1,2-Ferrocenediylazaphosphinines 2: A new class of nucleophilic catalysts for ring-opening of epoxides

Article abstract of DOI:10.1055/s-2003-38758

1,2-Ferrocenediylazaphosphinines (1a-c) have been successfully employed as a new class of nucleophilic catalysts for ring-opening of a range of epoxides, their catalytic efficiency in terms of regioselectivity as well as chemical yield comparing well with the existing catalysts in the literature. In contrast, low enantiomeric excesses have been obtained from the reactions of meso-epoxides catalyzed by (R)-1.

Full text of DOI:10.1055/s-2003-38758

LETTER
849
1,2-Ferrocenediylazaphosphinines 2: A New Class of Nucleophilic Catalysts
for Ring-Opening of Epoxides1
1
, 2-Ferrocene
d
e
iylazaphos
u
phinines ng Hwan Paek, Sang Chul Shim, Chan Sik Cho, Tae-Jeong Kim*
Department of Industrial Chemistry, Kyungpook National University, Taegu 702-701, Korea
Fax +82(53)9506594; E-mail: tjkim@knu.ac.kr
Received 22 November 2002
Abstract: 1,2-Ferrocenediylazaphosphinines (1a–c) have been
Me
successfully employed as a new class of nucleophilic catalysts for
ring-opening of a range of epoxides, their catalytic efficiency in
terms of regioselectivity as well as chemical yield comparing
well with the existing catalysts in the literature. In contrast, low
enantiomeric excesses have been obtained from the reactions of
meso-epoxides catalyzed by (R)-1.
Fe
R = H (1a)
R = Me (1b)
R = Ph (1c)
N
O
P
Ph2
R
Figure 1
Key words: ferrocenediylazaphosphinines, nucleophilic catalysts,
ring-opening, epoxides
SiCl3) ether with acid, cleanly affords a chlorohydrin
(Table 1).8 Representative procedure for Table 1, includ-
We have recently shown that the reaction of 1-(a-amino- ing monitoring the background reaction: a solution was
alkyl)-2-diphenylphosphinoferrocene with glyoxals re- prepared of hex-1-ene oxide (0.148 g, 1.48 mmol) and
sults in, via an unusual cyclization, 1,2-ferrocene- TMSCl (0.230 ml, 1.81 mmol) in CH2Cl2 (4.52 ml) in a
diylazaphosphinines (1, Figure 1) rather than the expected Schlenk tube. After degassing the solution by the freeze-
bisferrocenyldiimine products.1 These compounds consti- and-thaw method three times, a portion of this stock solu-
tute a new class of planar chiral ferrocenes that are of in- tion was transferred to a 5 mL screw-capped vial (back-
tense current research interest in the field of asymmetric ground reaction), and 1.69 mL of the stock solution (0.49
catalysis.2,3 For instance, Fu has demonstrated that fer- mmol of epoxide, 0.60 mmol of TMSCl) was transferred
rocenes of the type (p-heterocycle)FeCp can be success- to a Schlenk tube containing catalyst 1a (11.3 mg, 0.025
fully employed not only as ligands but also as mmol). The two reactions were allowed to proceed at
nucleophilic catalysts in a number of asymmetric catalytic room temperature and monitored by GC. After 70 min-
reactions.3h,4 In particular, he recently established that a utes, the catalyzed reaction was complete and the back-
phosphaferrocene catalyzes the ring opening of epoxides ground reaction had not proceeded (<5% conversion). For
with TMSCl,5 and more recently the same author reported the catalyzed reaction, the solvent was removed in vacuo,
very high enantiomeric excesses (up to 98% ee) for the and the TMS ether was treated with HCl (1 M in Et2O) for
enantioselective ring opening of meso-epoxides catalyzed 1 hour at room temperature. The resulting chlorohydrins
by a series of ferrocene-fused planar-chiral N-oxides.6 were purified by flash chromatography on silica gel (20%
Denmark had earlier accomplished an enantiomeric ex- Et2O/pentane), and their NMR identifications were made
cess up to 87% ee for the same reaction by employing a by comparison with literature data.5–7
chiral HMPA as a nucleophilic catalyst.7 In this regard, it
is worth noting that our new compounds 1a–c are power-
All reactions went to completion with inversion of config-
uration at the carbon undergoing substitution as verified
ful ligands in a Cu-catalyzed cyclopropanation of styrene
by comparison with the known chlorohydrins.5–7 Cis-stil-
to achieve a complete diastereocontrol.1 We have thus
bene oxide reacted at the lowest rate than any of the sub-
strates employed (runs 1–3 vs. 4–15). This is most likely
due to hindered approach to the TMS-nucleophile inter-
reasoned that our new compound 1, carrying in principle
two donor sites of sp2-nitrogen and carbonyl oxygen,
would also function as a suitable nucleophilic catalyst for
mediate (cf. Figure 2). The regioselectivity in the reaction
the same reaction.
of unsymmetrical epoxides is governed by both steric and
In this communication, we describe our discovery that a electronic effects. This is illustrated in the high level but
variety of epoxides undergo ring-opening with TMSCl (or opposite sense of regioselectivity in the opening of sty-
SiCl4) by the action of 1a–c as Lewis base catalysts. Thus, rene oxide (runs 7–9) and other terminal epoxides (runs
treatment of an epoxide with 1.2 equivalents of TMSCl 10–18). For example, in the case of alkene oxide [R1 = Bu,
(or SiCl4) and 5 mol% of 1 in CH2Cl2 at room tempera- Bn, CH2Cl, CH2OBn, CH2OC(O)Pr], displacement oc-
ture, followed by deprotection of the resulting TMS (or curs preferentially at the less hindered carbon (runs 10–
18), barring an overriding electronic effect observed with
styrene oxide (runs 7–9). Fu and Denmark independently
have also made parallel observations.5,7 Structural change
in the catalyst plays also a certain role in the regioselectiv-
Synlett 2003, No. 6, Print: 05 05 2003.
Art Id.1437-2096,E;2003,0,06,0849,0851,ftx,en;U13302ST.pdf.
© Georg Thieme Verlag Stuttgart · New York

850
S. H. Paek et al.
LETTER
ity, thus 1a giving the highest and in some cases a perfect
regiocontrol (runs 7, 10, and 13). The catalyst 1c also ex-
hibits near perfect regiocontrol for some substrates carry-
ing other functional groups (runs 16–19).
Me
Si
Me
Si
Me
CpFe
Me
Me
Me
CpFe
Me
Me
N
N
O
O
Cl
Cl
P
P
Ph2
Ph2
R
R
(D)
(C)
Our working hypothesis for the catalytic cycle may be the
same as that proposed by others.7 As such, the first step in
the catalytic cycle is expected to be the formation of a cat-
ionic complex C or D between TMSCl (or SiCl4) and 1
(Figure 2). Donor-stabilized silyl cations are ubiquitous,9
and the structure and reactivity of resulting pentacoordi-
nate silicon compounds are now well documented.10 Sub-
sequent complexation of the epoxide to the silicon cation
followed by nucleophilic attack with the chloride ion in an
SN2 fashion will eventually lead to ring-opening.
Figure 2
Encouraged by the exceptionally high regioselectivity of
rac-1, we have decided to perform the asymmetric ring
opening of meso-epoxides by employing (R)-1 only to
find disappointingly low enantiomeric excesses (% ee) re-
gardless the types of catalysts or the substrates employed
(Table 2). Lowering the reaction temperature to –78 °C
has little effect on the enantiomeric excess.
Table 1 Ring-Opening of Epoxides with TMSCl in the Presence of 1
In summary, we now put into entry 1,2-ferrocenediylaza-
phosphinines as a new class of nucleophilic catalysts for
ring-opening of a variety of epoxides. Their catalytic effi-
ciency in terms of regioselectivity as well as chemical
yield compares well with such catalysts as phosphafer-
rocene and HMPA although enantioselectivity is far less
efficient. Studies on fine-tuning of the catalysts to im-
prove their enantioselection are currently underway and
will be the subject of our future communication.
OH
Cl
O
H+
+
1
R1
R2
R2
R2
OH
+ TMSCl
R1
R1
Cl
(A)
(B)
Run Epoxide
1
1
t (min) Yield Ratio
(%)a
(A:B)b
1a
1b
1c
600
600
600
95
95
95
–
–
–
O
cis-
2
3
Ph
Ph
Table 2 Asymmetric Ring-Opening of meso-Epoxides Catalyzed
by (R)-1a
4
5
6
1a
1b
1c
60
60
60
97
98
97
–
–
–
O
OH
OH
Cl
or
cis-
[H+]
O
TMSCl/(R)-1
R
*
R
R
R
*
CH2Cl2
Cl
(R,R)-1
7
8
9
1a
1b
1c
15
15
15
98
98
98
6:94
O
(S,S)-2
10:90
Ph
H
13:87c
Entry Epoxide
Catalyst
Temp.
(°C)
Product ee (%)
(config.)
R1=Ph; R2=H
O
10
11
12
1a
1b
1c
70
70
70
97
97
98
100:0
92:8
93:7
Bu
H
1
2
3
4
(R)-1a
(R)-1a
(R)-1b
(R)-1c
r.t.
–78
r.t.
r.t.
1
1
1
1
1.1 (R,R)
5.4 (R,R)
<1 (R,R)
5.7 (R,R)
O
Ph
Ph
R1=Bu; R2=H
O
13
14
15
1a
1b
1c
60
60
60
98
97
97
100:0
99:1
96:4
Bn
H
5
6
7
8
(R)-1a
(R)-1a
(R)-1b
(R)-1c
r.t.
–78
r.t.
r.t.
2
2
2
2
4.0 (S,S)
6.8 (S,S)
2.3 (S,S)
3.6 (S,S)
O
R1=Bn; R2=H
O
16
1c
150
93
100:0
H
Cl
a (R) Refers to the planar chirality in the catalyst 1.
R1=CH2Cl; R2=H
17
1c
180
91
96:4
O
H
Bn
O
Acknowledgment
R1=CH2OBn; R2=H
TJK gratefully acknowledges financial support from KOSEF (Grant
# R01-2000-000-00044-0).
18
19
1c
1c
90
89
98
100:0
90:10
O
O
H
nPr
O
R1=CH2OC(O)Pr; R2=H
References
120
trans-
O
(1) For part 1 in the series, see: Hwang, G.-H.; Ryu, E.-S.; Park,
D.-K.; Shim, S. C.; Cho, C. S.; Kim, T.-J.; Jeong, J. H.;
Cheong, M. Organometallics 2001, 20, 5784.
(2) For recent reviews on planar chiral ferrocenes, see:
(a) Richards, C. J.; Locke, A. J. Tetrahedron: Asymmetry
1998, 9, 2377. (b) Kagan, H. B.; Riant, O. Advances in
Ph
Me
a Overall isolated yield.
b Determined by GC.
c The ratio (A:B) becomes 0:100 in the presence of SiCl4.
Synlett 2003, No. 6, 849–851 ISSN 1234-567-89 © Thieme Stuttgart · New York

LETTER
Asymmetric Synthesis, Vol. 2; Hassner, A., Ed.; JAI Press
1,2-Ferrocenediylazaphosphinines
851
C.; Tweddell, J.; Fu, G. C. J. Org. Chem. 1998, 63, 2794.
(h) Ruble, J. C.; Fu, G. C. J. Org. Chem. 1996, 61, 7230.
(i) Tanaka, K.; Qiao, S.; Tobisu, M. M.; Lo, M.-C.; Fu, G. C.
J. Am. Chem. Soc. 2000, 122, 9870. (j) Shintani, R.; Lo, M.
M.-C.; Fu, G. C. Org. Lett. 2000, 2, 3695. (k) Rios, R.;
Liang, J.; Lo, M. M.-C.; Fu, G. C. Chem. Commun. 2000,
377. (l) Lo, M. M.-C.; Fu, G. C. J. Am. Chem. Soc. 1998,
120, 10270. (m) Qiao, S.; Fu, G. C. J. Org. Chem. 1998, 63,
4168.
Inc.: Greenwich CT, 1997, 189.
(3) For recent reviews and papers on heterocyclic analogues of
planar chiral ferrocenes such as phospha- and azaferrocenes,
see: (a) Ganter, C. J. Chem. Soc., Dalton Trans. 2001, 3541.
(b) Ganter, C.; Kaulin, C.; Englert, U. Organometallics
1999, 18, 5444. (c) Ganter, C.; Glinsböckel, C.; Ganter, B.
Eur. J. Inorg. Chem. 1998, 1163. (d) Ganter, C.; Brassat, L.;
Glinsböckel, C.; Ganter, B. Organometallics 1997, 16,
2862. (e) Ogasawara, M.; Yoshida, K.; Hayashi, T.
Organometallics 2001, 20, 1014. (f) Ogasawara, M.;
Yoshida, K.; Hayashi, T. Organometallics 2001, 20, 3913.
(g) Qiao, S.; Hoic, D. A.; Fu, G. C. Organometallics 1998,
17, 773. (h) Fu, G. C. Acc. Chem. Res. 2000, 33, 412; and
references therein.
(4) (a) Lo, M. M.-C.; Fu, G. C. J. Am. Chem. Soc. 2002, 124,
4572. (b) Hodous, B.; Fu, G. C. J. Am. Chem. Soc. 2002,
124, 1578. (c) Tao, B.; Ruble, J. C.; Hoic, D. A.; Fu, G. C. J.
Am. Chem. Soc. 1999, 121, 5091. (d) Hodous, B.; Fu, G. C.
J. Am. Chem. Soc. 1999, 121, 2637. (e) Ruble, J. C.; Fu, G.
C. J. Am. Chem. Soc. 1998, 120, 11532. (f)Liang, J.;Ruble,
J. C.; Fu, G. C. J. Org. Chem. 1998, 63, 3154. (g) Ruble, J.
(5) Garrett, C. E.; Fu, G. C. J. Org. Chem. 1997, 62, 4534.
(6) Tao, B.; Lo, M. M.-C.; Fu, G. C. J. Am. Chem. Soc. 2001,
123, 353.
(7) Denmark, S. E.; Barsanti, P. A.; Wong, K.-T.; Stavenger, R.
A. J. Org. Chem. 1998, 63, 2428.
(8) The same reaction, however, did not proceed at all by
replacing TMSCl with TMSCN for an unknown reason, as
confirmed by quantitative recovery of the starting compound
on GC after 24 h.
(9) For a recent example, see: Kalikhman, I.; Gostevskii, B.;
Girshberg, O.; Krivonos, S.; Kost, D. Organometallics 2002,
21, 2551.
(10) (a) Holmes, R. R. Chem. Rev. 1996, 96, 927. (b) Chuit, C.;
Corriu, R. J. P.; Young, J. C. Chem. Rev. 1993, 93, 1371.
(c) Haaland, A. Angew. Chem., Int. Ed. Engl. 1989, 28, 992.
Synlett 2003, No. 6, 849–851 ISSN 1234-567-89 © Thieme Stuttgart · New York