Nickel-Catalyzed Activation of Acyl C-O Bonds of Methyl Esters

Article abstract of DOI:10.1002/anie.201511486

We report the first catalytic method for activating the acyl C-O bonds of methyl esters through an oxidative-addition process. The oxidative-addition adducts, formed using nickel catalysis, undergo in situ trapping to provide anilide products. DFT calculations are used to support the proposed reaction mechanism, to understand why decarbonylation does not occur competitively, and to elucidate the beneficial role of the substrate structure and the Al(OtBu)₃ additive on the kinetics and thermodynamics of the reaction.

Full text of DOI:10.1002/anie.201511486

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201511486
German Edition: DOI: 10.1002/ange.201511486
Nickel Catalysis
À
Nickel-Catalyzed Activation of Acyl C O Bonds of Methyl Esters
Liana Hie, Noah F. Fine Nathel, Xin Hong, Yun-Fang Yang, Kendall N. Houk,* and
Neil K. Garg*
Abstract: We report the first catalytic method for activating the
À
acyl C O bonds of methyl esters through an oxidative-addition
process. The oxidative-addition adducts, formed using nickel
catalysis, undergo in situ trapping to provide anilide products.
DFT calculations are used to support the proposed reaction
mechanism, to understand why decarbonylation does not
occur competitively, and to elucidate the beneficial role of the
substrate structure and the Al(OtBu)3 additive on the kinetics
and thermodynamics of the reaction.
C
atalytic methods that rely on the activation of carbon–
heteroatom bonds have transformed the way chemists build
molecules of importance.[1] Although decades of research
have mainly focused on the coupling of halide and sulfonate
derivatives, particularly on aryl systems, recent efforts have
been put forth to catalytically activate functional groups that
have traditionally been considered inert in cross-coupling
reactions.[2] One such endeavor involves couplings of pivalate
esters that proceed by the nickel-mediated activation of aryl
Figure 1. Known methods for the catalytic activation of esters and our
approach for the activation of methyl esters (without decarbonylation).
C O bonds (Figure 1).[3] In contrast, the cleavage of the acyl
À
À
C O bond of esters remains underdeveloped. Seminal efforts
conversion of aryl methyl esters into anilides,[7–9] in addition to
computational insights.
À
in ester acyl C O bond cleavage include Yamamotoꢀs
stoichiometric studies of ester reactivity,[4] Itamiꢀs coupling
of phenolic esters, which proceeds with loss of the carbonyl
moiety in the form of CO,[5] and Chataniꢀs Suzuki–Miyaura
coupling of activated pyridyl esters.[6] To the best of our
knowledge, no transition-metal-catalyzed couplings of simple
esters, such as readily available methyl esters, have been
reported.
Our decision to pursue ester into anilide conversion was in
part driven by this transformation being the reverse of one
that we recently reported,[10] and it was therefore considered
to be both challenging and conceptually interesting. Methyl 1-
naphthoate (5) was selected as the substrate for our initial
studies (Figure 2).[11] We surveyed a range of reaction
With the aim of developing non-decarbonylative cou-
plings of simple esters using non-precious-metal catalysis, we
considered the sequence outlined in Figure 1. Ni-catalyzed
À
activation of the acyl C O bond of ester 1 would furnish
oxidative-addition adduct 3. Subsequent ligand exchange by
trapping with a nucleophile would provide acyl nickel species
4. Finally, reductive elimination would furnish product 2 and
regenerate the requisite Ni0 catalyst. Despite the simplicity of
this strategy and the abundance of methyl esters, no such
process has been discovered. Herein, we report the validation
of this approach, as demonstrated by the nickel-catalyzed
Figure 2. Conversion of ester 5 into amide 7. Ni(cod)2 =bis(1,5-
cycloactadiene)nickel(0), SIPr=1,3-bis(2,6-diisopropylphenyl)imidazoli-
din-2-ylidene.
parameters, including the choice of amine coupling partner,
ligand, solvent, temperature, concentration, and additives.
When N-methylaniline (6) was used as the coupling partner,
in conjunction with Ni/SIPr in toluene at 608C, only trace
amounts of amide product 7 was observed. This finding is
consistent with the overall reaction (i.e., ester 5+amine 6!
amide 7+methanol) being energetically uphill, which would
be expected based on our recent studies.[10] However, the
addition of Al(OtBu)3 was found to have a critical beneficial
effect and led to the formation of amide 7 in 89% yield.[12] As
described below, we propose that Al(OtBu)3 benefits the
[*] L. Hie, N. F. Fine Nathel, X. Hong, Y.-F. Yang, Prof. K. N. Houk,
Prof. N. K. Garg
Department of Chemistry and Biochemistry
University of California, Los Angeles
Los Angeles, CA 90095 (USA)
E-mail: houk@chem.ucla.edu
neilgarg@chem.ucla.edu
Homepage: http://www.chem.ucla.edu/houk
http://garg.chem.ucla.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201511486.
2810
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 2810 –2814

Angewandte
Communications
Chemie
reaction both kinetically and thermodynamically. Also, it
coupling partners, as noted earlier, no reaction occurred in the
should be emphasized that the reaction does not proceed in
absence of Ni(cod)2, SIPr, or Al(OtBu3). Therefore, these
results support the notion that nickel catalysis is indeed
the absence of Ni/SIPr.[13]
À
We next examined variations in both coupling partners
(Figure 3).[14] 1- And 2-naphthyl substrates bearing fluoride,
methoxy, and morpholino substituents were tolerated, as
shown by the formation of anilides 8–11, respectively. It is
notable that ortho substitution did not hinder reactivity and
operative in the methyl ester acyl C O bond-cleavage
process.
Given that decarbonylation is not observed in the nickel-
catalyzed conversion of esters into amides, we examined the
competing pathways that would stem from the putative
oxidative-addition intermediate 22 using DFT methods
(Figure 4).[18] Ligand exchange[19] to give 23 is thought to
occur through a two-step process, with a small barrier of
4.9 kcalmolÀ1 relative to oxidative-addition intermediate 22.
In contrast, the barrier for decarbonylation of 22 to give 25
Figure 3. Scope of methodology (R’=1-naphthyl). Bn=benzyl.
that the methoxy group did not undergo activation by nickel
under these reaction conditions. The reaction could also be
performed in the presence of a furan heterocycle to give
anilide 12. The coupling of a phenanthrene derivative
proceeded smoothly to furnish 13 in 74% yield. In contrast,
attempts to employ non-extended aromatic substrates were
less successful, as shown by the formation of 14 in only modest
yield. Extended aromatic substrates are frequently necessary
[2,15]
À
to enable nickel-mediated C O bond cleavage,
although
this effect is still not well understood.[16] With regard to the
aniline coupling partner,[17] N-benzyl- and N-butyl-substituted
anilines could be coupled, as shown by the formation of
amides 15 and 16, respectively. Substitution on the arene of
the aniline coupling partner was also well tolerated. For
example, methoxy- and fluoride-containing substrates under-
went the coupling reaction to give amides 17–19. An aniline
bearing a furan moiety could also be used, as demonstrated by
the formation of 20. Finally, the use of the cyclic aniline
derivative indoline gave the corresponding amide product 21
in 61% yield. Whereas this first-generation variation of our
new method requires the use of aryl esters and aniline
Figure 4. DFT calculations show the relative ease of ligand exchange
and reductive elimination compared to disfavorable decarbonylation
pathways. Al(OMe)3 is used as a model for Al(OtBu)3 and R=
1-naphthyl. Dipp=2,6-diisopropylphenyl.
Angew. Chem. Int. Ed. 2016, 55, 2810 –2814
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2811

Angewandte
Communications
Chemie
was calculated to be 17.0 kcalmolÀ1 relative to 22. Further-
more, we examined the activation barriers for reductive
elimination and decarbonylation of 23. The barrier for
reductive elimination to give 24 is 15.4 kcalmolÀ1 more
favorable than decarbonylation to give 26, which is consistent
with amide bond formation taking place. Moreover, the high
barriers for decarbonylation are consistent with prior compu-
tational studies.[18c] The transition states for oxidative addition
(TS1), ligand exchange (TS2), and reductive elimination
(TS3) are depicted in Figure 4 (see the Supporting Informa-
tion for the full computed catalytic cycle).
DFT calculations were also used to probe the beneficial
influence of the Al(OtBu)3 additive on the Ni-catalyzed ester
into amide conversion (Figure 5). Without the additive, the
Figure 6. Effects of distortion of the ester aluminum additive complex
on the thermodynamics of the amidation based on substrate.
Al(OMe)3 is used as a model for Al(OtBu)3.
distortion of the ester–Al(OR)3 complex from steric hin-
drance facilitates and controls the thermodynamics of the
amidation. In the ester–Al(OR)3 complexes, the carbonyl and
arene moieties are nearly co-planar in all cases (138 or 28) to
maintain conjugation. In the case of methyl 1-naphthoate,
steric repulsion between the naphthyl group and the acyl
moiety distorts the highlighted angle to 122.78, which is about
48 larger than the corresponding angles of the complexes with
methyl 2-naphthoate and methyl benzoate. This renders the
Al(OR)3 complex with methyl 1-naphthoate less stable than
the other two complexes.[24] The amide–Al(OR)3 complexes
are all relatively nonplanar, and each possesses a similar
C-C-C(O) angle of 122.3–122.78. This is due to reduced
arene–carbonyl conjugation; amide conjugation prevails, and
the arenes and attached carbonyl groups are easily twisted out
of planarity to minimize steric effects. Therefore, the stability
of the amide–Al(OR)3 complex is minimally affected by the
identity of the arene attached to the carbonyl group.[25] The
steric repulsion seen in the ester–Al(OR)3 complex of methyl
1-naphthoate makes reactions of these substrates thermody-
namically most favorable. This insight into ester destabiliza-
tion is expected to guide future reaction discovery efforts.
An attractive aspect of employing simple methyl esters in
this method is that esters are generally stable to a variety of
reaction conditions. As such, they are well suited for use in
multistep synthesis. To probe this feature, we conducted the
reaction sequence shown in Scheme 1. First, proline-derived
ester 30 was united with 31 in a Buchwald–Hartwig cou-
Figure 5. Effect of the additive on the thermodynamics of amidation
and the kinetic barrier for oxidative addition as determined by DFT
calculations. Al(OMe)3 is used as a model for Al(OtBu)3 and R=
1-naphthyl.
amidation of ester 5 with aniline 6 is endergonic by
4.9 kcalmolÀ1. However, upon addition of the aluminum
additive, the amidation becomes almost thermoneutral.[20]
This is due to the greater Lewis basicity of the carbonyl
oxygen atom of the amide compared to that of the ester,
which therefore drives the equilibrium towards amide com-
plex 28.[21] The additive is also thought to have a beneficial
kinetic influence with regard to the rate-determining oxida-
tive-addition step. In the absence of the additive, the kinetic
barrier for oxidative addition is computed to be 33.2 kcal
molÀ1 relative to [Ni(SIPr)2] 29.[22] With the additive, however,
the oxidative addition becomes significantly more facile, with
a kinetic barrier of 26.8 kcalmolÀ1.[23]
With insight into the beneficial role of the Al(OtBu)3
additive, we questioned why certain substrates performed,
whereas others proved more challenging in the nickel-
catalyzed amidation. Key results are shown in Figure 6.
Experimentally, methyl 1-naphthoate undergoes amidation in
higher yields than methyl 2-naphthoate and methyl benzoate
(89% vs. 53% and 15% yield, respectively). This agrees with
the computed trends in the Gibbs free energy for the
amidation of each substrate. Calculations reveal that the
pling.[26] This C N bond formation occurred smoothly without
À
disturbing either of the ester motifs. Treatment of the coupled
product with TFA led to selective tert-butyl ester cleavage to
give 32. This set the stage for sequential amide bond forming
2812
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 2810 –2814

Angewandte
Communications
Chemie
Environment (XSEDE), which is supported by the National
Science Foundation (OCI-1053575), as well as the UCLA
Institute of Digital Research and Education (IDRE). These
studies were also supported by shared instrumentation grants
from the NSF (CHE-1048804) and the National Center for
Research Resources (S10RR025631).
À
Keywords: C O activation · density functional calculations ·
methyl esters · nickel catalysis · oxidative addition
How to cite: Angew. Chem. Int. Ed. 2016, 55, 2810–2814
Angew. Chem. 2016, 128, 2860–2864
[1] a) L. Jiang, S. L. Buchwald in Metal-Catalyzed Cross-Coupling
Reactions, 2nd ed. (Eds.: A. Meijere, F. Diederich), Wiley-VCH,
Weinheim, 2004, pp. 699 – 760; b) H. C. Shen in Application of
Transition Metal Catalysis in Drug Discovery and Development:
An Industrial Perspective (Eds.: M. L. Crawley, B. M. Trost),
Wiley, Hoboken, 2012, pp. 25 – 96.
[2] a) S. Z. Tasker, E. A. Standley, T. F. Jamison, Nature 2014, 509,
299 – 309; b) T. Mesganaw, N. K. Garg, Org. Process Res. Dev.
2013, 17, 29 – 39; c) B. M. Rosen, K. W. Quasdorf, D. A. Wilson,
N. Zhang, A.-M. Resmerita, N. K. Garg, V. Percec, Chem. Rev.
2011, 111, 1346 – 1416; d) B.-J. Li, D.-G. Yu, C.-L. Sun, Z.-J. Shi,
Chem. Eur. J. 2011, 17, 1728 – 1759.
[3] For initial studies, see: a) B.-T. Guan, Y. Wang, B.-J. Li, D.-G. Yu,
Z.-J. Shi, J. Am. Chem. Soc. 2008, 130, 14468 – 14470; b) K. W.
Quasdorf, X. Tian, N. K. Garg, J. Am. Chem. Soc. 2008, 130,
14422 – 14423.
Scheme 1. Multistep synthesis using mild catalytic ester activation and
À
sequential site-selective C N bond-forming processes. BINAP=2,2’-
bis(diphenylphosphino)-1,1’-binaphthalene, dba=dibenzylideneace-
tone, DCC=N,N’-dicyclohexylcarbodiimide, HOBt=1-hydroxybenzo-
triazole, TFA=trifluoroacetic acid.
[4] T. Yamamoto, J. Ishizu, T. Kohara, S. Komiya, A. Yamamoto, J.
Am. Chem. Soc. 1980, 102, 3758 – 3764.
reactions. The first involved a conventional DCC coupling
with valine-derived amino ester 33 and furnished peptide 34.
With the methyl ester still intact, a nickel-catalyzed amidation
was performed to deliver dipeptide 35. The tert-butyl ester
was not disturbed in this process, and the stereochemical
integrity was preserved at both epimerizable centers. Aside
[5] a) K. Amaike, K. Muto, J. Yamaguchi, K. Itami, J. Am. Chem.
Soc. 2012, 134, 13573 – 13576; b) H. Xu, K. Muto, J. Yamaguchi,
C. Zhao, K. Itami, D. G. Musaev, J. Am. Chem. Soc. 2014, 136,
14834 – 14844; c) K. Muto, J. Yamaguchi, D. G. Musaev, K. Itami,
Nat. Commun. 2015, 6, 7508 – 7515.
[6] H. Tatamidani, F. Kakiuchi, N. Chatani, Org. Lett. 2004, 6, 3597 –
3599.
[7] For a review on amides, see: The Amide Linkage: Structural
Significance in Chemistry, Biochemistry and Materials Science
(Eds.: A. Greenberg, C. M. Breneman, J. F. Liebman), Wiley-
VCH, Weinheim, 2000.
[8] For reviews on catalyzed amide formation, see: a) C. L. Allen,
J. M. J. Williams, Chem. Soc. Rev. 2011, 40, 3405 – 3415; b) V. R.
Pattabiraman, J. W. Bode, Nature 2011, 480, 471 – 479.
[9] The conversion of carboxylic acids into amides is used com-
monly, but relies on the use of stoichiometric coupling agents,
which generates waste. Alternatives have been highly sought
after as described in the following select examples: a) C.
Gunanathan, Y. Ben-David, D. Milstein, Science 2007, 317,
790 – 792; b) J. W. Bode, S. S. Sohn, J. Am. Chem. Soc. 2007, 129,
13798 – 13799; c) H. U. Vora, T. Rovis, J. Am. Chem. Soc. 2007,
129, 13796 – 13797; d) L. U. Nordstrøm, H. Vogt, R. Madsen, J.
Am. Chem. Soc. 2008, 130, 17672 – 17673; e) R. M. Al-Zoubi, O.
Marion, D. G. Hall, Angew. Chem. Int. Ed. 2008, 47, 2876 – 2879;
Angew. Chem. 2008, 120, 2918 – 2921.
[10] L. Hie, N. F. Fine Nathel, T. K. Shah, E. L. Baker, X. Hong, Y.-F.
Yang, P. Liu, K. N. Houk, N. K. Garg, Nature 2015, 524, 79 – 83.
[11] As shown in Figure 6, theory predicts that substrate 5 should
prove most suitable.
[12] On gram scale, this coupling could be performed with 2.5 mol%
of Ni to give amide 7 in 50% yield; see the Supporting
Information for details.
À
from highlighting the mildness of the acyl C O bond
activation and illustrating the potential of esters as cross-
coupling partners, this sequence demonstrates that conven-
À
tional and new C N bond-forming methods can be strategi-
cally merged to build linkages in a predictable and chemo-
selective manner.
À
In summary, this study has established that the acyl C O
bonds of simple esters may be activated using nickel catalysis.
This finding is expected to prompt the further exploration of
simple esters in non-decarbonylative cross-coupling processes
that rely on non-precious-metal catalysis.
Acknowledgements
We are grateful to Boehringer Ingelheim, DuPont, Bristol-
Myers Squibb, the Camille and Henry Dreyfus Foundation,
the A. P. Sloan Foundation, the UCLA Gold Shield Alumnae,
the University of California, Los Angeles, and the NIH-
NIGMS (GM36700 to K.N.H.) for financial support. We are
grateful to the NIH (F31 GM101951-02, N.F.F.N.), Bristol-
Myers Squibb (L.H.), the ACS Division of Organic Chemistry
(L.H.), and the Foote family (L.H. and X.H.) for fellowship
support. Computations were performed with resources made
available by the Extreme Science and Engineering Discovery
[13] The use of other ligands (e.g., mono- and bidentate phosphines,
bidentate pyridyl, pybox, and many other N-heterocyclic car-
benes) in place of SIPr also led to no reaction or low conversions.
Angew. Chem. Int. Ed. 2016, 55, 2810 –2814
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2813

Angewandte
Communications
Chemie
IPr, however, can be used in place of SIPr to give comparable
yields of products.
[14] The mass balance in most reactions is unreacted starting
material.
[19] Related ligand-exchange processes presumably take place in the
nickel-catalyzed amination of methyl ethers; see: M. Tobisu, A.
Yasutome, K. Yamaka, T. Shimasaki, N. Chatani, Tetrahedron
2012, 68, 5157 – 5161.
[20] The variation between the calculated thermoneutral reaction
free energy and the observed yields may partially be attributed
to differences between the actual experimental conditions and
those used for the DFT calculations.
[15] a) H. M. Wisniewska, E. C. Swift, E. R. Jarvo, J. Am. Chem. Soc.
2013, 135, 9083 – 9090; b) Q. Zhou, H. D. Srinivas, S. Dasgupta,
M. P. Watson, J. Am. Chem. Soc. 2013, 135, 3307 – 3310;
c) B. L. H. Taylor, M. R. Harris, E. R. Jarvo, Angew. Chem. Int.
Ed. 2012, 51, 7790 – 7793; Angew. Chem. 2012, 124, 7910 – 7913;
d) D.-G. Yu, Z.-J. Shi, Angew. Chem. Int. Ed. 2011, 50, 7097 –
7100; Angew. Chem. 2011, 123, 7235 – 7238; e) D. G. Yu, B.-J. Li,
S.-F. Zheng, B.-T. Guan, B.-Q. Wang, Z.-J. Shi, Angew. Chem. Int.
Ed. 2010, 49, 4566 – 4570; Angew. Chem. 2010, 122, 4670 – 4674;
f) M. Tobisu, T. Shimasaki, N. Chatani, Angew. Chem. Int. Ed.
2008, 47, 4866 – 4869; Angew. Chem. 2008, 120, 4944 – 4947, and
references therein.
[21] Al(OtBu)3 may also serve to absorb the methanol that is
generated, thus promoting the forward reaction. The complex-
ation between Al(OMe)3 and methanol was calculated to be
exergonic by 11.0 kcalmolÀ1
.
[22] A nickel toluene/NHC complex can also be considered as the
resting stage of the catalyst; see: Y. Hoshimoto, Y. Hayashi, H.
Suzuki, N. Ohashi, S. Ogoshi, Organometallics 2014, 33, 1276 –
1282.
[23] For further discussion, see the Supporting Information.
[24] Distortion of bond lengths was also examined, but found to be
insignificant in all cases.
[16] The favorable reactivity seen with p-extended systems may be
related to pre-complexation of the nickel catalyst, as well as to
the thermodynamic factors discussed here.
[25] The twisting out of planarity has a minimal effect on the amide–
Al(OR)3 stability; rather, electron donation from the amide
nitrogen atom is the primary stabilizing factor.
[26] J. P. Wolfe, S. L. Buchwald, Tetrahedron Lett. 1997, 38, 6359 –
6362.
[17] The non-catalyzed reaction of anilines with ester 5 is sluggish
and not observed under our reaction conditions.
[18] For computational studies of nickel-catalyzed decarbonylative
ester couplings, see Refs. [5b,c] and: a) X. Hong, Y. Liang, K. N.
Houk, J. Am. Chem. Soc. 2014, 136, 2017 – 2025; b) Z. Li, S.-L.
Zhang, Y. Fu, Q.-X. Guo, L. Liu, J. Am. Chem. Soc. 2009, 131,
8815 – 8823; c) Q. Lu, H. Yu, Y. Fu, J. Am. Chem. Soc. 2014, 136,
8252 – 8260.
Received: December 10, 2015
Published online: January 25, 2016
2814
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 2810 –2814