Ni-catalyzed direct carboxylation of benzyl halides with CO2

Article abstract of DOI:10.1021/ja311045f

A novel Ni-catalyzed carboxylation of benzyl halides with CO₂ has been developed. The described carboxylation reaction proceeds under mild conditions (atmospheric CO₂ pressure) at room temperature. Unlike other routes for similar means, our method does not require well-defined and sensitive organometallic reagents and thus is a user-friendly and operationally simple protocol for assembling phenylacetic acids.

Full text of DOI:10.1021/ja311045f

Communication
pubs.acs.org/JACS
Ni-Catalyzed Direct Carboxylation of Benzyl Halides with CO2
Thierry Leon, Arkaitz Correa, and Ruben Martin*
́
Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans 16, 43007 Tarragona, Spain
S
* Supporting Information
Scheme 2. Synthesis of Phenylacetic Acids
ABSTRACT: A novel Ni-catalyzed carboxylation of
benzyl halides with CO2 has been developed. The
described carboxylation reaction proceeds under mild
conditions (atmospheric CO2 pressure) at room temper-
ature. Unlike other routes for similar means, our method
does not require well-defined and sensitive organometallic
reagents and thus is a user-friendly and operationally
simple protocol for assembling phenylacetic acids.
arbon dioxide (CO2) is abundant, inexpensive, nonflam-
C
mable, and attractive as an environmentally friendly
chemical reagent.1 Indeed, the fixation of CO2 holds great
promise for revolutionizing approaches toward the elaboration
of chemicals of industrial significance. In this regard, metal-
catalyzed carboxylation protocols have become excellent
alternatives to the classical methods for preparing carboxylic
acids.1 Illustrative examples are the direct carboxylation of
arylmetal species and other carbon nucleophiles such as alkynes
and activated C−H bonds.2 Recently, we3a and others3b
described the carboxylation of aryl bromides and chlorides
with CO2 en route to benzoic acids. Unfortunately, these
protocols cannot be utilized to access aliphatic backbones,
namely phenylacetic acids. The interest in these compounds
arises from the fact that a wide number of complex molecules
such as vancomycin, carbenicillin, ibuprofen or diclofenac,
among many others, display significant biological activity
(Scheme 1).
catalytic protocol without the need for sensitive metal
complexes would be highly appreciated. Despite the impressive
preparative scope when alkyl halides are utilized as electrophiles
in cross-coupling reactions,6 to the best of our knowledge the
direct carboxylation of alkyl halides has not yet been described
in the literature.7,8 Herein we report the first catalytic
carboxylation of primary, secondary, and tertiary benzyl halides
utilizing CO2 as the source of carbon, thus exploring an
unrecognized opportunity to prepare phenylacetic acids from
commercially available and cheap starting materials.
We initially explored the carboxylation of 4-tert-butylbenzyl
chloride (1a) as the model substrate using nickel catalysts
under 1 atm CO2 (Table 1). After considerable experimenta-
tion, we found that the combination of NiCl2·glyme (10 mol
%), PCp3·HBF4 (Cp = cyclopentyl) (20 mol %), Zn dust as the
reducing agent, and MgCl2 as the additive in N,N-
dimethylformamide (DMF) under 1 atm CO2 provided the
best results, giving rise to 4-tert-butylphenylacetic acid (2a) in
70% isolated yield (entry 1). Control experiments in the
absence of either metal, ligand, or MgCl2 confirmed that these
species play a critical role in the carboxylative coupling reaction
(entries 2−4). Furthermore, Zn dust was found to be essential
for the reaction to occur (entry 5).9,10 Other nickel sources
such as Ni(acac)2 (entry 6) and the structurally related ligand
PCy3 (Cy = cyclohexyl) (entry 7) provided 2a in lower yields,
thus showing the subtleties of our protocol. It is noteworthy
that the use of other additives such as tetraethylammonium
iodide (TEAI),11 MgBr2 (entries 8 and 9) or solvents (entry
11) was not beneficial. Interestingly, the carboxylation was
completely suppressed when the reaction was conducted at a
slightly higher temperature (entry 10); in line with these
Scheme 1. Significance of Phenylacetic Acids
In recent years, metal-catalyzed hydrocarboxylation of
styrene derivatives (Scheme 2, route a)4 and carboxylation of
dialkylzinc species (Scheme 2, route b)5 have provided direct
access to phenylacetic acids. Unfortunately, however, the
limitation to unsubstituted styrenes (R1 = H, route a), the
low α/β selectivity observed in some cases (route a), and the
use of sensitive metal complexes in stoichiometric amounts
(e.g., Grignard reagents or pyrophoric Et2Zn) (route a) or the
need for well-defined dialkylzinc derivatives (route b)
constitute significant drawbacks to be overcome. Thus, the
development of a user-friendly and operationally simple
Received: November 9, 2012
Published: January 9, 2013
© 2013 American Chemical Society
1221
dx.doi.org/10.1021/ja311045f | J. Am. Chem. Soc. 2013, 135, 1221−1224

Journal of the American Chemical Society
Communication
a
allowing for orthogonal modifications via conventional cross-
coupling reactions.12 In most instances, the less expensive
benzyl chlorides provided higher yields than the corresponding
benzyl bromides (2f, 2g). Propargyl halides also underwent the
carboxylation, although in lower yields (2m).13,14 Overall, we
believe these results show not only the excellent activity and
functional group compatibility but also the robustness of our
carboxylative reaction.
The low propensity of secondary alkyl halides to undergo
oxidative addition and the innate proclivity of the initially
formed alkylmetal species toward unproductive β-hydride
elimination makes the coupling of secondary alkyl halides
particularly challenging.6 Indeed, the reaction of 3a under the
optimized conditions for primary benzyl halides (Table 2)
resulted in recovery of the starting material. Pleasingly, we
found that the combination of NiCl2(PCy3)2 (10 mol %),
tetrabutylammonium iodide (TBAI) (20 mol %), Zn dust as
the reducing agent, and N,N-dimethylacetamide (DMA) as the
solvent at room temperature under 1 atm CO2 was optimal for
secondary benzyl halides, furnishing 4a in 58% yield (Table
3).15 As shown in Table 3, α-alkyl and α-aromatic substitutions
Table 1. Ni-Catalyzed Carboxylation of 1a with CO2
b
entry
change from standard conditions
yield of 2a (%)
c
1
2
none
74 (70 )
without NiCl2·glyme
without PCp3·HBF4
without MgCl2
0
3
20
28
0
4
5
without Zn
6
Ni(acac)2 instead of NiCl2·glyme
PCy3 instead of PCp3
43
62
0
7
d
8
TEAI instead of MgCl2
9
MgBr2 instead of MgCl2
42
0
10
11
12
40 °C instead of rt
e
DMI instead of DMF
21
43
10 atm instead of 1 atm CO2
a
Standard conditions: 1a (0.5 mmol), NiCl2·glyme (10 mol %),
PCp3·HBF4 (20 mol %), MgCl2 (2 equiv), Zn (5 equiv), CO2 (1 atm)
and DMF (0.50 M) at r.t. HPLC yield using anisole as an internal
standard. Isolated yield. TEAI = tetraethylammonium iodide. DMI
b
c
d
e
Table 3. Carboxylation of Secondary and Tertiary Alkyl
ab
,
Halides
= 1,3-dimethyl-2-imidazolidinone.
striking observations, a higher CO2 pressure had a deleterious
effect on the reaction outcome (entry 12).
Encouraged by these results, we next turned our attention to
the preparative scope of this reaction (Table 2). Gratifyingly, a
a b
,
Table 2. Carboxylation of Primary Alkyl Halides
a
Reaction conditions: 3 (0.5 mmol), NiCl2(PCy3)2 (10 mol %), TBAI
(20 mol %), Zn (2.5 mmol), CO2 (1 atm), and DMA (0.50 M) at rt.
b
TBAI = tetrabutylammonium iodide. Isolated yields (average of at
c
d
least two independent runs) are shown. 50 °C. 40 °C.
on the alkyl halide backbone, regardless of the electronic effects
on the aryl ring, were equally tolerated (4a−g). Cyclic systems
were similarly reactive under our catalytic protocol, affording 4e
and 4f in 51 and 60% yield, respectively. Notably, tertiary alkyl
halide 3g could be utilized as well, affording 4g in 55% yield.
Interestingly, we showed that ibuprofen (4d) was within reach
from readily available and cheap starting materials.
Taking into consideration the crucial role of Zn and the
tendency of alkylmetal species to undergo β-hydride elimi-
nation, we wondered whether our coupling reaction proceeded
via in situ formation of alkylzinc species16 or styrene
derivatives.4 Interestingly, the reaction of benzylzinc bromide
(5) and styrene (6) under the optimized reaction conditions in
Tables 2 and 3 resulted in either no conversion or exclusive
formation of toluene;17 not even traces of the corresponding
phenylacetic acids 2b or 4a were observed in the crude reaction
mixtures. These results tacitly suggest that 5 and 6 are not
competent as reaction intermediates. We tentatively propose a
catalytic cycle consisting of an initial reduction of the Ni(II)
precatalyst promoted by Zn followed by oxidative addition to
a
b
As for Table 1, entry 1. Isolated yields (averages of at least two
c
independent runs) are shown. Using the alkyl bromide.
wide range of substituted benzyl halides bearing both electron-
withdrawing and electron-donating groups could be carboxy-
lated in moderate to good yields. As is evident from the results
compiled in Table 2, our carboxylative protocol showed an
excellent chemoselectivity profile. Thus, alkenes (2i), ketones
(2j), esters (2k), aryl halides (2e, 2h) and silyl ethers (2l) were
perfectly accommodated and remained intact. Additionally, the
process was not hampered by ortho substitution (2g, 2h).
Particularly noteworthy was the exquisite selectivity in the
presence of aryl halides3 (2e, 2h) or styrenes (2i),4 thus
1222
dx.doi.org/10.1021/ja311045f | J. Am. Chem. Soc. 2013, 135, 1221−1224

Journal of the American Chemical Society
Communication
the corresponding alkyl halide, thus delivering an η3- and η1-
bound nickel complexes II and III (Scheme 3).18 It is expected
excellent chemoselectivity profile. The transformation repre-
sents a straightforward alternative to other existing method-
ologies that require the preparation of well-defined and
sensitive organometallic reagents. Further mechanistic studies
as well as the development of an enantioselective carboxylation
reaction and the extension to unactivated alkyl halides are
currently underway in our laboratories.
Scheme 3. Proposed Catalytic Cycle
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, spectral data, and crystallographic
data (CIF). This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
rmartinromo@iciq.es
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the ICIQ Foundation, the European Research
Council (ERC-277883), and MICINN (CTQ2009-13840 and
CTQ2012-34054) for financial support. Johnson Matthey,
Umicore, and Nippon Chemical Industrial are acknowledged
for gifts of metal and ligand sources. R.M. and A.C. thank
MICINN for RyC and JdC Fellowships.
that II and III would indeed be in rapid equilibrium as
suggested by Jamison in a recent disclosure.19 Zn would then
mediate the generation of Ni(I) species IV, setting the stage for
CO2 insertion en route to V. A final treatment with Zn would
regenerate the active Ni(0)L2 species and provide zinc
carboxylate VI, which upon hydrolytic workup would deliver
the expected product. While the role of MgCl2 remains to be
elucidated, at present we speculate that this additive might act
as either a Lewis acid5a to facilitate CO2 insertion or as an
activator of Zn dust.20
REFERENCES
■
(1) For recent reviews dealing with the use of CO2, see: (a) Tsuji, Y.;
Fujihara, T. Chem. Commun. 2012, 48, 9956. (b) Cokoja, M.;
Bruckmeier, C.; Rieger, B.; Herrmann, W. A.; Kuhn, F. E. Angew.
Chem., Int. Ed. 2011, 50, 8510. (c) Martin, R.; Kleij, A. W.
ChemSusChem 2011, 4, 1259. (d) Huang, K.; Sun, C.-L.; Shi, Z.-J.
Chem. Soc. Rev. 2011, 40, 2435. (e) Correa, A.; Martin, R. Angew.
Chem., Int. Ed. 2009, 48, 6201. (f) Sakakura, T.; Choi, J.-C.; Yasuda, H.
Chem. Rev. 2007, 107, 2365.
̈
While unraveling all of the basic features of the catalytic
scenario shown in Scheme 3 was rather complicated, we set out
to explore the reactivity of the putative oxidative addition
species II (Scheme 3). Following the methodology described
by Carmona,18 we prepared 7 in high yield by reacting benzyl
chloride with Ni(COD)2 (COD = 1,5-cyclooctadiene) and
PCp3 in Et2O. The structure of 7 in the solid state (depicted in
Scheme 3) clearly shows the η3 hapticity of the Ni complex.17
Gratifyingly, we found that complex 7 did not undergo
carboxylation in the absence of Zn but on the contrary
afforded 2a in 71% yield in the presence of Zn dust (Scheme 3,
bottom).17,21 Moreover, we found that the coupling of alkyl
halides was completely inhibited by the addition of radical
scavengers such as 2,2,6,6-tetramethylpiperidine N-oxyl
(TEMPO) or butylated hydroxytoluene (BHT).22 Further-
more, the coupling of optically pure (R)-3a (94% ee) resulted
in racemic 4a. Although the lability of the initially generated
organometallic intermediate might account for this result, the
mechanistic hypothesis depicted in Scheme 3 involving single-
electron transfer processes seems more plausible.
(2) For selected examples, see: (a) Inomata, H.; Ogata, K.;
Fukuzawa, S.-I.; Hou, Z. Org. Lett. 2012, 14, 3986. (b) Kikuchi, S.;
Sekine, K.; Ishida, T.; Yamada, T. Angew. Chem., Int. Ed. 2012, 51,
6989. (c) Zhang, L.; Cheng, J.; Carry, B.; Hou, Z. J. Am. Chem. Soc.
2012, 134, 14314. (d) Li, S.; Yuan, W.; Ma, Y. Angew. Chem., Int. Ed.
2011, 50, 2578. (e) Fujihara, T.; Xu, T.; Semba, K.; Terao, J.; Tsuji, Y.
Angew. Chem., Int. Ed. 2011, 50, 523. (f) Mizuno, H.; Takaya, J.;
Iwasawa, N. J. Am. Chem. Soc. 2011, 133, 1251. (g) Boogaerts, I. I. F.;
Nolan, S. P. J. Am. Chem. Soc. 2010, 132, 8858. (h) Zhang, L.; Cheng,
J.; Ohishi, T.; Hou, Z. Angew. Chem., Int. Ed. 2010, 49, 8670.
(i) Yeung, C. S.; Dong, V. M. J. Am. Chem. Soc. 2008, 130, 7826.
(j) Ochiai, H.; Jang, M.; Hirano, K.; Yorimitsu, H.; Oshima, K. Org.
Lett. 2008, 10, 2681. (k) Ohishi, T.; Nishiura, M.; Hou, Z. Angew.
Chem., Int. Ed. 2008, 47, 5792. (l) Ukai, K.; Aoki, M.; Takaya, J.;
Iwasawa, N. J. Am. Chem. Soc. 2006, 128, 8706.
(3) (a) Correa, A.; Martin, R. J. Am. Chem. Soc. 2009, 131, 15974.
(b) Fujihara, T.; Nogi, K.; Xu, T.; Terao, J.; Tsuji, Y. J. Am. Chem. Soc.
2012, 134, 9106.
(4) For selected examples, see: (a) Greenhalgh, M. D.; Thomas, S. P.
J. Am. Chem. Soc. 2012, 134, 11900. (b) Williams, C. M.; Johnson, J.
B.; Rovis, T. J. Am. Chem. Soc. 2008, 130, 14936. (c) Hoberg, H.;
Peres, Y.; Kruger, C.; Tsay, Y.-H. Angew. Chem., Int. Ed. Engl. 1987, 26,
771.
In summary, we have described a novel, direct, and efficient
Ni-catalyzed carboxylation of primary, secondary, and tertiary
benzyl halides with CO2 via C(sp3)−halide activation. We
believe that our protocol constitutes a user-friendly and
operationally simple coupling reaction for the preparation of
phenylacetic acids under mild reaction conditions with an
(5) (a) Metzger, A.; Bernhardt, S.; Manolikakes, G.; Knochel, P.
Angew. Chem., Int. Ed. 2010, 49, 4665. (b) Bernhardt, S.; Metzger, A.;
Knochel, P. Synthesis 2010, 3802.
(6) For recent reviews, see: (a) Jana, R.; Pathak, T. P.; Sigman, M. S.
Chem. Rev. 2011, 111, 1417. (b) Rudolph, A.; Lautens, M. Angew.
1223
dx.doi.org/10.1021/ja311045f | J. Am. Chem. Soc. 2013, 135, 1221−1224

Journal of the American Chemical Society
Communication
Chem., Int. Ed. 2009, 48, 2656. (c) Frisch, A. C.; Beller, M. Angew.
Chem., Int. Ed. 2005, 44, 674.
(7) For examples of carbonylation of benzyl halides using CO
pressure, see: (a) Liu, Q.; Zhang, H.; Lei, A. Angew. Chem., Int. Ed.
2011, 50, 10788. (b) Jones, R. V. H.; Lindsell, W. E.; Palmer, D. D.;
Preston, P. N.; Whitton, A. J. Tetrahedron Lett. 2005, 46, 8695.
(c) Mizushima, E.; Hayashi, T.; Tanaka, M. Top. Catal. 2004, 29, 163.
(d) Trzeciak, A. M.; Ziolkowski, J. J. Mol. Catal. A: Chem. 2000, 154,
93.
(8) For a formal alkyl carboxylation event initiated by in situ
formation of alkylborates, see: Ohmiya, H.; Tanabe, M.; Sawamura, M.
Org. Lett. 2011, 13, 1086. While this manuscript was under revision, a
Pd-catalyzed carboxylative coupling of benzyl chlorides with
allyltributylstannane was recently published: Feng, X.; Sun, A.;
Zhang, S.; Yu, X.; Bao, M. Org. Lett. 2013, 15, 108.
(9) Reproducibility issues were observed when the amount of Zn was
decreased to 1−2 equiv.
(10) For a selection of recent disclosures using Zn or Mn as reducing
agents in Ni-catalyzed reductive couplings of alkyl halides, see: (a) Yu,
X.; Yang, T.; Wang, S.; Xu, L.; Gong, H. Org. Lett. 2011, 13, 2138.
(b) Everson, D. A.; Shrestha, R.; Weix, D. J. J. Am. Chem. Soc. 2010,
132, 920. (c) Prinsell, M. R.; Everson, D. A.; Weix, D. J. Chem.
Commun. 2010, 46, 5743. (d) Goldup, S. M.; Leigh, D. A.; McBurney,
R. T.; McGonigal, P. R.; Plant, A. Chem. Sci. 2010, 1, 383.
(11) The use of TEAI as an additive has been found to be beneficial
in Ni-catalyzed reductive coupling reactions. See: (a) Iyoda, M.;
Otsuka, H.; Sato, K.; Nisato, N.; Oda, M. Bull. Chem. Soc. Jpn. 1990,
63, 80. (b) Iyoda, M.; Sakaitini, M.; Otsuka, H.; Oda, M. Chem. Lett.
1985, 127.
(12) Metal-Catalyzed Cross-Coupling Reactions, 2nd ed.; de Meijere,
A., Stang, P. J., Eds.; Wiley-VCH: Weinheim, Germany, 2004.
(13) Mixtures of both alkynyl and allenyl carboxylic acids are typically
observed in other coupling reactions. For example, see: Satoh, T.;
Nakamura, A.; Iriuchijima, A.; Hayashi, Y.; Kubota, K.-I. Tetrahedron
2001, 57, 9689.
(14) The use of cinnamyl chloride as the coupling partner resulted in
the targeted carboxylic acid in 18% yield as a 1:1 mixture of linear and
branched regioisomers. Significant amounts of dehalogenated and
homocoupling products were observed in the crude reaction mixtures.
(15) The formation of ∼10% α-methyl styrene via β-hydride
elimination was observed by GC analysis of the crude reaction mixture.
(16) (a) Krasovskiy, A.; Duplais, C.; Lipshutz, B. H. Org. Lett. 2010,
12, 4742. (b) Czaplik, W. M.; Mayer, M.; von Wangelin, A. J. Angew.
Chem., Int. Ed. 2009, 48, 607. (c) Krasovskiy, A.; Duplais, C.; Lipshutz,
B. H. J. Am. Chem. Soc. 2009, 131, 15592.
(17) See the Supporting Information for details.
(18) (a) Carmona, E.; Paneque, M.; Poveda, M. L. Polyhedron 1989,
8, 285. (b) Carmona, E.; Marin, J. M.; Paneque, M.; Poveda, M. L.
Organometallics 1987, 6, 1757.
(19) Matsubara, R.; Gutierrez, A. C.; Jamison, T. F. J. Am. Chem. Soc.
2011, 133, 19020.
(20) Wu, F.; Lu, W.; Qian, Q.; Ren, Q.; Gong, H. Org. Lett. 2012, 14,
3044.
(21) The reaction of 1b with CO2 (1 atm) utilizing 7 (10 mol %), Zn
(5 equiv), PCp3·HBF4 (10 mol %), and MgCl2 (2 equiv) in DMF
(0.50 M) delivered 2b in 61% isolated yield, thus reinforcing the
notion that 7 is catalytically competent in our protocol.
(22) Pryor, W. A.; Gu, J.-T.; Church, D. F. J. Org. Chem. 1985, 50,
185.
1224
dx.doi.org/10.1021/ja311045f | J. Am. Chem. Soc. 2013, 135, 1221−1224