One-pot Unsymmetrical Ketone Synthesis Employing a Pyrrole-Bearing Formal Carbonyl Dication Linchpin Reagent

Article abstract of DOI:10.1002/anie.201502894

A one-pot procedure for the synthesis of unsymmetrical ketones utilizing a pyrrole-bearing carbonyl linchpin reagent (carbonyl linchpin N,O-dimethylhydroxylamine pyrrole; CLAmP) is reported. In contrast to other carbonyl dielectrophile equivalents, CLAmP enables the synthesis of ketones from a variety of organolithium and Grignard reagents. The electrophilic nature of CLAmP enables the addition of less reactive as well as thermally unstable nucleophiles. CLAmP was designed to form kinetically stable tetrahedral intermediates upon the addition of organometallic nucleophiles. Evidence for the existence of persistent tetrahedral intermediates was obtained through in situ IR studies.

Full text of DOI:10.1002/anie.201502894

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201502894
German Edition: DOI: 10.1002/ange.201502894
Acylation
One-pot Unsymmetrical Ketone Synthesis Employing a Pyrrole-
Bearing Formal Carbonyl Dication Linchpin Reagent**
Stephen T. Heller, James N. Newton, Tingting Fu, and Richmond Sarpong*
Abstract: A one-pot procedure for the synthesis of unsym-
metrical ketones utilizing a pyrrole-bearing carbonyl linchpin
reagent (carbonyl linchpin N,O-dimethylhydroxylamine pyr-
role; CLAmP) is reported. In contrast to other carbonyl
dielectrophile equivalents, CLAmP enables the synthesis of
ketones from a variety of organolithium and Grignard
reagents. The electrophilic nature of CLAmP enables the
addition of less reactive as well as thermally unstable
nucleophiles. CLAmP was designed to form kinetically stable
tetrahedral intermediates upon the addition of organometallic
nucleophiles. Evidence for the existence of persistent tetrahe-
dral intermediates was obtained through in situ IR studies.
U
nsymmetrical ketones are ubiquitous building blocks in
chemical synthesis. Retrosynthetically, the ketone carbonyl
may be thought of as one of three possible synthons—the
carbonyl dication, dianion, or carbene (i.e., carbon monox-
ide)—and could serve as a linchpin upon which two carbon
ligands can be appended (Scheme 1). Sequential attachment
of carbon ligands through multiple steps is the most popular
approach to ketone synthesis. However, the ketone linkage
can conceivably be forged with a single synthetic operation (a
one-pot synthesis) by combining the two carbon ligands with
a synthetic equivalent of one of the above synthons. The
preparation of unsymmetrical ketones by this approach would
be synthetically empowering, but is particularly challenging as
the carbonyl synthetic equivalent must exhibit differential
reactivity toward the two carbon ligands.
Scheme 1. Retrosynthetic analysis of the ketone group: Synthons and
synthetic equivalents.
practice by the groups of Tietze[2] and Smith[3] by using the
sequential alkylation of dithianes such as 1 (Scheme 1).[4]
Surprisingly, a carbonyl dication equivalent, which enables
a general synthesis of unsymmetrical ketones, has not been
developed,[5] though several groups have reported tactics and
reagents with a very narrow scope. Herein we report that N-
methoxy-N-methyl-1H-pyrrole-1-carboxamide (2; CLAmP)
is a versatile electrophilic carbonyl linchpin reagent for the
synthesis of unsymmetrical ketones.
In order to effect the selective formation of unsym-
metrical ketones, an electrophilic carbonyl linchpin must
feature two nucleofuges, each of which generate differentially
stable tetrahedral intermediates upon sequential monoaddi-
tion of organometallic nucleophiles. Specifically, initial reac-
tion of a carbonyl dication synthon (e.g., 2) with the first
nucleophile equivalent should afford a tetrahedral intermedi-
ate that persists throughout the addition but whose rate of
collapse can be modulated by experimental conditions.
Controlled collapse of the first tetrahedral intermediate
generates a new carbonyl electrophile (acylium ion synthon)
that is available for reaction with a second organometallic
nucleophile. Ideally, persistence of the tetrahedral intermedi-
ate derived from the second addition prevents overalkylation
at the electrophilic carbon, preserving the ketone oxidation
state.
Of the three possible carbonyl synthons, only two have
been successfully employed in one-pot syntheses of unsym-
metrical ketones. Transition-metal-catalyzed carbonylative
cross-coupling is essentially a one-pot ketone synthesis
involving carbon monoxide as the carbonyl linchpin.[1] The
carbonyl dianion synthetic equivalent, has been reduced to
[*] J. N. Newton, T. Fu, R. Sarpong
Department of Chemistry, University of California
Berkeley, CA 94720 (USA)
E-mail: rsarpong@berkeley.edu
S. T. Heller
Department of Chemistry, Willamette University
Salem, OR 97301 (USA)
[**] We are grateful to the Toste group (UC Berkeley) for the use of their
ReactIR instrument and to Prof. Jason Hein (UC Merced) for helpful
discussions on the acquisition of ReactIR data. We thank the
National Science Foundation for Predoctoral Fellowships for S.T.H.
and J.N.N., and the McNair Scholars program for an undergraduate
fellowship for T.F. This work was supported by a CAREER Award
(0643264) to R.S. from the National Science Foundation.
Several tactics that enable controlled monoaddition of
organometallic nucleophiles to carbonyls have been reported.
The groups of Weinreb,[7] Meyers,[8] and Mukaiyama[9] each
developed carboxylic acid derivatives bearing chelating
nucleofuges that retard the collapse of the initially formed
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201502894.
Angew. Chem. Int. Ed. 2015, 54, 9839 –9843
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9839

.
Angewandte
Communications
tetrahedral intermediates. It has been shown that some
carbonyl electrophiles with poor nucleofuges (e.g., amides)
also form unusually stable tetrahedral intermediates.[10]
However, the same structural features that give rise to poor
nucleofugality typically also reduce the inherent electrophi-
licity of the carbonyl group. For instance, the ideal electro-
philic carbonyl linchpin, CO2, forms unsymmetrical ketones
from only the most reactive organolithium derivatives.[11]
Exceptions to this general trend are N-acylpyrroles, which
Evans et al. showed to form stable tetrahedral intermediates
upon reaction with either organolithium or Grignard
reagents.[12]
The groups of Reich, Hlasta, and Lee have all reported
potential ꢀdoubleꢁ electrophilic carbonyl linchpin reagents
employing chelative stabilization (3–5; Scheme 2), but the
Scheme 3. Scope of Weinreb amides synthesis from organometallic
reagents and 2. Yields within parentheses are for the isolated
compounds, with the exception of 12, the yield of which is based on
1H NMR analysis using an internal standard. THF=tetrahydrofuran.
Scheme 2. Carbonyl dication synthetic equivalents.
properties of the first departing nucleofuge have severely
limited their use. The symmetrical urea 3[13] reacts only at
a relatively high temperature with strongly nucleophilic
Grignard and organolithium reagents and does not undergo
in situ elimination, even under forcing conditions. Therefore,
it cannot be employed in a one-pot unsymmetrical ketone
synthesis. The unsymmetrical urea 4[14] reacts only with
organolithium species to yield a nearly 1:1 mixture of N,N-
dimethylamide and Weinreb amide products, the former of
which is so weakly electrophilic that only the most reactive
organolithium nucleophiles (e.g., nBuLi) may be used in the
second addition step. The carbamate 5[15] reacts unselectively
with organolithium reagents, presumably because of the
relatively high nucleofugality of the 2-pyridoxy group.
synthesis of Weinreb amides, particularly in cases involving
thermally unstable or relatively unreactive organometallic
nucleophiles.
In principle, both the type of organometallic nucleophile
as well as the order in which it is appended to the carbonyl
group of 2 could impact the outcome of the ketone synthesis.
To better understand these factors, we synthesized 4’-meth-
oxyvalerophenone (13; Table 1) using several permutations of
nucleophile type and order of addition. Unlike the use of 5,
the best results with 2 were obtained when an aryllithium was
used as the first nucleophile (entries 1 versus 2, and 3 versus
4). Unlike the use of 4, the nature of the second nucleophile
does not appear to strongly influence the overall yield of
ketones, though it appears that organolithium nucleophiles
The carbonyl dication synthetic equivalent 2 combines the
desirable nucleofugal properties of Weinreb amides and N-
acylpyrroles, thus allowing facile one-pot synthesis of unsym-
metrical ketones. This carbonyl bis-electrophile reacts rapidly
with different organometallic reagents to generate a semi-
persistent tetrahedral intermediate at À788C, and thus
permits selective monoaddition of the first nucleophile.
Upon warming to ambient temperature, the putative inter-
mediate readily collapses to exclusively generate a Weinreb
amide in situ, which can react with a suite of organometallic
nucleophiles to append the second carbon ligand.
Table 1: Impact of nucleophile type and order of addition on unsym-
metrical ketone synthesis.
Entry R1 M
R2 M
Yield [%][a] Reaction conditions[b]
We first studied the reactivity and selectivity afforded by 2
in the presence of a single organometallic species to yield
Weinreb amides (Scheme 3). In contrast to 5 and 4, relatively
non-nucleophilic lithiated picoline and dithiane species, as
well as Grignard reagents can be employed to synthesize
Weinreb amides. Although our primary goal was to identify
a modular synthesis of unsymmetrical ketones, the synthesis
of Weinreb amides is also synthetically useful. Thus, 2 should
find use as a complement to 3 for one-carbon homologative
1
2
3
4
5
6
ArLi
AlkylLi
ArylLi
ArMgBr
94
63
54
À788C, 1.5 h; À788C, 0.5 h
À788C, 1 h; À788C, 1 h
À788C, 1 h; 238C, 8 h
À788C, 1 h; 08C, 3 h
08C, 5 h; 238C, 8 h
AlkylLi
AlkylLi
ArLi
AlkylMgCl ArLi
AlkylMgCl ArMgBr
AlkylMgCl 80
36
75
238C, 6 h; 238C, 8 h
[a] Yield of isolated product. [b] Reaction conditions are reported in the
following format: conditions for first nucleophile; conditions for second
nucleophile.
9840 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9839 –9843

Angewandte
Chemie
perform somewhat better (entries 1 versus 4, and 2 versus 3).
Moreover, Grignard reagents may be used as the first
nucleophile that is added to 2, as the use of two different
Grignard reagents is well-tolerated (entry 6). There are two
important caveats: using sp2-hybridized Grignard reagents as
the first nucleophile with 2 leads to poor selectivity,[16] and the
use of organolithium species following the addition of any
Grignard reagent impairs the second addition step (entry 5).
It is possible that in these cases, the lithium reagents react
with magnesium salt byproducts to produce lithium magnesi-
ate species, which may not be viable nucleophiles. In addition,
bimetallic tetrahedral intermediates, which are less likely to
undergo collapse, may form. Salt effects of this type have
recently been discussed.[17] The former caveat can be avoided
simply by using the sp2-hybridized Grignard reagent as the
second nucleophile, while the latter is readily solved by
introducing the organolithium nucleophile first.
With a better understanding of the reactivity of 2, the
substrate scope of the one-pot unsymmetrical ketone syn-
thesis was investigated further (Table 2). A variety of carbon
nucleophiles can be utilized in the second addition step,
including dithiane anions (entry 1), lithiophosphonates
(entry 2), acetylides (entry 3), enolates (entry 4), and pico-
linyl anions (entry 5). Unlike 3, the reagent 2 reacts rapidly
with organolithium nucleophiles at low temperature, thus
enabling the use of unstable organometallic species such as (2-
fluorophenyl)lithium, which would otherwise decompose at
noncryogenic temperatures.[18] Heteroaromatic nucleophiles
are generally well tolerated (entries 7–9). Highly functional-
ized ketones can also be prepared using 2. For instance, 19,
a key intermediate in the Boehringer Ingelheim synthesis of
an NNRT inhibitor (HIV therapeutic),[19] was obtained in
good yield. In general, the unsymmetrical ketones were
formed with greater than 20:1 selectivity over the possible
Table 2: Scope of unsymmetrical ketone synthesis using 2.
Entry First
nucleophile
(R1M)
Second
Product
Yield Reaction conditions[c]
[%][a]
nucleophile
(R2M)[b]
1
2
3
4
5
14 63
15 82
16 96
17 84
18 78
À788C, 1 h; À788C, 1 h, then warm to 08C
À788C, 1 h; À788C, 1 h, then warm to 08C
À788C, 1 h; À308C!08C over 3 h
À788C, 1.5 h; À788C, 2 h
À788C, 2 h; À788C, 3 h, then warm to 238C
À788C, 1 h; À788C, 3 h, then warm to 238C, R2M added
6
7
19 67
20 60
as a solution in Et2O
À788C, 1 h; 238C, 8 h
8
21 58
22 64
23 79
À788C, 2 h; 238C, 24 h
9
À788C, 1 h; À788C, 1 h, then warm to 238C
08C, 1 h, then 238C for 24 h; 08C, 1 h, then 238C for 24 h
10
[a] Yield of isolated product. [b] Upon completion of the addition of the first nucleophile, the reaction mixture was warmed to 238C and stirred for
30 min prior to the addition of the second nucleophile. [c] Reaction conditions are reported as: conditions for first nucleophile; conditions for second
nucleophile.
Angew. Chem. Int. Ed. 2015, 54, 9839 –9843
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9841

.
Angewandte
Communications
symmetrical congeners (except when alkyllithium nucleo-
philes were added first), and only trace amounts of the
protonated organometallic nucleophiles were detected.
It also proved possible to telescope post-functionalization
chemistry onto the unsymmetrical ketone synthesis sequence
(Scheme 4). The ynone 25 is efficiently prepared in one pot
Figure 1. Persistence of lithio-tetrahedral intermediates studied by
in situ IR. A) IR Surface of the collapse of 31. B) IR Surface of the
collapse of 32. C,D) Plots of the disappearance of the tetrahedral
intermediates versus time.
Scheme 4. Applications of the CLAmP unsymmetrical ketone synthesis
to telescoped syntheses and the preparation of natural products.
TBAF=tetra-n-butylammonium fluoride, TIPS=triisopropylsilyl.
over about 5 hours at À788C[21] In contrast, the tetrahedral
intermediate 31 had a significantly longer lifetime, as it had
not completely collapsed even after 40 hours at À788C.
Warming the reaction mixture to room temperature resulted
in rapid and complete collapse of 31.
from 1,3-dibromobenzene and 1-lithio-1-hexyne. Addition of
methylhydrazine to the reaction vessel following ketone
synthesis afforded the pyrazole 26 in 89% overall yield. This
set of transformations constitutes a four-component coupling
of 2, 24, 1-hexyne, and methylhydrazine.
In conclusion, we have reported a carbonyl dication
synthetic equivalent, 2, which to our knowledge is the first
carbonyl linchpin reagent that reacts with a wide variety of
organometallic nucleophiles without the need for transition-
metal catalysis. This versatile reagent can be used to prepare
Weinreb amides or unsymmetrical ketones using a one-pot
protocol. The success of these reactions hinges on the ability
of 2 to form persistent tetrahedral intermediates upon the
addition of the first organometallic reagent. Lifetimes of
these tetrahedral intermediates range from a few hours to
a few days at À788C, depending on the nucleophile. As
a consequence of the broad product scope and the ease of the
synthetic operations, this procedure should serve as a useful
complement to existing unsymmetrical ketone synthesis
methodology.
Natural products that contain ketones are also accessible
through short synthetic sequences which exploit the simplify-
ing power of the one-pot ketone synthesis. For instance,
verticillatine B[20] (30; Scheme 4, ) was efficiently prepared by
using 2 to couple 28 with isobutylmagnesium chloride.
In designing 2, we speculated that addition of organome-
tallic nucleophiles would result in a persistent tetrahedral
intermediate resistant to nucleophilic attack. Experimental
support for this hypothesis was obtained by monitoring the
addition of organolithium nucleophiles to 2 using in situ IR
spectroscopy. Carbonyl stretches corresponding to the Wein-
reb amide products (33, 12) appeared and intensified over
several hours after the consumption of 2 (Figure 1A and B),
and indirectly supports the existence of persistent tetrahedral
intermediates arising from 2. Interestingly, the tetrahedral
intermediates that form from 2 and either PhLi or nBuLi (31
and 32, respectively), have dramatically different lifetimes
(Figure 1C and D). The tetrahedral intermediate 32 collapsed
Keywords: acylation · amides · IR spectroscopy · ketones ·
synthetic methods
How to cite: Angew. Chem. Int. Ed. 2015, 54, 9839–9843
Angew. Chem. 2015, 127, 9977–9981
9842 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9839 –9843

Angewandte
Chemie
[7] S. Nahm, S. M. Weinreb, Tetrahedron Lett. 1981, 22, 3815.
[1] A large volume of research has been conducted in this area. For
a recent review of the field, see: X.-F. Wu, H. Neumann, M.
Beller, Chem. Soc. Rev. 2011, 40, 4986 – 5009; A cross-coupling
approach to unsymmetrical ketones involving S-phenyl carbon-
ochloridothioate, a synthetic equivalent of CO: C. Cardellicchio,
V. Fiandanese, G. Marchese, L. Ronzini, Tetrahedron Lett. 1985,
26, 3595 – 3598.
[2] L. F. Tietze, H. Giessler, J. A. Gewert, U. Jakobi, Synlett 1994,
511 – 512.
[3] A. B. Smith III, A. M. Boldi, J. Am. Chem. Soc. 1997, 119, 6925 –
6926.
[4] For overviews of this tactic, see: a) A. B. Smith III, S. M. Pitram,
A. M. Boldi, M. J. Gaunt, C. Sfouggatakis, W. H. Moser, J. Am.
Chem. Soc. 2003, 125, 14435 – 14445; b) M. Yus, C. Najera, F.
Foubelo, Tetrahedron 2003, 59, 6147 – 6212; c) A. B. Smith III,
S. M. Condon, J. A. McCauley, Acc. Chem. Res. 1998, 31, 35 – 46.
[5] The synthesis of symmetrical ketones from carbonyl dication
synthetic equivalents and organometallic nucleophiles is well-
explored: a) from phosgene and organomanganese nucleo-
philes: G. Cahiez, J. F. Normant, Bull. Soc. Chim. Fr. 1977,
570; b) From phosgene and organocadmium nucleophiles: E. K.
Euranto, L. Puupponen, Acta Chem. Scand. 1970, 24, 363 – 364;
c) from 1,1’-carbonyldiimidazole and Grignard reagents: D.
Bottalico, F. Vito, M. Giuseppe, A. Punzi, Synthesis 2009,
2316 – 2318; d) from CO2 and organolithium nucleophiles: G.
Zadel, E. Breitmaier, Angew. Chem. Int. Ed. Engl. 1992, 31,
1035 – 1036; Angew. Chem. 1992, 104, 1070 – 1071; e) from S,S-
dimethyl dithiocarbonate and cuprates: C.-D. Chen, J.-W.
Huang, M. Leung, H. Li, Tetrahedron 1998, 54, 9067 – 9078;
f) from N-carboethoxypiperidine: G. K. S. Prakash, C. York, Q.
Liao, K. Kotian, G. A. Olah, Heterocycles 1995, 40, 79 – 83;
g) from 4,4-dimethyl-2-oxazoline: B. Tsutomu, A. Suzuki, Synth.
Commun. 1983, 13, 367 – 371; h) from CO and organolithium
nucleophiles: M. Ryang, S. Tsutsumi, Bull. Chem. Soc. Jpn. 1962,
35, 1121 – 1124.
[8] A. I. Meyers, D. L. Comins, Tetrahedron Lett. 1978, 19, 5179.
[9] T. Mukaiyama, M. Araki, H. Takei, J. Am. Chem. Soc. 1973, 95,
4763.
[10] a) F. Wingler, Methoden Org. Chem., Vol. 7/2a, 4th ed. 1952,
Houben-Weyl, 1973, p. 581; b) N,N-dimethylamides react with
Grignard reagents to afford ketones: G. A. Olah, G. K. S.
Prakash, M. Arvanaghi, Synthesis 1984, 228 – 230.
[11] See Ref. [5,d].
[12] D. A. Evans, G. Borg, K. A. Scheidt, Angew. Chem. Int. Ed. 2002,
41, 3188 – 3191; Angew. Chem. 2002, 114, 3320 – 3323.
[13] Developed by Hlasta and Reich independently: a) D. J. Hlasta,
J. J. Court, Tetrahedron Lett. 1989, 30, 1773 – 1776; b) W. L.
Whipple, H. J. Reich, J. Org. Chem. 1991, 56, 2911 – 2912.
[14] See ref. [13,a].
[15] N. R. Lee, J. I. Lee, Synth. Commun. 1999, 29, 1249.
[16] The tetrahedral intermediate formed during the reaction may
not be sufficiently persistent at the temperatures required for
addition of aryl Grignard reagents.
[17] a) For a discussion of magnesiates, see: G. Wittig, F. J. Meyer, G.
Lange, Justus Liebigs Ann. Chem. 1951, 571, 167; b) For salt
effects on organometallics, see: E. Hevia, R. E. Mulvey, Angew.
Chem. Int. Ed. 2011, 50, 6448 – 6450; Angew. Chem. 2011, 123,
6576 – 6578.
[18] H. Gilman, R. D. Gorisch, J. Am. Chem. Soc. 1956, 78, 2217 –
2222.
[19] X. Wang, L. Zhang, X. Sun, H. Lee, D. Krishnamurthy, J. A.
O’Meara, S. Landry, C. Yoakim, B. Simoneau, N. Yee, C. H.
Senanayake, Org. Process Res. Dev. 2012, 16, 561 – 566.
[20] Initial attempts to deprotonate the indole NH of 27 prior to the
lithium–halogen exchange to eliminate these protecting group
manipulations were unfruitful.
[21] See the Supporting Information for details.
Received: April 2, 2015
[6] See Ref. [3].
Published online: July 3, 2015
Angew. Chem. Int. Ed. 2015, 54, 9839 –9843
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9843