Transition-metal-free formal decarboxylative coupling of ?±-oxocarboxylates with ?±-bromoketones under neutral conditions: A simple access to 1,3-diketones

Article abstract of DOI:10.1002/anie.201409361

A transition-metal-free formal decarboxylative coupling reaction between ?±-oxocarboxylates and ?±-bromoketones to synthesize 1,3-diketone derivatives is presented. In this reaction, a broad scope of substrates can be employed, and neither a metal-based reagent nor an additional base is required. DFT calculations reveal that this reaction proceeds through a coupling followed by decarboxylation mechanism and the ?±-bromoketone unprecedentedly serves as a nucleophile under neutral conditions. The rate-determining step is an unusual hydrogen-bond-assisted enolate formation by thermolysis.

Full text of DOI:10.1002/anie.201409361

Angewandte
Chemie
DOI: 10.1002/anie.201409361
Decarboxylative Coupling
Transition-Metal-Free Formal Decarboxylative Coupling of
a-Oxocarboxylates with a-Bromoketones under Neutral Conditions:
A Simple Access to 1,3-Diketones**
Zhen He, Xiaotian Qi, Shiqing Li, Yinsong Zhao, Ge Gao,* Yu Lan,* Yiwei Wu, Jingbo Lan, and
Jingsong You*
Abstract:
A
transition-metal-free formal decarboxylative
hybridized carbon derivatives, including allyls,[2d,e] (hetero)-
arenes,[2f–j] enamides,[2k] and formamides,[2l] have been suc-
cessfully acylated using this strategy. However, no example
involving sp3-hybridized carbon centers has been reported.
coupling reaction between a-oxocarboxylates and a-bromo-
ketones to synthesize 1,3-diketone derivatives is presented. In
this reaction, a broad scope of substrates can be employed, and
neither a metal-based reagent nor an additional base is
required. DFT calculations reveal that this reaction proceeds
through a coupling followed by decarboxylation mechanism
and the a-bromoketone unprecedentedly serves as a nucleo-
phile under neutral conditions. The rate-determining step is
an unusual hydrogen-bond-assisted enolate formation by
thermolysis.
T
À
ransition-metal-catalyzed decarboxylative coupling reac-
tions have recently attracted much attention for their use in
C C bond formation reactions.[1] Among them, a-oxocarbox-
ylic acids have emerged as a new acyl anion reagent.[2] The
acyl anions[3] are generated in situ by extrusion of carbon
dioxide facilitated by the use of transition metals (TM).
Compared to the masked acyl reagents, such as dithianes,[4]
protected cyanohydrins,[5] and hydrazones,[6] where the car-
bonyl functionality must be unmasked after the reaction,
a-oxocarboxylic acids are stable, easy to prepare,[7] and need
no post-reaction treatment. These advantages make them
convenient and clean acyl anion equivalents, which comple-
ment the traditional acyl cation equivalents, such as acyl
chlorides and anhydrides. In 2008, Goossen et al. reported the
synthesis of diarylketones through the Pd/Cu-catalyzed
decarboxylative coupling of a-oxocarboxylates with aryl
bromides [Eq. (1)].[2a] After this seminal work, various sp2-
a-Bromoketones are commonly used C(sp3) electrophiles
in organic synthesis.[8] We envisioned that the decarboxylative
coupling of a-oxocarboxylate salts with a-bromoketones
should afford 1,3-diketone derivatives, which are an impor-
tant class of organic compounds. 1,3-Diketone derivatives are
starting materials in Knoevenagel reactions and as inter-
mediates for the synthesis of heterocycles. Many naturally
occurring 1,3-diketones exhibit antimicrobial, antiviral, and
antifungal activities.[9] They also serve as bidentate ligands in
metal complexes for catalysis and luminescent materials.[10]
The main strategy to construct this useful scaffold is the well-
established enolate chemistry. The typical example is the
classical Claisen condensation reaction between an ester and
a ketone, in which a hard enolate[11] is generated by the
deprotonation of the a-hydrogen atom on the ketone using
a strong base. The rigorous reaction conditions required and
hence limited substrate scope are major drawbacks for this
reaction. Recent improvements include the use of soft
enolates in the presence of a Lewis acid and a weak base,[12]
tin enolates[13] through a radical process, and kinetic zinc
enolates using bis(iodozincio)methane.[14] The decarboxyla-
tive coupling strategy would certainly provide an attractive
alternative route to 1,3-diketone compounds. Herein,
we report that a-oxocarboxylates undergo reaction with
a-bromoketones to afford 1,3-diketones in the absence of
a transition-metal catalyst [Eq. (2)]. DFT calculations sug-
gested a unique mechanism involving coupling followed by
decarboxylation.
[*] Z. He, S. Li, Y. Zhao, Prof. Dr. G. Gao, Y. Wu, Prof. Dr. J. Lan,
Prof. Dr. J. You
Key Laboratory of Green Chemistry and Technology of Ministry of
Education, College of Chemistry, and State Key Laboratory of
Biotherapy, West China Medical School, Sichuan University
29 Wangjiang Road, Chengdu 610064 (P. R. China)
E-mail: gg2b@scu.edu.cn
jsyou@scu.edu.cn
X. Qi, Prof. Dr. Y. Lan
School of Chemistry and Chemical Engineering
Chongqing University, Chongqing 400030 (P. R. China)
E-mail: lanyu@cqu.edu.cn
[**] This work was financially supported by the NSFC (nos. 21172159,
21025205, 21321061, 21372266, and J1103315) and the NCET-13-
0384. We thank the group of Prof. Aiwen Lei for help with
mechanism studies.
Our investigation started with the cross-coupling reaction
between potassium 2-oxo-2-phenylacetate (1a) and 2-bromo-
1-phenylethanone (2a) in the presence of PdCl2 (2 mol%),
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201409361.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Optimization of reaction conditions.[a]
Entry
Catalyst
Solvent
T [8C]
t [h]
Yield[b] [%]
1
2
3
4
KBr
KCl
KI
NMP
NMP
NMP
NMP
NMP
NMP
DMF
DMA
toluene
DMSO
NMP
NMP
NMP
NMP
NMP
140
140
140
140
140
140
140
140
140
140
120
160
140
140
150
6
6
6
6
6
6
6
6
6
6
6
6
8
8
6
65
38
trace
63
82
86
69
73
66
N.D.
54
88
92
88
NaBr
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
5
6[c]
7
Scheme 1. Initial investigations of the reaction. NMP=N-methyl-2-
pyrrolidone; Phen=phenanthroline; Cy=cyclohexyl.
8
9
10
11
12
13
14[d]
15[e]
CuBr (15 mol%), and P(Cy)3·HBF4 (6 mol%) in N-methyl-2-
pyrrolidone (NMP; Scheme 1).[2a] After heating at 1408C for
6 h, the desired 1,3-diphenylpropane-1,3-dione (4aa) was
obtained in a low yield (< 30%, Scheme 1a). Surprisingly, the
control experiment without Pd, Cu, and ligand gave an even
better yield of 55% (Scheme 1b). Meanwhile, the formation
of compound 2-oxo-2-phenylethyl 2-oxo-2-phenylacetate
(3aa) was detected within minutes of the start of the reaction.
To gain a better understanding of this decarboxylative
coupling reaction, the ester 3aa was then separately synthe-
sized[15] as the starting material for further investigations.
However, heating 3aa in NMP alone did not result in the
formation of 4aa (Scheme 1c). When KBr, the inorganic
product of the metathesis of 1a and 2a, was added, 4aa was
again obtained (Scheme 1d). It suggested that KBr catalyzed
the intramolecular decarboxylation of 3aa to give the 1,3-
diketone product 4aa. The ICP–AES analysis (inductively
coupled plasma atomic emission spectroscopy) showed that
the concentrations of transition metals, such as Pd, Ag, Rh,
Ru, and Cu, in the reaction system were lower than the
detection limits of the machine, confirming that no transition
metal was involved. Thus, this reaction was quite attractive
because it was not only a decarboxylative coupling reaction of
an a-oxocarboxylate with an sp3-hybridized carbon, but also
a TM-free version. A radical pathway[16] was also precluded
since the reaction was not affected significantly in the
presence of a radical scavenger, such as 2,2,6,6-tetramethyl-
piperidine-N-oxide (TEMPO) or 3,5-di-tert-4-butylhydroxy-
toluene (BHT; see Supporting Information).
We then set out to optimize the reaction conditions
(Table 1). 4aa was isolated in 65% yield in the decarbox-
ylation of 3aa in the presence of KBr (10 mol%; entry 1).
Using KCl as the catalyst only gave 38% yield of 4aa and the
majority of the 3aa reagent was recovered (entry 2). The use
of KI resulted in a complicated mixture and only a trace
amount of 4aa (entry 3). When different cations were used,
tetrabutylammonium bromide (TBAB) resulted in an
enhanced yield of 82%, whereas NaBr gave a similar result
as KBr (entries 4 and 5). The use of 1.0 equivalent of TBAB
further increased the yield to 86% (entry 6). These results
suggested that the bromide anion might be the actual catalyst
and the TBA cation probably served as the phase-transfer
reagent and the intermediate stabilizer.[17] It was also noted
73
[a] The reactions were carried out with 3aa (0.50 mmol) in solvent
(1 mL) at 1408C in the present of a catalyst (10 mol%, unless otherwise
stated) under N2. [b] Yield of isolated product. [c] TBAB (1.0 equiv;
tetrabutylammonium bromide) was used. [d] NMP was used as
purchased. [e] One-pot reactions between 1a and 2a in stoichiometric
amounts. N.D.: not detected. Trace: detected by GC–MS. DMA=
dimethylacetamide. DMF=dimethylformamide.
that using less potassium ion was beneficial, comparing
entry 1 with the direct coupling of 1a with 2a (Scheme 1b).
Other aprotic solvents were also tested in the reaction but
gave rise to lower yields (entries 7–9), but no product was
detected in dimethylsulfoxide (DMSO; entry 10). Only
a small improvement of the yield was found at 1608C whereas
a significantly decreased yield was obtained at 1208C
(entries 11 and 12). Finally, extending the reaction time to
8 hours resulted in an optimal yield of 4aa of 92% (entry 13).
The use of unpurified NMP only slightly influenced the
reaction (entry 14). More practically, the yield of the direct-
coupling reaction of 1a with 2a could be improved to 73% in
the presence of TBAB (10 mol%) at 1508C for 6 h (entry 15).
With the optimized conditions in hand, a series of
potassium a-oxocarboxylates 1 and a-bromoketones 2 were
subjected to the TBAB-catalyzed decarboxylative reaction
(Table 2, method a). In general, the one-pot procedure
delivered 4 in yields ranging from 52% to 81%. The aryl
a-oxocarboxylates 1 (R1 = aryl) with an electron-withdrawing
substituent on the phenyl ring gave satisfactory yields
(entries 2–6) except for the reaction employing 1 with an
ortho-bromo substituent (entry 4). 1g with an electron-
donating methoxy group required 18-crown-6 (1.0 equiv) to
achieve comparable yields (entries 7, 23, and 24). The heter-
oaryl a-oxocarboxylate containing a thienyl ring produced the
desired 1,3-diketone 4ha in 53% yield (entry 8). The alkyl a-
oxocarboxylate derivatives 1 (R1 = alkyl) gave complicated
mixtures and very low yields (entries 9 and 10).
For a-bromoketones 2, both aryl and alkyl groups were
suitable (R2 = aryl and alkyl). The para- and meta-methoxy
substrates afforded products 4ag and 4ak in 72% and 81%
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Table 2: Substrate scope of the decarboxylative reaction.[a]
oaryl-functionalized 1,3-diketone derivative 4hh was isolated
in 52% yield (entry 26).
Reactions starting from the ester derivatives 3 were also
conducted (Table 2, method b). The corresponding yields
were generally higher than the one-pot processes, probably
because of the absence of the deleterious potassium ion.
Additionally, the reactions of 3ia and 3ja (R1 = Me and iPr,
R2 = phenyl) afforded the 1,3-diketones 4ia and 4ja in 52%
and 49% yields, respectively (entries 9 and 10). It should also
be mentioned that the yield of 4at could be improved to 81%
using method b by elevating the reaction temperature to
1708C (entry 21), but limited yield improvements were found
for other substrates after increasing the reaction temperature.
The preliminary mechanistic study showed that the
reaction was triggered by the SN2 attack of a bromide anion
Entry
R1
R2
4
Yield[b] [%]
1
2
3
4
5
6
7
8
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4aa
4ba
4ca
4da
4ea
4 fa
4ga
4ha
4ia
73 (92)
66 (87)
73 (82)
54 (58)
78
62 (81)
74[c] (86)
53
– (52)
– (49)
72 (74)
81 (91)
56 (54)
75 (76)
75 (91)
64 (67)
69 (71)
64
p-BrPh
m-BrPh
o-BrPh
p-ClPh
p-CF3Ph
p-MeOPh
2-thienyl
Me
iPr
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
À
toward the methylene carbon of the ester 3 to cleave its C O
bond. When a hydrogen atom on the methylene carbon was
substituted by a methyl group (3aa’), which hindered the
bromide attack, no reaction occurred [Eq. (3)]. When 3aa
was heated together with TBAB (10 mol%) in 1,2-dichloro-
ethane (DCE), compounds 2-chloroethyl 2-oxo-2-phenyl-
acetate (5) and 2-chloro-1-phenylethanone (6) were obtained
in 79% and 57% yields, respectively [Eq. (4), a]. Without
TBAB, however, only the starting material 3aa was recovered
[Eq. (4), b]. These results indicated that bromide first
attacked the methylene carbon of 3aa to generate 2-bromo-
1-phenylethanone (2a) and the a-oxocarboxylate anion. Then
esterification of the a-oxocarboxylate anion with DCE
formed the ester 5 and released a chloride anion. Finally,
halide exchange between the chloride anion and 2a gave
product 6. Interestingly, the reaction in the presence of
tetrabutylammonium iodide (TBAI) afforded only trace
amounts of 5 and 6, and 3aa was almost fully recovered
[Eq. (4), c].[18] However, the use of KI as the catalyst led to
a complicated reaction mixture in NMP, in which the more
nucleophilic iodide probably preferred to attack the carbonyl
carbon rather than the methylene carbon of 3aa (Table 1,
entry 3). Other nucleophiles, such as Et3N and PPh3, were not
effective. Therefore, it is assumed that the nucleophilicity of
bromide should be key to this reaction. The kinetic isotopic
effect (KIE) experiment gave a kH/kD value of 2.6, indicating
9
Ph
Ph
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
4ja
p-MeOPh
m-MeOPh
o-MeOPh
1-naphthyl
benzo[d][1,3]dioxole
p-MeOC(O)Ph
p-NO2Ph
p-CNPh
p-HOPh
PhCH CH-
tBu
p-BrPh
p-MeOPh
p-BrPh
4ag
4ak
4al
4am
4an
4ao
4ap
4aq
4ar
4as
4at
4bb
4gg
4gb
4eg
4hh
58
=
73 (72)
74 (62, 81[d])
58 (79)
57[c] (61)
69[c] (66)
70
p-BrPh
p-MeOPh
p-MeOPh
p-ClPh
2-thienyl
p-MeOPh
2-thienyl
52 (50)
[a] Method a: reagents 1 (0.50 mmol) and 2 (0.50 mmol) in NMP (2 mL)
in the presence of TBAB (10 mol%) at 1508C for 6 h under N2.
Method b: reagent 3 (0.50 mmol) in NMP (1 mL) at 1408C for 8 h in the
presence of TBAB (10 mol%) under N2. [b] Yields obtained using
method a given outside parentheses, yields obtained using method b
given inside parentheses. [c] 18-crown-6 (0.50 mmol) was added.
[d] 1708C.
À
that the cleavage of the methylene C H bond was involved in
the rate-determining step [Eq. (5); see also the Supporting
Information].
yields, respectively, whereas the ortho-methoxy substrate
gave 4al in 56% yield (entries 11–13). When R2 was the bulky
1-naphthyl group, the yield of 4am reached 75% (entry 14).
The sensitive functionalities, such as benzo[d][1,3]dioxole,
para-MeOC(O)Ph, para-NO2Ph, and para-CNPh, were all
tolerated (entries 15–18). Strikingly, phenol was also well-
tolerated, and the phenol-containing 1,3-diketone 4ar was
achieved in 58% yield (entry 19). Additionally, the a,b-
unsaturated 1,3-diketone 4as was synthesized in 73% yield
(entry 20). When alkyl a-bromoketone 2t was used, 1-tert-
butyl-3-phenylpropane-1,3-dione (4at) was obtained in 74%
yield (entry 21). As well as the monofunctional 1,3-diketones,
the bifunctional 1,3-diketones could also be accessed con-
veniently. The symmetrical and unsymmetrical 1,3-diketones
with electron-withdrawing and electron-donating phenyl
were obtained in 57–70% yields (entries 22–25).The diheter-
The next decarboxylation step was elusive as the sponta-
neous decarboxylation of an a-oxocarboxylate anion seemed
unlikely to occur without the help of any other reagent under
the current conditions.[1b,d] To shed light on this process, the
density functional theory (DFT) method B3LYP with a stan-
dard 6-31 + G(d) basis set was employed. The calculation
suggested that the reaction was a two-stage process as shown
À
in Figure 1. In the first stage, the C O bond of the ester 3aa
was cleaved by the SN2 attack of the bromide anion to form
a-bromoketone 2a and a-oxocarboxylate anion 8 through the
transition state 7-TS with an energy barrier of 11.2 kcalmolÀ1
(path I).
In the second stage, the keto–enol tautomerization of 2a
tended to form enol 2a’ as a result of the stabilization by the
formation of an eight-membered ring 9 through two hydrogen
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
=
higher in energy) formed the C C double bond to give 15,
which was the more stable isomer of 1,3-diketone 4aa. An
alternative potential route, denoted path II, was also deter-
mined by DFT calculations. Path II involves the direct
nucleophilic attack of a-bromoketone 2a towards a-oxo-
carboxylate anion 8 via the transition state 16-TS. The relative
free energy of 16-TS turned out to be 33.9 kcalmolÀ1, which
was 6.2 kcalmolÀ1 higher than that of 12-TS. Therefore,
path II was thermodynamically disfavored as compared with
path I.[19] Based on the DFT calculations, a proposed
mechanism is shown in Scheme 2.
Several key points should be made regarding these
proposed pathways. First, the crucial step was the dissociation
of complex 9, which resulted in an unusual proton transfer
from enol 2a’ to a-oxocarboxylate 8 to generate the more
basic naked enolate 10 and the more acidic a-oxocarboxylic
acid 11. The process was believed to be a thermolysis
facilitated by hydrogen bonding. Second, the nucleophilic
attack via transition state 12-TS was the rate-determining
step, which had the highest energy of 27.7 kcalmolÀ1 relative
to 3aa. Third, the effect of the TBA cation was to stabilize the
as-generated naked enolate anion.[20] Finally, the nucleophilic
addition of the a-bromoketone to the carbonyl carbon of the
a-oxocarboxylate gave the intermediate 13, which had an
a-hydroxycarboxylate structure with an a-bromoketone sub-
stituent. It had a significantly lower energy barrier of only
5.4 kcalmolÀ1 to achieve decarboxylation, compared with
transition-metal-assisted decarboxylation.[2a,b] A comparable
reported example is the thiazolium-substituted a-hydroxy-
carboxylate intermediate reported by Scheidt and co-work-
ers.[21] CO2 was easily extruded from this inter-
mediate under mild conditions to generate the
active Breslow intermediate, which subsequently
underwent addition to the a,b-unsaturated 2-acyl
imidazoles to afford 1,4-dicarbonyl compounds.
bonds, which was 3.2 kcalmolÀ1 lower in energy. Next, an
unusual thermolytic dissociation of the complex 9 generated
the naked enolate anion 10 and the a-oxocarboxylic acid 11
with an energy barrier of 23.5 kcalmolÀ1. The nucleophilic
attack of 10 toward the carbonyl carbon in 11 took place
immediately via the transition state 12-TS to form the
intermediate 13 through an energy barrier of only 0.6 kcal
molÀ1. Next, a synergetic decarboxylation and bromide
leaving process via the transition state 14-TS (5.4 kcalmolÀ1
In summary, we report herein a transition-
metal-free decarboxylative coupling of a-oxo-
carboxylates with a-bromoketones to synthesize
1,3-diketone compounds. The reaction was self-
catalyzed by the bromide anion but performed
better in the presence of a catalytic amount of
TBAB. It featured mild and neutral conditions
without any metal or extra base, and also
tolerated a broad spectrum of functional groups.
This reaction is unique in two aspects: a) the
transition-metal-free formal decarboxylative cou-
pling of a C(sp3)-hybridized substrate proceeded
through a coupling followed by decarboxylation
pathway, and b) a-bromoketone, a common elec-
trophile, acted as a nucleophile through its
enolate form by thermolysis. Note, the mecha-
nism of coupling followed by decarboxylation
significantly lowers the energy barrier for decar-
boxylation and may serve as a new strategy to
conduct metal-free decarboxylative coupling
reactions. Other applications using this strategy
are currently under investigation in our labora-
tory.
Figure 1. Free energy profile of the decarboxylative coupling reaction.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[3] D. Seebach, Angew. Chem. Int. Ed. Engl. 1969, 8,
639; Angew. Chem. 1969, 81, 690.
[4] A. B. Smith III, C. M. Adams, Acc. Chem. Res.
2004, 37, 365.
[5] G. Stork, L. Maldonad, J. Am. Chem. Soc. 1971, 93,
5286.
[6] J. E. Baldwin, R. M. Adlington, J. C. Bottaro, J. N.
Kolhe, M. W. D. Perry, A. U. Jain, Tetrahedron
1986, 42, 4223.
[7] a) K. Wadhwa, C. Yang, P. R. West, K. C. Deming,
S. R. Chemburkar, R. E. Reddy, Synth. Commun.
2008, 38, 4434; b) S. Hu, D. C. Neckers, J. Org.
Chem. 1996, 61, 6407.
[8] N. De Kimpe, R. Verhꢂ, a-Haloketones, a-Hal-
oaldehydes and a-Haloimines, Wiley, New York,
1988, p. 1.
[9] a) A. V. Kelꢃin, Curr. Org. Chem. 2003, 7, 1691;
b) A. V. Kelꢃin, A. Maioli, Curr. Org. Chem. 2003,
7, 1855.
[10] P. A. Vigato, V. Peruzzo, S. Tamburini, Coord.
Chem. Rev. 2009, 253, 1099.
Scheme 2. Proposed mechanism for the bromide-catalyzed formation of 4aa.
[11] For a Pd-catalyzed CO insertion into hard enolates,
see: T. M. Gøgsig, R. H. Taaning, A. T. Lindhardt,
T. Skrydstrup, Angew. Chem. Int. Ed. 2012, 51, 798; Angew.
Chem. 2012, 124, 822.
Received: September 22, 2014
Revised: November 2, 2014
[12] D. Lim, F. Fang, G. Zhou, D. M. Coltart, Org. Lett. 2007, 9, 4139.
[13] K. Miura, M. Tojoni, N. Fujisawa, A. Hosomi, I. Ryu, Angew.
Chem. Int. Ed. 2004, 43, 2423; Angew. Chem. 2004, 116, 2477.
[14] M. Sada, S. Matsubara, J. Am. Chem. Soc. 2010, 132, 432.
[15] Esters 3 could be obtained in a quantitative yield by the reaction
of 1 with 2 in DMSO at room temperature in 30 min.
[16] Metal-free decarboxylative reactions of a-amino acids through
radical mechanism have been reported. See: a) Q. Wang, C.
Wan, Y. Gu, J. Zhang, L. Gao, Z. Wang, Green Chem. 2011, 13,
578; b) Y. Yan, Z. Wang, Chem. Commun. 2011, 47, 9513.
[17] Tetrabutylammonium halides have been used as the additive in
several reactions. For selected examples, see: a) T. Jeffery,
Tetrahedron 1996, 52, 10113; b) L. J. Goossen, J. Paetzold,
Angew. Chem. Int. Ed. 2004, 43, 1095; Angew. Chem. 2004,
116, 1115; c) Y. Wei, I. Deb, N. Yoshikai, J. Am. Chem. Soc. 2012,
134, 9098; d) F. Chen, X. Huang, X. Li, T. Shen, M. Zou, N. Jiao,
Angew. Chem. Int. Ed. 2014, 53, 10495; Angew. Chem. 2014, 126,
10663.
Published online: && &&, &&&&
Keywords: cross-coupling · decarboxylation · enolates ·
.
homogeneous catalysis · synthetic methods
[1] For selected reviews, see: a) O. Baudoin, Angew. Chem. Int. Ed.
2007, 46, 1373; Angew. Chem. 2007, 119, 1395; b) L. J. Goossen,
N. Rodrꢀguez, K. Goossen, Angew. Chem. Int. Ed. 2008, 47, 3100;
Angew. Chem. 2008, 120, 3144; c) J. D. Weaver, A. Recio III,
A. J. Grenning, J. A. Tunge, Chem. Rev. 2011, 111, 1846; d) N.
Rodrꢀguez, L. J. Goossen, Chem. Soc. Rev. 2011, 40, 5030.
[2] For selected recent examples, see: a) L. J. Goossen, F. Rudolphi,
C. Oppel, N. Rodriguez, Angew. Chem. Int. Ed. 2008, 47, 3043;
Angew. Chem. 2008, 120, 3085; b) R. Shang, Y. Fu, J.-B. Li, S.-L.
Zhang, Q.-X. Guo, L. Liu, J. Am. Chem. Soc. 2009, 131, 5738;
c) F. Rudolphi, B. Song, L. J. Goossen, Adv. Synth. Catal. 2011,
353, 337; d) N. Rodrꢀguez, F. Manjolinho, M. F. Grꢁnberg, L. J.
Goossen, Chem. Eur. J. 2011, 17, 13688; e) F. Manjolinho, M. F.
Grꢁnberg, N. Rodrꢀguez, L. J. Goossen, Eur. J. Org. Chem. 2012,
4680; f) P. Fang, M. Li, H. Ge, J. Am. Chem. Soc. 2010, 132,
11898; g) M. Li, H. Ge, Org. Lett. 2010, 12, 3464; h) C. Pan, H.
Jin, X. Liu, Y. Cheng, C. Zhu, Chem. Commun. 2013, 49, 2933;
i) L. Yu, P. Li, L. Wang, Chem. Commun. 2013, 49, 2368; j) K.
Yang, C. Zhang, P. Wang, Y, Zhang, H. Ge, Chem. Eur. J. 2014,
20, 7241; k) H. Wang, L. N. Guo, X. H. Duan, Org. Lett. 2012, 14,
4358; l) D. Li, M. Wang, J. Liu, Q. Zhao, L. Wang, Chem.
Commun. 2013, 49, 3640.
À
[18] The DFT calculation showed that the cleavage of the C O bond
by iodide is more difficult than by bromide. See the Supporting
Information.
[19] The SN2 attack of bromide toward the enol form of 3aa was also
calculated (path III). The activation energy was as high as
61.1 kcalmolÀ1, which was not favored. See the Supporting
Information.
[20] a) I. Kuwajima, E. Nakamura, J. Am. Chem. Soc. 1975, 97, 3257;
b) I. Kuwajima, E. Nakamura, Acc. Chem. Res. 1985, 18, 181.
[21] M. C. Myers, A. R. Bharadwaj, B. C. Milgram, K. A. Scheidt, J.
Am. Chem. Soc. 2005, 127, 14675.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Decarboxylative Coupling
No metal needed: 1,3-Diketone com-
pounds are synthesized from a-oxo-
carboxylates and a-bromoketones using
a bromide-catalyzed, mild, neutral, and
transition-metal-free procedure which
tolerates a wide spectrum of functional
groups. DFT calculations suggest
a mechanism involving coupling followed
by decarboxylation. NMP=N-methyl-2-
pyrrolidone; TBAB=tetrabutylammo-
nium bromide.
Z. He, X. Qi, S. Li, Y. Zhao, G. Gao,*
Y. Lan,* Y. Wu, J. Lan,
J. You*
&&&& — &&&&
Transition-Metal-Free Formal
Decarboxylative Coupling ofa-
Oxocarboxylates with a-Bromoketones
under Neutral Conditions:A Simple
Access to 1,3-Diketones
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!