Highly stereoselective synthesis of C2-chiral and meso nitroxides from an optically active pyrrolidine

Article abstract of DOI:10.1016/S0957-4166(00)00181-6

Starting from (2R,5R)-2,5-bis(methoxymethyl)pyrrolidine 1, hydroxylamine cis-3 was synthesized with high stereoselectivity by successive oxidation and addition of PhMgBr. By using PhLi, trans (C₂-chiral) pyrrolidine nitroxide trans-7 was obtained from nitrone 5 derived from hydroxylamine 3. The cis (meso) counterpart cis-7 was produced along with trans-7 when PhMgBr was employed in place of PhLi. Moreover, cis-7 was also obtained selectively by using PhLi and Et₂AlCl with nitrone 5. The change of stereochemical bias observed when EtMgBr and/or nitrone 10 bearing an ethyl group were employed is also discussed. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00181-6

Tetrahedron: Asymmetry 11 (2000) 2339±2346
Highly stereoselective synthesis of C2-chiral and meso
nitroxides from an optically active pyrrolidine
Tomohiro Shibata, Kouhei Uemae and Yukio Yamamoto*
Graduate School of Human and Environmental Studies, Kyoto University, Sakyo-ku, Yoshida, Kyoto 606-8501, Japan
Received 22 March 2000; accepted 9 May 2000
Abstract
Starting from (2R,5R)-2,5-bis(methoxymethyl)pyrrolidine 1, hydroxylamine cis-3 was synthesized with
high stereoselectivity by successive oxidation and addition of PhMgBr. By using PhLi, trans (C2-chiral)
pyrrolidine nitroxide trans-7 was obtained from nitrone 5 derived from hydroxylamine 3. The cis (meso)
counterpart cis-7 was produced along with trans-7 when PhMgBr was employed in place of PhLi.
Moreover, cis-7 was also obtained selectively by using PhLi and Et2AlCl with nitrone 5. The change of
stereochemical bias observed when EtMgBr and/or nitrone 10 bearing an ethyl group were employed is
also discussed. # 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
Optically active nitroxides have been employed as chiral catalysts or reagents in the ®eld of
asymmetric synthesis, including oxidations of alcohols, trapping prochiral carbon radicals, and
living polymerization.1 They were also synthesized as the key components of paramagnetic chiral
liquid crystals.2 In particular, C2-chiral nitroxides attract much attention because of their
potential ability for excellent stereoregulation. Several C2-chiral nitroxides have been prepared
and they have been recently well reviewed by Braslau.3 We now describe the synthesis of a new
homochiral pyrrolidine nitroxide starting from an optically active pyrrolidine 1 bearing two
symmetrically substituted methoxymethyl groups. Although our synthetic strategy is based
fundamentally upon the reported procedure for the trans dimethyl analogue by Einhorn,4 we
achieved highly stereoselective syntheses of both the optically active trans (C2-chiral) and the cis
(meso) products. Recently, Braslau reported the synthesis of racemic trans-isoindoline nitroxide
and the cis counterpart from an amine through a hydroxynitrone and from a phthalimide,
respectively.3
* Corresponding author.
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00181-6

2340
T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
2. Results and discussion
Oxidation of (2R,5R)-2,5-bis(methoxymethyl)pyrrolidine5 1 with MeReO3/urea H2O2 gave
nitrone 2.6 Without puri®cation, it was converted to hydroxylamine cis-3 by the action of
PhMgBr in THF (Scheme 1). In order to assess the cis/trans ratio of 3, an aliquot of 3 was
reduced with Zn/HCl to give pyrrolidine cis-4. The GC±MS analysis showed that cis-4 contained
a small amount (<5%) of the trans counterpart. This contamination might lower the ee of
nitrone 5 and therefore the ee of nitroxide trans-7. However, it proved not to be a problem
because the ee of trans-7 was easily increased by recrystallization. Hydroxylamine cis-3 can be
puri®ed by silica gel column chromatography, but cis-3 should not be puri®ed because it was
susceptible to air oxidation to nitrone 5. The cis geometry of phenyl group to hydrogen atom at
position 5 in hydroxylamine cis-3 was proved by the NOE experiment as depicted in Scheme 1.
Accordingly, the 2R con®guration of cis-3 was assigned based on the 5R con®guration which was
not a€ected by the oxidation and the Grignard reaction. The stereochemistry of the addition
agrees with the model presented by Keana (Scheme 3, model A) that an organometallic reagent
attacks the pyrrolidine nitrone from the less hindered site.7 Hydoxylamine cis-3 was oxidized with
air in the presence of Cu(II) ion to nitrone 5 which was treated with two equivalent amounts of
PhLi in THF to give hydroxylamine trans-6. This hydroxylamine was also susceptible to air oxi-
dation and often contained a trace amount of nitroxide 7 which interfered with the NMR mea-
surement for the unambiguous assignment of the con®guration of trans-6. Without puri®cation,
trans-6 was converted to nitroxide trans-7 by air oxidation in the presence of Cu(II) ion.
Nitroxide trans-7 was obtained in an overall yield of 19% from amine 1 after silica gel chromato-
graphy followed by recrystallization from hexane. The ee of trans-7 proved to be >98% by chiral
HPLC analysis on Chiralcel OD-H. A portion of nitroxide trans-7 was converted to amine trans-8
in order to carry out the unambiguous structure con®rmation by the NMR measurement.
.
Scheme 1. Synthesis of C2-chiral pyrrolidine nitroxide

T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
2341
In order to assess the stereoselectivity of PhLi attack on nitrone 5, a portion of crude
hydroxylamine 6 was reduced with Zn/HCl to give amine 8. It was shown by the GC±MS
measurement that the trans content of 8 was >98%. When PhMgBr was used in place of PhLi,
the content of the trans form decreased to 67% in contrast to the high trans selectivity reported
with the nitrone having Me group.4 From the hydroxylamine thus obtained, nitroxide cis-7 and
amine cis-8 were also prepared. The cis (meso) con®guration of these compounds was established
based on the observation that their speci®c rotations were essentially zero while the trans (chiral)
counterparts trans-7 and trans-8 had signi®cant values. Finally, the single crystal X-ray structure
of nitroxide trans-7 unambiguously established the trans geometry.
As to nitrones bearing ether functions, the change of stereochemistry in addition reactions of
organometallics has been reported when Lewis acids are added into the reaction mixture.8 In
particular, Et2AlCl was reported to change the stereochemistry signi®cantly.8c In this context, we
examined the e€ect of Et2AlCl and found that the stereochemical bias of the addition of PhLi to
nitrone 5 was reversed by the additive (Scheme 2). The cis content increased to 72% when the
reaction was conducted with an equimolar amount of Et2AlCl and the value ®nally reached 92%
with 4 equivalents of the Lewis acid. These values were also determined by the GC±MS analysis
of amine 8. The addition of Et2AlCl did not change the cis/trans ratio in the reaction by PhMgBr.
Scheme 2. Inversion of cis/trans selectivity with Et2AlCl
In previous reports,8 the chelation of the Lewis acids with two oxygen atoms of nitrone and
ether functions is proposed. Accordingly, the change of the steric course is ascribed to the change
of the nitrone conformation generated by the chelate formation. In the present synthesis of
hydroxylamine trans-6, when Et2AlCl is absent, it is rational that PhLi attacked from the
opposite direction of the phenyl group in the pyrrolidine ring because the phenyl group can be
considered to be larger than the methoxymethyl group as depicted in Scheme 3 (model B).
When the Lewis acid coordinates to oxygen atom (1) in methoxymethyl group, the upper side
in Scheme 3 becomes more crowded (model D). That is, the chelating methoxymethyl group is
considered to be larger than the phenyl group in the pyrrolidine ring. Consequently, PhLi attacks
from the same direction of the phenyl group in the ring to yield cis-6. An NMR titration experiment
indicated that the Lewis acid coordinated to both the two ether oxygen atoms (1) and (2) in
nitrone 5. The other chelation structure formed with oxygen atom (2) and nitrone oxygen atom is
considered not to a€ect the stereochemistry of the reaction. This estimation agrees with the fact
that the cis selectivity was improved when Et2AlCl dosage increased. The above chelation model
also rationalizes the lower trans selective addition of PhMgBr to nitrone 5 (model C). As to the
nitrone bearing ether functions, the similar chelation model has been proposed in which a molecule
of the Grignard reagent coordinates to the oxygen atoms in the nitrone and another molecule of
the Grignard reagent attacks on the nitrone from the less hindered side.9

2342
T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
Scheme 3. Direction of PhLi attack on nitrone
Next, we aimed to clarify the stereochemistry of the addition reaction to the nitrone bearing Et
group which is considered to be less bulky than the methoxymethyl group. We examined the reaction
with EtMgBr and found that the yields were comparable with those using PhMgBr (Scheme 4)
although the lithium counterpart did not give the desired compounds. We here designate the
isomers as trans- and cis-form which have the two methoxymethyl groups on the opposite side
and on the same side of the pyrrolidine ring, respectively, regardless of the spatial relation of the
other substituents. The stereochemical bias in the reaction between nitrone 2 and EtMgBr was
almost same as the reaction with PhMgBr. That is, hydroxylamine cis-9 was obtained as the
major isomer (96%) whose geometry was proved by the NOE experiment as depicted in Scheme
4. The addition of PhMgBr to nitrone 10 gave hydroxylamine cis-11 as the major product. The
cis/trans ratio was also determined by GC±MS after converting it to amine 12. The geometry of
12 was also assessed by the NOE experiment where the correlation between Ph and Et groups was
observed. This result is consistent with model C in Scheme 3 because the Et group becomes much
smaller than the methoxymethyl group coordinated with PhMgBr. On the other hand, EtMgBr
Scheme 4. Comparison of Et and Ph groups on Grignard reagents attack

T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
2343
attacked less selectively on nitrone 5 bearing a Ph group to give hydroxylamine 110. Although
hydroxylamines 11 and 110 as well as amines 12 and 120 are depicted di€erently in Scheme 4, they
are identical compounds and their cis/trans ratios are di€erent. When nitrone 10 with an Et group
was treated with EtMgBr, a ca. 50:50 mixture of cis and trans isomers of hyroxylamine 13 was
1
obtained whose ratio was assessed by its H NMR. It can be concluded that only the organo-
metallic reagent bearing the bulky Ph group discriminates the bulkiness of the substituents at the
opposite site of the pyrrolidine ring.
3. Experimental
3.1. General
All mp's were measured on a Yanagimoto micro melting point apparatus and are uncorrected.
1
1H NMR and 13C NMR spectra were recorded with a JEOL alpha-500 (at 500 MHz to H, 126
MHz to 13C) and a JEOL-JNM-EX-270 (at 270 MHz to 1H, 68 MHz to 13C). J Values are given
in hertz. IR spectra were recorded with a SHIMADZU FTIR-8600PC. ESR spectra were recor-
ded with a JEOL JES-FE1XG. Optical rotation was measured with a JASCO DIP-1000 (with a 1
dm cell). HRMS were done with a JEOL JMS-DX-300. HPLC analyses were run on a JASCO
880-PU chromatographic system with an 875-UV detector (280 nm). GC±MS analyses were done
on a SHIMADZU GCMS-QP5050A. Reactions with organometallics were conducted under
argon atmosphere. The elemental analyses were performed by Kyoto University elemental ana-
lysis center.
3.2. (R)-3,4-Dihydro-2,5-bis(methoxymethyl)-2H-pyrrole 1-oxide 2
.
A mixture of MeReO3 (73 mg, 0.29 mmol), urea H2O2 complex (10.7 g, 112 mmol) in MeOH
(50 mL) was stirred at room temperature for 10 min and then cooled to 0ꢀC. (2R,5R)-2,5-
Bis(methoxymethyl)pyrrolidine 1 (2.30 g, 14.4 mmol) in MeOH (20 mL) was then added dropwise
to this reaction mixture. After stirring at 0ꢀC for 1 h and at room temperature for 12 h, the
mixture was evaporated. The yellowish solid residue was washed with CHCl3 (50 mL). The
®ltrate was concentrated to give 2 as a yellow oil (2.34 g) which was employed for the next step
without puri®cation. Compound 2: ‰ꢀŠD25=+4.3 (c 1.1, CHCl3); ꢁH (270 MHz, CDCl3) 2.22 (dd, J
3.8, 11.3, 2H), 2.6±2.9 (m, 2H), 3.37 (s, 3H), 3.39 (s, 3H), 3.45±3.6 (m, 1H), 4.0±4.2 (m, 2H), 4.38
(s, 2H); ꢁC (68 MHz, CDCl3) 20.0, 28.3, 59.2, 59.3, 66.9, 70.4, 73.1, 146.5; ꢂmax (neat)/cm^1 2827,
1661, 1117, 910.
3.3. (2R,5R)-1-Hydroxy-2,5-bis(methoxymethyl)-2-phenylpyrrolidine cis-3
A solution of PhBr (4.28 g, 27 mmol) in anhydrous THF (4 mL) was added to a suspension of
Mg turnings (0.66 g, 28 mg-atom) in anhydrous THF (16 mL). This mixture was re¯uxed for 1 h
to give PhMgBr solution. This solution was cooled at ^78ꢀC, and a solution of nitrone 2 (2.33 g,
14 mmol) in THF (14 mL) was added dropwise to it. The temperature was raised slowly to room
temperature, and stirring was maintained overnight. The reaction mixture was then poured into a
saturated NH4Cl solution (30 mL). The mixture was extracted with CHCl3 (30 mLÂ3). The
combined organic extracts were dried over anhydrous MgSO4, and the solvent was removed

2344
T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
under reduced pressure. The cis, trans mixture (3.36 g, cis:trans 20:1) was obtained. Pure cis-3
(2.95 g, 81% overall yield from 1) was obtained by silica gel column chromatography (75%
EtOAc in hexane). cis-3: Rf=0.7 (75% EtOAc in hexane); ‰ꢀŠ2D5=^4.7 (c 1.4, CHCl3); ꢁH (500 MHz,
CDCl3) 1.7±1.8 (m, 1H), 1.9±2.1 (m, 1H), 2.1±2.2 (m, 1H), 2.3±2.4 (m, 1H), 3.0±3.1 (m, 1H), 3.32 (s,
3H), 3.34 (s, 3H), 3.4±3.6 (m, 2H), 3.7±4.0 (m, 2H), 7.2±7.5 (m, 5H); ꢁC (126 MHz, CDCl3) 23.8,
30.0, 59.0, 59.4, 63.6, 72.8, 74.4, 78.2, 125.4, 127.2, 127.8, 128.2, 139.0; ꢂmax (neat)/cm^1 3377,
2826, 1113; HRFABMS, m/z found 252.1608 (M+H)+, calcd for C14H21NO3 252.1601 (M+H)+.
3.4. (R)-3,4-Dihydro-2,5-bis(methoxymethyl)-2-phenyl-2H-pyrrole 1-oxide 5
To a solution of hydroxylamine cis-3 (1.71 g, 6.8 mmol) in methanol (8 mL), NH4OH (28%,
.
1.7 mL) and Cu(OAc)2 H2O (207 mg, 1.1 mmol) were added. Air was bubbled into the solution
until a persistent deep blue color was observed (5 min). After evaporation, saturated NaHCO3
(20 mL) was added to the residue. The mixture was extracted with CHCl3 (20 mLÂ3). The
combined extracts were dried over anhydrous MgSO4 and evaporation gave 5 as an oil (0.49 g)
which was employed for the next step without further puri®cation. Compound 5: ‰ꢀŠ2D5=+85.9 (c
1.0, CHCl3); ꢁH (500 MHz, CDCl3) 2.1±2.4 (m, 2H), 2.5±3.0 (m, 2H), 3.41 (s, 3H), 3.44 (s, 3H),
3.6±4.3 (m, 2H), 4.5 (br s, 2H), 7.2±7.5 (m, 5H); ꢁC (126 MHz, CDCl3) 26.1, 28.7, 59.2, 59.3, 67.1,
73.6, 83.0, 125.6, 127.9, 128.5, 139.1, 146.8; ꢂmax (neat)/cm^1 2823, 1732, 1600, 1117.
3.5. (2R,5R)-2,5-Bis(methoxymethyl)-2,5-diphenylpyrrolidin-1-oxy radical trans-7
To a solution of PhLi in cyclohexane±diethyl ether (0.86 M, 14 mL, 12 mmol) and THF (10 mL),
nitrone 5 (1.51 g, 6.0 mmol) in THF (20 mL) was added dropwise at ^78ꢀC. The temperature was
raised slowly to room temperature, and stirring was maintained overnight. The reaction mixture
was then poured into saturated NH4Cl (30 mL). The mixture was extracted with CHCl3 (30
mLÂ3). The combined organic extracts were dried over anhydrous MgSO4, and the solvent was
removed under reduced pressure to give trans-6 (2.22 g). To a solution of trans-6 (2.22 g) in
.
methanol (30 mL), NH4OH (28%, 2 mL) and Cu(OAc)2 H2O (226 mg, 1.13 mmol) were added.
Air was bubbled into the solution until a persistent deep blue color was observed (5 min). After
evaporation, saturated NaHCO3 (40 mL) was added to the residue. The mixture was extracted
with CHCl3 (30 mLÂ3). The combined extracts were dried over anhydrous MgSO4 and
evaporation gave crude trans-7 (2.15 g). Pure trans-7 (0.33 g, overall yield of 19% from 1) was
obtained by silica gel column chromatography (10% EtOAc in hexane) followed by recrystal-
lization from hexane. trans-7: Rf=0.5 (10% EtOAc in hexane); mp 134±136ꢀC; ‰ꢀŠ25 +144.6 (c 1.5,
D
CHCl3); >98% ee by chiral HPLC [Chiralcel OD-H (4 mmÂ25 cm), ¯ow rate 0.5 mL/min, 280
nm], retention time/min 24.8 [(R,R)-trans-7], 27.5 [(S,S)-trans-7]; ꢂmax (®lm)/cm^1 2893, 1452,
1103; HRFABMS, m/z found 326.1755 (M+), calcd for C20H24NO3 326.1757 (M+). Anal. found:
C, 73.31; H, 7.40; N, 4.21; calcd for C26H24NO3: C, 73.59; H, 7.41; N, 4.29; ESR (toluene) 3 lines,
aN=13.90 G, g=2.006.
3.6. cis-2,5-Bis(methoxymethyl)-2,5-diphenylpyrrolidin-1-oxy radical cis-7
To a solution of Grignard reagent obtained from PhBr (0.30 g, 1.9 mmol), Mg turnings (46 mg,
1.9 mg-atom) and anhydrous THF (4 mL), nitrone 5 (0.24 g, 0.96 mmol) in THF (6 mL) was
added dropwise at ^78ꢀC. The same work-up as described above gave cis-6 (0.36 g). From cis-6

T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
2345
(0.36 g), crude cis-7 (0.36 g) was obtained according to the same procedure for trans-7. On silica
gel column chromatography (10% EtOAc in hexane), cis-7 (28 mg, overall yield of 7% from 1)
eluted after trans-7 (56 mg, overall yield of 14% from 1). cis-7: Rf=0.3 (10% EtOAc in hexane);
mp 49±51ꢀC; ‰ꢀŠD25=0.0 (c 1.0, CHCl3); ꢂmax (Nujol)/cm^1 1111, 698; HRFABMS, m/z found
326.1742 (M+), calcd for C20H24NO3 326.1757 (M+). Anal. found: C, 73.47; H, 7.31; N, 4.05;
calcd for C26H24NO3: C, 73.59; H, 7.41; N, 4.29; ESR (toluene) 3 lines, aN=13.81 G, g=2.007.
3.7. (2R,5R)-2,5-Bis(methoxymethyl)-2,5-diphenylpyrrolidine trans-8
A mixture of nitroxide trans-7 (72 mg), water (19 mL), concentrated HCl (5 mL) and Zn
powder (100 mg, 1.6 mg-atom) was re¯uxed under vigorous stirring until the yellow color of the
nitroxide had disappeared (1 h). After cooling, the mixture was made alkaline (pH>12) with
NaOH (5%, 50 mL) and then extracted with ether (30 mLÂ3). The combined extracts were dried
over anhydrous MgSO4 and evaporated to a€ord trans-8 (72 mg, 100%). An analytical sample
was obtained by silica gel column chromatography (10% EtOAc in hexane). trans-8: ‰ꢀŠ2D5=+73.3
(c 0.87, CHCl3); ꢁH (500 MHz, CDCl3) 1.3±1.9 (br s, 1H), 2.1±2.2 (m, 2H), 2.2±2.4 (m, 2H), 3.14
(s, 6H), 3.25±3.36 (m, 4H), 7.2±7.3 (m, 2H), 7.3±7.4 (m, 4H), 7.5±7.6 (m, 4H); ꢁC (126 MHz,
CDCl3) 35.4, 59.2, 68.5, 80.7, 126.3, 126.4, 127.9, 147.7; ꢂmax (neat)/cm^1 3319, 2883, 1447, 1111;
HRFABMS, m/z found 312.1971 (M+H)+, calcd for C20H26NO2 312.1965 (M+H)+.
3.8. cis-2,5-Bis(methoxymethyl)-2,5-diphenylpyrrolidine cis-8
From cis-7 (46 mg), cis-8 (44 mg, 100%) was obtained by the same procedure for trans-8. An
analytical sample was obtained by silica gel column chromatography (10% EtOAc in hexane).
cis-8: ‰ꢀŠ2D5=0.0 (c 1.0, CHCl3); ꢁH (500 MHz, CDCl3) 1.26 (s, 1H), 1.9±2.1 (m, 2H), 2.2±2.3 (m,
2H), 3.33 (s, 6H), 3.50 (s, 4H), 7.0±7.65 (m, 10H); ꢁC (126 MHz, CDCl3) 34.9, 59.3, 68.4, 80.5,
126.0, 126.7, 127.6, 147.6; ꢂmax (neat)/cm^1 3348, 3323, 2882, 1445, 1109; HRFABMS, m/z found
312.1969 (M+H)+, calcd for C20H26NO2 312.1965 (M+H)+.
3.9. (2R*,5S*)-1-Hydroxy-2-ethyl-2,5-bis(methoxymethyl)pyrrolidine cis-9
From 1 (550 mg), cis-9 (548 mg, 78%) was obtained by the same procedure for cis-3 where
EtMgBr was employed in place of PhMgBr. cis-9: ꢁH (500 MHz, CDCl3) 0.9 (t, J 7.5, 3H), 1.4±1.5
(m, 3H), 1.6±1.7 (m, 1H), 1.8±1.9 (m, 1H), 1.9±2.0 (m, 1H), 3.2±3.3 (m, 1H), 3.28 (d, J 9.5, 1H),
3.35 (s, 3H), 3.37 (s, 3H), 3.43 (dd, J 5.5, 9.3, 1H), 3.44 (d, J 9.5, 1H), 3.55 (dd, J 5.5, 9.3, 1H); ꢁC
(126 MHz, CDCl3) 9.1, 20.1, 22.8, 27.1, 59.2, 59.3, 63.5, 68.9, 75.7, 77.8.
3.10. (2R*,5R*)-2-Ethyl-2,5-bis(methoxymethyl)-5-phenylpyrrolidine cis-12
From cis-9 (105 mg), cis-12 (37 mg, 27%) was obtained through 10 by the same procedure for
cis-8. cis-12: ꢁH (500 MHz, CDCl3) 0.8 (t, J 7.5, 3H), 1.4±1.5 (m, 2H), 1.5±1.6 (m, 1H), 1.7±1.8 (m,
1H), 2.0±2.1 (m, 1H), 2.1±2.2 (br, 1H), 2.2±2.3 (m, 1H), 3.25 (d, J 9.0, 1H), 3.26 (s, 3H), 3.32 (d, J
9.0, 3H), 3.3±3.4 (m, 2H), 3.39 (s, 3H), 7.1±7.2 (m, 1H), 7.2±7.3 (m, 2H), 7.5±7.6 (m, 2H); ꢁC (126
MHz, CDCl3) 9.0, 31.4, 33.2, 34.6, 59.2, 59.2, 65.1, 68.3, 77.8, 80.5, 126.2, 126.4, 126.6, 127.7,
147.7; HRFABMS (the corresponding nitroxide radical), m/z found 278.1754 (M+), calcd for
C16H24NO3 278.1757.

2346
T. Shibata et al. / Tetrahedron: Asymmetry 11 (2000) 2339±2346
3.11. General procedure for determination of stereoselectivity of reactions yielding hydroxylamines
3, 6, 11, and 13 Ð determination of cis/trans ratio of amines 4, 8, and 12
The reactions of nitrones with organometallics, PhMgBr, PhLi or EtMgBr, were conducted as
described in the synthesis of cis-3. As to the reaction with Et2AlCl, it was added to the solution of
nitrones before the addition of organometallics. The cis/trans ratio of hydroxylamine 13 was
1
assessed by its H NMR. A portion of hydroxylamines 3, 6, and 11 were reduced to amines 4, 8,
and 12, respectively, by the procedure described in trans-8. The crude products were analyzed by
GC±MS to determine the cis/trans ratio [column: DB-5MS (J & W, 30 mÂ0.25 mm), initial
temperature 50ꢀC for 5 min, increasing rate 10ꢀC/min for 20 min, ®nal temperature 250ꢀC, ¯ow
rate 1.7 mL/min]. Retention time/min 19.86 (cis-4), 19.92 (trans-4), 25.15 (cis-8), 25.18 (trans-8),
14.02 (trans-12), 14.13 (cis-12).
Acknowledgements
We thank Dr. R. Tamura at Kyoto University for helpful discussions. This research was
partially supported by a Grant-in-Aid for Scienti®c Research on Priority Area (A) (No.
10146227) from the Ministry of Education, Science, Sports, and Culture of Japan.
References
1. (a) Ma, Z.; Huang, Q.; Bobbitt, J. M. J. Org. Chem. 1993, 58, 4837. (b) Rychnovsky, S. D.; McLernon, T. L.;
Rajapakse, H. J. Org. Chem. 1996, 61, 1194. (c) Braslau, R.; Burrill II, L. C.; Mahal, L. K.; Wedeking, T. Angew.
Chem., Int. Ed. Engl. 1997, 36, 237. (d) Naik, N.; Braslau, R. Tetrahedron 1998, 54, 667 (review). (e) de Nooy,
A. E. J.; Basemer, A. C.; van Bekkum, H. Synthesis 1996, 1153 (review).
2. Tamura, R.; Suzuki, S.; Azuma, N.; Matsumoto, A.; Toda, F.; Ishii, Y. J. Org. Chem. 1995, 60, 6820.
3. Braslau, R.; Chaplinski, V.; Goodson, P. J. Org. Chem. 1998, 63, 9857 and references cited therein.
4. Einhorn, J.; Einhorn, C.; Ratajczak, F.; Gautier-Luneau, I.; Pierre, J.-L. J. Org. Chem. 1997, 62, 9385.
5. (a) Yamamoto, Y.; Ohmori, H.; Sawada, S. Synlett 1991, 319. (b) Yamamoto, Y.; Hoshino, J.; Fujimoto, Y.;
Ohmoto, J.; Sawada, S. Synthesis 1993, 298.
6. (a) Murray, R. W.; Iyanar, K.; Chen, J.; Wearing, J. T. J. Org. Chem. 1996, 61, 8099. (b) Goti, A.; Nannelli, L.
Tetrahedron Lett. 1996, 37, 6025.
7. Lee, T. D.; Birrell, G. B.; Keana, J. F. W. J. Am. Chem. Soc. 1978, 100, 1618.
8. (a) Schade, W.; Reissig, H.-U. Synlett 1999, 632. (b) Merino, P.; Castillo, E.; Merchan, F. L.; Tejero, T.
Tetrahedron: Asymmetry 1997, 8, 1725. (c) Dondoni, A.; Franco, S.; Junquera, F.; Merchan, F. L.; Merino, P.;
Tejero, T.; Bertolasi, V. Chem. Eur. J. 1995, 1, 505. (d) Dondoni, A.; Junquera, F.; Merchan, F. L.; Merino, P.;
Tejero, T. Synthesis 1994, 1450. (e) Dondoni, A.; Franco, S.; Merchan, F. L.; Merino, P.; Tejero, T. Tetrahedron
Lett. 1993, 34, 5475. (f) Merino, P.; Castillo, E.; Franco, S.; Merchan, F. L.; Tejero, T. Tetrahedron 1998, 54,
12301. (g) Bloch, R. Chem. Rev. 1998, 98, 1407 (review). (h) Dondoni, A.; Perrone, D.; Rinaldi, M. J. Org. Chem.
1998, 63, 9252.
9. (a) Chang, Z.-Y.; Coates, R. M. J. Org. Chem. 1990, 55, 3464. (b) Chang, Z.-Y.; Coates, R. M. J. Org. Chem.
1990, 55, 3475.