Ag/ThioClickFerrophos-catalyzed enantioselective mannich reaction and amination of glycine schiff base

Article abstract of DOI:10.1021/jo2003727

The AgOAc/ThioClickFerrophos complex catalyzed the asymmetric Mannich reaction of glycine Schiff base with N-tosylimines effectively to give a mixture of syn and anti adducts (syn/anti = 60/40-70/30) at high yields with high enantioselectivities (up to 98% ee). The complex also catalyzed the asymmetric amination of glycine Schiff base with di-tert-butyl azodicarboxylate with high enantioselectivity.

Full text of DOI:10.1021/jo2003727

NOTE
pubs.acs.org/joc
Ag/ThioClickFerrophos-Catalyzed Enantioselective Mannich Reaction
and Amination of Glycine Schiff Base
Kazumi Imae, Kenta Shimizu, Kenichi Ogata, and Shin-ichi Fukuzawa*
Department of Applied Chemistry, Institute of Science and Engineering, Chuo University, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551,
Japan
S Supporting Information
b
ABSTRACT: The AgOAc/ThioClickFerrophos complex catalyzed the asym-
metric Mannich reaction of glycine Schiff base with N-tosylimines effectively to
give a mixture of syn and anti adducts (syn/anti = 60/40ꢀ70/30) at high yields
with high enantioselectivities (up to 98% ee). The complex also catalyzed the
asymmetric amination of glycine Schiff base with di-tert-butyl azodicarboxylate
with high enantioselectivity.
he asymmetric Mannich reaction of glycine Schiff base with
imines is of current interest because it can provide an
T
efficient route to the preparation of optically active R,β-diamino
acids, which are versatile building blocks in organic synthesis and
have medicinal and biological significance.1 Therefore, chiral
metal complex catalysts and organocatalysts have been
developed2 for the asymmetric Mannich reaction of glycine
Schiff base since Jørgensen’s group first reported the Cu(I)/
phosphino-oxazoline complex-catalyzed reaction.3 Wu and
Hou’s Cu(I)/ FcPHOX,4 Wang’s Cu(I)/TF-BiphamPhos,5 Car-
retero’s Cu(I)/Fesulphos,6 and Feng’s Cu(II)/N,N-dioxide7 are
representative, effective chiral metal catalysts for the reaction.
Recently, we reported that CuOAc/ClickFerrophos (L2)
complexes exhibited highly exo stereoselectivity and excellent
enantioselectivity in the asymmetric 1,3-dipolar cycloaddition of
azomethine ylides with a vinyl sulfone.8 We also succeeded in
proceeding a highly endo and enantioselective reaction with R-
enones and R,β-unsaturated esters by using AgOAc/ThioClick-
Ferrophos complexes (L6).9 Zhou et al. recently reported the
first silver-catalyzed and highly enantioselective Mannich reac-
tion of glycine Schiff base by using the AgOAc/FcPHOX
complex.10 Encouraged by Zhou’s success with the silver-cata-
lyzed reaction, we tried to create AgOAc/ClickFerrophos and
ThioClickFerrophos-complex-catalyzed asymmetric Mannich
reactions of glycine Schiff base with N-tosylimine 2.
Figure 1. Diphosphines and P,S-ligands.
1,5-diphosphine to ClickFerrophos, was also examined for
comparison. The glycine Schiff base 1 and phenyl tosylimine
2a (1.2 equiv) were used as the model substrate. The reaction
was carried out in THF at room temperature with a loading of 3
mol % of the silver complex derived from a silver salt and the
ligand. The reaction proceeded to give a mixture of syn 3 and anti
4 diastereomers: the diastereomeric ratio and enantiomeric
excess (% ee) of each isomer were determined by chiral HPLC.
The results are summarized in Table 1. The use of Taniaphos L1
gave a mixture of (2S,3R)-syn-3 and (2S,3S)-anti-4 in the ratio of
72/28 with low % ee values (entry 1). It took 16 h for the
We initially focused on the exploration of appropriate Click-
Ferrophos and ThioClickFerrophos ligands and on an examina-
tion of reaction parameters such as silver salts and solvent. We
chose L2ꢀL6 as ligands (Figure 1), which were revealed to be
effective at producing high enantio- and diastereoselectivity in
the asymmetric reductive aldol reaction,11 allylic substitution,12
and 1,3-dipolar reaction with azomethine ylides in our previous
works.8,9 Taniaphos ligand (L1),13 which is structurally similar to
Received: February 27, 2011
Published: March 16, 2011
r
2011 American Chemical Society
3604
dx.doi.org/10.1021/jo2003727 J. Org. Chem. 2011, 76, 3604–3608
|

The Journal of Organic Chemistry
NOTE
Table 1. Optimization of the Mannich Reaction of Glycine
Table 2. Scope of Mannich Reaction of Glycine Schiff Base
Schiff Basea
with Respect to Tosyliminesa
entry [Ag]
ligand time (h) yield (%)b syn/antic ee (%)dsyn/anti
1
2
3
4
5
6
7
AgOAc L1
8
16
16
3
99
99
98
99
99
99
99
99
94
88
99
99
65
99
72/28
74/26
67/33
63/37
48/52
60/40
60/40
66/34
69/31
67/33
60/40
64/36
72/28
46/54
58/36
AgOAc L2
AgOAc L3
AgOAc L4
AgOAc L5
AgOAc L6
AgOAc ent-L6
ꢀ87/ꢀ70
ꢀ17/0
ꢀ42/ꢀ44
87/85
entry
imine (Ar)
2a (Ph)
yield (%)b
syn/antic
ee (%)c
4
1
2
3
4
5
6
7
8
9
99
99
99
99
99
99
99
99
99
60/40
68/32
71/29
75/25
58/42
58/42
56/44
60/40
57/43
98/98
98/97
98/97
95/98
95/98
98/98
97/96
85/94
91/91
3
98/98
2b (o-FC6H4)
3
ꢀ98/ꢀ98
97/95
2c (o-ClC6H4)
2d (o-BrC6H4)
2e (o-MeC6H4)
2f (p-ClC6H4)
2g (p-CF3C6H4)
2h (p-MeC6H4)
2i (p-MeOC6H4)
8e AgOAc L6
9f AgOAc L6
10g AgOAc L6
11 AgOTf L6
12 AgBF4 L6
13 AgPF6 L6
14h CuClO4 L6
5
19
20
24
24
24
18
97/95
93/98
98/95
86/95
33/65
96/96
a Glycine Schiff base (0.10 mmol), tosylimine (0.12 mmol), AgOAc
(0.003 mmol, 3 mol %), ligand (0.003 mmol, 3 mol %). b Isolated yield.
c Determined by HPLC (AD-H, OD-H, IA, and IB).
a Glycine Schiff base (0.10 mmol), tosylimine (0.12 mmol), [Ag] (0.003
mmol, 3 mol %), ligand (0.0033 mmol, 3.3 mol %). b Isolated yield.
c Determined by 1H NMR. d Determined by HPLC (Chiralpak AD-H).
e The reaction was carried out in diethyl ether. f The reaction was carried
out in toluene. g The reaction was carried out in CH2Cl2. h MS-4A and
triethyl amine (4.5 mol %) were added.
triethylamine: it yielded a 1:1 mixture of 3 and 4 with 96% ee
values after an 18ꢀh reaction. Thus, we concluded that the
optimal catalyst and solvents were AgOAc/L6 and THF, respec-
tively, with the goal of completing the reaction in the shortest
time (3 h) and obtaining the highest enantioselectivities for syn
and anti adducts. However, the high diastereoselectivity could
not be achieved by using the catalyst in the presence or absence of
MS-4A.
complex of AgOAc/(S,Rp)-L2 to complete the reaction that
produced a mixture of the opposite enantiomer of syn/anti
adducts quantitatively, % ee of (2R,3S)-syn and (2R,3R)-anti
adducts being 87% and 70%, respectively (entry 2). The cyclo-
hexyl derivative L3 also gave a mixture of 3 and 4 as adducts with
poor enantioselectivities (17%, and 0% ee, respectively): syn/anti
= 67/33 (entry 3). We found that derivative L6 was the most
appropriate ligand for the production of high % ee for both syn
and anti adducts, although diastereoselectivity was low; (2S,3R)-
3 and (2S,3S)-4 were both obtained at 98% ee (entry 6).14
Additionally, the reaction could be completed in a much shorter
time (3 h) than that with L2. Ligand L4, in which P- and S-donor
atoms were opposite to those in L5, was not effective at
producing the enantiomer of 3 and 4 at low enantioselectivities
(entry 4). It must be noted that similarly configured syn (and
anti) adduct(s) were obtained with the use of L2 and L6, where
central and planar chirality were opposite to each other; L2 and
L6 are configured as (S,Rp) and (R,Sp), respectively. The use of
the enantiomer of L6 produced the enantiomer of 3 and 4.
Having established ThioClickFerrophos L6 as the optimal
ligand, we investigated the effects of the metal precursor and
solvent in order to further optimize the reaction conditions. The
use of other solvents such as diethyl ether, toluene, and CH2Cl2
required longer reaction time (5ꢀ20 h) to complete the reaction,
and diastereo- and enantioselectivity was almost the same as
THF (entries 8ꢀ10). Other silver salts such as AgOTf, AgPF6,
and AgBF4 took longer reaction time to complete the reaction:
AgPF6 gave a low yield and enantioselectivity (entries 11ꢀ13).
The copper complex as CuClO4 was tested as a metal complex
precursor combining L6 in the presence of MS-4A and
The scope of the Mannich reaction with glycine Schiff base
was evaluated by using various aryl tosylimines (2aꢀi) by
altering the nature of the substituents on the phenyl ring under
the optimized conditions. The substitution pattern of the phenyl
ring in the tosylimines had an effect on the transformation. The
electron-withdrawing substituents in the ortho position tended to
produce higher syn selectivity (syn/anti = 70/30) than those with
unsubstituted tosylimine 2a (Table 2, entries 2ꢀ4); however, the
o-methyl group gave a similar ratio to 2a. For these substrates, the
% ee values of the adducts remained at as high a level as that of 2a.
para-Electron-donating substituents had a slight, negative effect
on enantioselectivity (entries 8 and 9), whereas para-electron-
withdrawing substituents had little effect on enantioselectivity; %
ee was similar to that of unsubstituted 2a (entries 6 and 7). In
either substrate, the syn/anti ratio was moderate, at 60/40.
Next, we applied the AgOAc/L6 complex to R-amination of
the glycine Schiff base 1 with azodicarboxylate 5 (Scheme 1).15
After optimization experiment with di-tert-butyl azodicarboxy-
late 5 (R = t-Bu), we could conclude that optimal conditions were
that the reaction occurred at 0 °C for 3 h in diethyl ether. The
optimal conditions would produce (ꢀ)-6 at the highest enan-
tioselectivity (96% ee). The same level of enantioselectivity was
obtained at even lower temperature (ꢀ20 °C). Solvents such as
THF, toluene, and CH2Cl2 could be used as a solvent, yielding
enantioselectivities of 93%, 93%, and 96% ee, respectively. The
3605
dx.doi.org/10.1021/jo2003727 |J. Org. Chem. 2011, 76, 3604–3608

The Journal of Organic Chemistry
NOTE
Scheme 1. Asymmetric Amination of Glycine Schiff Base
1Hanti, J = 6.5 Hz), 4.96 (dd, 1Hanti, J = 6.4, 7.7 Hz), 5.45 (dd, 1Hsyn, J =
2.5, 8.6 Hz), 5.84 (d, 1H, J = 7.8 Hz, NHsyn), 6.39 (d, 2Hsyn, J = 8.0 Hz),
6.40 (d, 1H, J = 7.5 Hz, NHanti), 6.75ꢀ7.82 (m, 16Hsyn, 18Hanti). 13
C
NMR (CDCl3) δ 21.38 (Memin), 21.41 (Memaj), 52.2 (CminNH), 52.3
(CmajNH), 53.7 (d, J = 2.4 Hz, CmajNd), 55.2 (d, J = 1.1 Hz, CminNd),
68.0 (OMemaj), 68.3 (OMemin), 115.0maj (d, J = 21.5 Hz), 115.1min (d,
J = 21.9 Hz), 123.8maj (d, J = 3.6 Hz), 123.9min (d, J = 3.4 Hz),
126.2ꢀ138.7 (several Ar signals), 142.98maj, 143.03min, 159.0 (d, J =
246.1 Hz, CmajF), 169.3anti, 169.9syn, 173.2maj (d, J = 43.4 Hz). 19F NMR
(CDCl3) δ 119.3syn, 120.5anti. HPLC (Daicel Chiralpak AD-H, hexane/
i-PrOH 90:10, flow rate 1.0 mL/min, 254 nm): tR = 25.9 min (minor
syn), 36.9 min (minor anti), 42.3 min (major anti), 47.8 min (major syn).
HRMS calcd for C30H28FN2O4S [M þ H]þ 531.1748, found 531.1766.
Methyl 3-(o-chlorophenyl)-2-[(diphenylmethylene)amino]-
3-(tosylamino)propanoate syn-3c þ anti-4c: 1H NMR (400 MHz,
CDCl3) δ 2.24 (s, 3H, Meanti), 2.35 (s, 3H, Mesyn), 3.44 (s, 3H, OMesyn),
3.48 (s, 3H, OMeanti), 4.30 (d, 1Hsyn, J = 2.0 Hz), 4.48 (d, 1Hanti, J = 5.8
Hz), 4.85 (dd, 1Hanti, J = 5.8, 6.5 Hz), 5.54 (dd, 1Hsyn, J = 2.0, 8.1 Hz), 6.28
(d, 2Hsyn, J = 6.5 Hz), 6.29 (d, 1H, J = 6.0 Hz, NHsyn), 6.47 (d, 1H, J = 7.5
Hz, NHanti), 6.80ꢀ7.82 (m, 16Hsyn, 18Hanti). 13C NMR (CDCl3) δ 21.40
(Memin), 21.41 (Memaj), 52.0 (CminNH), 52.3 (CmajNH), 56.2 (CmajNd),
66.5 (OMemaj), 66.6 (OMemin), 126.5ꢀ143.5 (several Ar signals), 169.4syn,
170.0anti, 172.6anti, 173.6syn. HPLC (Daicel Chiralpak IB, hexane/i-PrOH
90:10, flow rate 1.0 mL/min, 254 nm): tR = 9.8 min (minor syn), 10.2 min
(minor anti)-isomer, 11.8 min (major anti), 18.4 min (major syn). HRMS
calcd for C30H27ClN2O4SNa [M þ Na]þ 569.1278, found 569.1295.
Methyl 3-(o-bromophenyl)-2-[(diphenylmethylene)amino]-
3-(tosylamino)propanate syn-3d þ anti-4d:3,4 1H NMR (400
MHz, CDCl3) δ 2.24 (s, 3H, Meanti), 2.36 (s, 3H, Mesyn), 3.45 (s, 3H,
use of other silver salts resulted in lower yield and enantioselec-
tivity: AgBF4, 93% (85% ee); AgOTf, 43% (78% ee); AgPF6, 26%
(84% ee). The ethyl and tert-butyl esters of azodicarboxylate 5
gave almost the same level of enantioselectivity and the reaction
with p-methylbenzophenone imino ester (1: Ar = p-tolyl) gave
lower %ee values.
In conclusion, AgOAc/ThioClickFerrophos L6 complex cata-
lyzed the asymmetric Mannich reaction of glycine Schiff base with
tosylimines. The syn and anti adducts were obtained with high
enantioselectivities (up to 98% ee) but with a low to moderate syn
selectivity. The complex was also effective for the enantioselective R-
amination of glycine Schiff base with azodicarboxylate.
OMeanti), 3.46 (s, 3H, OMesyn), 4.33 (d, 1Hsyn, J = 2.1 Hz), 4.50 (d, 1Hanti
,
J = 5.2 Hz), 4.80 (dd, 1Hanti, J = 6.3, 5.5 Hz), 5.50 (dd, 1Hsyn, J = 2.2, 8.1 Hz),
6.26 (d, 2Hsyn, J = 6.3 Hz), 6.40 (d, 1H, J = 4.2 Hz, NHanti), 6.48 (d, 1H, J =
7.8 Hz, NHsyn), 6.81ꢀ7.83 (m, 16Hsyn, 18Hanti). 13C NMR (CDCl3)
δ 21.40 (Meanti), 21.41 (Mesyn), 52.0 (CantiNH), 52.3 (CsynNH), 58.4
(CsynþantiNd), 66.48 (OMeanti), 66.51 (OMesyn), 122.3ꢀ143.1 (several Ar
signals), 169.4syn, 169.8anti, 172.5anti, 173.6syn. HPLC (Daicel Chiralpak OD-
H, hexane/i-PrOH 90/10, flow rate 0.8 mL/min, 254 nm): tR = 11.4 min
(2R,3R)-isomer, 12.7 min (2R,3S)-isomer, 17.2 min (2S,3R)-isomer,
27.4 min (2S,3S)-isomer. HRMS calcd for C30H27BrN2O4SNa [M þ Na]þ
613.0773, found 613.0794.
’ EXPERIMENTAL SECTION
Typical Experimental Procedure for the Mannich-Type
Reaction of Glycine Schiff Base. All reactions were carried out
under nitrogen atmosphere with oven-dried glassware. In a 20-mL
Schlenk tube containing a stirring bar, ThioClickFerrophos L6 (2.07
mg, 0.0033 mmol) and AgOAc (0.50 mg, 0.0030 mmol) were dissolved
in THF (2.5 mL) and the mixture was stirred at room temperature for 30
min under nitrogen. Glycine Schiff base 1 (25.3 mg, 0.10 mmol) was
added then followed by phenyl tosylimine 2a (31.1 mg, 0.12 mmol). The
resulting solution was stirred at the same temperature for 3 h and then
filtered through Celite and concentrated. The 1H NMR measurement of
the crude product showed the presence of a diastereomeric mixture of
adducts (syn/anti = 60/40). The residue was purified by preparative
TLC (n-hexane/EtOAc = 4/1) to afford a mixture of syn-3a and anti-4a
Methyl 2-[(diphenylmethylene)amino]-3-(o-methylphenyl)-
3-(tosylamino)propanoate syn-3e þ anti-4e:5 1H NMR (400
MHz, CDCl3) δ 1.97 (s, 3H, Meanti), 2.04 (s, 3H, Mesyn), 2.26 (s, 3H,
Meanti), 2.33 (s, 3H, Mesyn), 3.51 (s, 3H, OMesyn), 3.53 (s, 3H, OMeanti),
4.06 (d, 1Hsyn, J = 2.0 Hz), 4.31 (d, 1Hanti, J = 6.6 Hz), 4.86 (dd, 1Hanti
,
1
J = 6.3 Hz), 5.40 (dd, 1Hsyn, J = 2.1, 8.4 Hz), 6.00 (d, 1H, J = 6.3 Hz,
NHanti), 6.20 (d, 2Hsyn, J = 4.6 Hz), 6.41 (d, 1H, J = 7.7 Hz, NHsyn),
6.79ꢀ7.82 (m, 16Hsyn, 18Hanti). 13C NMR (CDCl3) δ 18.5 (Mesyn),
18.9 (Meanti), 21.4 (Mesynþanti), 52.1 (CantiNH), 52.4 (CsynNH), 55.96
(CsynNd), 55.97 (CantiNd), 67.4 (OMesyn), 68.4 (OMeanti),
as a white solid, yield 51.0 mg (99%). H NMR (400 MHz, CDCl3)
δ 2.27 (s, 3H, Meanti), 2.34 (s, 3H, Mesyn), 3.49 (s, 3H, OMeanti), 3.51 (s,
3H, OMesyn), 4.14 (d, 1Hsyn, J = 2.5 Hz), 4.36 (d, 1Hanti, J = 5.7 Hz), 4.71
(dd, 1Hanti, J = 5.8, 7.6 Hz), 5.17 (dd, 1Hsyn, J = 2.4, 8.6 Hz), 5.80 (d, 1H,
J = 7.5 Hz, NHanti), 6.34 (d, 2Hsyn, J = 6.7 Hz), 6.40 (d, 1H, J = 8.6 Hz,
NHsyn), 6.85ꢀ7.83 (m, 17Hsyn, 19Hanti). 13C NMR (CDCl3) δ 21.4
(Me), 52.1 (CantiNH), 52.3 (CsynNH), 59.4 (CsynNd), 60.0 (CantiNd),
69.4 (OMeanti), 70.0 (OMesyn), 126.5ꢀ142.9 (several Ar signals),
169.6syn, 170.0anti, 172.1anti, 173.1syn. The enantiomeric excess (98%
and 98% ee values) and absolute configuration were determined by
HPLC by reference to the literature.3,4 HPLC (Daicel Chiralpak AD-H,
hexane/i-PrOH 90:10, flow rate 1.0 mL/min, 254 nm): tR = 21.5 min
(2R,3S)-isomer, 28.1 min (2R,3R)-isomer, 33.0 min (2S,3S)-isomer,
45.1 min (2S,3R)-isomer. HRMS calcd for C30H28N2O4SNa [M þ Na]þ
535.1668, found 535.1664.
122.2ꢀ138.6 (several Ar signals), 142.8syn, 142.9anti, 169.8syn, 170.3anti
,
172.3syn, 173.2anti. HPLC (Daicel Chiralpak AD-H, hexane/i-PrOH
85:15, flow rate 1.0 mL/min, 254 nm): tR = 10.4 min (2R,3S)-isomer,
15.4 min (2R,3R)-isomer, 16.6 min (2S,3S)-isomer, 24.0 min (2S,3R)-
isomer. HRMS calcd for C31H31N2O4S [M þ H]þ 527.2000, found
527.2003.
Methyl 3-(p-chlorophenyl)-2-[(diphenylmethylene)amino]-
3-(tosylamino)propanate syn-3f þ anti-4f:5 1H NMR (400 MHz,
CDCl3) δ 2.30 (s, 3H, Meanti), 2.37 (s, 3H, Mesyn), 3.50 (s, 3H þ 3H,
OMesynþanti), 4.13 (d, 1Hsyn, J = 2.2 Hz), 4.35 (d, 1Hanti, J = 5.7 Hz), 4.69
(dd, 1Hanti, J = 5.7, 7.3 Hz), 5.11 (dd, 1Hsyn, J = 2.1, 8.3 Hz), 5.75 (d, 1H,
J = 7.4 Hz, NHanti), 6.38 (d, 1H, J = 8.0 Hz, NHsyn), 6.45 (d, 2Hsyn, J = 7.1
Hz), 6.79ꢀ7.82 (m, 16Hsyn, 18Hanti). 13C NMR (CDCl3) δ 21.4 (Me),
52.2 (CantiNH), 52.4 (CsynNH), 58.9 (CsynNd), 59.3 (CantiNd), 69.0
Methyl 2-[(diphenylmethylene)amino]-3-(o-fluorophenyl)-
3-(tosylamino)propanoate syn-3b þ anti-4b: 1H NMR (400
MHz, CDCl3) δ 2.24 (s, 3H, Meanti), 2.35 (s, 3H, Mesyn), 3.44 (s, 3H,
OMesyn), 3.52 (s, 3H, OMeanti), 4.21 (d, 1Hsyn, J = 2.4 Hz), 4.42 (d,
3606
dx.doi.org/10.1021/jo2003727 |J. Org. Chem. 2011, 76, 3604–3608

The Journal of Organic Chemistry
NOTE
(OMesyn), 69.6 (OMeanti), 126.5ꢀ138.6 (several Ar signals), 143.1syn
,
determined by HPLC (Daicel Chiralpak AD-H, hexane/i-PrOH 90:10,
flow rate 1.0 mL/min, 254 nm): tR = 11.8 min (major), 13.1 min
(minor). [R]29D ꢀ40.87 (c 0.3, CHCl3). 1H NMR (CDCl3, 300 MHz)
δ 1.21ꢀ1.51 (m, 18H), 3.72 (s, 3H), 6.08 (br s, 1H), 6.75 (br s, 1H),
7.31ꢀ7.66 (m, 10H). 13C NMR (CDCl3) δ 27.9, 28.0, 52.6, 73.4, 80.7,
81.8, 127.9, 128.2, 128.4, 128.9, 129.2, 129.9, 130.7, 132.3, 135.7, 138.9,
154.1, 154.7, 169.0. HRMS cacld for C26H34N3O6 [M þ H]þ 484.2447,
found 484.2442.
143.2anti, 169.4syn, 169.7anti, 173.4syn, 173.5anti. HPLC (Daicel Chiralpak
OD-H, hexane/i-PrOH 85:15, flow rate 0.6 mL/min, 254 nm): tR = 10.9
min (2R,3R)-isomer, 12.3 min (2R,3S)-isomer, 13.4 min (2S,3S)-
isomer, 22.7 min (2S,3R)-isomer. HRMS calcd for C30H27ClN2O4SNa
[M þ Na]þ 569.1278, found 569.1281.
Methyl 2-[(diphenylmethylene)amino]-3-(tosylamino)-3-
(p-trifluoromethylphenyl)propanoate syn-3g þ anti-4g:7 1H
NMR (400 MHz, CDCl3) δ 2.27 (s, 3H, Meanti), 2.34 (s, 3H, Mesyn),
3.50 (s, 3H, OMeanti), 3.52 (s, 3H, OMesyn), 4.16 (d, 1Hsyn, J = 2.1 Hz),
4.40 (d, 1Hanti, J = 5.5 Hz), 4.79 (dd, 1Hanti, J = 5.7, 7.5 Hz), 5.19 (dd,
1Hsyn, J = 1.9, 8.2 Hz), 5.81 (d, 1H, J = 7.6 Hz, NHanti), 6.38 (d, 2Hsyn, J =
7.2 Hz), 6.46 (d, 1H, J = 8.0 Hz, NHsyn), 6.88ꢀ7.82 (m, 16Hsyn, 18Hanti).
13C NMR (CDCl3) δ 21.29 (Meanti), 21.32 (Mesyn), 52.2 (CantiNH),
52.5 (CsynNH), 59.1 (CsynNd), 59.4 (CantiNd), 68.9 (OMeanti), 69.4
(OMesyn), 124.9ꢀ143.3 (several Ar signals), 169.3syn, 169.5anti, 173.7syn
(d, J = 24.0 Hz). HPLC (Daicel Chiralpak AD-H, hexane/i-PrOH 95:5,
flow rate 1.0 mL/min, 254 nm): tR = 26.0 min (2R,3S)-isomer, 36.3 min
(2R,3R)-isomer, 66.8 min (2S,3R)-isomer, 75.0 min (2S,3S)-isomer.
HRMS calcd for C31H27F3N2O4SNa [M þ Na]þ 603.1542, found
603.1547.
Diethyl 1-{1-[(diphenylmethylene)amino]-2-methoxy-2-
oxoethyl}hydrazine-1,2dicarboxylate 6b: 1H NMR (CDCl3,
300 MHz) δ 1.18ꢀ1.34 (m, 6H), 3.75 (s, 3H), 4.15ꢀ4.34 (m, 4H),
6.06 (br s, 1H), 6.94 (br s, 1H), 7.30ꢀ7.46 (m, 8H), 7.66 (d, 2H, J = 8.2
Hz). 13C NMR (CDCl3) δ 14.0, 14.3, 52.7, 61.8, 62.8, 74.3, 127.6, 127.9,
128.1, 128.3, 129.0, 129.1, 129.9, 130.8, 135.5, 138.7, 155,2, 155.7, 168.6,
173.8. HPLC (Daicel Chiralpak AD-H, hexane/i-PrOH 90:10, flow rate
0.7 mL/min, 254 nm): tR = 25.4 min (minor), 27.3 min (major). [R]30
D
ꢀ45.19 (c 0.1, CHCl3). HRMS calcd for C22H26N3O6 [M þ H]þ
428.1821, found 428.1837.
Di-tert-butyl 1-{1-[(di-p-tolylmethylene)amino]-2-meth-
oxy-2-oxoethyl}hydrazine-1,2-dicarboxylate 6c: 1H NMR
(CDCl3, 300 MHz) δ 1.24ꢀ1.51 (m, 18H, t-Bu), 2.35 (s, 3H), 2.40
(s, 3H), 3.71 (br s, 3H), 6.08 (s, 1H), 6.75 (s, 1H), 7.09ꢀ7.26 (m, 8H),
7.54 (d, 2H, J = 8.2 Hz). 13C NMR (CDCl3) δ 21.3, 21.6, 27.9, 28.0,
52.6, 80.6, 81.6, 127.8, 128.5, 128.8, 129.0, 129.2, 130.1, 132.9, 135.1,
136.5, 138.7, 141.0, 142.8, 154.2, 154.7, 166.2, 169.1. HPLC (Daicel
Chiralpak AD-H, hexane/i-PrOH 90:10, flow rate 1.0 mL/min,
254 nm): tR = 10.0 min (minor), 11.5 min (major). [R]30D ꢀ43.37 (c
0.2, CHCl3). HRMS calcd for C28H38N3O6 [M þ H]þ 512.2760, found
512.2729.
Methyl 2-[(diphenylmethylene)amino]-3-(p-methylphenyl)-
3-(tosylamino)propanoate syn-3h þ anti-4h:5 1H NMR (400
MHz, CDCl3) δ 2.25 (s, 3H, Meanti), 2.26 (s, 3H, Mesyn), 2.28 (s, 3H,
Meanti), 2.35 (s, 3H, Mesyn), 3.48 (s, 3H, OMesyn), 3.49 (s, 3H, OMeanti),
4.14 (d, 1Hsyn, J = 2.3 Hz), 4.35 (d, 1Hanti, J = 5.7 Hz), 4.66 (dd, 1Hanti, J =
5.7, 7.5 Hz), 5.10 (dd, 1Hsyn, J = 2.0, 8.5 Hz), 5.74 (d, 1H, J = 7.5 Hz,
NHanti), 6.34 (d, 1H, J = 8.2 Hz, NHsyn), 6.40 (d, 2Hsyn, J = 6.9 Hz),
6.86ꢀ7.82 (m, 16Hsyn, 18Hanti). 13C NMR (CDCl3) δ 20.98 (Mesyn),
21.04 (Meanti), 21.4 (Meanti), 29.7 (Mesyn), 52.1 (CantiNH), 52.3
(CsynNH), 59.3 (CsynNd), 59.8 (CantiNd), 69.4 (OMeanti), 70.0
(OMesyn), 126.4ꢀ142.9 (several Ar signals), 169.7syn, 170.1anti, 172.96syn,
173.01anti. HPLC (Daicel Chiralpak IA, hexane/i-PrOH 85:15, flow rate
0.8 mL/min, 254 nm): tR = 13.6 min (2R,3S)-isomer, 17.8 min (2R,3R)-
isomer, 22.6 min (2S,3R)-isomer, 23.9 min (2S,3S)-isomer. HRMS calcd
for C31H31N2O4S [M þ H]þ 527.2000, found 527.2011.
’ ASSOCIATED CONTENT
Supporting Information. 1H and 13C NMR spectra
S
b
(PDF) and HPLC analytical dada for the products (3a and
4aꢀ3i and 4i) of Mannich reaction. This material is available free
of charge via the Internet at http://pubs.acs.org.
Methyl 2-[(diphenylmethylene)amino]-3-(p-methoxyph-
enyl)-3-(tosylamino)propanoate syn-3i þ anti-4i:3,5 1H NMR
(400 MHz, CDCl3) δ 2.29 (s, 3H, Meanti), 2.35 (s, 3H, Mesyn), 3.50 (s,
3H þ 3H, OMesynþanti), 3.74 (s, 3H þ 3H, OMesynþanti), 4.13 (d, J = 2.4
Hz, 1Hsyn), 4.35 (d, J = 5.8 Hz, 1Hanti), 4.68 (dd, 1Hanti, J = 5.7, 7.4 Hz),
5.10 (dd, 1Hsyn, J = 2.3, 8.4 Hz), 5.70 (d, 1H, J = 7.7 Hz, NHanti), 6.32 (d,
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: orgsynth@kc.chuo-u.ac.jp.
1H, J = 8.8 Hz, NHsyn), 6.45 (d, 2Hsyn J = 7.2 Hz), 6.64ꢀ7.82 (m, 16Hsyn
,
18Hanti). 13C NMR (CDCl3) δ 21.4 (Mesynþanti), 52.1 (CantiNH), 52.3
(CsynNH), 55.2 (OMeanti), 55.3 (OMesyn), 59.0 (OMesyn), 59.4
(OMeanti), 69.4 (OMeanti), 70.0 (OMesyn), 113.45 (OMeanti), 113.50
’ ACKNOWLEDGMENT
This study was financially supported by a Grant-in-Aid, No.
22550044, for Scientific Research from the Japan Society for the
Promotion of Science (JSPS) and a Chuo University Grant for
Special Research.
(OMesyn), 113.5ꢀ142.9 (several Ar signals), 158.8syn, 159.0anti, 169.7syn
,
170.0anti, 173.0anti, 173.1syn. HPLC (Daicel Chiralpak IA, hexane/i-
PrOH 85:15, flow rate 0.8 mL/min, 254 nm): tR = 16.7 min (2R,3S)-
isomer, 25.0 min (2R,3R)-isomer, 29.8 min (2S,3S)-isomer, 48.6 min
(2S,3R)-isomer. HRMS calcd for C31H31N2O5S [M þ H]þ 543.1948,
found 543.1939.
’ REFERENCES
(1) For a review, see: Viso, A.; de la Pradilla, R. F.; García, A.; Flores,
A. Chem. Rev. 2005, 105, 3167–3196.
(2) For reviews, see: (a) Arrayꢀas, R. G.; Carretero, J. C. Chem. Soc.
Rev. 2009, 38, 1940–1948. (b) Nꢀajera, C.; Sansano, J. M. Chem. Rev.
2007, 107, 4584–4671.
(3) Bernardi, L.; Gothelf, A. S.; Hazell, R. G.; Jørgensen, K. A. J. Org.
Chem. 2003, 68, 2583–2591.
(4) Yan, X.-X.; Peng, Q.; Li, Q.; Zhang, K.; Yao, J.; Hou, X.-L.; Wu,
Y.-D. J. Am. Chem. Soc. 2008, 130, 14362–14363.
(5) Liang, G.; Tong, M.-C.; Tao, H.; Wang, C.-J. Adv. Synth. Catal.
General Procedure for the Amination Reaction of Glycine
Schiff Base. In a 20-mL Schlenk tube containing a stirring bar,
ThioClickFerrophos L6 (4.7 mg, 0.0075 mmol) and AgOAc (1.1 mg,
0.0070 mmol) were dissolved in Et2O (1.3 mL) and the mixture was
stirred at room temperature for 30 min under nitrogen. The mixture was
cooled to 0 °C, and then a Et2O (1.2 mL) solution of Glycine Schiff base
1 (Ar = Ph) (58 mg, 0.23 mmol) was added followed by di-tert-butyl
azodicarboxylate (0.12 mL, 0.27 mmol, 2.2 M). The resulting solution
was stirred at the same temperature for 3 h and then filtered through
Celite and concentrated. The crude product was purified by preparative
TLC (n-hexane/EtOAc = 2/1) to afford 6a (Ar = Ph, R= t-Bu) as a
colorless oil, yield 98 mg (99%). The enantiomeric excess (96% ee) was
2010, 352, 1851–1855.
(6) (a) Hernꢀandez-Toribio, J.; Arrayꢀas, R. G.; Carretero, J. C. J. Am.
Chem. Soc. 2008, 130, 16150–16151. (b) Hernꢀandez-Toribio, J.;
Arrayꢀas, R. G.; Carretero, J. C. Chem.—Eur. J. 2010, 16, 1153–1157.
3607
dx.doi.org/10.1021/jo2003727 |J. Org. Chem. 2011, 76, 3604–3608

The Journal of Organic Chemistry
NOTE
(7) Shang, D.; Liu, Y. L.; Zhou, X.; Liu, X.; Feng, X. Chem.—Eur. J.
2009, 15, 3678–3681.
(8) Fukuzawa, S.-i.; Oki, H. Org. Lett. 2008, 10, 1747–1750.
(9) (a) Oura, I.; Shimizu, K.; Ogata, K.; Fukuzawa, S.-i. Org. Lett.
2010, 12, 1752–1755. (b) Shimizu, K.; Ogata, K.; Fukuzawa, S.-i.
Tetrahedron Lett. 2010, 51, 5068–5070.
(10) Chen, Q.-A.; Zeng, W.; Wang, D.-W.; Zhou, Y.-G. Synlett 2009,
2236ꢀ2241.
(11) Kato, M.; Oki, H.; Ogata, K.; Fukuzawa, S.-i. Synlett
2009, 1299–1301.
(12) Kato, M.; Nakamura, T.; Ogata, K.; Fukuzawa, S. Eur. J. Org.
Chem. 2009, 5232–5238.
(13) Ireland, T. T.; Grossheimann, G.; Wieser- Jeunesse, C.;
Knochel, P. Angew. Chem., Int. Ed. 1999, 38, 3212–3215; 2008, 47, 3666.
(14) The use of AgOAc/PPh3 afforded the product in a ratio of syn/anti =
68/32. The ratio was similar to that with ClickFerrophos ligands. The
diastereoselectivity may be governed by product stability not the ligand.
(15) The silver-catalyzed amination of glycine Schiff base has been
reported: Chen, Q.-A.; Zeng, W.; Zhou, Y.-G. Tetrahedron Lett. 2009,
50, 6866–6868.
3608
dx.doi.org/10.1021/jo2003727 |J. Org. Chem. 2011, 76, 3604–3608