Chemoselective synthesis of ketones and ketimines by addition of organometallic reagents to secondary amides

Article abstract of DOI:10.1038/nchem.1268

The development of efficient and selective transformations is crucial in synthetic chemistry as it opens new possibilities in the total synthesis of complex molecules. Applying such reactions to the synthesis of ketones is of great importance, as this motif serves as a synthetic handle for the elaboration of numerous organic functionalities. In this context, we report a general and chemoselective method based on an activation/addition sequence on secondary amides allowing the controlled isolation of structurally diverse ketones and ketimines. The generation of a highly electrophilic imidoyl triflate intermediate was found to be pivotal in the observed exceptional functional group tolerance, allowing the facile addition of readily available Grignard and diorganozinc reagents to amides, and avoiding commonly observed over-addition or reduction side reactions. The methodology has been applied to the formal synthesis of analogues of the antineoplastic agent Bexarotene and to the rapid and efficient synthesis of unsymmetrical diketones in a one-pot procedure. Macmillan Publishers Limited. All rights reserved.

Full text of DOI:10.1038/nchem.1268

ARTICLES
PUBLISHED ONLINE: 21 FEBRUARY 2012 | DOI: 10.1038/NCHEM.1268
Chemoselective synthesis of ketones and
ketimines by addition of organometallic reagents
to secondary amides
´
William S. Bechara, Guillaume Pelletier and Andre B. Charette
*
The development of efficient and selective transformations is crucial in synthetic chemistry as it opens new possibilities in
the total synthesis of complex molecules. Applying such reactions to the synthesis of ketones is of great importance, as
this motif serves as a synthetic handle for the elaboration of numerous organic functionalities. In this context, we report
a general and chemoselective method based on an activation/addition sequence on secondary amides allowing the
controlled isolation of structurally diverse ketones and ketimines. The generation of a highly electrophilic imidoyl triflate
intermediate was found to be pivotal in the observed exceptional functional group tolerance, allowing the facile addition of
readily available Grignard and diorganozinc reagents to amides, and avoiding commonly observed over-addition or
reduction side reactions. The methodology has been applied to the formal synthesis of analogues of the antineoplastic
agent Bexarotene and to the rapid and efficient synthesis of unsymmetrical diketones in a one-pot procedure.
hemical synthesis in nature is defined by selective and redox- from a bench-stable secondary (28) amide (4) (Fig. 1c). This inter-
economical transformations that give rise to complex natural mediate may play a dual role: (i) enabling a 1,2-addition of organo-
C
products1–4. The development of synthetic processes with metallics at a faster rate than other functionalities and (ii)
such outstanding chemoselectivity remains a priority for organic preventing further addition by initially forming a stable 28 ketimine
chemists in efforts to avoid the accumulation of unnecessary (6), which unmasks the ketone moiety (3) after subsequent acidic
steps and to allow the extrusion of protecting groups in total work-up. Indeed, the lower electrophilicity of ketimines, as well as
synthesis5–8. However, developing widely applicable conditions their isomerization to an unreactive enamine moiety in basic
that selectively target a defined functional group is still challenging,
most particularly in C–C bond forming reactions9.
a
• 1,2-addition of
organometallics
The assembly of the ketone moiety by means of a C–C
bond forming reaction (Fig. 1a) is often featured as a key step in
the retrosynthetic analysis of natural and pharmaceutically relevant
molecules10–14. Accordingly, there are abundant examples of ketones
synthesized by direct 1,2-addition of organometallic reagents onto
electrophilic carboxylic acid derivatives. Unfortunately, these
nucleophilic substitutions suffer from a number of limitations,
including precarious further additions, poor chemoselectivity,
excessive use of acylating reagents or tedious procedures15,16. To
circumvent these problems, the successful and highly useful trans-
formation of N,O-dimethylhydroxyamides (1) to a variety of
functionalized ketones (3) following reaction with organolithium
O
X = Cl, OCOR,
• Friedel-Crafts
• Cross-coupling
SR, SeR, OPyr
R1
X
• Carbonylative coupling
Z
Z = Cl, Br, I, OTf
R1
b
Weinreb
MXn
O
O
i. R2Li
ii. H3O+
O
OMe
OMe
R1
N
R1
or
R2MgX
N
R1
R2
Me
R2
Me
1
2
3
or Grignard reagents was developed by Weinreb (Fig. 1b)17–19
.
R1 is restricted to non-electrophilic substituents
The efficiency of this operation is attributed to the exceptional
stability of the five-membered-ring chelate intermediate (2).
Despite the utility and generality of the Weinreb ketone syn-
thesis20,21, no comprehensive chemoselectivity study has yet been
reported for this process22,23. Interestingly, as pointed out in
the pioneering work of Weinreb17 and others23, the addition of
Grignard reagents to N,O-dimethylhydroxyamides containing an
ester functionality is not trivial, because it results in polyalkylated
adducts. Ideally, one would like to obtain a variety of ketones
from amides while being able to include unprotected and electro-
iii. H3O+
c
This work
O
OTf
R3
R3
R2
OTf
i. Tf2O
ii. R2mMXn
N
R3
R1
N
R1
N
Activating
agent
R1
H
H
4
5
6
R1 contains aldehydes, ketones, esters, amides
philic functional groups (such as esters, ketones, aldehydes, alkyl Figure 1 | Strategies for the conversion of carboxylic acid derivatives
halides), which are often avoided when using hard nucleophiles24,25
One chemoselective and conceptually distinct approach consists
of the in situ generation of a highly electrophilic intermediate (5)
.
to ketones. a, Synthesis of ketones by C–C bond forming reactions.
b, Conversion of Weinreb amides (1) to ketones (3). c, Sequential
activation/addition on 28 amides (4) to ketimines (6) and ketones (3).
´
´
´
´
Department of Chemistry, Universite de Montreal, PO Box 6128, Station Downtown, Montreal, Quebec, Canada H3C 3J7.
e-mail: andre.charette@umontreal.ca
*
228
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
conditions, prevents over-addition at low temperatures26,27
.
observed by 1H NMR analysis of the crude mixture. Stringent exper-
Furthermore, such a strategy would be considerably more versatile, imental precautions were unnecessary, as up to 4 equiv. of Grignard
as it would not be restricted to a certain substituent on the nitrogen reagent could be used without compromising the excellent
amide component. Considering the vast utility of structurally efficiency and chemoselectivity of the reaction (entry 9). It should
diverse amides as directing28–30 and protecting groups25, here we be noted that
a diluted dichloromethane (DCM) medium
report the development of general conditions allowing the con- (0.044 M) and rapid addition of the nucleophile are recommended
trolled conversion of an array of 28 amides (4) to ketones (3) and as they avoid homocoupling of the enamine intermediate with the
ketimines (6) with unprecedented functional group tolerance.
activated imidoyl triflate.
Standard optimized conditions (Supplementary pages S5–S14)
were applied to a large variety of secondary amides (4) with
Results and discussion
Our group has been interested in developing general methods for readily available EtMgBr for a practical set-up procedure to evaluate
the controlled reduction of 28 and 38 amides to imines, aldehydes the functional group tolerance (Table 2). High levels of chemoselec-
and amines in the presence of a variety of functional groups31,32
.
tivity were achieved with amides bearing substituents known to
Cognizant of the Lewis basicity of amides, these reductions were react with hard nucleophilic organometallic reagents (such as
enabled by chemoselective activation with triflic anhydride esters, aryl and alkyl halides, nitriles, azides, azo and ketones;
(Tf2O)33–36 and the successive addition of triethylsilane (Et3SiH) entries 1–3, 5, 7–9 and 14). Tertiary and Weinreb amides were
and/or Hantzsch ester hydride (HEH) as poor hydride donors. also tolerated due to chemoselective activation of secondary
For the reduction of 28 amides, the incorporation of 2-fluoropyri- amides after a longer reaction period with Tf2O (entries 11 and
dine (2-FPyr) as a 2-halopyridine base buffer before Tf2O activation 12)32. The concept of selective monoalkylation of a doubly activated
was found to be mandatory to achieve acceptable conversion N-aryl-N′-alkyl secondary diamide was also explored with substrate
and stability for the imidoyl triflate33. Inspired by our success, we 4m (entry 13). It is important to note that the addition of 2 equiv. of
next investigated the possibility of developing a sequential activation/ EtMgBr to activated N-aryl secondary amides at –78 8C resulted in
addition with various hard organometallic nucleophiles for the low conversion to the desired ketones. Because 4m is poorly soluble
synthesis of ketones (3). To directly address the problem of alkylat- in DCM, it required a prolonged activation time and an excess of
ing N,O-dimethylhydroxyamides in the presence of an electrophilic EtMgBr (3.5 equiv.) to selectively alkylate the activated N-alkyl
ester moiety, we decided that amide 4a bearing a methyl ester func- segment, furnishing a reasonable 60% yield of 3m. Interestingly,
tionality should be our chemoselectivity benchmark (Table 1). From starting amide 4m was the major contaminant observed by
a chemoselective standpoint, the use of hard organometallic 1H NMR in the crude mixture, indicating that partial Tf2O acti-
nucleophiles is counter-intuitive and contrasts with the choice of vation occurred. The methodology was also applied with success
Et3SiH and HEH in the reduction series, which were shown to be to heterocyclic and more complex amides with a drug-like frame-
inert towards many electrophilic functional groups32. Initial work37,38 containing Lewis basic pyridine, sulfone or pyrimidine
attempts with EtCeCl2, Et3In or EtLi as alkylating reagents resulted moieties (entries 16, 17 and 21). Enolizable aliphatic-activated
in moderate and sluggish conversion to the corresponding ketone 3a amides were tolerated, where competitive ketenimine or enamine for-
(entries 1–3, Table 1). However, Et2CuMgI and Et2Zn provided mation could have been an issue (entries 19–22). To support this
good results within 1 h of reaction (82% yield, entries 4 and 5). observation, an enantioenriched secondary amide 4v derived from
The addition of a commercially available EtMgBr solution in Et2O commercially available (S)-Roche’s ester was submitted to the reac-
was found to be more practical and provided ketone 3a in a repro- tion conditions and showed no racemization after hydrolysis (entry
ducible 87% yield after hydrolysis (entry 6). Shorter reaction times 22). Moreover, the reaction could be scaled up without significant
and maintaining the reaction at –78 8C were optimal conditions loss of efficiency (see entry 20 for a 13.5 mmol scale). Hindered
for the formation of 3a (99% yield, entry 8) as no over-alkylation amides also reacted smoothly using this two step activation/
on the ketimine intermediate nor on the ester functionality was alkylation protocol (75%, entry 23)—although successful activation
required warming of the reaction mixture to room temperature.
When the process involved more sensitive functional groups,
diorganozinc reagents were ideal alternatives and provided a
broader scope of compatible functional groups. By using the less
Table 1 | 1,2-Addition with various organometallic reagents
on activated secondary amides to form ketones.
nucleophilic Et2Zn, detrimental over-alkylation was suppressed,
in contrast to EtMgBr addition for azide, ketone, aldehyde and aryl
i. Tf2O (1.1 equiv.)
2-FPyr (1.1 equiv.)
O
O
acetate (compare entries 5, 9, 10 and 15), and the chemoselectivity
was expanded to amides carrying an aryl nitro functionality (entry 4).
After studying the functional group tolerance of the reaction
with EtMgBr and Et2Zn, we considered the reactivity of a range of
different Grignard and diorganozinc reagents by simultaneously
varying the amide N-substituent (Table 3). Indeed, the nitrogen
alkyl branch (R3) can be varied without affecting the efficiency of
the reaction as N-i-Pr, N-Bn, N-allyl and N-(þ)-dehydroabiethyl
behaved similarly to N-Cy amides. Following a subsequent
optimization with various Grignard reagents (Supplementary page
S14), we observed that the addition of more hindered nucleophiles
than EtMgBr should be performed at 0 8C or –20 8C instead of
–78 8C to obtain ideal conversions to the ketones. Using these
conditions, commercially available Grignard reagents containing
primary or secondary alkyl (entries 1–4 and 6–8), alkenyl
(entry 5) or aryl (entries 9–12) groups led to isolation of the
corresponding ketones in good to excellent yields after hydrolysis.
2-Furanylmagnesium bromide prepared from the respective
DCM (0.044 M)
–78 to 0 °C, 20 min
Cy
N
Et
H
ii. EtmMXn (x equiv.)
iii. HCl (0.5 M)/THF
65 °C, 2 h
MeO2C
MeO2C
4a
3a
Entry
EtmMXn
Equiv.
EtmMXn addition
conditions
Yield 3a (%)*
1
EtCeCl2
Et3In
EtLi
2
1
278 8C to rt, 1 h
278 8C to rt, 1 h
278 8C to rt, 1 h
278 8C to rt, 1 h
278 8C to rt, 1 h
278 8C to rt, 1 h
278 8C, 25 min
278 8C, 25 min
278 8C, 25 min
278 8C, 25 min
15
2
3
4
5
6
7
8
9
10
30
63
82
82
87
74
99
99
85
2
2
2
2
1.5
2
4
6
Et2CuMgI
Et2Zn
EtMgBr
EtMgBr
EtMgBr
EtMgBr
EtMgBr
*Yields were obtained by 1H NMR analysis of the crude mixture using Ph3CH as an internal standard.
rt, room temperature.
.
lithiated precursor with MgBr2 OEt2 also reacted well with 4a,
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
229
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
Table2 | Nucleophilic 1,2-addition with EtMgBr or Et2Zn on activated secondary amides for the preparation of
1-propanone derivatives.
i. Tf2O (1.1 equiv.), 2-FPyr (1.1 equiv.)
O
O
DCM (0.044 M), –78 to 0 °C, 20 min
R3
R1
N
R1
Et
ii. EtmMXn (2.0 equiv.), –78 °C, 25 min
iii. HCl (0.5 M)/THF, 65 °C, 2 h
H
4
3
Entry
Amide
EtmMXn
Yield 3 (%)*
Entry
Amide
EtmMXn
Yield 3 (%)*
60§
O
O
1
4a
EtMgBr
96
13
4m
EtMgBr
Cy
Cy
N
N
H
H
H
N
MeO2C
Ph
O
O
O
O
2
3
4
4b
4c
4d
EtMgBr
EtMgBr
Et2Zn
74
92
14
15
16
4n
4o
4p
EtMgBr
88
Cy
Cy
N
Cl
N
H
H
Br
O
EtMgBr
Et2Zn
55
75†
Cy
Cy
Cy
N
H
N
H
NC
O2N
OAc
O
Ph
62†
EtMgBr
EtMgBr
99‡
N
O
N
H
Cy
S
N
H
N
CF3
O
Cl
5
6
7
4e
4f
EtMgBr
Et2Zn
72
17
4q
98
80†
Cy
O
N
H
N
Cy
N
N
N
H
S
O
O
S
EtMgBr
EtMgBr
EtMgBr
72
84
72
O
Cy
N
H
Me2N
O
O
4g
18
19
20
4r
4s
4t
EtMgBr
EtMgBr
EtMgBr
95
84
Cy
Cy
N
Ph
N
H
H
EtO2C
O
O
i-Pr
8
9
4h
4i
Cy
OMe
N
Ph
N
H
H
N
Ph
N
O
O
EtMgBr
Et2Zn
61
87
79†
84ꢀ
Cy
O
Cy
N
N
H
H
Cl
O
O
O
10
4j
EtMgBr
Et2Zn
6
21
4u
EtMgBr
96
80†
Cy
Cy
N
H
N
H
H
N
N
F3C
O
N
O
O
11
4k
4l
EtMgBr
EtMgBr
94‡
22
23
4v
EtMgBr
EtMgBr
94}
Cy
Bn
Et
N
TBDPSO
N
H
H
N
Et
O
O
O
12
87‡
4w
75#
Cy
Cy
Me
N
N
H
H
N
MeO
O
*Isolated yields on 1 mmol scale. †Et2Zn (1.5 equiv.) added at 0 8C and stirred for 2 h at room temperature. ‡Amide activated at 278 8C for 1 h, then 220 8C for 1 h and 0 8C for 10 min. §i. Tf2O (2.2 equiv.), 2-FPyr
(2.2 equiv.), 278 8C to room temperature for 40 min. ii. EtMgBr (3.5 equiv.), 278 8C, 25 min. ꢀIsolated yield on 13.5 mmol scale. }Amide 4v and ketone 3v were found to be .93% e.e. #Amide activated at
278 8C for 10 min, then room temperature for 10 min.
230
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
Table 3 | Addition of various Grignard and diorganozinc reagents to N-alkyl branched amides
i .Tf2O (1.1 equiv.), 2-FPyr (1.1 equiv.)
O
O
DCM (0.044 M), –78 to 0 °C, 20 min
R3
R1
N
R1
R2
ii. R2mMXn (2.0 equiv.), 0 °C, 25 min
iii. HCl (0.5 M)/THF, 65 °C, 2 h
H
4
3
Entry
Amide
R2mMXn
Yield 3 (%)* Entry
Amide
R2mMXn
Yield 3 (%)*
O
O
1
MgCl
4x
4y
86†
8
4a
4a
4a
75
8
Cy
Cy
Cy
N
H
N
H
MgBr
MeO2C
MeO2C
MeO2C
MeO2C
i-Pr
O
O
89‡
MgCl
2
94
9
N
H
MgBr
H
O
MeO
10
82‡
N
H
MgBr
N
H
MeO2C
MeO2C
MeO2C
MeO2C
MeO2C
O
Ph
O
O
O
O
F
3
4
5
6
4z
4aa
4aa
4a
90
90
87
92
11
4a
4a
4a
4k
99‡
MgCl
MgBr
Cy
Cy
Cy
N
Ph
N
H
H
MgBr
MgBr
MeO2C
MeO2C
MeO2C
O
O
Cl
12
13
14
98‡
N
N
H
H
68‡,§
MgBr
O
MgBr
N
H
N
H
O
NC
70‡,§
Ph
MgCl
Cy
Cy
• LiCl
MgCl
N
Et
N
N
H
H
Et
O
O
O
7
4a
92
15
4j
80‡,ꢀ
Cy
Cy
MeO
Zn
MgCl
N
H
N
H
2
H
MeO2C
O
*Isolated yields on 1 mmol scale. †n-decylMgCl was added at 220 8C and stirred for 25 min. ‡Hydrolysis performed for 3 h. §3.0 equiv. of Grignard reagent added. ꢀ(PMP)2Zn (1.5 equiv.) was added at 0 8C and
stirred for 2 h at room temperature.
affording a 68% yield (entry 13). An example of a functionalized generation of ligands40 and a-chiral amine species41. Although
Grignard reagent prepared from 4-bromobenzonitrile and enolizable ketimines can be difficult to separate by chromatography
39
i-PrMgCl LiCl according to Knochel’s procedure was added because of their fast hydrolysis under acidic conditions26,
.
successfully to activated amide 4k (70% yield, entry 14), thus sufficiently stable ketimines were isolated after an aqueous basic
demonstrating the high chemoselectivity of the reaction on both work-up with Na2CO3 (Table 4). In the case of the ketimine
reaction components. Moreover, the addition of diarylzinc species derived from amide 4ac, a mixture of stereoisomers (18:1, E:Z)
on an aldehyde-containing amide (4j) provided the desired ketone was obtained in 84% combined yield (entry 2). Moreover, it is
in good yield without further addition on the aldehyde moiety.
possible to isolate the amine branch following an acid/base work-
We were also interested in controlling the isolation outcome to up after hydrolysis of the corresponding ketimine (see entry 1 and
ketimines, given that they are renowned building blocks for the Supplementary page S69).
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.
231

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
To contrast the synthetic utility of our methodology with estab-
lished ketone syntheses, the formal synthesis of analogues (7–9) of
the antineoplastic agent Bexarotene was explored (Fig. 2a)42. The
tetrahydronaphthalene (7) and dihydronaphthalenes (8,9) have
been identified by Eli Lilly pharmaceuticals as promising retinoid-
X receptors (RXR) agonists for the treatment of non-insulin depen-
dent diabetes (type II) mellitus (NIDDM)42,43. Initial trials by Eli
Lilly were performed using the acyl chloride 11 and Weinreb
amide 12 containing an ester functionality affording the desired
ketone 10 with low yields due to over-addition on the ester
moiety with the lithium reagent (13a) or rapid oxidation of the
Grignard nucleophile (13b) (Fig. 2b, entries 1–3)42. To gauge the
chemoselectivity of our methodology with the previous results,
the addition of the Grignard reagent 13b to the activated secondary
amide 4a was performed and provided the desired ketone 10 with an
excellent 95% yield without addition to the ester component
(Fig. 2b, entry 4).
Ongoing efforts have been directed towards the synthesis of
unsymmetrical diketones44,45. In this regard, the Weinreb ketone
synthesis allows a direct access to these units by successive additions
with two different organometallic nucleophiles in a one-pot pro-
cedure46,47. Unfortunately, this operation was shown to generally
work with C2-symmetric substrates and is not chemoselective for
a particular amide19. To address this problem, the combination of
Weinreb amides and secondary amides was explored with a view
to achieving the one-pot synthesis of unsymmetrical diketones
(Fig. 2c). We hypothesized that a double addition could be
Table 4 | Secondary ketimine isolation after basic work-up.
i. Tf2O (1.1 equiv.), 2-FPyr (1.1 equiv.)
O
R3
R2
DCM (0.044 M), –78 to 0 °C, 20 min
N
R3
R1
N
ii. R2MgBr (2.0 equiv.), 0 °C, 25 min
iii. aq. Na2CO3
R1
H
4
6
Entry
Amide
R2MgBr
Yield 6 (%)*
82† 6a
O
1
4ab
OMe
OMe
Ph
N
H
MgBr
O
2
3
4ac
4a
84‡ 6b
Cy
N
MgBr
H
MeO
O
OTBS
88§ 6c
Cy
N
H
MeO2C
MgBr
*Isolated yields on 1 mmol scale. †Imine 6a was also hydrolysed in acidic conditions to recuperate the
amine in 85% yield after an acid/base work-up. ‡Isolated as a mixture of stereoisomers (18:1 E:Z
ratio). §The alkynyl Grignard reagent was added at 0 8C and stirred for 40 min.
a
O
X
O
R
MX
CO2Me
CO2H
CO2Me
Acylating agent
(11,12, or 4a)
X = cis-NOH 7 (tetrahydronaphthalene)
10
MX = Li
MgBr
13a
13b
cis-NOH
8 (dihydronaphthalene)
9 (dihydronaphthalene)
CH2
b
Acylating agent
(R =)
Yield 10
Entry
MX
*
(%)
ND†
53
1
2
3
Cl
Cl
11
Li
MgBr
Li
13a
13b
13a
11
12
4a
CO2H
MeO(Me)N
16
95‡
Bexarotene
4
(Cy)HN
MgBr
13b
c
O
O
Cy
One-pot reaction
Me
N
Ph
H
N
Et
MeO
O
O
14
97%
4l
i. Tf2O (1.1 equiv.), 2-FPyr (1.1 equiv.)
DCM (0.044 M), –78 to 0 °C, 20 min
PhMgBr (2.0 equiv.), 0 °C, 25 min
iii. HCl (0.5 M)/THF
65 °C, 2 h
Cy
Cy
N
N
ii. EtMgBr (2.0 equiv.)
rt, 1 h
Me
Ph
Me
N
Ph
Et
N
MeO
MeO
O
O
BrMg
Figure 2 | Applications of the chemoselective secondary amides transformation to ketones. a, Formal synthesis of Bexarotene analogues 7–9 with different
1,2-addition strategies. b, Comparison of results obtained by Eli Lilly using acyl chloride 11 and Weinreb amide 12 with the present methodology. *Isolated
yields. †Unidentified multiple products obtained. ‡Using reaction conditions illustrated in Table 3 on 1 mmol scale with 3.0 equiv. of Grignard reagent and
performing the hydrolysis for 5 h. c, One-pot synthesis of an unsymmetrical diketone. Diamide 4l is converted to diketone 14 by successive addition of two
organomagnesium reagents followed by hydrolysis.
232
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
The aqueous layer was extracted with DCM (2×) and the organic layers were
combined. The organic layer was dried over anhydrous Na2SO4, filtered over a
sintered funnel and evaporated to dryness. The ketones were purified by flash
chromatography over silica gel (IscoTeledyne RediSep Rf Gold 24g) using a gradient
of 0–80% EtOAc/hexanes. Caution: manipulation of neat diorganozinc reagents
should be performed with care with the exclusion of moisture and air in flasks,
syringe and needles.
performed by the successive addition of a Grignard reagent to the
activated amide at a low temperature followed by addition of the
second nucleophile at room temperature. This strategy was envi-
sioned after observing the chemoselective transformation of amide
4l to the corresponding ketone 3l in the presence of a Weinreb
amide moiety (Table 2, entry 12). Using this approach, ketone 14
was synthesized in 97% isolated yield by the successive addition
of PhMgBr at 0 8C and EtMgBr at room temperature. The addition
of these two Grignard reagents can be reversed without affecting
the efficiency of the reaction (95% yield for 14; Supplementary
pages S76–S77).
Received 5 October 2011; accepted 9 January 2012;
published online 21 February 2012
References
1. Clardy, J. & Walsh, C. Lessons from natural molecules. Nature 432,
829–837 (2004).
Summary
2. Chau, M., Jennewein, S., Walker, K. & Croteau, R. Taxol biosynthesis: molecular
cloning and characterization of a cytochrome P450 taxoid 7b-hydroxylase.
Chem. Biol. 11, 663–672 (2004).
3. Hubbard, B. K. & Walsh, C. T. Vancomycin assembly: nature’s way. Angew.
Chem. Int. Ed. 42, 730–765 (2003).
4. Schwecke, T. et al. The biosynthetic gene cluster for the polyketide
immunosuppressant rapamycin. Proc. Natl Acad. Sci. USA 92, 7839–7843 (1995).
5. Trost, B. M. Selectivity: a key to synthetic efficiency. Science 219, 245–250 (1983).
6. Burns, N. Z., Baran, P. S. & Hoffmann, R. W. Redox economy in organic
synthesis. Angew. Chem. Int. Ed. 48, 2854–2867 (2009).
The exceptional chemoselectivity feature studied in the present
methodology allows the controlled conversion of secondary
amides to structurally diverse ketones and ketimines. A sequential
activation/addition with Tf2O is efficiently performed by the
addition of readily available Grignard and diorganozinc reagents.
The success of this approach is attributed to the generation of a
very reactive intermediate, which reacts in a 1,2-addition at a
faster rate and permits control of over-addition by the formation
of a ketimine intermediate. The method obviates the need to
prepare a specific acylating reagent to synthesize ketones. Our
results therefore showcase the first comprehensive study to take
advantage of the unique chemoselectivity of the transformation.
Other applications illustrating the synthetic potential of this meth-
odology have been disclosed through the synthesis of Bexarotene
analogues and unsymmetrical diketones. The development of this
protocol should allow the late-stage and diversified functionaliza-
tion of secondary amides in total synthesis while avoiding the use
of unnecessary protecting groups. Efforts are presently focused on
the application of this concept to tertiary amides and peptide
derivatives. The results will be reported in due course.
7. Young, I. S. & Baran, P. S. Protecting-group-free synthesis as an opportunity for
invention. Nature Chem. 1, 193–205 (2009).
8. Trost, B. M. & Dong, G. B. Total synthesis of bryostatin 16 using atom-
economical and chemoselective approaches. Nature 456, 485–488 (2008).
9. Afagh, N. A. & Yudin, A. K. Chemoselectivity and the curious reactivity
preferences of functional groups. Angew. Chem. Int. Ed. 49, 262–310 (2010).
10. Huckins, J. R., de Vicente, J. & Rychnovsky, S. D. Synthesis of the C1–C52
fragment of amphidinol 3, featuring a b-alkoxy alkyllithium addition reaction.
Org. Lett. 9, 4757–4760 (2007).
11. Chen, C.-y. et al. Catalytic, enantioselective synthesis of taranabant, a novel,
acyclic cannabinoid-1 receptor inverse agonist for the treatment of obesity.
Org. Proc. Res. Dev. 11, 616–623 (2007).
12. Olah, G. A. Friedel–Crafts and Related Reactions Vol. 1, Ch. 11 (Interscience,
1963–1965).
13. Milstein, D. & Stille, J. K. Mild, selective, general method of ketone synthesis
from acyl chlorides and organotin compounds catalyzed by palladium. J. Org.
Chem. 44, 1613–1618 (1979).
Methods
14. Brunet, J.-J. & Chauvin, R. Synthesis of diarylketones through carbonylative
coupling. Chem. Soc. Rev. 24, 89–95 (1995).
General procedure for the synthesis of ketones with Grignard reagents on a
1 mmol scale. To a flame-dried and argon-flushed 50 ml round-bottom flask
equipped with a stir-bar and a septum was added the amide (1.0 mmol, 1.0 equiv.).
The amide was diluted with anhydrous DCM (22.5 ml, 0.044 M), and
2-fluoropyridine (107.0 mg, 95 ml, 1.1 mmol, 1.1 equiv.) was added to the solution.
The solution was then cooled to –78 8C using an acetone/dry-ice cooling bath and
stirred for 2 min. Tf2O (310.1 mg, 185 ml, 1.1 mmol, 1.1 equiv.) was added dropwise
using a syringe at –78 8C and the reaction was stirred for 10 min. The solution was
warmed to 0 8C using a water/ice bath and the reaction was stirred for 10 min.
A solution of Grignard reagent in Et2O or THF (2.0 mmol, 2.0 equiv.) was added
in one portion to the reaction at –78 8C (or 0 8C), and the reaction was stirred for
25 min at –78 8C (or 0 8C). The reaction was quenched by the addition of 8 ml of an
aqueous solution of HCl (0.5 M) and 8 ml THF. The biphasic mixture was gently
heated to 65 8C (keeping the flask ventilated for DCM evaporation) for 2 h to ensure
complete hydrolysis to the ketone. The biphasic mixture was then transferred to a
60 ml separation funnel using DCM and the layers were separated. The aqueous
layer was extracted with DCM (2×) and the organic layers were combined.
The organic layer was dried over anhydrous Na2SO4, filtered over a sintered
funnel and evaporated to dryness. The ketones were purified by flash
chromatography over silica gel (IscoTeledyne RediSep Rf Gold 24g) using a
gradient of 0–80% EtOAc/hexanes.
15. Dieter, R. K. Reaction of acyl chlorides with organometallic reagents: a banquet
table of metals for ketone synthesis. Tetrahedron 55, 4177–4236 (1999).
16. Katritzky, A. R., Le, K. N. B., Khelashvili, L. & Mohapatra, P. P. Alkyl,
unsaturated, (hetero)aryl, and N-protected a-amino ketones by acylation of
organometallic reagents. J. Org. Chem. 71, 9861–9864 (2006).
17. Nahm, S. & Weinreb, S. M. N-Methoxy-N-methylamides as effective acylating
agents. Tetrahedron Lett. 22, 3815–3818 (1981).
18. Sibi, M. P. Chemistry of N-methoxy-N-methylamides. Applications in synthesis.
A review. Org. Prep. Proced. Int. 25, 15–40 (1993).
19. Balasubramaniam, S. & Aiden, I. S. The growing synthetic utility of the Weinreb
amide. Synthesis, 3707–3738 (2008).
20. Sengupta, S., Mondal, S. & Das, D. Amino acid derived morpholine amides for
nucleophilic a-amino acylation reactions: a new synthetic route to enantiopure
a-amino ketones. Tetrahedron Lett. 40, 4107–4110 (1999).
21. Comins, D. L. & Brown, J. D. Directed lithiation of tertiary b-amino benzamides.
J. Org. Chem. 51, 3566–3572 (1986).
22. Murphy, J. A. et al. Direct conversion of N-methoxy-N-methylamides (Weinreb
amides) to ketones via a nonclassical Wittig reaction. Org. Lett. 7,
1427–1429 (2005).
23. Calosso, M. et al. Enantioselective synthesis of 2,3-disubstituted piperidines.
Lett. Org. Chem. 4, 4–6 (2007).
24. Comins, D. L. The synthetic utility of a-amino alkoxides. Synlett 615–625 (1992).
25. Wuts, P. G. M. & Greene, T. W. Greene’s Protective Groups in Organic Synthesis
Ch. 4, Ch. 7 (Wiley, 2007).
General procedure for the synthesis of ketones with diorganozinc reagents on
a 1 mmol scale. To a flame-dried and argon-flushed 50 ml round-bottom flask
equipped with a stir-bar and a septum was added the amide (1.0 mmol, 1.0 equiv.).
The amide was diluted with anhydrous DCM (22.5 ml, 0.044 M), and
2-fluoropyridine (107.0 mg, 95 ml, 1.1 mmol, 1.1 equiv.) was added to the solution.
The solution was then cooled to –78 8C using an acetone/dry-ice cooling bath and
stirred for 2 min. Tf2O (310.1 mg, 185 ml, 1.1 mmol, 1.1 equiv) was added dropwise
using a syringe at –78 8C and the reaction was stirred for 10 min. The solution was
warmed to 0 8C using a water/ice bath and the reaction stirred for 10 min. Neat
diorganozinc reagent (1.5 mmol, 1.5 equiv.) was added in one portion to the
reaction at 0 8C. The reaction was warmed to room temperature and the reaction was
stirred for 2 h at room temperature. The reaction was quenched by the addition of
8 ml of an aqueous solution of HCl (0.5 N) and 8 ml THF. The biphasic mixture was
gently warmed to 65 8C (keeping the flask ventilated for DCM evaporation) for 2 h
to ensure complete hydrolysis to the ketone. The biphasic mixture was then
transferred to a 60 ml separation funnel using DCM and the layers separated.
26. Steinig, A. G. & Spero, D. M. Amines via nucleophilic 1,2-addition to ketimines.
Construction of nitrogen-substituted quaternary carbon atoms. A review. Org.
Prep. Proced. Int. 32, 205–234 (2000).
27. Reingruber, R. & Bra¨se, S. 1,2-Addition of trialkylaluminium reagents to N-
diphenylphosphinoyl-ketimines in the absence of any additional reagents.
Chem. Commun. 105–107 (2008).
28. Chen, Q., Ilies, L. & Nakamura, E. Cobalt-catalyzed ortho-alkylation of
secondary benzamide with alkyl chloride through directed C–H bond activation.
J. Am. Chem. Soc. 133, 428–429 (2011).
29. Yoo, E. J., Ma, S., Mei, T.-S., Chan, K. S. L. & Yu, J. Q. Pd-catalyzed
intermolecular C–H amination with alkylamines. J. Am. Chem. Soc. 133,
7652–7655 (2011).
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
233
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1268
ARTICLES
30. Marcoux, D. & Charette, A. B. Trans-directing ability of amide groups in
cyclopropanation: application to the asymmetric cyclopropanation of alkenes
with diazo reagents bearing two carboxy groups. Angew. Chem. Int. Ed. 47,
10155–10158 (2008).
31. Barbe, G. & Charette, A. B. Highly chemoselective metal-free reduction of
tertiary amides. J. Am. Chem. Soc. 130, 18–19 (2008).
32. Pelletier, G., Bechara, W. S. & Charette, A. B. Controlled and chemoselective
reduction of secondary amides. J. Am. Chem. Soc. 132, 12817–12819 (2010).
33. Medley, J. W. & Movassaghi, M. Direct dehydrative N-pyridinylation of amides.
J. Org. Chem. 74, 1341–1344 (2009).
34. Myers, A. G., Tom, N. J., Fraley, M. E., Cohen, S. B. & Madar, D. J. A convergent
synthetic route to (þ)-dynemicin A and analogs of wide structural variability.
J. Am. Chem. Soc. 119, 6072–6094 (1997).
43. Canan Koch, S. S. et al. Synthesis of retinoid X recetor-specific ligands that
are potent inducers of adipogenesis in 3T4-LI cells. J. Med. Chem. 42,
742–750 (1999).
44. Hudlicky, T. & Price, J. D. Anionic approaches to the construction of
cyclopentanoids. Chem. Rev. 89, 1467–1486 (1989).
¨
˙
45. Ozkay, Y., I¸sikdag˘, I., Incesu, Z. & Akalin, G. Synthesis of 2-substituted-N-[4-
(1-methyl-4,5-diphenyl-1H-imidazole-2-yl) phenyl]acetamide derivatives and
evaluation of their anticancer activity. Eur. J. Med. Chem. 45, 3320–3328 (2010).
46. Prasad, K. R. & Penchalaiah, K. Total synthesis of (–)-anamarine. J. Org. Chem.
76, 6889–6893 (2011).
47. Sibi, M. P., Sharma, R. & Paulson, K. L. N,N′-Dimethoxy-N,N′-
dimethylethanediamide: a useful a-oxo-N-methoxy-N-methylamide and
1,2-diketone synthon. Tetrahedron Lett. 33, 1941–1944 (1992).
35. Movassaghi, M. & Hill, M. D. Single-step synthesis of pyrimidine derivatives.
J. Am. Chem. Soc. 128, 14254–14255 (2006).
Acknowledgements
36. Charette, A. B. & Grenon, M. Spectroscopic studies of the electrophilic activation of
amides with triflic anhydride and pyridine. Can. J. Chem. 79, 1694–1703 (2001).
37. Bartoli, J. et al. New azole antifungals. 2. Synthesis and antifungal activity of
heterocyclecarboxamide derivatives of 3-amino-2-aryl-1-azolyl-2-butanol.
J. Med. Chem. 41, 1855–1868 (1998).
This work was supported by the Natural Science and Engineering Research Council of
Canada (NSERC), the Canada Research Chair Program, the Canada Foundation for
Innovation, the FQRNT Centre in Green Chemistry and Catalysis (CGCC) and the
´
´
´
Universite de Montreal. W.S.B. and G.P. are grateful to NSERC, FQRNT and the Universite
´
de Montreal for postgraduate scholarships. The authors would like to thank A. Lemire and
38. Jamieson, C., Miller, D. D., Rami, H. K. & Thompson, M. Preparation of
piperidinecarboxamide and cyclohexanecarboxamide derivatives as vanilloid
receptor modulators. International patent WO2005016915 (24 February 2005).
39. Krasovskiy, A. & Knochel, P. A LiCl-mediated Br/Mg exchange reaction for the
preparation of functionalized aryl- and heteroarylmagnesium compounds from
organic bromides. Angew. Chem. Int. Ed. 43, 3333–3336 (2004).
40. Small, B. L., Brookhart, M. & Bennett, A. M. A. Highly active iron and cobalt
catalysts for the polymerization of ethylene. J. Am. Chem. Soc. 120,
4049–4050 (1998).
´
´
´
P. Lavallee (Universite de Montreal), P. Lapointe (IRIC) and M. Grenon (Lundbeck) for
supplying starting materials and reagents.
Author contributions
W.S.B. and G.P. carried out the experimental work, organized the research, analysed data
and wrote the manuscript. All the authors contributed to the design of the experiments and
editing of the manuscript.
41. Charette, A. B. Chiral Amine Synthesis: Methods, Developments and Applications Additional information
Ch. 1 (Wiley, 2010).
The authors declare no competing financial interests. Supplementary information and
42. Faul, M. M., Ratz, A. M., Sullivan, K. A., Trankle, W. G. & Winneroski, L. L.
Synthesis of novel retinoid X receptor-selective retinoids. J. Org. Chem. 66,
5772–5782 (2001).
chemical compound information accompany this paper at www.nature.com/
naturechemistry. Reprints and permission information is available online at http://www.
nature.com/reprints. Correspondence and requests for materials should be addressed to A.B.C.
234
NATURE CHEMISTRY | VOL 4 | MARCH 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.