Horner-Wadsworth-Emmons modification for Ramirez gem-dibromoolefination of aldehydes and ketones using P(Oi-Pr)3

Article abstract of DOI:10.1055/s-2008-1032045

A simple procedure for the use of triisopropylphosphite in the Ramirez olefination is described. This reagent is equally or more reactive than PPh ₃ toward aldehydes and ketones in the gem-dibromoolefination of aldehydes and ketones. Under the

Full text of DOI:10.1055/s-2008-1032045

LETTER
413
Horner–Wadsworth–Emmons Modification for Ramirez
gem-Dibromoolefination of Aldehydes and Ketones Using P(Oi-Pr)3
Modificationfor
R
u
amirez gem-
a
Dibrom
o
O
n
lefination of
-
A
ldehyd
Q
e
s
and
K
etones ing Fang, Olga Lifchits, Mark Lautens*
Davenport Chemistry Laboratories, Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, ON, M5S 3H6, Canada
Fax +1(416)9468185; E-mail: mlautens@chem.utoronto.ca
Received 28 September 2007
The most common way to prepare gem-dibromoolefins is
Abstract: A simple procedure for the use of triisopropylphosphite
by using the Ramirez olefination, a reaction involving an
in the Ramirez olefination is described. This reagent is equally or
more reactive than PPh3 toward aldehydes and ketones in the gem-
dibromoolefination of aldehydes and ketones. Under the reaction
aldehyde (or ketone), CBr4 and PPh3.5 Modified methods
to cope with the reactive Lewis acidic byproduct Br2PPh3
conditions, an a-bromoketone gave a novel tribromomethyl-substi- using scavengers such as zinc powder or Et3N were sub-
sequently developed.1,6 However, the utility of this meth-
tuted oxirane product in good yield. The substrate-dependent nature
of this reaction suggests that this Horner–Wadsworth–Emmons
od towards less reactive ketones remains low. Although
equivalent of the Ramirez gem-dibromoolefination reaction pro-
there are alternative non-ylide approaches which use or-
ceeds through an ionic mechanism.
ganolithium and Grignard reagents, or a copper-catalyzed
Key words: olefination, phosphite, Horner–Wadsworth–Emmons
reaction, aldehydes, ketones
redox chemistry between hydrazones and CBr4, multiple
steps are usually required.7
We have recently published a series of papers on the use
of gem-dihalovinylanilines for the synthesis of indole de-
gem-Dibromoolefins serve as important synthetic inter-
rivatives (Scheme 1).8 Although the one-pot, two-step
mediates in a variety of transformations, including Corey–
procedure for preparation of the substrate gave a good
Fuchs alkyne synthesis,1 trisubstituted olefin synthesis via
yield, there are several drawbacks in large-scale prepara-
stereoselective cross-couplings,2 and carbocycle or het-
tions. The significant amount of triphenylphosphine oxide
erocycle synthesis via tandem cross-couplings.3 Although
waste generated in the reaction is difficult to remove, re-
gem-dihaloolefin units are rarely found in natural prod-
ucts, they have been used to enhance the potency and sta-
sulting in a tedious and solvent-consuming purification
process. Due to the crystalline nature of triphenylphos-
bility of biologically active compounds such as the
phine oxide, solid products cannot be purified by recrys-
common household insecticide deltamethrin.4
tallization and require column chromatography which is
impractical on a large scale.
R3
R3
C–N/Suzuki
R1
Pd/L
N
R4
R
O
R1
NO2
R1
C–N/Heck
Pd/L
1. CBr4, PPh3
2. reduction
N
R4
R
R
R1
C–N/Sonogashira
Pd/Cu/L
R3
N
R4
X
R1
X
R3
NH
R4
C–N/C-N
Cu/L
R1
Cbz
N
N
R
O
Scheme 1 Indole synthesis via tandem cross couplings
SYNLETT 2008, No. 3, pp 0413–0417
x
x.
x
x.
2
0
0
7
Advanced online publication: 16.01.2008
DOI: 10.1055/s-2008-1032045; Art ID: S07607ST
© Georg Thieme Verlag Stuttgart · New York

414
Y.-Q. Fang et al.
LETTER
To address these issues, we examined alternative phos- appears to be sensitive to steric hindrance in ketones, with
phorus reagents such as alkylphosphines and phosphites, hindered ortho substituents significantly slowing down
since their byproducts are oils, and are therefore easier to the reaction as demonstrated in the case of 2¢-nitroace-
separate. A number of reagents were screened for the tophenone (Table 2, entry 9). Interestingly, while the re-
gem-dibromoolefination of 2-nitrobenzaldehyde (1a) and action of acyl cyanide 1j under P(Oi-Pr)3 gave the desired
the results are shown in Table 1.
2j in poor yield, PPh3-mediated reaction afforded the
product in excellent yield (Table 2, entry 10).
Table 1 Ramirez Olefination with Various Phosphorus Reagents
Table 2 Scope of the P(Oi-Pr)3-Mediated Ramirez Olefination of
Aldehydes and Ketones
CHO
Br
CBr4, P reagent
CH2Cl2, 0 °C
Br
R
NO2
NO2
O
Br
1a
2a
CBr4, P(Oi-Pr)3
R1
R1
R
Br
Yield (%)a Comment
CH2Cl2, 0 °C
Entry
Reagent
2
1
2
PPh3
95
<40
48
57
92
0
excessive Ph3P(O) waste
incomplete conversion
side reaction
Entry Starting material
Product
Yield (%)a
PBu3
P(Oi-Pr)3 PPh3
3
P(OMe)3
P(OEt)3
P(Oi-Pr)3
P(Ot-Bu)3
P(OPh)3
P(OEt)2OH
NaH2PO2
PCl3
Br
O
1
92
95
4
side reaction
Br
NO2
NO2
5
clean, fast
1a
2a
Br
6
no reaction
O
2
76
94
88
85
Br
7
13
0
incomplete conversion
no reaction
O
O
1b
2b
8
Br
O
9
0
no reaction
3
Br
10
0
no reaction
2c
1c
a Isolated yield using 1a (1.0 mmol), CBr4 (1.5 mmol), and phospho-
Me
Me
rus reagent (3 mmol).
Br
O
4
73
73
<40b
<40b
Br
It was found that tri-n-butylphosphine was less efficient
than PPh3, resulting in a moderate yield of the desired
product 2a (Table 1, entry 2). Considering the air-sensi-
tive nature of these reagents, no further alkylphosphines
were examined. We were encouraged to find that the gen-
erally inexpensive and less air sensitive trialkyl phosphi-
tes (Table 1, entries 3–7) showed greater reactivity.
P(OMe)3 and P(OEt)3 gave the desired product in moder-
ate yields with complete conversion of the starting mate-
rial. To our delight, reaction with the bulkier P(Oi-Pr)3
was very clean and rapid, giving the desired product in
92% yield. Further increasing the bulkiness of the reagent
using P(Ot-Bu)3, however, resulted in no reaction. Other
phosphites such as triphenylphosphite and P(OEt)2OH
were similarly unsuccessful, and inorganic phosphorus
sources such as PCl3 and sodium hyperphosphite were
completely inactive.
1d
2d
Me
Me
Br
O
5
Br
O2N
O2N
2e
1e
Ph
Ph
Br
6
98
69
O2N
O2N
O
Br
F
F
1f
2f
Me
Me
Br
62c
44c
<20b
O
7
8
Br
MeO
MeO
With the optimal triisopropylphosphite in hand, we exam-
ined the reaction scope by comparing it to the PPh3-medi-
ated olefination (Table 2).9 In general, the olefination of
aldehydes gave comparable yields for both reagents
(Table 2, entries 1–3). However, upon examining the
scope of ketones, we found that P(Oi-Pr)3 was more reac-
tive than PPh3, although incomplete conversion was ob-
served in some cases (Table 2, entries 4–9). The reaction
1g
2g
Br
Br
O
<5b
Ph
Ph
1h
2h
Synlett 2008, No. 3, 413–417 © Thieme Stuttgart · New York

LETTER
Modification for Ramirez gem-Dibromoolefination of Aldehydes and Ketones
415
Table 2 Scope of the P(Oi-Pr)3-Mediated Ramirez Olefination of
Aldehydes and Ketones (continued)
R
O
Br
CBr4, P(Oi-Pr)3
R1
R1
R
Br
CH2Cl2, 0 °C
2
Scheme 2 P(Oi-Pr)3-mediated oxirane formation
Entry Starting material
Product
Yield (%)a
P(Oi-Pr)3 PPh3
Olefination of alkynyl aldehyde 5 on the other hand re-
sulted in a mixture of the desired gem-dibromoolefin 6
and a vinyl phosphate 7 (Scheme 3).11 In contrast, a PPh3-
mediated reaction did not produce compound 7, giving
only 6 in 90% yield. The gem-dibromovinyl phosphate
structure is known to exhibit insecticidal activity.12
Me
Me
Br
Br
O
9
Trace
Trace
Br
Br
NO2
CN
NO2
1i
2i
2j
O
Br
CHO
Br
Br
CN
CBr4, P(Oi-Pr)3
10
21
91
Br
O
+
Ph
Ph
Oi-Pr
P
CH2Cl2, –20 °C
O
Br
Br
Ph
Oi-Pr
1j
5
6
47%
7
50%
a Isolated yield.
b Conversion measured by 1H NMR.
c Incomplete conversion
Scheme 3 P(Oi-Pr)3-mediated Ramirez olefination of alkynyl alde-
hyde
Two possible mechanistic pathways for this P(Oi-Pr)3-
mediated olefination are shown in Scheme 4: an ylide
pathway (A) which is analogous to PPh3-mediated
reaction5 and an ionic pathway (B). Given the substrate-
An unexpected epoxide product 4 was obtained in good
yield when a-bromoacetophenone (3) was treated with
CBr4/P(Oi-Pr)3 (Scheme 2).10 In contrast, the reaction
with CBr4/PPh3 gave a 1:5 mixture of the starting material
and a,a-dibromoacetophenone.
Ylide pathway (A)
R'-CHO
P(OR)3
Br
CBr4
Br
–
+
Br
Br
R'
Br –
Br2C-P(OR)3
+
+
P(OR)3
Br
P(OR)3
10
Br2P(OR)3
Ionic pathway (B)
O
OR
OR
(RO)3PHBr
P
Br
O
(RO)3P
Ph
H
Br
O
Br
Br
Ph
O
P
OR
OR
Br
9
7
RBr
Br –
Ph
R' =
OR
R'-CHO
RO
O
OR
O –
P+
CBr4
+
Br
CBr3–
R'
R'
Br
Br
P(OR)3
CBr3
BrP(OR)3
R'
Br
P(OR)3
8
+
Br –
Br
BrP(OR)3
+
(O)P(OR)3
+
11
for R' = CH2Br
BrP(OR)3
+
O–
Br3C
Ph
O
Br
Ph
CBr3
4
Scheme 4 Possible mechanistic pathways for phosphite-mediated Ramirez olefination
Synlett 2008, No. 3, 413–417 © Thieme Stuttgart · New York

416
Y.-Q. Fang et al.
LETTER
dependent behavior of this reaction, we favor the ionic
pathway B, which is consistent with all the evidence.
CHO
NO2
Br
(O)P(Oi-Pr)3
+
Br2P(Oi-Pr)3
CBr4, P(Oi-Pr)3
+
Br
CH2Cl2, 0 °C
NO2
87%
Initial attack of phosphite to CBr4 results in a tribromo-
mixture
–
1a
methyl anion (CBr3 ), which probably exists as a close ion
pair with the generated phosphonium or as a pentavalent
phosphorus species. Attack of the aldehyde or ketone by
1. HCl (6 M, reflux)
2. NaHCO3
–
the CBr3 anion results in an alkoxide, which in the case
Br
of a-bromoacetophenone (3) gives epoxide 4. This also
explains why acyl cyanide afforded low yield under our
conditions since cyanide is a good leaving group under
Br
NO2
2a
–
CBr3 attack. Otherwise, the alkoxide reacts with the
pure according to 1H NMR
phosphonium to form 8. The second molecule of phos-
phite attacks the bromide in 8 to give the normal olefina-
tion product. The intermediate 8 can also undergo an
Arbuzov dealkylation to form a phosphate 9. In the case
of the alkynyl substrate 5, the phosphate further under-
goes an E2-elimination to afford the observed vinyl phos-
Scheme 5 Preparation of 2a without chromatographic separation
Acknowledgment
This work was supported by NSERC (Canada), Merck Frosst Cana-
phate 7, presumably due to the high acidity of the da, and the University of Toronto. We also thank Dr. Alan Lough
for X-ray structure determination. Y.-Q. F. thanks the Ontario Go-
propargylic proton.
vernment and the University of Toronto for financial support in the
form of postgraduate scholarships (OGS, OGSST).
Shifting the mechanism from an ylide pathway to anionic
pathways is presumably due to the change of relative sta-
bility of the phosphonium ion 10 and 11. Due to their
strong electron-withdrawing inductive while poor elec-
References and Notes
tron-donating effects of oxygens on the phosphorus, the
positive charge of phosphonium ion 10 is localized only
on phosphorus, destabilizing the intermediate. However,
for the bromophosphonium 11, its stability is improved
through resonance structures (Figure 1).
(1) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769.
(2) Kumada and Negishi coupling: (a) Minato, A.; Suzuki, K.;
Tamao, K. J. Am. Chem. Soc. 1987, 109, 1257. (b) Minato,
A. J. Org. Chem. 1991, 56, 4052. (c) Zeng, X.; Qian, M.;
Hu, Q.; Negishi, E. Angew. Chem. Int. Ed. 2004, 43, 2259.
Suzuki coupling: (d) Roush, W. R.; Moriarty, K. J.; Brown,
B. B. Tetrahedron Lett. 1990, 31, 6509. (e) Evans, D. A.;
Starr, J. T. J. Am. Chem. Soc. 2003, 125, 13531.
RO
RO
RO
+
Br
+
P
+
CBr3
P
Br
P
RO
RO
RO
RO
RO
RO
(f) Molander, G. A.; Yokoyama, Y. J. Org. Chem. 2006, 71,
2493. (g) Reed, M. A.; Chang, M. T.; Snieckus, V. Org. Lett.
2004, 6, 2297. Hydride reduction: (h) Uenishi, J.;
Kawahama, R.; Yonemitsu, O.; Tsuji, J. J. Org. Chem. 1998,
63, 8965.
10
destablized
11
stabilized
Figure 1 Comparison of possible phosphonium ion intermediates
(3) (a) Soderquist, J. A.; Leon, G.; Colberg, J. C.; Martinez, I.
Tetrahedron Lett. 1995, 36, 3119. (b) Shen, W.; Wang, L.
J. Org. Chem. 1999, 64, 8873. (c) Wang, L.; Shen, W.
Tetrahedron Lett. 1998, 39, 7625. (d) Thielges, S.; Meddah,
E.; Bisseret, P.; Eustache, J. Tetrahedron Lett. 2004, 45, 907.
(4) (a) Elliot, M.; Farnham, A. W.; Janes, N. F.; Needham, P. H.;
Pulman, D. A. Nature (London) 1974, 248, 710. (b) Elliot,
M.; Janes, N. F. Chem. Soc. Rev. 1978, 7, 473.
(5) Ramirez, F.; Desal, N. B.; McKelvie, N. J. Am. Chem. Soc.
1962, 84, 1745.
(6) Grandjean, D.; Pale, P.; Chuche, J. Tetrahedron Lett. 1994,
35, 3529.
Since the byproducts from triisopropylphosphite-mediat-
ed reaction are oils, the gem-dibromoolefinated products
such as 2 can be purified by recrystallization. A conve-
nient alternative protocol is a simple treatment of the
crude reaction mixture with acid, which hydrolyzes triiso-
propylphosphate into nonhazardous phosphoric acid and
isopropanol (Scheme 5), giving analytically pure product
2 in good yield after a simple acid/base workup.
(7) (a) Korotchenko, V. N.; Shastin, A. V.; Nenajdenko, V. G.;
Balenkova, E. S. J. Chem. Soc., Perkin Trans. 1 2002, 883.
(b) Rezaei, H.; Normant, J. F. Synthesis 2000, 1, 109.
(8) (a) Fang, Y.-Q.; Lautens, M. Org. Lett. 2005, 7, 3549.
(b) Lautens, M.; Fang, Y. WO 06047888, 2006; Chem.
Abstr. 2006, 144, 468020. (c) Fayol, A.; Fang, Y.-Q.;
Lautens, M. Org. Lett. 2006, 8, 4203. (d) Nagamochi, M.;
Fang, Y.-Q.; Lautens, M. Org. Lett. 2007, 9, 2955.
(e) Yuen, J.; Fang, Y.-Q.; Lautens, M. Org. Lett. 2006, 8,
653. (f) Fang, Y.-Q.; Yuen, J.; Lautens, M. J. Org. Chem.
2007, 72, 5152. (g) Fang, Y.-Q.; Karisch, R.; Lautens, M.
J. Org. Chem. 2007, 72, 1341.
In summary, we have developed a Horner–Wadsworth–
Emmons equivalent of the Ramirez olefination of ketones
and aldehydes using triisopropylphosphite. In general, the
reactivity of triisopropylphosphite toward gem-dibro-
moolefination is comparable to PPh3 with aldehydes and
higher than PPh3 with ketones. This phosphite-mediated
reaction likely proceeds through an ionic mechanism rath-
er than an ylide intermediate observed in PPh3-mediated
reactions.
(9) General Procedure for Ramirez Olefination
To a solution of 1a (0.151 g, 1 mmol) and CBr4 (0.49 g, 1.5
mmol) in CH2Cl2 (4 mL) was added dropwise a solution of
Synlett 2008, No. 3, 413–417 © Thieme Stuttgart · New York

LETTER
Modification for Ramirez gem-Dibromoolefination of Aldehydes and Ketones
417
the corresponding phosphine or phosphite (3.0 mmol) in
mmol), CBr4 (3.54 g, 10.8 mmol), and P(Oi-Pr)3 (5.3 mL,
21.6 mol). The product was obtained as a white crystalline
solid (1.5 g, 80%). The single crystal suitable for X-ray
determination was obtained by slow diffusion of hexanes in
a solution of 4 in CH2Cl2. Rf = 0.70 (10% EtOAc in hexane);
mp 98–100 °C. IR (neat): 3017, 1582, 1447 cm–1. 1H NMR
(400 MHz, CDCl3): d = 7.80–7.82 (2 H, m), 7.36–7.42 (3 H,
CH2Cl2 (1 mL) at 0 °C. After 30 min, the reaction was
warmed to r.t., quenched with NaHCO3, extracted with Et2O,
and dried over MgSO4. The product was isolated using
column chromatography, and yields are shown in Table 2.
Preparation of 2a Using CBr4/P(Oi-Pr)3 without
Chromatographic Purification
To a solution of 1a (2.0 g, 13.2 mmol) and CBr4 (6.6 g, 19.9
mmol) in CH2Cl2 (50 mL) was added dropwise a solution of
P(Oi-Pr)3 (7.2 mL, 29 mmol) in CH2Cl2 (10 mL) at 0 °C.
After 30 min, the reaction was warmed to r.t., quenched with
NaHCO3, and extracted with Et2O. After removal of solvent,
the mixture was taken into concd HCl (12 M, 15 mL) and
AcOH (15 mL) and heated to reflux overnight. The mixture
was cooled to r.t., diluted with H2O (50 mL), neutralized
with Na2CO3, and extracted with Et2O. After drying over
MgSO4, solvent was removed under vacuum to give an off-
white solid (3.51 g, 87%). The 1H NMR spectrum of the
product was identical to the authentic sample.
Compound 2f: The general procedure was followed using 1f
(0.135 g, 0.5 mmol), CBr4 (0.250 g, 0.75 mmol), and P(Oi-
Pr)3 (0.37 mL, 1.5 mol). The product was obtained as a
slightly yellow solid (0.207 g, 98%). Rf = 0.41 (10% EtOAc
in hexane); mp 63–65 °C. IR (neat): 3079, 2921, 1624, 1528,
1347, 1254, 1099 cm–1. 1H NMR (400 MHz, CDCl3): d =
8.28–8.36 (2 H, m), 7.47–7.49 (2 H, m), 7.28–7.38 (4 H, m).
13C NMR (100 MHz, CDCl3): d = 162.4 (JC–F = 262 Hz),
144.4, 131.8, 129.7, 128.7, 126.8 (JC–F = 4.6 Hz), 126.3 (JC–
F = 10.7 Hz), 123.5, 122.1, 117.6 (JC–F = 24.5 Hz), 104.1,
98.9, 87.2. HRMS (EI): m/z calcd for C16H8Br2FNO2 [M]+:
422.8906; found: 422.8907.
m), 3.84 (1 H, d, J = 5.0 Hz), 3.12 (1 H, d, J = 5.0 Hz). 13
C
NMR (100 MHz, CDCl3): d = 133.3, 131.2, 129.4, 127.7,
67.8, 55.4, 45.8. HRMS (EI): m/z calcd for C9H7Br3O [M]+:
367.8047; found: 367.8041.
(11) The general procedure for Ramirez olefination was followed
using phenylpropynal (0.130 g, 1 mmol), CBr4 (0.498 g, 1.5
mmol), and P(Oi-Pr)3 (0.74 mL, 3 mol) in CH2Cl2 at –20 °C.
The crude material was separated by chromatography to give
6 (0.136 g, 47%) and 7 (0.232 g, 50%).
Compound 6: 1H NMR (400 MHz, CDCl3): d = 7.33–7.52 (5
H, m), 6.77 (1 H, s).
Compound 7: Rf = 0.22 (25% EtOAc in hexane). IR (neat):
2980, 2934, 2211, 1572, 1488, 1375, 1280, 1158 cm–1. 1H
NMR (400 MHz, CDCl3): d = 7.50–7.57 (2 H, m), 7.33–7.42
(3 H, m), 4.84 (2 H, septet, J = 6.2 Hz), 1.39 (12 H, t, J = 6.0
Hz). 13C NMR (100 MHz, CDCl3): d = 131.8, 129.8, 128.6,
121.4, 99.7, 91.4 (d, JC–P = 11.8 Hz), 81.3 (d, JC–P = 2.4 Hz),
79.1 (d, JC–P = 4.3 Hz), 74.2 (d, JC–P = 6.3 Hz), 23.8 (d,
J
C–P = 4.6 Hz), 23.7 (d, JC–P = 5.4 Hz). 31P NMR (300
MHz, CDCl3): d = –9.44. ESI-HRMS: m/z calcd for
C16H20O4PBr2 [M + H]+: 464.9460; found: 464.9458.
(12) Sledzinski, B.; Kroczynski, J.; Zwierzak, A.; Cieslak, L.;
Majda, A. DE 2338847, 1974; Chem. Abstr. 1974, 83,
54610.
(10) Product 4: The general procedure for Ramirez olefination
was followed using a-bromoacetophenone (1.08 g, 5.4
Synlett 2008, No. 3, 413–417 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.