often utilized in the synthesis of natural products and mation of intermediate A was supported by NMR
bioactive compounds.[1] Considerable effort has been analysis of a mixture of R’3SiCl and pyridine N-oxide
directed toward regioselective reactions with halogens derivative. Since A was the likely active species in the
to synthesize halohydrin compounds.[2] Halohydrins reaction, we expected that the active species can acti-
are useful synthetic intermediates and are found in vate oxirane compounds. Herein, we describe the or-
the structures of halogenated natural products.[3] ganocatalytic regioselective chlorosilylation of oxirane
Direct methods have been reported for constructing derivatives using 4-methoxypyridine N-oxide and
halohydrins from oxiranes using various halogen sour- bulky silyl reagents (Scheme 1c).
ces such as metal halides,[4] ammonium halides,[5] hy-
We chose glycidyl phenyl ether (1) as the standard
drohalides[6] and elemental halogens[7] but these meth- substrate for catalyst screening (Table 1). Treatment
ods typically suffer from substrate incompatibility, of 1 with 1.7 equiv. TBDPSCl in CH2Cl2 at room tem-
low-to-moderate regioselectivity, high cost, or poor perature for 24 h did not produce chlorosilylated
stereoselectivity.
product 2 (entry 1). The addition of 30 mol% PPh3
These efforts have yielded several catalytic methods also was ineffective, although PPh3 has been reported
for promoting the insertion of silyl halides into oxir- as an effective catalyst for the chlorotrimethylsilyla-
anes. For example, the combinations TMSX-PPh3,[8] tion of oxirane compounds[8] (entry 2). The use of
TMSX-pyridine,[9] TMSX-CoCl2,[10] and TMSX-Et3N- 30 mol% dimethylaminopyridine (DMAP) as the cat-
NaBr[11] served as effective and mild reagents for alyst yielded only a small amount of 2 and chlorohy-
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1886
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
Table 1. Catalyst screening for the chloro-tert-butyldiphenyl-
silylation reaction.
Entry Catalyst Yield of 2[a] Yield of 3[a] Recovery of 1[a]
[%]
[%]
[%]
1
2
3
4
5
6
7
8
none
PPh3
DMAP
4
5
6
0
0
9
11
51
73
6
7
56
61
70
65
30
0
16
19
15
19
24
11
DMAPO 40
PPYO
32
50
30
[a]
Isolated yield.
Scheme 1. Reactions catalyzed by pyridine N-oxides.
Several additives were examined to suppress the
drin 3 (entry 3). In contrast, using pyridine N-oxide production of 3. Since undesired product 3 was pro-
derivatives instead of PPh3 or DMAP provided 2 in duced by a Brønsted acid generated unintentionally
significant yield. Of the catalysts tested (4–6, in situ, we expected that the production of 3 could be
DMAPO, and PPYO),[16] the N-oxide derivatives prevented using a water scavenger or mild base to
bearing an electron-donating group at the 4-position trap the acid. When MS 4Š was added as a water
of the pyridine ring were found to be effective for the scavenger, the product 2 was obtained in good yield
chlorosilylation reaction (entries 4–8) and 4-methoxy- (80%), but 3 was also produced in 19% yield
pyridine N-oxide (6) showed the highest activity. The (entry 5). The use of Na2SO4 significantly inhibited
insertion of TBDPSCl proceeded with complete re- the generation of 3 and 2 was obtained in excellent
gioselectivity to afford 2 as a single isomer in good yield (entry 6). We also examined various bases. Inor-
yield. We determined the structure of 2 by its trans- ganic bases such as NaHCO3 did not work effectively
formation into the known compound 3 by deprotec- (entry 7). Trialkylamines such as Et3N or DIPEA re-
tion of the TBDPS group.
sulted in no reaction. Bulky pyridine derivatives such
Next, we investigated the optimal reaction condi- as 2,4,6-collidine and 2,6-tert-butyl-4-methylpyridine
tions (Table 2). Treatment of 1 with 30 mol% 6 and were tested as proton capture agents (entries 8 and 9)
1.7 equiv. TBDPSCl in CHCl3 or 1,2-dichloroethane and 2,6-tert-butyl-4-methylpyridine showed reactivity
at room temperature resulted in moderate yield (en- similar to that obtained using Na2SO4. We chose
tries 1 and 2). The use of CH2Cl2 as the solvent result- Na2SO4 as the optimum additive because of the short
ed in a 73% yield of product 2 (entry 3). DMF afford- reaction time. A minimum amount of 20 mol% cata-
ed 2 in low yield, although this solvent often works ef- lyst 6 was required to provide a high yield of 2 (en-
fectively for various silyl etherification reactions[15] tries 10 and 11) and 30 mol% 6 was chosen as the op-
(entry 4). Using toluene and ether-type solvents timal amount based on the reaction time.
(Et2O or THF) resulted in no observable reaction.
We examined the versatility of the catalyst system
We therefore selected CH2Cl2 as the optimal solvent. for various silyl reagents (Table 3). The reactions pro-
ceeded smoothly at room temperature and gave the
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1887
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
Table 2. Optimization of the chloro-tert-butyldiphenylsilylation reaction.
Entry Additive
Solvent
CHCl3
ClCH2CH2Cl 30
CH2Cl2
DMF
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
X (mol%) Time Yield of 2[a] [%] Yield of 3[a] [%] Recovery of 1[a] [%]
1
none
30
24 h 57
24 h 57
24 h 73
24 h 27
24 h 80
36 h 95
24 h 32
50 h 35
50 h 95
60 h 71
48 h 94
11
10
19
67
19
5
26
25
0
2
none
3
none
30
30
30
30
30
30
30
10
20
4
5
6
none
MS 4Š
Na2SO4
0
0
0
49
55
0
0
0
[b]
[b]
[b]
7
8
9
10
11
NaHCO3
17
4
4
26
3
2,4,6-collidine[c]
2,6-di-t-Bu-4-Me-pyridine[c] CH2Cl2
[b]
Na2SO4[b]
CH2Cl2
CH2Cl2
Na2SO4
[a]
[b]
[c]
Isolated yield.
500 wt% was used.
2 equiv. was used.
Table 3. Scope of screened silyl reagents for the catalyst
system.
Entry R’3SiCl
Time Yield of Yield of Recovery of
2[a] [%]
3[a] [%]
1[a] [%]
1
2
3
4
5
TESCl
TBSCl
TBDPSCl 36 h 95
Ph3SiCl
TIPSCl
22 h 77
48 h 81
18
4
5
0
15
0
0
0
0
24 h 97
24 h
0
62
[a]
Isolated yields.
chlorosilylated product 2 and a small amount of halo-
hydrin 3 when a less hindered silyl reagent such as
TESCl or TBSCl was used (entries 1 and 2). The un-
desired formation of 3 was due to proton release from
the decomposed silyl reagent. As shown in entries 3
Figure 1. NMR study of 4-methoxypyridine N-oxide and
TBDPSCl.
and 4, the chlorosilylation of 1 using a bulkier silyl re- A mixture of TBDPSCl and 20 mol% 6 in CDCl3 at
1
agent such as TBDPSCl or Ph3SiCl proceeded without room temperature gave H NMR spectrum B. Com-
production of halohydrin 3, whereas TIPSCl was inef- parison of B with the spectra of 6 alone (A) and
fective (entry 5).
TBDPSCl alone (C) showed that both the aromatic
We conducted an NMR study to investigate the protons and 4-OMe group of 6 shifted downfield and
mechanism of the chlorosilylation reaction (Figure 1). a new peak corresponding to the t-Bu group appeared
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1888
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
at 1.09 ppm. TBDPS etherification of 3 was examined
to investigate if this chlorosilylation is a concerted re-
action. Treatment of 3 with 30 mol% 6 and 1.7 equiv.
TBDPSCl in CH2Cl2 at room temperature for 24 h re-
sulted in no reaction (Scheme 2), thus demonstrating
that silylation is accompanied by ring-opening.
The results of these experiments led us to propose
a reaction mechanism for the catalytic regioselective
chlorosilylation of oxirane by 4-methoxypyridine N-
oxide (Figure 2). First, R’3SiCl and 6 form intermedi-
ate A. The formation of A is supported by NMR
analysis of a mixture of TBDPSCl and 6.[17] The inter-
mediate A coordinates with the oxygen of oxirane
and activates it to open the epoxy ring, then chloride
anion attacks from the less sterically hindered posi-
tion. TBDPS etherification proceeds upon ring-open-
ing to give the product 2, and 6 is regenerated to com-
plete the catalytic cycle.
Figure 2. A possible mechanism for chloro-tert-butyldiphe-
nylsilylation by 4-methoxylpyridine N-oxide.
We explored the use of various oxirane derivatives
to expand the scope of this reaction (Table 4). Glycid-
yl ethers R=OPh (1a), O-allyl (1b), O-ethyl (1c), and
O-benzyl (1d) were treated under the optimized con-
ditions to afford TBDPS ethers in 95%, 89%, 94%, taining no oxygen atom other than that in the epoxy
and 91% yields, respectively (entries 1–4). Benzoylat- ring. 2-Hexyloxirane (1f), 2-(chloromethyl)oxirane
ed glycidyl (1e) was converted to 2e in 85% yield (1g) and 2-benzyloxirane (1h) were converted to the
(entry 5). Furthermore, we examined substrates con- products in high yields with good regioselectivities
(entries 6–8). The reaction of cyclohexene oxide (1i)
provided the product 2i in moderate yield and the
combination of Ph3SiCl with 6 afforded 2j in high
yield (entry 9). The reaction of styrene oxide (1k)
showed the opposite regioselectivity and produced 2k
due to the high stability of the carbocation at the ben-
zylic position (entry 10). This result suggests a Lewis
acidic nature of the transient pentacoordinated silicon
atom and thus supports the mechanistic scenario pre-
sented in Figure 2.
Scheme 2. Experiment to investigate the reaction mecha-
nism.
Table 4. Scope of the chloro-tert-butyldiphenylsilylation reaction.
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1889
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
We demonstrated the utility of chlorosilylation by Experimental Section
using the reaction in the synthesis of carvedilol (7)
(Scheme 3). Carvedilol is a b1-, b2-, and a1-adrenore- Typical Procedure for the Chlorosilylation Reaction
ceptor blocker drug for the treatment of hyperten-
sion, stable angina pectoris, and congestive heart fail-
To a stirred solution of glycidyl phenyl ether (1) (20.0 mg,
0.133 mmol, 1.0 equiv.), catalyst
6 (5.0 mg, 0.040 mmol,
ure.[18] An alternative and efficient synthesis of carve-
dilol was recently reported in which an O-protected
chlorohydrin was used as a key intermediate to im-
prove the overall yield.[19] We applied the chlorosilyla-
tion reaction to this synthetic route. 4-Glycidyloxycar-
bazole (8)[20] underwent regioselective opening of the
oxirane ring upon treatment with TBDPSCl in the
presence of 30 mol% catalyst 6 and Na2SO4 to pro-
duce the silylated chlorohydrin 9 in 92% yield. Cou-
30 mol%) and Na2SO4 (100 mg, 500 wt%) in CH2Cl2
(0.3 mL) was added TBDPSCl (62.2 mg, 0.226 mmol,
1.7 equiv.). The mixture was stirred at room temperature for
36 h, quenched with 1M aqueous HCl (10 mL), and extract-
ed with AcOEt (10 mL”3). The combined extracts were
washed with water (10 mL) and saturated brine (10 mL),
then dried and concentrated under reduced pressure. The
residue was purified by column chromatography on silica
gel (EtOAc/hexane, 1:30) to provide the silylated product 2
(yield: 53.8 mg, 95%) and chlorohydrin 3 yield: 1.2 mg
(5%).
pling of
9 with 2-(2-methoxyphenoxy)ethylamine
(10)[20] under basic conditions provided the O-silylat-
ed carvedilol 11, which was deprotected to afford car-
vedilol (7). The present approach realizes an access to
carvedilol in 3 steps with an overall yield of 27%
from 8, whereas the conventional method required 4
steps in 14% overall yield from 8.[19]
Acknowledgements
This work was supported by Keio Gijuku Academic Devel-
opment Funds.
In conclusion, we have developed a method for the
chlorosilylation of various oxirane derivatives using
a bulky silyl reagent, Na2SO4, and 4-methoxypyridine
N-oxide. The reaction proceeds under mild conditions
with excellent regioselectivity. We applied this reac-
tion to the synthesis of carvedilol. Our group is cur-
rently developing an asymmetric version of this
method using chiral pyridine N-oxide catalysts.
References
[1] a) A. R. Rodríguez, B. W. Spur, Tetrahedron Lett. 2003,
44, 7411–7415; b) S. E. Fangour, A. Guy, J.-P. Vidal, J.-
C. Rossi, T. Durand, J. Org. Chem. 2005, 70, 989–997;
c) G. Hernµndez-Torres, J. Mateo, A. Urbano, M. C.
CarreÇo, Eur. J. Org. Chem. 2013, 6259–6262; d) S. S.
Dachavaram, K. B. Kalyankar, S. Das, Tetrahedron
Lett. 2014, 55, 5629–5631; e) R. W. Bates, T. G. Lek,
Synthesis 2014, 46, 1731–1738.
[2] a) E. Mincione, G. Ortaggi, A. Sirna, J. Org. Chem.
1979, 44, 1569–1570; b) G. Palumbo, C. Ferreri, R.
Caputo, Tetrahedron Lett. 1983, 24, 1307–1310; c) R. D.
Dawe, T. F. Molinski, J. V. Turner, Tetrahedron Lett.
1984, 25, 2061–2064; d) R. Caputo, C. Ferreri, S. Nov-
iello, G. Palumbo, Synthesis 1986, 499–501; e) T. W.
Bell, J. A. Ciaccio, Tetrahedron Lett. 1986, 27, 827–830;
f) J. A. Ciaccio, K. J. Addess, T. W. Bell, Tetrahedron
Lett. 1986, 27, 3697–3700; g) Y. Guindon, M. Therien,
Y. Girard, C. Yoakim, J. Org. Chem. 1987, 52, 1680–
1686; h) N. N. Joshi, M. Srebnik, H. C. Brown, J. Am.
Chem. Soc. 1988, 110, 6246–6248; i) J. S. Bajwa, R. C.
Anderson, Tetrahedron Lett. 1991, 32, 3021–3024; j) P.
Bovicelli, E. Mincione, G. Grtaggi, Tetrahedron Lett.
1991, 32, 3719–3722; k) M. Chini, P. Crotti, C. Gardelli,
F. Macchia, Tetrahedron 1992, 48, 3805–3812; l) A. Sol-
ladiØ-Cavallo, P. Lupattelli, C. Marsol, T. Isarno, C.
Bonini, L. Caruso, A. Maiorella, Eur. J. Org. Chem.
2002, 1439–1444.
[3] a) J. D. Martin, J. M. Palazon, C. Perez, J. L. Ravelo,
Pure Appl. Chem. 1986, 58, 395–406; b) J. D. Martin, C.
Perez, J. L. Ravelo, J. Am. Chem. Soc. 1986, 108, 7801–
7811.
Scheme 3. Synthesis of carvedilol.
[4] a) J.-M. Bartas-Yacoubou, N. Maduike, S. Kyere, L.
Doan, D. L. Whalen, Tetrahedron Lett. 2002, 43, 3781–
3782; b) G. Righi, G. Pescatore, F. Bonadies, C. Bonini,
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1890
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
Tetrahedron 2001, 57, 5649–5656; c) H. Kotsuki, T. Shi-
manouchi, R. Ohshima, S. Fujiwara, Tetrahedron 1998,
54, 2709–2722; d) F. Azzena, F. Calvani, P. Crotti, C.
Gardelli, F. Macchia, M. Pineschi, Tetrahedron 1995,
51, 10601–10626; e) J. S. Bajwa, R. C. Anderson, Tetra-
hedron Lett. 1991, 32, 3021–3024.
[13] K. Yoshida, Y. Fujino, Y. Itatsu, H. Inoue, Y. Kanoko,
K. Takao, Tetrahedron Lett. 2016, 57, 627–631.
[14] For recent reports, see: a) I. Shiina, H. Ushiyama, Y.
Yamada, Y. Kawakita, K. Nakata, Chem. Asian J. 2008,
3, 454–461; b) J. I. Murray, R. Woscholski, A. C. Spivey,
Chem. Commun. 2014, 50, 13608–13611; c) J. I. Murray,
R. Woscholski, A. C. Spivey, Synlett 2015, 985–990;
d) J. I. Murray, A. C. Spivey, Adv. Synth. Catal. 2015,
357, 3825–3830; e) K. Ishihara, Y. Lu, Chem. Sci. 2016,
7, 1276–1280.
[5] a) M. Chini, P. Crotti, C. Gardelli, F. Macchia, Tetrahe-
dron 1992, 48, 3805–3812; b) M. Onaka, K. Sugita, H.
Takeuchi, Y. Izumi, J. Chem. Soc. Chem. Commun.
1988, 1173–1174.
[15] Zipse and co-workers have reported a mechanistic
analysis for the Lewis base-catalyzed silylation of alco-
hols: a) P. Patschinski, C. Zhang, H. Zipse, J. Org.
Chem. 2014, 79, 8348–8357; b) P. Patschinski, H. Zipse,
Org. Lett. 2015, 17, 3318–3321.
[16] PPYO was prepared by the described procedure: A. R.
Katritzky, D. Rasala, F. Brito-Palma, J. Chem. Res. (S)
1988, 42–43.
[17] There is a possibility of involvement of pentacoordi-
nate silicon (hypervalent) species in this transforma-
tion: R. Hrdina, F. Opekar, J. Roithovµ, M. Kotora,
Chem. Commun. 2009, 2314–2316.
[18] K. Beattie, G. Phadke, J. Novakovic, Profiles Drug
Subst. Excip. Relat. Methodol. 2013, 38, 113–157.
[19] B. Anandkumar, R. Buchi Reddy, L. Gangaiah, G.
Madhusudhan, K. Mukkanti, Der Pharma Chemica
2011, 3, 620–626.
[6] a) L.-X. Gao, A. Murai, Chem. Lett. 1989, 357–358;
b) L.-X. Gao, A. Murai, Chem. Lett. 1991, 1503–1504.
[7] a) H. Sharghi, M. M. Eskandari, R. Ghavami, J. Mol.
Catal. A: Chem. 2004, 215, 55–62; b) H. Sharghi, M. M.
Eskandari, Tetrahedron 2003, 59, 8509–8514; c) H.
Sharghi, M. M. Eskandari, Synthesis 2002, 1519–1522;
d) M. I. Konaklieva, M. L. Dahl, E. Turos, Tetrahedron
Lett. 1992, 33, 7093–7096.
[8] G. C. Andrews, T. C. Crawford, L. G. Contillo Jr, Tetra-
hedron Lett. 1981, 22, 3803–3806.
[9] a) S. D. Stamatov, J. Stawinski, Tetrahedron Lett. 2006,
47, 2543–2547; b) S. D. Stamatov, J. Stawinski, Eur. J.
Org. Chem. 2008, 2635–2643.
[10] J. Iqbal, M. A. Khan, Chem. Lett. 1988, 1157–1158.
[11] J. Iqbal, M. A. Khan, S. Ahmad, Synth. Commun. 1989,
19, 641–644.
[12] K. Yoshida, K. Takao, Tetrahedron Lett. 2014, 55, 6861–
6863.
[20] 4-Glycidyloxycarbazole (8) and 2-(2-methoxyphen-
oxy)ethylamine (10) were purchased from TCI.
Adv. Synth. Catal. 2016, 358, 1886 – 1891
1891
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim