Aminobromination of Unsaturated Phosphonates

Article abstract of DOI:10.1021/jo034535x

Unexpected syn β-amino-α-bromination of unsaturated phosphonates was observed under typical Sharpless AA reaction conditions with excess N-bromoacetamide.

Full text of DOI:10.1021/jo034535x

reaction using R,â-unsaturated phosphonates as sub-
strates. In the course of probing the reaction conditions,
we observed the novel syn-â-amino-R-bromination of
electron-rich unsaturated phosphonates where N-bro-
moacetamide was used as the nitrogen/bromine sources.
Am in obr om in a tion of Un sa tu r a ted
P h osp h on a tes
Xin Qi,† Sang-Hyeup Lee,‡ J i Young Kwon,§
Youngmee Kim,§ Sung-J in Kim,§ Yoon-Sik Lee,*,‡ and
J uyoung Yoon*,§
In the Sharpless AA reactions with vinyl phospho-
nates,2 the corresponding N-chloro-N-sodioamides were
used as nitrogen sources to introduce N-p-toluenesulfonyl
and ethoxycarbonyl amino groups. The yields of â-amino-
R-hydroxy phosphonates varied from 21 to 53% depend-
ing on the starting materials. In addition, the formation
of diols and regioisomers was reported under these
reaction conditions. In our case, N-bromoacetamide was
used as the nitrogen source following a typical AA
procedure for cinnamate ester4c to prepare the corre-
sponding aminohydroxylation product from the vinyl
phosphonate 1. However, in our initial attempt using 1.1
equiv of N-bromoacetamide, no starting material 1 was
consumed in the presence of different kinds of 1:1
aqueous solvents such as tert-butyl alcohol, 1-propanol,
and acetonitrile and a prolonged reaction time (2 days),
although the color of the reaction mixture changed from
green to light yellow in all cases. Once 3.5 equiv of
N-bromoacetamide was added, all of the starting vinyl
phosphonate disappeared within 2 h and we observed an
unexpected syn-â-amino-R-bromination product instead
of aminohydroxylation product (Scheme 1).
School of Chemical Engineering,
Dalian University of Technology, Dalian 116012,
People’s Republic of China, School of Chemical Engineering,
Seoul National University, Seoul 151-742, Korea, and
Department of Chemistry, Ewha Womans University,
Seoul 120-750, Korea
jyoon@ewha.ac.kr; yslee@snu.ac.kr
Received April 27, 2003
Abstr a ct: Unexpected syn â-amino-R-bromination of un-
saturated phosphonates was observed under typical Sharp-
less AA reaction conditions with excess N-bromoacetamide.
The synthesis of R- and/or â-substituted phosphonic
acids is an important area of research, particularly in
connection with the search for a biologically active
surrogate for the corresponding carboxylic acids and
phosphoric acids. Interesting inhibitory activities toward
renin and HIV protease have been observed with phos-
phonic acid derivatives.1 The syntheses of both racemic
and optically active phosphonates have been reported.
Recently, the Sharpless2 and Sisti3 groups have applied
the asymmetric aminohydroxylation (AA) reaction de-
scribed by Sharpless et al.4 to unsaturated phosphonates
to afford vicinal aminohydroxyl groups using osmium and
cinchona alkaloid ligands in low to moderate yields.
In our previous paper,5 we demonstrated the asym-
metric syntheses of R-substituted-â-amino acid deriva-
tives utilizing the nucleophilic ring-opening reaction of
both cis- and trans-oxazoline-5-carboxylates. These ox-
azoline-5-carboxylates were easily prepared from syn-
acetylamino alcohols, which can be obtained via the
acetamide-based Sharpless AA reaction using R,â-
unsaturated carboxylates as starting materials. To ex-
tend this oxazoline ring-opening methodology for the
preparation of R-substituted â-amino phosphonates, we
attempted to use the acetamide-based Sharpless AA
The identification of the unexpected product 2 was first
revealed via MS spectroscopy in which a prominent
molecular ion peak at 436.0888 (M + H) accompanied
by a 438.0870 peak (M + H + 2) with an almost equal
intensity supported the presence of an aminobromination
product rather than the anticipated aminohydroxylation
product at 374. In addition, the presence of the
[MeOPhCHNHCOCH3]+ fragment ion peak at 178 and
the absence of the [MeOPhCHBr]+ fragment ion peak at
199 and 201 showed the â-amino-R-bromo regioselectiv-
ity. This regiochemistry was further characterized by
1H-1H COSY analysis of the diisopropyl phosphonate 2
as follows: While the sextet peak at 5.60 ppm showed
strong correlations with both the amide group (6.50 ppm)
and the peak at 4.09 ppm and weak correlation with aryl
protons (7.21 ppm), the doublet-doublet peak at 4.09
ppm with a coupling constant of 12.2 Hz had only one
correlation with the sextet peak and should be more
likely to belong to the C-1 proton, which attached to the
same carbon atom with a bromine atom. 1H-13C HMQC
2D NMR analysis confirmed this interpretation that the
carbon atom correlating with the proton at 4.09 ppm has
a chemical shift of 49 ppm and a distinct geminal
phosphorus-carbon coupling constant of 153 Hz.
† Dalian University of Technology.
‡ Seoul National University.
§ Ewha Womans University.
(1) (a) Patel, D. V.; Rielly-Gauvin, K.; Ryono, D. E. Tetrahedron Lett.
1990, 31, 5587. (b) Stowasser, B.; Budt, K.-H.; J ian-Qi, L.; Peyman,
A.; Ruppert, D. Tetrahedron Lett. 1992, 33, 6625. (c) Wester, R. T.;
Chamber, R. J .; Green, M. D.; Murphy, W. R. Bioorg. Med. Chem. Lett.
1994, 4, 2005. (d) Patel, D. V.; Rielly-Gauvin, K.; Ryono, D. E.; Free,
C. A.; Rogers, W. L.; Smithy, S. A.; DeForrest, J . M.; Oehle, R. S.;
Petrillo, E. W., J r. J . Med. Chem. 1995, 38, 4557.
(2) Thomas, A. A.; Sharpless, K. B. J . Org. Chem. 1999, 64, 8379.
(3) Cravotto, G.; Giovenzana, G. B.; Pagliarin, R.; Palmisano, G.;
Sisti, M. Tetrahedron: Asymmetry 1998, 9, 745.
(4) (a) Li, G.; Chang, H.-T.; Sharpless, K. B. Angew. Chem., Int. Ed.
Engl. 1996, 35, 451. (b) Rudolph, J .; Sennhenn, P. C.; Vlaar, C. P.;
Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1996, 35, 2810. (c)
Bruncko, M.; Schlingloff, G. Sharpless, K. B. Angew. Chem., Int. Ed.
Engl. 1997, 36, 1483.
(5) (a) Lee, S.-H.; Yoon, J .; Nakamura, K.; Lee, Y.-S. Org. Lett. 2000,
2, 1243. (b) Lee, S.-H.; Qi, X.; Yoon, J .; Nakamura, K.; Lee, Y.-S.
Tetrahedron 2002, 58, 2777.
Although the alcoholic solvents (tert-butyl alcohol and
1-propanol) or acetonitrile were commonly used in Sharp-
less AA reactions,2,4 a much simpler reaction mixture was
obtained using acetonitrile in our case. The amount of
N-bromoacetamide was also an important factor; further
experiments revealed that more than 2.0 equiv of aceta-
mide was necessary while reducing the amount of the
acetamide to 1.6 equiv significantly retarded the reactiv-
10.1021/jo034535x CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/21/2003
9140
J . Org. Chem. 2003, 68, 9140-9143

SCHEME 1. Un exp ected Am in obr om in a tion of Un sa tu r a ted P h osp h on a te
TABLE 1. Resu lts of th e Am in obr om in a tion of Ar yl-Su bstitu ted Un sa tu r a ted P h osp h on a te Diester s
a
The numbers in parentheses represent the yield after a single recrystallization (>99% purity as evidenced by 31P NMR). Noncrystalline
products 6 [(DHQ)2PHAL], 6 [(DHQD)2PHAL], 8 [(DHQ)2PHAL], and 8 [(DHQD)2PHAL] were isolated in 95, 89, 92, and 88% purity (as
evidenced by 31P NMR7), respectively.
ity of this reaction. There was no reaction in the absence
of the osmium catalyst.
rich and para-substituted 2-phenyl vinyl phosphonates
(R ) p-methoxyphenyl, 3,4-dimethoxyphenyl, and p-N,N-
dimethylaminophenyl) gave â-amino-R-bromo phospho-
nates (Table 1). On the other hand, only starting material
or hydrolysis products were obtained from the rest of the
vinyl phosphonates under the same reaction conditions.
The yields of â-amino-R-bromo phosphonates varied from
21% to 35%. The main byproduct was the corresponding
syn-â-hydroxy-R-brominated adduct in most cases.7
To assign the stereochemistry of the â-amino-R-bro-
mophosphonates and confirm the regiochemistry afore-
mentioned, a suitable single crystal of compound 4 was
submitted for X-ray crystal determination. The X-ray
structure (see the Supporting Information) confirmed the
syn stereochemistry and regiochemistry.
On the basis of this interesting result, various unsat-
urated aryl phosphonates with different substitution
groups (R ) phenyl, p-chlorophenyl, o-methoxyphenyl,
m-methoxyphenyl, p-methoxyphenyl, 3,4-dimethoxyphe-
nyl, p-N,N-dimethylaminophenyl, p-tolyl, 2-furyl, 3-furyl,
p-nitrophenyl, and naphthyl), prepared using a modified
Horner-Wittig reaction,6 were subjected to the same
conditions to investigate the applicability of this reaction.
Among the vinyl phosphonates examined, only electron-
(6) Rambaud, M.; Vecchio, A.; Villieras, J . Synth. Commun. 1984,
14, 833.
(7) The â-hydroxy-R-bromo regiochemistry of the hydroxybrominated
adduct was determined using the MS spectroscopy and 1H NMR
coupling constants by an analysis similar to that used in the identifica-
tion of our â-amino-R-brominated adduct. The syn stereoselectivity was
determined as follows: the epoxide was synthesized from diethyl
[1-bromo-2-hydroxy-2-(p-methoxyphenyl)ethyl]phosphonate and the
latter was confirmed as a cis isomer based on the coupling constants
in 1H NMR spectrum. The J P-H and J H-H of hydrogen at the C-2
position (CHP) were 28.13 and 4.47 Hz, respectively, which were
consistent with 28.24 and 4.49 Hz in (()-cis-diethyl 1,2-epoxy-2-
phenylethylphosphonate. (a) Cristau, H. J .; Yangkou-Mbianda, X.;
Geze, A.; Beziat, Y.; Gasc, M. B. J . Organomet. Chem. 1998, 571, 189.
(b) Cristau, H. J .; Pirat, J .-L.; Drag, M.; Kafarski, P. Tetrahedron Lett.
2000, 41, 9781. The detailed results and mechanism will appear
elsewhere. However, we believe the mechanistic pathway for the
formation of this syn-â-hydroxy-R-brominated adduct might be similar
to that of our syn-â-amino-R-brominated adduct and closely related to
the formation of commonly observed dihydroxylated side products in
the Sharpless AA reactions.
However, most surprisingly, almost no enantioselec-
tivity was observed in our reactions using (DHQ)2PHAL
or (DHQD)2PHAL as a chiral ligand. Changes in the
addition sequence, the amount of base (0.5 equiv), or the
amount of potassium osmate dihydrate (0.02 or 0.08
(8) The noncrystalline products were always accompanied by small
amounts of phosphorous byproducts as shown by 31P NMR. Similar
phenomena were previously reported.1,6b No further attempts were
made to identify these compounds in this paper.
(9) (a) Li, G.; Wei, H.-X.; Kim, S. H.; Neighbors, M. Org. Lett. 1999,
1, 395. (b) Li, G.; Wei, H.-X.; Kim, S. H. Org. Lett. 2000, 2, 2249. (c)
Wei, H.-X.; Kim, S. H.; Li, G. Tetrahedron 2001, 57, 3869. (d) Li, G.;
Wei, H.-X.; Kim, S. H. Tetrahedron 2001, 57, 8407.
J . Org. Chem, Vol. 68, No. 23, 2003 9141

SCHEME 2. P r op osed Mech a n ism for th e Am in obr om in a tion Rea ction a
a
L ) ligand.
mmol) did not help to increase the ee, even though there
were discernible differences in the distribution of the
products.
Sharpless AA reaction condition with excess N-bromoac-
etamide as the nitrogen/bromine sources. Attempts to
vary the primary amide, by using amides other than
acetamide, to elucidate the reaction mechanism and an
effort to increase the yield and ee of this aminobromina-
tion reaction are currently in progress.
Recently, Li et al. reported a novel anti-R-amino-â-
halogenation of cinnamic esters using N,N-dichloro-p-
toluenesulfonamide as the chlorine/nitrogen sources.9 In
their reports, N-tosyl-N-chloroaziridinium was proposed
as a possible intermediate to explain the regio- and
stereochemical outcome. Also, a recent report from the
Sudalai group10 illustrated the transition-metal-catalyzed
regio- and stereoselective aminobromination of olefins
with TsNH2 and NBS as the nitrogen and bromine
sources, respectively. anti-â-Amino-R-bromo adducts were
mainly obtained in this case via the bromonium ion
intermediate.
On the contrary, the fact that we obtained only syn-
â-amino-R-brominated phosphonates under our reaction
conditions clearly excludes the involvement of aziri-
dinium or bromonium intermediate. To explain the regio-
and stereoselectivity of the resulting syn-â-amino-R-
brominated phosphonates, a catalytic pathway is pro-
posed as shown in Scheme 2. Possibly, in our cases, the
second catalytic cycle illustrated in the Sharpless paper3b
was promoted so that both enantiomers were produced.
After the first olefin substitution, the second olefin may
be promoted to bind to osmium to go to the ligand-
independent second cycle of reaction under our reaction
conditions. The osmium(VI) bisazaglycolates (II, Scheme
2) are supposed to be attacked by N-bromoacetamide
anion during the reoxidation process in our case. Con-
comitantly, the releasing bromides attack at the R-posi-
tion of glycolates from the osmium side to produce the
syn-â-amino-R-brominated phosphonates.
Exp er im en ta l Section
Gen er a l P r oced u r e for Am in obr om in a tion Rea ction of
Vin yl P h osp h on a tes. To a magnetically stirred solution of
vinyl phosphonate (1 mmol) and 39 mg of (DHQ)2PHAL or
(DHQD)2PHAL ligand (0.05 mmol) in 7 mL of acetonitrile were
added water (7 mL), LiOH‚H2O (42.8 mg, 1.02 mmol), and
potassium osmate dihydrate (14.7 mg, 0.04 mmol) to afford a
pale pinkish solution. After 10 min of stirring in an ice-water
bath, N-bromoacetamide (320 mg, 2.3 mmol) was added in one
portion which caused the color of the reaction mixture to change
to green immediately. With vigorous stirring, the color gradually
changed to yellow during 30 min. The reaction was monitored
by TLC (EtOAc/hexane ) 3:1). (A change in color to clear yellow
did not always mean the consumption of the olefin starting
materials. When 1.1 equiv of N-bromoacetamide was used, the
reaction mixture also changed to yellow with all of the olefin
remaining as shown by TLC). After the end of the reaction,
sodium sulfite (500 mg) was added, the mixture was stirred for
another 1 h at room temperature, and ethyl acetate (10 mL) was
added. The aqueous phase was then extracted with ethyl acetate
(3 × 20 mL). The combined organic layer was washed with brine
(20 mL) and dried over anhydrous sodium sulfate. After con-
centration in vacuo, the crude residue was purified first by flash
chromatography (1% MeOH/EtOAc) to remove most of the side
product and then by prerarative TLC with 2% MeOH/chloroform
to afford the aminobromination product.
Diisop r op yl [2-Acetyla m in o-1-br om o-2-(p-m eth oxyp h e-
n yl)eth yl]p h osp h on a te (2). A yield of 114 mg (0.262 mmol,
26%) of crude white solid was obtained with (DHQ)2PHAL. After
recrystallization with EtOAc/hexane, 81 mg (0.187 mmol, 19%)
In conclusion, we demonstrated a new â-amino-R-
bromination of unsaturated phosphonates under typical
of white, fluffy, crystalline solid was obtained: mp 152-3 °C;
-1
IR (KBr) 3415, 2952, 2926, 2852, 1663, 1610 cm
;
1H NMR
(CDCl3, 600 MHz) δ 1.08 (d, J ) 6.2 Hz, 3H), 1.31 (d, J ) 6.2
Hz, 3H), 1.34-1.40 (m, 6H), 2.09 (s, 3H), 3.79 (s, 3H), 4.09 (dd,
(10) Thakur, V. V.; Talluri, S. K.; Sudalai, A. Org. Lett. 2003, 5,
861.
9142 J . Org. Chem., Vol. 68, No. 23, 2003

1
J ) 2.6, 12.2 Hz, 1H), 4.77-4.83 (m, 2H), 5.60 (sextet, J ) 2.5,
7.9 Hz, 1H), 6.50 (d, J ) 7.8 Hz, 1H), 6.86 (d, J ) 8.7 Hz, 2H),
7.22 (d, J ) 8.6 Hz, 2H); 13C NMR (CDCl3, 150 MHz) δ 23.5,
23.8 (d, J ) 6.1 Hz), 24.0 (d, J ) 5.7 Hz), 24.4-24.7 (m), 49.0 (d,
J ) 153 Hz), 51.9 (d, J ) 1.6 Hz), 55.5, 73.1 (d, J ) 7.3 Hz), 73.4
(d, J ) 7.5 Hz), 113.9, 127.8, 131.1 (d, J ) 12.6 Hz), 159.3, 169.5;
31P NMR (CDCl3, 202 MHz) δ 16.33; HRMS (FAB) m/ z )
436.0888 [M + H]+ and 438.0870 [M + 2 + H]+ (1:1), calcd for
2937, 1660, 1595 cm-1; H NMR (CDCl3, 600 MHz) δ 1.31 (d, J
) 6.2 Hz, 3H), 1.34 (d, J ) 6.2 Hz, 3H), 1.38-1.40 (m, 6H), 2.09
(s, 3H), 3.86 (s, 3H), 3.89 (s, 3H), 4.11 (dd, J ) 2.6, 12.2 Hz,
1H), 4.78-4.83 (m, 2H), 5.58 (sextet, J ) 2.4, 8.1 Hz, 1H), 6.49
(d, J ) 7.7 Hz, 1H), 6.81-6.87 (m, 3H); 13C NMR (CDCl3, 150
MHz) δ 23.5, 23.9 (d, J ) 5.9 Hz), 24.0 (d, J ) 5.6 Hz), 24.4-
24.5 (m), 48.9 (d, J ) 153 Hz), 52.3, 56.2, 56.3, 73.1 (d, J ) 7.3
Hz), 73.5 (d, J ) 7.4 Hz), 110.7, 111.3, 119.1, 131.7 (d, J ) 12.3
Hz), 149.0, 149.2, 169.5; 31P NMR (CDCl3, 202 MHz) δ 16.28;
HRMS (FAB) m/ z 466.0994 [M + H]+ and 468.0976 [M + 2 +
H]+ (1:1), calcd for C18H30Br NO6P 466.0986.
C
17H28BrNO5P ) 436.0881. Anal. Calcd for C17H27Br NO5 P: C,
46.80; H, 6.24; N, 3.21. Found: C, 47.08; H, 6.32; N, 3.18.
A yield of 110 mg (0.252 mmol, 25%) of crude white solid was
obtained with (DHQD)2PHAL. After recrystallization with EtOAc/
hexane, 80 mg (0.183 mmol, 18%) of white, fluffy, crystalline
solid was obtained.
A yield of 140 mg (0.30 mmol, 30%) of a viscous, yellow oil
with 11% impurities (by 31P NMR) was obtained with
(DHQD)2PHAL.
Dieth yl [2-Acetyla m in o-1-br om o-2-(p-m eth oxyp h en yl)-
eth yl]p h osp h on a te (4). A yield of 118 mg (0.29 mmol, 29%) of
crude white solid was obtained with (DHQ)2PHAL. After re-
crystallization with EtOAc/hexane, 86 mg (0.21 mmol, 21%) of
white crystalline solid was obtained: mp 116-7 °C; IR (KBr)
3279, 2983, 2927, 1659, 1611, 1514 cm-1; 1H NMR (CDCl3, 600
MHz) δ 1.32 (t, J ) 14.1 Hz, 3H), 1.38 (t, J ) 14.1 Hz, 3H), 2.10
(s, 3H), 3.80 (s, 3H), 4.13-4.21 (m, 2H), 4.21-4.24 (m, 3H), 5.63
(sextet, J ) 2.6, 8.6 Hz, 1H), 6.48 (d, J ) 8.0 Hz, 1H), 6.87 (d, J
) 8.7 Hz, 2H), 7.22 (d, J ) 8.6 Hz, 2H); 13C NMR (CDCl3, 150
MHz) δ 16.7, 23.6, 48.1 (d, J ) 152 Hz), 51.9, 55.5 (d, J ) 4.0
Hz), 64.2 (d, J ) 6.8 Hz), 64.5 (d, J ) 6.8 Hz), 114.1, 114.6, 127.9,
130.9, 159.5, 169.6; 31P NMR (CDCl3, 202 MHz) δ 18.24; HRMS
(FAB) m/ z 408.0551 [M + H]+ and 410.0551 [M + 2 + H]+ (1:
1), calcd for C15H24BrNO5 P ) 408.0569. Anal. Calcd for C15H23
BrNO5P: C, 44.13; H, 5.68; N, 3.43. Found: C, 44.50; H, 5.69;
N, 3.40.
Diisop r op yl [2-Acetyla m in o-1-br om o-2-(4′-N,N-d im eth y-
la m in op h en yl)eth yl]p h osp h on a te (8). A yield of 107 mg
(0.238 mmol, 24%) of a viscous, yellow oil with 8% impurity (by
31P NMR) was obtained with (DHQ)2PHAL: 1H NMR (CDCl3,
600 MHz) δ 1.26 (d, J ) 7.4 Hz, 3H), 1.32 (d, J ) 6.6 Hz, 3H),
1.34-1.38 (m, 6H), 2.09 (s, 3H), 2.78 (s, 6H), 4.06 (dd, J ) 2.6,
12.2 Hz, 1H), 4.76-4.82 (m, 2H), 5.52-5.55 (m, 1H), 6.52 (d, J
) 7.6 Hz, 1H), 7.02 (d, J ) 8.1 Hz, 1H), 7.18 (d, J ) 8.3 Hz, 1H),
7.48 (s, 1H); 13C NMR (CDCl3, 150 MHz) δ 23.5, 23.9 (d, J ) 5.7
Hz), 24.0 (d, J ) 5.6 Hz), 24.4-24.8 (m), 44.4, 48.2 (d, J ) 153
Hz), 51.9, 55.1, 73.2 (d, J ) 7.0 Hz), 73.6 (d, J ) 7.2 Hz), 119.2,
120.4, 126.7, 132.2, 151.7, 169.5; 31P NMR (CDCl3, 202 MHz) δ
16.05; HRMS (FAB) m/ z 447.1033 [M - H]+ and 449.1044 [M
+ 2 - H]+ (1:1), calcd for C18H29BrN2O4P 447.1040.
A yield of 96 mg (0.214 mmol, 21%) of a viscous, yellow oil
with 12% impurities (by 31P NMR) was obtained with
(DHQD)2PHAL.
The single crystal was grown in a NMR test tube during about
10 days with 0.2 mL of ethyl acetate and 2 mL of hexane.
A yield of 114 mg (0.28 mmol, 28%) of crude white solid was
obtained with (DHQD)2PHAL. After recrystallization with EtOAc/
hexane, 86 mg (0.211 mmol, 21%) of white crystalline solid was
obtained.
Diisop r op yl [2-Acetyla m in o-1-br om o-2-(3′,4′-d im eth ox-
yp h en yl)eth yl]p h osp h on a te (6). A yield of 144 mg (0.31
mmol, 31%) of a viscous, yellow oil with 5% impurities (by 31P
NMR) was obtained with (DHQ)2PHAL: IR (KBr) 3293, 2981,
Ack n ow led gm en t. We thank Professor Myung Ho
Hyun for the chiral HPCL data. Financial support from
the Brain Korea 21 Program is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Data for the X-ray
crystal structure of 4 (1R*,2S*). This material is available free
of charge via the Internet at http://pubs.acs.org.
J O034535X
J . Org. Chem, Vol. 68, No. 23, 2003 9143