An improved procedure for the diastereoselective addition of triorganozincates to N-(tert-butanesulfinyl)imines: use of catalytic dialkylzinc

Article abstract of DOI:10.1016/j.tetlet.2009.01.136

The addition of triorganozincates to (R)-N-(tert-butanesulfinyl)benzaldimine has been performed with very good results by using a catalytic amount of Me₂Zn (0.15 equiv) to generate the organozincate. Yields and/or diastereoselectivities of the

Full text of DOI:10.1016/j.tetlet.2009.01.136

Tetrahedron Letters 50 (2009) 3198–3201
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An improved procedure for the diastereoselective addition
of triorganozincates to N-(tert-butanesulfinyl)imines:
use of catalytic dialkylzinc
*
*
Raquel Almansa, David Guijarro , Miguel Yus
Departamento de Química Orgánica, Facultad de Ciencias and Instituto de Síntesis Orgánica (ISO), Universidad de Alicante, Apdo. 99, 03080 Alicante, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The addition of triorganozincates to (R)-N-(tert-butanesulfinyl)benzaldimine has been performed with
very good results by using a catalytic amount of Me2Zn (0.15 equiv) to generate the organozincate. Yields
and/or diastereoselectivities of the formed a-branched sulfinamides improve in comparison with the val-
ues obtained in the same reactions carried out with an excess of a previously formed triorganozincate. On
the other hand, the transfer of the two alkyl groups of a dialkylzinc reagent to the imine has been
achieved by using 0.5 equiv of dialkylzinc and 1.5 equiv of MeMgBr to generate the trialkylzincate. In
both methods, good to excellent yields and diastereomeric ratios (up to 98:2) have been obtained.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 5 January 2009
Accepted 28 January 2009
Available online 1 February 2009
Dedicated to Professor Luis Castedo on
occasion of his 70th birthday
Keywords:
Sulfinylimine
Triorganozincate
Diastereoselective addition
Catalytic dialkylzinc
Chiral primary amine
Triorganozincates1 are a class of zinc reagents which are more
reactive than the most common dialkylzincs and alkylzinc halides.
They have found several synthetic applications, such as addition to
by employing a catalytic amount of the precursor dialkylzincs and
the diastereoselective addition of those triorganozincates to N-
(tert-butanesulfinyl)imines.
In our previous work, we had tried to carry out the addition of
EtMe2ZnMgBr10 to N-(tert-butanesulfinyl)benzaldimine 1 (Scheme
1) by utilizing 0.5 equiv of Me2Zn to generate the mixed trialkyl-
zincate.8b We assumed that Me2Zn was regenerating after the
transfer of the ethyl group from the trialkylzincate to the imine,
and we hoped that the formed Me2Zn would react with another
molecule of EtMgBr, regenerating in this way the trialkylzincate
aldehydes and ketones,2 conjugate addition to
a,b-unsaturated
ketones,1b,3 cross-coupling,4 nucleophilic substitution4b,5 and halo-
gen-zinc exchanges with alkyl,5,6 alkenyl4a and aryl halides.5
Organozincates have also been used as nucleophiles in the addition
to imines bearing a chiral auxiliary derived from a phenethyl-
amine, an
a
-aminoester or an O-protected b-aminoalcohol.7 The
participation of trialkylzincates in the zinc chloride-catalyzed
addition of Grignard reagents to N-phenyl- and N-tosylimines has
also been postulated.2b Recently, we have reported on the use of
triorganozincates as very efficient nucleophiles to perform the
addition of alkyl or alkenyl groups to N-(tert-butanesulfi-
nyl)imines,8 a class of imines that have been shown to be versatile
substrates for asymmetric synthesis,9 especially for the prepara-
tion of chiral primary amines.9b,c,f By using an excess of a dialkyl-
zinc and a Grignard reagent to generate the triorganozincate,
very good yields and diastereoselectivities were observed in the
addition products, which could be easily transformed into the cor-
responding enantiomerically enriched amines by desulfinylation of
the nitrogen atom under mild reaction conditions. Herein, we pres-
ent our preliminary results on the generation of triorganozincates
and closing
a possible catalytic cycle. Unfortunately, when
O
S
O
S
O
S
i,ii
N
t-Bu
+
HN
t-Bu
HN
t-Bu
Ph
H
Ph
R3
Ph
R3
1
2
3
a: R3 = Et
b: R3 = n-C5H11
c: R3 = i-Pr
d: R3 = CH2=CH
* Corresponding authors. Fax: +34 965903549.
E-mail addresses: dguijarro@ua.es (D. Guijarro), yus@ua.es (M. Yus).
Scheme 1. Reagents and conditions: (i) R1MgBr, R22Zn (cat.), THF, ꢀ78 °C. (ii) NH4Cl
(aq.).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.01.136

R. Almansa et al. / Tetrahedron Letters 50 (2009) 3198–3201
3199
Table 1
Diastereoselective addition of triorganozincates to N-(tert-butanesulfinyl)benzaldimine 1 using a catalytic amount of the dialkylzinc reagent
Ent.
Proced.
R1MgBr
R22Zn
Time (h)
Product
Yieldb (%)
R1
Equiv
R2
Equiv
No.a
R3
2:3 ratioc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Ad
Be
Be
Be
Ci
Et
Et
Et
Et
Et
Et
Et
Et
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.125
1.3
1.3
1.3
1.3
1.5
1.5
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
0.5
0.5
12
1
1
1
1
1.75
3.25
2.25
2
2
2
2a
2a
2a
2a
2a
2a
2a
2a
2a
2b
2c
2d
2a
2c
Et
Et
Et
Et
Et
Et
Et
Et
47
90
94:6
97:3
90:10
93:7
97:3
97:3
97:3
97:3
97:3
93:7
94:6
98:2
97:3
94:6
0.25f
0.1g
0.5
77
20h
87
Ci
0.25f
0.1g
0.125
0.15
0.15
0.15
0.15
0.5
88
Ci
54h
78h
93
Dj
Ek
Ek
Ek
Ek
Fl
Et
Et
n-C5H11
i-Pr
CH2=CH
Me
Me
n-C5H11
i-Pr
CH2=CH
Et
99
83
2
1
1
94
84m
90m
Fl
i-Pr
0.5
i-Pr
a
Product number corresponding to the major diastereoisomer.
The crude reaction mixture only showed the mixture of diastereoisomers 2 + 3 (300 MHz 1H NMR) without any noticeable by-product. The yields are calculated according
b
to the amount of crude mixture that was obtained after work-up.
c
Diastereomeric ratio determined from the crude reaction mixture by HPLC using a ChiralCel OD-H column. The absolute configuration of the major diastereoisomer was
deduced by removal of the N-sulfinyl group and comparison of the sign of the specific rotation of the free primary amine with the reported data.
d
EtMgBr (0.75 mmol) was added dropwise to a mixture of Me2Zn (0.25 mmol) and imine 1 (0.5 mmol) in THF (3 mL) at ꢀ78 °C, and the reaction mixture was stirred at that
temperature.
e
The solution of imine 1 (0.5 mmol) in THF (3 mL) was added dropwise over ca. 30 min to the mixture of Me2Zn (0.25 mmol) and EtMgBr (0.75 mmol) at ꢀ78 °C, and the
reaction mixture was stirred at that temperature.
f
The reaction was performed by following the indicated procedure, but by using 0.25 equiv of Me2Zn with regard to imine 1.
g
The reaction was performed by following the indicated procedure, but by using 0.1 equiv of Me2Zn with regard to imine 1.
h
Yield estimated by 1H NMR using diphenylmethane as an internal standard. The reaction was not complete.
i
A mixture of Me2Zn (0.25 mmol) and EtMgBr (0.25 mmol) was stirred for 10 min at room temperature. After cooling to ꢀ78 °C, a fraction (1 mL, 0.17 mmol) of the
solution of imine 1 (prepared by dissolving 0.50 mmol of the imine in 3 mL of THF) was added dropwise over ca. 5 min, and the mixture was stirred for additional 10 min.
Then, a fraction of EtMgBr (0.25 mmol) was added dropwise over ca. 5 min and the mixture was stirred for additional 10 min. This process of alternate addition of imine and
EtMgBr was repeated until the total amount of both reagents had been added.
j
A mixture of Me2Zn (0.25 mmol) and R1MgBr (0.75 mmol) was stirred for 10 min at room temperature. After cooling to ꢀ78 °C, a fraction (3 mL, 0.50 mmol) of the
solution of imine 1 (prepared by dissolving 2.0 mmol of the imine in 12 mL of THF) was added dropwise over ca. 20 min using a syringe pump. Immediately after the addition
was finished, a fraction of R1MgBr (0.50 mmol) was added dropwise over ca. 5 min, and the mixture was stirred for additional 10 min. This process of alternate addition of
imine and R1MgBr was repeated until the total amount of both reagents had been added.
k
Like procedure D, but the solution of imine 1 was prepared by dissolving 1.7 mmol of it in 10 mL of THF. Three fractions of 2.9 mL (0.50 mmol) and a last fraction of 1.3 mL
(0.20 mmol) of this solution were added. The time needed to add the last fraction was 9 min.
The solution of imine 1 (1.0 mmol) in THF (6 mL) was added with a syringe pump over ca. 40 min to the mixture of R22Zn (0.50 mmol) and MeMgBr (1.50 mmol) at ꢀ78 °C,
l
and the reaction mixture was stirred at that temperature.
Yield calculated for the stoichiometry 1 R22Zn ! 2 2.
m
EtMgBr was added dropwise to the mixture of Me2Zn and imine 1
in THF at ꢀ78 °C, the reaction was not complete after stirring for
12 h and afforded the expected sulfinamides in 47% yield and a
2a:3a ratio of 94:6 (procedure A, Table 1, entry 1). Since the
diastereoselectivity was quite good, we decided to carefully
screen the reaction conditions with the aim of improving the con-
version and reducing the reaction time and the amount of Me2Zn.
First, the order of addition of the reagents was changed in order
to assure the presence of the trialkylzincate in the reaction med-
ium when the imine was added. By slowly adding the imine to
the mixture of Me2Zn (0.5 equiv) and EtMgBr (1.5 equiv) at
ꢀ78 °C, a 90% yield and a diastereomeric ratio of 97:3 were ob-
tained (procedure B, Table 1, entry 2). The improvement in both
the conversion and the diastereoselectivity encouraged us to per-
form some tests with lower loadings of Me2Zn. When 0.25 and
0.1 equiv of the latter were used, maintaining the ratios between
the rest of the reagents and the addition rate, both yield and
diastereoselectivity were reduced (Table 1, entries 3 and 4). The
conversion decreases as the amount of Me2Zn is reduced. Accord-
ingly, it seemed that it was necessary to have a minimum amount
of the preformed triorganozincate in the reaction flask before the
addition of the imine in order for the reaction to work success-
fully. The presence of a large excess of the Grignard reagent with
regard to the amount of trialkylzincate should be also avoided,
since we had already proved that the results obtained with those
two organometallic species were different.8
In order to fulfil these two requirements, we thought about per-
forming the addition of the Grignard reagent and the imine in sev-
eral steps. The solution of the imine 1 (0.50 mmol in 3 mL of THF)
was divided into three fractions of 1 mL each. After the addition of
one of these fractions to the preformed trialkylzincate (0.25 mmol)
for ca. 5 min and stirring for additional 10 min, a fraction of EtMgBr
(0.25 mmol) was added dropwise for ca. 5 min and the mixture
was stirred for 10 min to allow the regeneration of the triorgano-
zincate before the entrance of a new fraction of the imine, by reac-
tion of the excess of EtMgBr with the Me2Zn formed after the
addition reaction. This process of alternate addition of imine and
EtMgBr was repeated until the total amount of both reagents had
been added, which afforded results almost same as for procedure
B (compare entries 2 and 5 in Table 1). This stepwise procedure
C (Table 1, footnote i) was repeated with 0.25 and 0.1 equiv of
Me2Zn. In the first case, the results were identical to the ones ob-
tained with 0.5 equiv of Me2Zn (compare entries 5 and 6 in Table
1), whereas the reaction with 0.1 equiv did not reach completion
(Table 1, entry 7). We were pleased to see that the diastereoselec-
tivity was as high as 97:3 in the three assays performed following
procedure C.
With the idea of adjusting Me2Zn loading to the minimum
amount that allowed for completion of the reaction, a slight mod-
ification in the addition procedure was made consisting in adding
the fractions of the solution of imine 1 for ca. 20 min by using a
syringe pump and starting the addition of a fraction of EtMgBr

3200
R. Almansa et al. / Tetrahedron Letters 50 (2009) 3198–3201
immediately after the end of the addition of the fraction of the
imine (procedure D, Table 1, entry 8, footnote j). This modification
allowed the use of 0.125 equiv of Me2Zn and 1.125 equiv of
EtMgBr, and gave a 78% yield and a 97:3 diastereomeric ratio. Since
the reaction was not complete, the amount of the imine was re-
duced in order to increase the number of equiv of Me2Zn and
EtMgBr to 0.15 and 1.3, respectively. This last modification led to
a 93% yield and a 2a:3a ratio of 97:3 in a reaction time of only
2 h (procedure E, Table 1, entry 9). Procedure E was chosen as
the optimum reaction conditions.
We then tested the transfer of some other alkyl groups to imine
1 (Scheme 1 and Table 1). Primary alkyls (Table 1, entries 9 and 10),
the isopropyl group (Table 1, entry 11) and the vinyl group (Table
1, entry 12) were effectively transferred to the imine, giving the ex-
pected sulfinamides in very good yields and diastereomeric ratios.
In all cases, yields were higher than the ones obtained when an ex-
cess of the previously formed zincate was used.8 The diastereose-
lectivity was of the order same as the ones observed in our
previous report, being even higher for the addition of the vinyl
group (98:2 in this catalytic addition against 95:5 in the stoichiom-
etric version8).
As was the case in the reactions with an excess of the triorgano-
zincate,8 in all cases the R1 group of the Grignard reagent was the
only one adding to the imine, the corresponding methylation prod-
uct being not detected in the crude mixtures. Thus, the methyl
group acted as a non-transferable group,11 which allowed the
regeneration of Me2Zn after the addition reaction and the complete
transfer of the R1 group. To the best of our knowledge, this is the
first time that a catalytically generated triorganozincate has been
used as nucleophile in a diastereoselective addition to an imine.
The very slow transfer rate of the methyl group gave us another
idea: we thought that by using 0.5 equiv of a dialkylzinc reagent
and an excess of MeMgBr, we would be able to transfer the two
R2 groups of the dialkylzinc to the imine. After the reaction be-
tween the initially generated R22MeZnMgBr and the imine, an R2
group would have been transferred and R2ZnMe would have been
formed. The reaction of the latter with MeMgBr would generate
R2Me2ZnMgBr, which would transfer the second R2 group to the
imine. Thus, the solution of imine 1 was added dropwise to the
stirred mixture of Et2Zn or i-Pr2Zn (0.5 equiv) and MeMgBr
(1.5 equiv) at ꢀ78 °C (procedure F, Table 1, entries 13 and 14, foot-
note l), and after only 1 h, yields of the expected addition products
of 84% (for 2a) and 90% (for 2c) were obtained, the diastereoselec-
tivities being exactly the same as the ones achieved with procedure
E (compare entry 9 with entry 13 and entry 11 with entry 14 in Ta-
ble 1). It is worth noting that these yields are based on the starting
amounts of the imine 1, and the fact that they are much higher
than 50% imply that two mmol of the addition products has been
formed for each mmol of R22Zn. To the best of our knowledge, this
is the first time that the two alkyl groups of a dialkylzinc reagent
have been transferred in an addition reaction. This feature makes
procedure F very useful from a synthetic point of view, especially
when the R22Zn is more easily accessible than the R1MgBr or when
the R group that one wishes to be transferred is functionalized.
After the addition of the triorganozincates to the imine, the 2:3
ratios were determined in the crude reaction mixtures by HPLC
using a ChiralCel OD-H column. In all the reactions that reached
completion, the only products that could be detected in the crude
reaction mixtures were the two diastereoisomers 2 and 3, which
were submitted to the desulfinylation procedure without further
purification. The removal of the N-sulfinyl group was easily
achieved by treatment of the crude reaction mixtures with HCl in
MeOH, affording the expected primary amines. By comparison of
the sign of their specific rotation with the data reported in the lit-
erature, the absolute configuration of the asymmetric carbon atom
of the major diastereoisomer of the addition products could be
determined. The enantiomeric excesses of the free primary amines
were determined by benzoylation of the nitrogen atom and analy-
sis of the obtained benzamides by HPLC using a ChiralCel OD-H
column. There was no loss of optical purity during the desulfinyla-
tion process: the ee values of the free amines matched perfectly
well with the corresponding 2:3 ratios.12
In conclusion, we have presented here two new and very effi-
cient procedures to perform the highly diastereoselective addition
of triorganozincates to N-(tert-butanesulfinyl)imines using a cata-
lytic amount of a dialkylzinc reagent as a precursor or the organo-
zincates. Reactions are fast and provide the expected sulfinamide
products in excellent yields and very high diastereomeric ratios.
The low transfer ability of the methyl group allows for (a) the suc-
cess of the addition reaction by using only a small amount
(0.15 equiv) of dimethylzinc, which reduces the reaction cost in
comparison with the same reaction carried out with an excess of
the triorganozincate, and (b) the transfer of both alkyl groups of
a dialkylzinc reagent to the imine via the corresponding mixed zin-
cate. Since the N-sulfinyl group can be easily removed, these meth-
odologies represent two new and very efficient procedures for the
asymmetric synthesis of primary amines. Further efforts to eluci-
date the mechanism of the reaction and to find synthetic applica-
tions of these methodologies are currently underway in our
laboratories.
Acknowledgements
This work was generously supported by the Spanish Ministerio
de Educación y Ciencia (MEC; Grant No. CONSOLIDER INGENIO
2010, CSD2007-00006 and CTQ200765218) and the Generalitat
Valenciana (GV/2007/036). R.A. thanks the Spanish Ministerio de
Educación y Ciencia for a predoctoral fellowship. We also thank
MEDALCHEMY S.L. for a gift of chemicals.
Supplementary data
Supplementary data (complete experimental details for proce-
dures E and F) associated with this article can be found, in the on-
line version, at doi:10.1016/j.tetlet.2009.01.136.
References and notes
1. (a) Harada, T.; Oku, A. In Organozinc Reagents: A Practical Approach; Knochel, P.,
Jones, P., Eds.; Oxford University Press: Oxford, 1999; pp 101–117; (b) Knochel,
P. Science of Synthesis 2004, 3, 5–90.
2. See, for instance: (a) Musser, C. A.; Richey, H. G., Jr. J. Org. Chem. 2000, 65, 7750–
7756; (b) Hatano, M.; Suzuki, S.; Ishihara, K. J. Am. Chem. Soc. 2006, 128, 9998–
9999; (c) Gosselin, F.; Britton, R. A.; Mowat, J.; O’Shea, P. D.; Davies, I. W. Synlett
2007, 2193–2196.
3. Lee, V. J. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I.,
Semmelhack, M. F., Eds.; Pergamon Press: Oxford, 1991; Vol. 4, section
1.2.2.1.4.
4. (a) Harada, T.; Katsuhira, T.; Hara, D.; Kotani, Y.; Maejima, K.; Kaji, R.; Oku, A. J.
Org. Chem. 1993, 58, 4897–4907; (b) Kondo, Y.; Takazawa, N.; Yamazaki, C.;
Sakamoto, T. J. Org. Chem. 1994, 59, 4717–4718; (c) Gauthier, D. R., Jr.;
Szumigala, R. H., Jr.; Dormer, P. G.; Armstrong, J. D., III; Volante, R. P.; Reider, P.
J. Org. Lett. 2002, 4, 375–378.
5. Kondo, Y.; Fujinami, M.; Uchiyama, M.; Sakamoto, T. J. Chem. Soc., Perkin Trans.
1 1997, 799–800.
6. Harada, T.; Kotani, Y.; Katsuhira, T.; Oku, A. Tetrahedron Lett. 1991, 32, 1573–
1576.
7. (a) Alvaro, G.; Pacioni, P.; Savoia, D. Chem. Eur. J. 1997, 3, 726–731; (b) Alvaro,
G.; Martelli, G.; Savoia, D. J. Chem. Soc., Perkin Trans. 1 1998, 775–784.
8. (a) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron: Asymmetry 2008, 19, 603–
606; (b) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron: Asymmetry 2008, 19,
2484–2491.
9. For reviews, see: (a) Davis, F. A.; Chen, B.-C. Chem. Soc. Rev. 1998, 27, 13–18; (b)
Ellman, J. A.; Owens, T. D.; Tang, T. P. Acc. Chem. Res. 2002, 35, 984–995; (c)
Ellman, J. A. Pure Appl. Chem. 2003, 75, 39–46; (d) Zhou, P.; Chen, B.-C.; Davis, F.
A. Tetrahedron 2004, 60, 8003–8030; (e) Morton, D.; Stockman, R. A.
Tetrahedron 2006, 62, 8869–8905; (f) Lin, G.-Q.; Xu, M.-H.; Zhong, Y.-W.; Sun,
X.-W. Acc. Chem. Res. 2008, 41, 831–840.

R. Almansa et al. / Tetrahedron Letters 50 (2009) 3198–3201
3201
10. This formula indicates only composition of the organometallic mixture and
implies neither the molecular formula nor the dominant species in solution.
For a study about the possible reactive species in a reaction between Et3ZnLi
and di-tert-butyl ketone, see: Maclin, K. M.; Richey, H. G., Jr. J. Org. Chem. 2002,
67, 4602–4604.
12. The determination of the relative proportion of 2 and 3 by HPLC using UV
detection is an estimate unless it has been proven that both diastereoisomers
have the same UV response. The fact that the enantiomeric ratios of the
free primary amines are equal to the corresponding 2:3 diaste-
reomeric ratios determined by HPLC indicates that the latter technique is
appropriate for estimating the diastereomeric ratios of this type of
sulfinamides.
11. The low transfer ability of the methyl group from triorganozincates has been
shown in other addition reactions. For instance, see Refs. 1b,3,7a.