Orthoacylimines: A new class of chiral auxiliaries for nucleophilic addition of organolithium reagents to imines

Article abstract of DOI:10.1021/jo026571m

A new class of orthoacylimine-derived chiral auxiliaries has been synthesized and tested in the nucleophilic addition of organolithium reagents to imines. The precursors can be prepared by an aza-Wittig reaction between the corresponding orthoacyl azide and a variety of aldehydes in the presence of trialkylphosphines. The nucleophilic addition of organolithium reagents led to the addition products in good yields and with good to excellent diastereoselectivities (from 85:15 to 99:1). The chiral, nonracemic secondary amines could be readily obtained under mild hydrolytic condition. Furthermore, the chiral auxiliary can be recovered in quantitative yield and reconverted to the starting orthoacyl azide precursor. This method was applied to the synthesis of (S)-t-leucine.

Full text of DOI:10.1021/jo026571m

Or th oa cylim in es: A New Cla ss of Ch ir a l Au xilia r ies for
Nu cleop h ilic Ad d ition of Or ga n olith iu m Rea gen ts to Im in es
Alessandro A. Boezio, Geoffrey Solberghe, Caroline Lauzon, and Andre´ B. Charette*
De´partement de Chimie, Universite´ de Montre´al, P.O. Box 6128, Station Downtown,
Montre´al, Que´bec, Canada H3C 3J 7
andre.charette@umontreal.ca
Received October 15, 2002
A new class of orthoacylimine-derived chiral auxiliaries has been synthesized and tested in the
nucleophilic addition of organolithium reagents to imines. The precursors can be prepared by an
aza-Wittig reaction between the corresponding orthoacyl azide and a variety of aldehydes in the
presence of trialkylphosphines. The nucleophilic addition of organolithium reagents led to the
addition products in good yields and with good to excellent diastereoselectivities (from 85:15 to
99:1). The chiral, nonracemic secondary amines could be readily obtained under mild hydrolytic
condition. Furthermore, the chiral auxiliary can be recovered in quantitative yield and reconverted
to the starting orthoacyl azide precursor. This method was applied to the synthesis of (S)-t-leucine.
Chiral amines are among the most important func-
tionality in organic chemistry. This class of compounds
are found in numerous natural1 and unnatural bioactive
products,2 and they are the key component of several
chiral ligands3 and auxiliaries.4 It is therefore not
surprising to see that several synthetic methodologies
F IGURE 1. Some examples of the most effective chiral
auxiliaries for nucleophilic addition to imines.
relying on the stereoselective nucleophilic addition to
imines have been developed for their preparation.5 Re-
cently, many effective catalytic systems have been de-
us to investigate whether we could develop a new method
that would allow the generation of chiral, nonracemic
amines as well as the complete recovery of the chiral
veloped for the preparation of R-chiral amines involving
the nucleophilic addition of dialkylzinc reagents.6 Al-
though several chiral auxiliaries are available for syn-
auxiliary or one of its precursors that could be easily
thesizing chiral amines (Figure 1), the most effective ones
reconverted into the required starting substrate. In light
of this, we were intrigued by the possibility of using
orthoacylimines 5 as precursors to chiral amines (Scheme
1). The substrate may show an increased electrophilicity
(1, 2)7 usually cannot be recovered upon cleavage or
require several steps to regenerate a suitable auxiliary
precursor.8 Others (3, 4)9 require some additional steps
for the regeneration of a suitable precursor.
One aspect of our program involving the development
of new routes to chiral, nonracemic secondary amines led
(6) (a) Hermanns, N.; Dahmen, S.; Bolm, C.; Bra¨se, S. Angew. Chem.,
Int. Ed. 2002, 41, 3692-3694. (b) Wei, C.; Li, C.-J . J . Am. Chem. Soc.
2002, 124, 5638-5639. (c) Dahmen, S.; Brase, S. J . Am. Chem. Soc.
2002, 124, 5940-5941. (d) Porter, J . R.; Traverse, J . F.; Hoveyda, A.
H.; Snapper, M. L. J . Am. Chem. Soc. 2001, 123, 984-985. (e) Porter,
J . R.; Traverse, J . F.; Hoveyda, A. H.; Snapper, M. L. J . Am. Chem.
Soc. 2001, 123, 10409-10410. (f) Frantz, D. E.; Fassler, R.; Tomooka,
C. S.; Carreira, E. M. Acc. Chem. Res. 2000, 33, 373-381. (g) Fujihara,
H.; Nagai, K.; Tomioka, K. J . Am. Chem. Soc. 2000, 122, 12055-12056.
(h) Boezio, A. A.; Charette, A. B. J . Am. Chem. Soc. 2003, 125, 1692-
1693.
(7) (a) Higashiyama, K.; Inoue, H.; Takahashi, H. Tetrahedron Lett.
1992, 33, 235-238. (b) Higashiyama, K.; Inoue, H.; Takahashi, H.
Tetrahedron 1994, 50, 1083-1092. (c) Davis, F. A.; Zhang, Y.; An-
demichael, Y.; Fang, T.; Fanelli, D. L.; Zhang, H. J . Org. Chem. 1999,
64, 1403-1406. (d) Liu, G.; Cogan, D. A.; Ellman, J . A. J . Am. Chem.
Soc. 1997, 119, 9913-9914. (e) Yang, T. K.; Chen, R. Y.; Lee, D. S.;
Peng, W. S.; J iang, Y. Z.; Mi, A. Q.; J ong, T. T. J . Org. Chem. 1994,
(1) In addition to well-known amino acids and chiral amines, several
natural products can be derived from simple chiral amines. For a recent
example, see: Davis, F. A.; Mohanty, P. K. J . Org. Chem. 2002, 67,
1290-1296.
(2) For recent examples, see: (a) Krishnamurthy, D.; Han, Z.; Wald,
S. A.; Senanayake, C. H. Tetrahedron Lett. 2002, 43, 2331-2333. (b)
Kobayashi, S.; Matsubara, R.; Kitagawa, H. Org. Lett. 2002, 4, 143-
145. (c) Hett, R.; Fang, Q. K.; Gao, Y.; Wald, S. A.; Senanayake, C. H.
Org. Process Res. Dev. 1998, 2, 96-99. (d) Fang, Q. K.; Senanayake,
C. H.; Han, Z.; Morency, C.; Grover, P.; Malone, R. E.; Butler, H.; Wald,
S. A.; Cameron, T. S. Tetrahedron: Asymmetry 1999, 10, 4477-4480.
(e) Spencer, C. M.; Foulds, D.; Peters, D. H. Drugs 1993, 46, 1055.
(3) (a) Lucet, D.; Le Gall, T.; Mioskowski, C. Angew. Chem., Int. Ed.
1998, 37, 2580-2627. (b) Alexakis, A.; Tomassini, A.; Chouillet, C.;
Roland, S.; Mangeney, P.; Bernardinelli, G. Angew. Chem., Int. Ed.
2000, 39, 4093-4095. (c) Bennani, Y. L.; Hanessian, S. Chem. Rev.
1997, 97, 3161-3195.
(4) (a) J uaristi, E.; Leon-Romo, J . L.; Reyes, A.; Escalante, J .
Tetrahedron: Asymmetry 1999, 10, 2441-2495. (b) Ager, D. J .;
Prakash, I.; Schaad, D. R. Chem. Rev. 1996, 96, 835-875.
(5) For excellent reviews, see: (a) Bloch, R. Chem. Rev. 1998, 98,
1407-1438. (b) Enders, D.; Reinhold, U. Tetrahedron: Asymmetry
1997, 8, 1895-1946. (c) J ohansson, A. Contemp. Org. Synth. 1995, 2,
393-408.
59, 914-921. For
a recent report on an improved and practical
auxiliary, see: (f) Han, Z.; Krishnamurthy, D.; Grover, P.; Fang, K.;
Senanayake, C. H. J . Am. Chem. Soc. 2002, 124, 7880-7881.
(8) Davis, F. A.; Reddy, R. E.; Szewczyk, J . M. J . Org. Chem. 1995,
60, 7037-7039.
(9) (a) Enders, D.; Schubert, H.; Nu¨bling, C. Angew. Chem., Int. Ed.
Engl. 1986, 25, 1109-1110. (b) Enders, D.; Nu¨bling, C.; Schubert, H.
Liebigs Ann./ Recl 1997, 1089-1100. (c) Laschat, S.; Kunz, H. J . Org.
Chem. 1991, 56, 5883-5889.
10.1021/jo026571m CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/27/2003
J . Org. Chem. 2003, 68, 3241-3245
3241

Boezio et al.
TABLE 1. Op tim iza tion of th e Aza -Wittiga
SCHEME 1
entry
R
time (h)
conversion (%)b
1
2
3
4
5
6
Ph
Ph
Ph
Me
Bu
Bu
4
20
48c
4
4
20
9
21
>95
>95 (89)d
69
>95
a
All reactions were carried out at room temperature using a
b
stoichiometric amount of phosphine. Conversions were deter-
mined by 1H NMR on the crude reaction mixture. c Reaction was
d
carried out under refluxing toluene. Isolated yield.
SCHEME 2
the azide group reaction directly by submitting the
reaction mixture to trimethylsilyl azide without the need
to isolate the ortho ester.
For example, (R,R)-1,2-diphenylethylene-1,2-diol (10a )
is converted into orthoacyl azide 11a in 60% overall yield
when trimethylorthoformate is used and in 75% overall
yield with triisopropylorthoformate (eq 1). A small amount
of the unreacted ortho ester accounts for the rest of the
mass and always remains present even if an excess of
the reagent is used. All our attempts to push the reaction
to completion were unsuccessful. Typically, the yield for
the orthoacyl azide formation is slightly higher with
triisopropylorthoformate, but it was more convenient and
practical to use trimethylorthoformate since the residual
ortho ester byproduct was more easily separated from
the orthoacyl azide on a larger scale.
compared to that of simple alkylimines since the amide
anion 6 obtained upon nucleophilic addition may undergo
ring opening to the more stable alkoxide 7 (pKa of ∼29
for the alcohol vs ∼35 for the amine).10 Furthermore, this
ring opening would lead to an O-alkylimidate 7 that
should be quite unsusceptible to further nucleophilic
attack. Protonation and mild acid hydrolysis would lead
to the secondary amine. In this paper, we report a method
for the preparation of novel chiral orthoacylimines 5 and
study their reactivity in the presence of nucleophiles. The
optimization of the chiral auxiliary structure to maximize
the diastereoselectivities of the addition was also carried
out. Finally, an application to the synthesis of t-leucine
is also presented.
Resu lts a n d Discu ssion
Our strategy for generating orthoacylimines 5 is
outlined in Scheme 2. We envisioned that the title
compounds could be obtained from the known orthoacyl
azides 11 and aldehydes 12 through a Staudinger pro-
cess.11 Since orthoacyl azides have been prepared from
the corresponding ortho esters, albeit in relatively low
yields (30-40%),12 our first goal was to optimize the
conversion of diols 10 to orthoacyl azide 11 and to develop
the Staudinger chemistry of the iminophosphorane re-
agent derived from 11.
The synthesis of orthoacyl azide has been accomplished
by taking an ortho ester and treating it with a large
excess of trimethylsilyl azide (10 equiv) without solvent.12
These conditions were not appropriate for our system,
and after optimization of the reaction conditions, we
found that the synthesis of orthoacyl azides could be
readily accomplished by treating a stoichiometric amount
of the diol with trimethyl- or triisopropylorthoformate in
the presence of p-toluenesulfonic acid in THF. The ortho
ester is typically formed in quantitative yield but we
found that it was preferable to carry the introduction of
The subsequent conversion of azide 11a into the
orthoacylimine derived from benzaldehyde using the aza-
Wittig reaction was then investigated. Attempts to carry
the aza-Wittig reaction with triphenylphosphine led to
a low yield of the desired imine even if the reaction
mixture was stirred for 20 h at room temperature (Table
1, entries 1 and 2). However, quantitative conversion
could be achieved with triphenylphosphine, but high
temperatures and long reaction times were necessary
(Table 1, entry 3). It was clear that under these condi-
tions, the iminophosphorane was formed but was not
sufficiently reactive to react with benzaldehyde. Con-
versely, the use of trialkylphosphines led to the rapid
quantitative conversion to the orthoacylimine (entries 4
and 6). The reaction with trimethylphosphine led to
quantitative formation of the imine within 4 h, whereas
that with tributylphosphine was complete after 20 h. The
orthoacylimine 5a was chromatographically stable (buff-
ered with Et3N) and could be isolated in high yield (89%).
In all cases, only one isomer of the imine was observed.
These reaction conditions were then applied to a wide
range of different diols and aldehydes, and the results
are shown in Table 2. Diols 10a -f were converted to their
corresponding orthoacyl azide 11 in yields ranging from
(10) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
(11) Stau¨dinger, H.; Meyer, J . Helv. Chem. Acta 1919, 2, 635-646.
(12) Hartmann, W.; Heine, H.-G. Tetrahedron Lett. 1979, 20, 513-
516.
3242 J . Org. Chem., Vol. 68, No. 8, 2003

Orthoacylimines: A New Class of Chiral Auxiliary
TABLE 2. Syn th esis of Or th oa cylim in e 5
TABLE 3. Op tim iza tion of th e Nu cleop h ilic Ad d ition
entry
5
MeM, solventa
drb
1
2
3
4
5
6
7
8
9
a
a
a
a
a
a
f
g
h
i
MeLi, toluene
MeLi, ether
MeLi, THF
52:48
75:25
79:21
91:9
nrc
MeLi, DME
MeMgBr, THF
Me2CuLi, ether
MeLi, DME
MeLi, DME
MeLi, DME
MeLi, DME
MeLi, DME
nr
74:26
60:40
81:19
dec
10
11
j
92:8
a
b
Unless otherwise noted, the conversions were >90%. Dias-
tereoselectivities were estimated by measuring the ee of the amine
obtained after cleavage (HCl, MeOH). c nr) no reaction. dec )
d
decomposition. 5f ) imine derived from 1,3-diphenylpropane-1,3-
diol.
a
Yield increases to 75% when HC(Oi-Pr)3 is used instead of
reoselectivities in this system were then optimized using
the orthoacylimine 5a derived from 1,2-diphenylethane-
1,2-diol (entries 1-4). It was noticed that the solvent
played a profound role, causing a high level of diaste-
reocontrol. The use of strong complexing solvents led to
higher diastereoselectivities, and DME was identified as
the optimal and most practical solvent among the most
complexing ones.15,16 With these conditions established,
several orthoacylimines derived from structurally similar
diols were tested (entries 7-11). Replacement of the
phenyl substituent by several other groups such as
naphthyl or benzyloxymethyl led to lower diastereose-
lectivities (entries 7 and 9). The only slight improvement
could be observed when 1,3-diphenylpropane-1,3-diol was
used as the chiral auxiliary precursor. Although the
selectivities were similar, we opted to use the simpler
auxiliary derived from 1,2-diphenylethane-1,2-diol to
define the synthetic scope of the reaction.
Several imines were prepared and tested with MeLi,
BuLi, and PhLi in DME (Table 4). Four classes of imines
were suitable for nucleophilic addition chemistry: the
arylimines, the R,â-unsaturated imines, heteroarylimines,
and alkylimines without enolizable protons. In most
cases, the yields were quite good and the selectivities
ranged from 85:15 to 99:1. The yields were also quite good
even with the imine derived from pivaldehyde (entries
6-9). In one case, the diastereoselectivities were out-
standing (entry 8), but the yield decreased to 54%. All
our attempts to improve the yield by increasing the
reaction temperature led to a slightly lower diastereose-
lectivities (entry 9). It should also be pointed out that in
all cases, the chiral diol could be recovered in high yield
(>95%) after mild acid hydrolysis.
b
HC(OMe)3. Conversion is >90%, but the product decomposed
upon purification. c 10f ) 1,3-Diphenylpropane-1,3-diol.
57 to 68% when trimethylorthoformate was used. The
reactions were very clean in all cases, and the intermedi-
ate ortho ester was the only byproduct. All the reaction
products were chromatographically stable and could be
separated from the residual ortho ester by chromatog-
raphy. The orthoacyl azides (11a -f) thus obtained were
then treated with several aldehydes in the presence of
trimethylphosphine to give the corresponding orthoa-
cylimines in high yields. The only exception was with the
imine derived from isobutyraldehyde for which a quan-
titative conversion was observed, but our attempts to
isolate it by chromatography on silica gel were not
successful due to the instability of the product (entry 5).13
Although the acyl azide 11a reacted with aliphatic,
unbranched aldehydes in quantitative yields, the lability
of the corresponding imine also prevented its isolation
in pure form.
With the orthoacylimines in hand, the next step was
to estimate their electrophilicity when submitted to
organometallic reagents and to optimize the structure of
the chiral auxiliary to maximize the diastereoselectivi-
ties. Some selected reaction conditions are shown in Table
3. Among the wide range of nucleophiles that were tested
(organolithium, Grignard, organocerium, organozinc, and
organocopper reagents), only the organolithium reagents
produced the desired addition products in high yields
(Table 3). This is somewhat surprising on the basis of
our mechanistic picture for this reaction. It appears that
orthoacylimines 5 are not significantly more electrophilic
than other alkylimines. Allylmagnesium bromide was the
only Grignard reagent to react cleanly but with a poor
level of diastereocontrol. The unique reactivity of allyl
Grignard reagents with imines compared to other Grig-
nard reagents has been observed before.14 The diaste-
(14) Kleinman, E. F.; Volkman, R. A. In Comprehensive Organic
Synthesis: Additions to C-X π-Bonds; Heathcock, C. H., Ed.; Perga-
mon: Oxford 1991; Vol. 2, Part 2, pp 975-1006.
(15) Additives such as DMPU, TMEDA, and HMPA were also tested,
but no increase in the diastereocontrol could be observed. The addition
of Lewis acids was also not effective in improving the selectivities.
(16) Although the role of DME is not clear, it certainly decreases
the aggregation state of methyllithium. See for example: Deacon, G.
B.; Forsyth, C. M.; Scott, N. M. J . Chem. Soc., Dalton Trans. 2001,
2494-2501.
(13) Although we could directly submit the crude imine to the
nucleophilic addition conditions, only decomposition occurred upon
treatment with alkyllithium reagents. This is presumably due to the
competitive enolization reaction of the alkylimine.
J . Org. Chem, Vol. 68, No. 8, 2003 3243

Boezio et al.
TABLE 4. Scop e of th e Ad d ition of Or ga n olith iu m
Rea gen ts
SCHEME 3. Syn th esis of t-Leu cin e
trifluoroacetamide 13, followed by Sharpless oxidation18
and deprotection, afforded t-leucine 14 in 68% overall
yield and without any racemization.
In conclusion, we have studied a novel class of chiral
auxiliaries for the nucleophilic addition to imines. The
main advantage of the method is the ease of recovery of
the chiral auxiliary after the addition and the mild
cleavage conditions for liberating the amine. Further
work is in progress to optimize the auxiliary structure.
Exp er im en ta l Section
Gen er a l P r oced u r e for th e Syn th esis of Or th oa cyl
Azid e 11a -f. To a solution of diol 10a (20.0 g, 93.4 mmol,
1.00 equiv) and p-TsOH‚H2O (888 mg, 4.67 mmol, 0.05 equiv)
in THF (950 mL, 0.1M) was added trimethylorthoformate or
triisopropylorthoformate (10.7 mL, 98.0 mmol, 1.05 equiv). The
reaction mixture was stirred at room temperature for 1 h, and
trimethylsily azide (62.0 mL, 466 mmol, 5.00 equiv) was added.
The clear solution was heated under reflux for 24 h (CAU-
TION: Care should be taken when manipulating trimethylsilyl
azide due to its explosive nature and toxicity). The reaction
mixture was cooled to room temperature, transferred to a
separatory funnel, and diluted with CH2Cl2, and the reaction
was quenched with saturated aqueous NaHCO3. The layers
were separated, and the aqueous layer was washed with three
portions of CH2Cl2. The organic solution was dried over
anhydrous Na2SO4, filtered, and concentrated under reduced
pressure. The crude residue was purified by flash chromatog-
raphy to afford the pure orthoacyl azide.
a
The enantiomeric ratios were determined by SFC on chiral
b
phase (chiralsel OD or OJ ). The stereochemistry (shown) was
established by comparison with authentic material. c Reaction was
run at -50 °C.
Gen er a l P r oced u r e for t h e Syn t h esis of Or t h oa cyl-
im in e 5a -k . To a solution of orthoacyl azide 11a (1.0032 g,
3.75 mmol, 1.00 equiv) and benzaldehyde (419 µL, 4.13 mmol,
1.10 equiv) in THF (38 mL, 0.1 M) was added trimethylphos-
phine (420 µL, 4.13 mmol, 1.10 equiv). Nitrogen evolution was
observed instantly upon the addition of trimethylphosphine.
The reaction mixture was stirred at room temperature for at
least 10 h and then concentrated under reduced pressure. The
crude residue was purified by flash chromatography to give
the pure orthoacylimine.
Gen er a l P r oced u r e for th e Nu cleop h ilic Ad d ition to
Or th oa cylim in es 5a -j. To a solution of orthoacylimine 5a
(525.9 mg, 1.60 mmol, 1.00 equiv) in DME (16 mL, 0.1 M) at
-78 °C was added dropwise MeLi (1.35 M) (salt free) (1.8 mL,
2.39 mmol, 1.50 equiv) at a rate that maintained the internal
temperature below -70 °C. The reaction mixture was stirred
for 2 h at -78 °C, and then the reaction was quenched by the
slow addition of methanol (2 mL). The mixture was then
warmed to room temperature, and 10 mL of a 50% (v/v) HCl/
MeOH solution was added. The resulting mixture was stirred
for at least 10 h or until TLC analysis showed complete
consumption of orthoacylamide 6a . Silica gel (ca. 5 g) was
added, and the mixture was concentrated under reduced
pressure (dry pack column). The residual silica gel was then
introduced on top of a prepacked silica gel flash chromatog-
raphy colomn and elution (15% MeOH/CH2Cl2) afforded diol
F IGURE 2. Model consistent with the observed relative
diastereoselection (S ) DME).
The relative diastereoselection can be explained by the
transition model shown in Figure 2. We believe that the
reaction proceeds via a six-membered transition structure
in which the equatorial and more accessible lone pair of
the dioxolane oxygen atom is involved in the chelation
with the monomeric organolithium‚DME complex. The
imine nitrogen substituent adopts the pseudoaxial con-
formation due to the anomeric stabilization (oxygen lone
pairs into σ*C-N). Transfer of the methyl group to the Re
face of the most stable conformation of the imine leads
to the (R)-R-chiral amine. The absolute configuration of
all the amines obtained are consistent with this model.
Amine 9h could be converted into t-leucine using a
simple three-step process (Scheme 3).17 Protection as the
(17) Bommarius, A. S.; Schwarm, M.; Stingl, K.; Kottenhahn, M.;
Huthmacher, K.; Drauz, K. Tetrahedron: Asymmetry 1995, 6, 2851-
2888.
(18) Carlsen, P. H. J .; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J . Org. Chem. 1981, 46, 3936-3938.
3244 J . Org. Chem., Vol. 68, No. 8, 2003

Orthoacylimines: A New Class of Chiral Auxiliary
10a (342.8 mg, 100%) in the first fractions followed by the pure
amine hydrochloride 9a as a white solid (252.2 mg, 100%).
(B2), and Boehringer Ingelheim for postgraduate fel-
lowships.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for all compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by the
E. W. R. Steacie Fund, NSERC, Merck Frosst Canada,
Boehringer Ingelheim (Canada), and the Universite´ de
Montre´al. A.A.B. is grateful to NSERC (PGF B), F.C.A.R.
J O026571M
J . Org. Chem, Vol. 68, No. 8, 2003 3245