Copper-Catalyzed Carbamoylation of Terminal Alkynes with Formamides via Cross-Dehydrogenative Coupling

Article abstract of DOI:10.1021/acscatal.5b02881

An efficient approach for direct carbamoylation of terminal alkynes with formamides affording propiolamides has been developed by copper-catalyzed oxidative cross coupling of C(sp)-H and C(sp²)-H bonds in the presence of a pincer ligand with tw

Full text of DOI:10.1021/acscatal.5b02881

Letter
pubs.acs.org/acscatalysis
Copper-Catalyzed Carbamoylation of Terminal Alkynes with
Formamides via Cross-Dehydrogenative Coupling
Jin-Ji Wu,† Yinwu Li,† Hai-Yun Zhou,‡ A-Hao Wen,† Chu-Chu Lun,† Su-Yang Yao,† Zhuofeng Ke,*,†
and Bao-Hui Ye*,†
†MOE Key Laboratory of Bioinorganic and Synthetic Chemistry, School of Chemistry and Chemical Engineering, Sun Yat-Sen
University, Xigang Xilu 135, Guangzhou 510275, China
‡Instrumental Analysis and Research Center, Sun Yat-Sen University, Xigang Xilu 135, Guangzhou 510275, China
S
* Supporting Information
ABSTRACT: An efficient approach for direct carbamoylation
of terminal alkynes with formamides affording propiolamides
has been developed by copper-catalyzed oxidative cross
coupling of C(sp)-H and C(sp2)-H bonds in the presence of
a pincer ligand with two imidazolyl groups. The catalytic
reaction is compatible with diverse functional groups but
sensitive to the electronic effect of terminal alkyne and the steric effect of formamides. KIE study indicates the cleavage of the
carbamoyl C−H bond affording formamide radical is the rate-determining step.
KEYWORDS: carbonylation, cross-dehydrogenative coupling, copper-catalyzed, propiolamide, aminocarbonylation
Recently, a new carbonylation of terminal alkynes has been
he oxidative carbonylation of terminal alkynes leads to the
1
Tdirect synthesis of ynones, ynoates, and ynamides, which exploited on the basis of the hypervalent iodine(III) reagent
(Scheme 1c).5,6 However, the substrates need to be
are important intermediates for the synthesis of natural
products and heterocyclic compounds.2 Traditionally, synthetic
preactivated and functionalized. Moreover, ynones can also be
synthesized by the addition of alkynyl to aldehyde, followed by
oxidation of the propargyl alcohols.7 A more straightforward
method for carbonylation of terminal alkynes is cross-
dehydrogenative coupling (CDC) of terminal alkynes and
carbonyl hydrogens of aldehydes, formates, or formamides.1
Although oxidative coupling of aromatic C(sp2)-H and C(sp)-
H bonds has been well-documented,8 directly CDC of formyl
C(sp2)-H and unactived terminal alkyne C(sp)-H bonds has
never been reported.1 This may be attributed to the lower
reactivity of the formyl C−H bond than the carbonyl group and
decarbonylation in harsh condition. Therefore, metal-catalyzed
activation of formyl C−H bond mostly leads to the addition of
formyl C−H bonds into alkynes, namely, hydroacylation,9
hydrocarboxylation,10 hydroesterification,11 or hydrocarbamoy-
lation.12 On the other hand, formamides are multipurpose
reagents and widely used in chemistry.13 Recent studies on the
aminocarbonylation with DMF have demonstrated that direct
activation of the formamide C−H bond is possible.14 We are
thus interested in carbamoylation of terminal alkynes with
readily available formamides using a copper complex as a
catalyst to provide an efficient and straightforward synthetic
approach to propiolamides (Scheme 1d).
methods for these carbonylate compounds are Pd-catalyzed
acyl Sonogashira reaction,3 or carbonylative Sonogashira
coupling (Scheme 1).4 However, the acyl chlorides are
moisture-sensitive, unstable and commercially limited; CO is
a highly toxic gas and hard to handle in the laboratory.
Scheme 1. General Approaches to Carbonylattion of Alkynes
We initially examined the CDC reaction of formamides and
alkynes via model compounds of N,N-diethylformamide (DEF,
Received: December 17, 2015
Revised: January 15, 2016
© XXXX American Chemical Society
1263
DOI: 10.1021/acscatal.5b02881
ACS Catal. 2016, 6, 1263−1267

ACS Catalysis
Letter
a
1a) and phenylacetylene (2a) at 60 °C, using 5 mol % of CuBr
and TBHP (1.5 equiv) as cocatalyst with and without LiOtBu
(0.6 equiv) as a base (Table S1, entries 1 and 2). However, no
desired product but unreacted 2a was detected. To promote the
copper-catalyzed performance, ligands 2,9-dimethyl-1,10-phe-
nanthroline (DMPhen), 2,2′:6′,2″-terpyridine (TPy), 2,2′-
biimidazole (H2BIm), and 2,2′-bibenzimidazole (H2BBIm)
were added, respectively, to form the Cu complex. A negative
result was observed, but a small amount of the homocoupling
product 4 was detected. To our delight, when ligand 2,6-
bis(benzimidazol-2′-yl)pyridine (H2BIP) or 2,6-bis-
(benzimidazol-2′-yl)-4-hydroxypyridine (H3OBIP) was used,
the reaction was triggered, affording 81% or 74% yield (entries
7 and 8). To elucidate the differences between the ligands, 2,6-
bis(1′-methylbenzimidazol-2′-yl)pyridine (BMIP) and 2,6-bis-
(1′-methylbenzimidazol-2′-yl)-4-hydroxypyridine (HOBMIP)
derived from H2BIP and H3OBIP by methylation were
examined, the negative result (entries 9 and 10) indicated the
nature of tridentate-chelate ligand and N−H groups in
imidazolyl moieties play key roles in the reaction. This may
be attributed to the subtle property of imidazole ligand that
would modulate its coordination capability via deprotonation
and protonation, further mediating the reactivity of metal
center.15 The control experiments indicated that copper salt,
ligand, base, and oxidant are indispensable for the reaction (see
Table S1).
Table 1. CDC Reaction of 1a with Terminal Alkynes
We next turned our attention to screen base, copper salt, and
oxidant as well as temperature. Screening of bases indicated
that the optimal amount of LiOtBu was 0.6 equiv to 2a (see
Table S1). The strong base LiOH was equally optimal (78%,
Table S2), and KOtBu and NaOH were moderate (64% and
73%). The weak bases K2CO3, Na3PO4, and DABCO were
inferior in the reaction. NaOAc and urotropine led to no
conversion. Both Cu(I) and Cu(II) salts were active for the
reaction (Table S3). Among them, CuCl2 gave a higher yield of
3aa (83%). Then, we evaluated the effect of oxidant; H2O2 and
DTBP gave no desired product, whereas CHP provided a
moderate yield of 3aa (47%, Table S4). The reaction worked
well at 40 to 90 °C and gave a higher yield at 60 °C. In
addition, the amount of substrates and catalyst as well as
reaction time were also optimized (Table S5). The best ratio is
1a:2a:Cu(II):ligand:oxidant = 32:1:0.05:0.06:1.5 with 0.6 equiv
of LiOtBu. Finally, variation of the reaction time indicated that
the reaction was completed in 1 h under the optimal
conditions.
With the optimal conditions in hand, we investigated the
scopes and limitations of the reaction. Under the optimal
conditions, a variety of terminal arylalkynes (2a−2n, Table 1)
bearing an electron-rich methyl or methoxyl group afforded
3ab−3ae smoothly in yields of 74−90%. The high yield in
ortho-methyl substituted alkyne (3ac) indicates that steric effect
of the alkyne terminus is insensitive to the reaction. Bearing an
electron-deficient group such as 4-F, 4-Cl, 4-Br, 4-CN, or 3-
COOMe at the aromatic ring also afforded 3af−3aj in
moderate yields of 47−78% by extending the reaction time to
2 h. Functional groups such as aryl bromide, aryl nitrile, and
ester sensitive to metal-catalysis remained intact in the reaction,
which could be readily used as sources for further derivations.
However, when pyridinyl alkynes were used, the reaction
became sluggish, affording 28% (3ak) and 29% (3al) yields.
Moreover, diethynyl benzenes also afforded ditopic products
3am and 3an in moderate yields. Thus, the terminal alkynes
with various substituents on aromatic ring are compatible, and
a
b
c
Isolated yields. Reaction in 2 h. Diethynyl benzene is 0.5 mmol.
d
e
Reaction in 4 h. 3 equiv of LiOtBu and 12 h.
moderate to good isolated yields have been afforded under the
optimal conditions. Efficient reactivity was also observed with
aliphatic alkynes (2o−2x). We were pleased to find that
alkylalkynes with long aliphatic chains were viable to furnish
propiolamides (3ao−3ar) in moderate yields of 49−55% by
prolonging the reaction time. Importantly, alkylalkynes bearing
internal alkyne (2s and 2t), ester (2u), ether (2v and 2w), and
N-Boc-piperidine (2x) functional groups were tolerated in the
reaction with moderated yields, providing opportunities for
further additional modification of the products. Gratifyingly,
Si(iPr)3 protected alkyne (2y) was a suitable coupling partner
and delivered 3ay in 45% yield, which has potential applications
in further synthesis.
Then, we turned our attention to formamides (Table 2).
When DMF was used to react with terminal alkynes, the
corresponding products 3ba, 3bb, 3bc, 3bd, and 3bh were
isolated in moderate yields of 32−52%, which are significantly
lower than those of DEF. To disclose the reasons, full analysis
of the reaction products of DMF and 2a was performed. In
1264
DOI: 10.1021/acscatal.5b02881
ACS Catal. 2016, 6, 1263−1267

ACS Catalysis
Letter
Table 2. CDC Reaction of Phenylacetylene with Various
a
Formamides
to formamide, formation of a progargylic alcohol intermediate,
followed by oxidation to produce propiolamide was also
precluded. According to Journet and co-workers’ report on the
addition of terminal alkyne anion to DMF, the intermediate α−
aminoalkoxide may collapse to α,β-acetylenic aldehydes and
other byproducts.19 However, these products were not detected
in our experiment. Moreover, 4-fluorobenzaldehydes and 2,3,4-
trimethoxybenzaldehyde were used to couple with 2a under the
modified conditions (see SI). The desired ynones had yields of
10% and 34%, respectively, indicating the electron-withdrawing
substitution, which may promote the nucleophilic addition of
carbonyl group, has a negative influence on the reaction. These
results exclude the possibility of the nucleophilic mechanism.
Finally, the kinetic isotope effect (KIE, see eqn 3) experiment
was conducted giving the KIE = 3.9. The high KIE value (see
SI) indicates that the cleavage of formyl C−H bond is the rate-
determining step in the reaction.
a
b
Isolated yields. Reaction in 12 h.
addition to 3ba being afforded in 45% yield, which has been
confirmed by single-crystal analysis (see SI), 5ba was also
isolated in 17% yield as a byproduct, demonstrating that a
competition reaction involving alkynylation of C(sp3)-H bond
adjacent to the nitrogen atom occurred under the experiment
conditions.16 When N,N-di-n-propylformamide and N,N-di-n-
butylformamide were used, the reaction became sluggish, and
low yields of 3ca and 3da afforded concomitant with the
unreacted 2a. Extending the reaction time to 12 h gave 3ca and
3da in yields of 53% and 40%, respectively. The bulky groups at
formamides such as N,N-diisopropylformamide and N-methyl-
formanilide failed to couple with 2a under the optimal
conditions, indicating that the reaction is highly sensitive to
the steric effect of formamides. In addition, the cyclo-
formamide N-formylmorpholine was also examined to couple
with 2a, affording the desired product 3ea in a moderate yield
of 36%.
On the basis of our findings and the literatures, two proposed
pathways were depicted in Scheme 2. Under basic condition,
Scheme 2. Proposed Mechanism of the Copper-Catalyzed
CDC Reaction
To understand the reaction mechanism, several control
experiments were carried out. First, the reaction of 1a and 2a
was conducted in the presence of the radical scavenger
TEMPO. No desired product 3aa was detected, indicating
that a radical mechanism might be involved in the coupling
process, which is often observed in Cu-TBHP catalyzed
oxidation reactions.14 Second, the reaction might proceed via
aminocarbonylation of alkyne with CO and amine which are
generated by decomposition of formamides at high temper-
ature.17 A competition experiment was conducted. When the
reaction of DMF with 2a was performed in the presence of
excess diethylamine (see eqn 1), 3ba, 5ba, and 3aa would be
generated. However, no 3aa was detected, indicating that the
mechanism involving the generation of CO and amine in situ
could be ruled out. Third, the reaction might involve a two-step
procedure, addition of formamide to alkyne, followed by α-H
or β-H elimination.18 In this case, the alkene intermediate
would be observed when the hydrogen elimination process
failed to deliver. Thus, 1-phenyl-1-propyne was introduced to
react with DEF. No reaction was observed (see eq 2). Thus, the
mechanism involving electrophilic attack of formamide to
alkyne followed by hydrogen elimination can be ruled out. On
the other hand, the mechanism of nucleophilic attack of alkynyl
complex CuH2BIP might be deprotonated and transformed to
CuHBIP (I), or further deprotonated to CuBIP (II). Species II
may promote the reaction via a SET pathway. Species II
coordinates with the terminal alkyne with the aid of base,
forming an alkynyl adduct V. In the meantime, TBHP
dissociates into a hydroxyl radical and a butoxy radical through
a homolytic cleavage. Species V was then converted to Cu(III)
species IV through a SET process with the aid of the hydroxyl
radical.14b,c On the other hand, the formamide radical,
analogous to the acyl radicals, which has been postulated in
1265
DOI: 10.1021/acscatal.5b02881
ACS Catal. 2016, 6, 1263−1267

ACS Catalysis
Letter
aminocarbonylation of azoles, alkene or phenol with
formamides,14 was afforded by butoxy radical through H
atom abstraction. The formamide radical would react with the
intermediate IV,14e delivering the product 3aa and species II. It
is inferred that the Cu(III) species IV is a key intermediate for
the C−C bond formation, in which the deprotonated ligand
BIP2− play a key role in stabilization of Cu(III) ion.
Alternatively, a mechanism triggered from species I via a
PCET pathway could also be possible. Species I coordinates
with the terminal alkyne in the presence of base, forming an
alkynyl adduct III. Species III was then converted to the
Cu(III) species IV intermediate through a concerted PCET
process with the aid of the hydroxyl radical,14e,20 in which the
ligand HBIP− is further deprotonated into BIP2− to stabilize
Cu(III) ion. Similarly, the C−C bond formed via the reaction
of formamide radical and alkynyl ligand from species IV,
delivering product 3aa and species II. To regenerate the
catalytic cycle, species II will be converted to species I by
protonation.
substrate scope and excellent functional group compatibility.
Preliminary mechanism studies revealed that the cleavage of
carbamoyl C−H bond affording formamide radical is the rate-
determining step.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b02881.
■
S
Experimental procedures, characterization data, the
kinetic isotope effect experiment, and DFT calculation
(PDF)
Crystallographic data of 3ba (CIF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail for B.Y.: cesybh@mail.sysu.edu.cn.
*E-mail for Z.K.: kezhf3@mail.sysu.edu.cn.
■
To further insight into the proposed mechanisms, DFT
calculation investigation was carried out with the M06
functional (see SI). Potential energy surfaces for both SET
and PCET pathways are depicted in Figure 1 and S73. The
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Natural Science
Foundation of China (21272284, 21473261, and 21571195).
We thank Prof. Xiao-Dan Zhao at SYSU for his helpfull
discussion.
REFERENCES
■
(1) (a) Liu, Q.; Zhang, H.; Lei, A. Angew. Chem., Int. Ed. 2011, 50,
10788−10799. (b) Chinchilla, R.; Najera, C. Chem. Rev. 2014, 114,
1783−1826.
(2) Muller, T. J. J. Top. Heterocycl. Chem. 2010, 25, 25−94.
̈
(3) Boersch, C.; Merkul, E.; Muller, T. J. J. Angew. Chem., Int. Ed.
̈
2011, 50, 10448−10452.
(4) (a) Wu, X.; Neumann, H.; Beller, M. Chem. Soc. Rev. 2011, 40,
4986−5009. (b) Gadge, S. T.; Khedkar, M. V.; Lanke, S. R.; Bhanage,
B. M. Adv. Synth. Catal. 2012, 354, 2049−2056.
Figure 1. Potential free energy profiles of the copper-catalyzed CDC
reaction in the SET pathway (ΔH/ΔG are given in kcal/mol).
(5) Huang, H.; Zhang, G.; Chen, Y. Angew. Chem., Int. Ed. 2015, 54,
7872−7876.
(6) (a) Wang, H.; Xie, F.; Qi, Z.; Li, X. Org. Lett. 2015, 17, 920−923.
(b) Ai, W.; Wu, Y.; Tang, H.; Yang, X.; Yang, Y.; Li, Y.; Zhou, B. Chem.
Commun. 2015, 51, 7871−7874. (c) Liu, X.; Yu, L.; Luo, M.; Zhu, J.;
Wei, W. Chem. - Eur. J. 2015, 21, 8745−8749. (d) Ouyang, X.-H.;
Song, R.-J.; Wang, C.-Y.; Yang, Y.; Li, J.-H. Chem. Commun. 2015, 51,
14497−14500.
(7) (a) Harigae, R.; Moriyama, K.; Togo, H. J. Org. Chem. 2014, 79,
2049−2058. (b) Yuan, J.; Wang, J.; Zhang, G.; Liu, C.; Qi, X.; Lan, Y.;
Miller, J. T.; Kropf, A. J.; Bunel, E. E.; Lei, A. Chem. Commun. 2015,
51, 576−579. (c) Tang, S.; Zeng, L.; Liu, Y.; Lei, A. Angew. Chem., Int.
Ed. 2015, 54, 15850−15854.
(8) (a) Wei, Y.; Zhao, H.; Kan, J.; Su, W.; Hong, M. J. Am. Chem. Soc.
2010, 132, 2522−2523. (b) Matsuyama, N.; Kitahara, M.; Hirano, K.;
Satoh, T.; Miura, M. Org. Lett. 2010, 12, 2358−2361. (c) Yang, L.;
Zhao, L.; Li, C.-J. Chem. Commun. 2010, 46, 4184−4186. (d) Haro, T.
D.; Nevado, C. J. Am. Chem. Soc. 2010, 132, 1512−1513. (e) Kim, S.
H.; Yoon, J.; Chang, S. Org. Lett. 2011, 13, 1474−1477. (f) Jie, X.;
Shang, Y.; Hu, P.; Su, W. Angew. Chem., Int. Ed. 2013, 52, 3630−3633.
(g) Feng, C.; Loh, T.-P. Angew. Chem., Int. Ed. 2014, 53, 2722−2726.
(h) Shang, M.; Wang, H.-L.; Sun, S.-Z.; Dai, H.-X.; Yu, J.-Q. J. Am.
Chem. Soc. 2014, 136, 11590−11593. (i) Wu, W.; Su, W. J. Am. Chem.
Soc. 2011, 133, 11924−11927.
calculation revealed that the SET pathway is relatively more
plausible. The further deprotonation of species I to II is
thermodynamically favored (ΔG = −45.3 kcal/mol) in basic
condition. The terminal alkyne coordinates to species II after
depronation, forming the alkynyl intermediate V spontaneously
(ΔG = −9.9 kcal/mol). Then, the SET step transfers species V
to IV, which is only uphill by 18.6 kcal/mol. In the whole cycle,
the H atom abstraction by butoxy radical is the rate-
determining step, with an activation free energy of 31.3 kcal/
mol, relative to species V. This is in agreement with our
observation of the KIE experiment. The C−C bond formation
between formamide radical and alkynyl species IV is found to
be barrierless with a large driving force due to the high
reactivity of the Cu(III) species and formamide radical.15,21 On
the contrary, the PCET pathway encounters a highly
endergonic step in regeneration of species I from II (Figure
S73), which is expected to be less feasible in the basic
condition.
In summary, we have developed the first catalytic amino-
carbonylation of terminal alkynes with formamides through
CDC of C(sp)-H and C(sp2)-H bonds in the presence of
CuH2BIP complex as a catalyst, which provides a direct and
efficient approach to synthesize propiolamides with a broad
(9) (a) Willis, M. C. Chem. Rev. 2010, 110, 725−748. (b) Shi, S.;
Wang, T.; Weingand, V.; Rudolph, M.; Hashmi, A. S. K. Angew. Chem.,
Int. Ed. 2014, 53, 1148−1151. (c) Chen, Q.-A.; Cruz, F. A.; Dong, V.
M. J. Am. Chem. Soc. 2015, 137, 3157−3160.
1266
DOI: 10.1021/acscatal.5b02881
ACS Catal. 2016, 6, 1263−1267

ACS Catalysis
Letter
(10) Hou, J.; Xie, J.-H.; Zhou, Q.-L. Angew. Chem., Int. Ed. 2015, 54,
6302−6305.
(11) Park, E. J.; Lee, J. M.; Han, H.; Chang, S. Org. Lett. 2006, 8,
4355−4358.
(12) (a) Nakao, Y.; Idei, H.; Kanyiva, K. S.; Hiyama, T. J. Am. Chem.
Soc. 2009, 131, 5070−5071. (b) Fujihara, T.; Katafuchi, Y.; Iwai, T.;
Terao, J.; Tsuji, Y. J. Am. Chem. Soc. 2010, 132, 2094−2098.
(13) Ding, S.; Jiao, N. Angew. Chem., Int. Ed. 2012, 51, 9226−9237.
(14) (a) Ju, J.; Jeong, M.; Moon, J.; Jung, H. M.; Lee, S. Org. Lett.
2007, 9, 4615−4618. (b) He, T.; Li, H.; Li, P.; Wang, L. Chem.
Commun. 2011, 47, 8946−8948. (c) Kumar, G. S.; Maheswari, C. U.;
Kumar, R. A.; Kantam, M. L.; Reddy, K. R. Angew. Chem., Int. Ed.
2011, 50, 11748−11751. (d) Li, X.; Li, B.; You, J.; Lan, J. Org. Biomol.
Chem. 2013, 11, 1925−1928. (e) Barve, B. D.; Wu, Y.-C.; El-Shazly,
M.; Chuang, D.-W.; Cheng, Y.-B.; Wang, J.-J.; Chang, F.-R. J. Org.
Chem. 2014, 79, 3206−3214. (f) Yang, X.-H.; Wei, W.-T.; Li, H.-B.;
Song, R.-J.; Li, J.-H. Chem. Commun. 2014, 50, 12867−12869.
(g) Kumar, G. S.; Kumar, R. A.; Kumar, P. S.; Reddy, N. V.;
Kumar, K. V.; Kantam, M. L.; Prabhakar, S.; Reddy, K. R. Chem.
Commun. 2013, 49, 6686−6688. (h) Ali, W.; Rout, S. K.; Guin, S.;
Modi, A.; Banerjee, A.; Patel, B. K. Adv. Synth. Catal. 2015, 357, 515−
522.
(15) Hashiguchi, B. G.; Bischof, S. M.; Konnick, M. M.; Periana, R. A.
Acc. Chem. Res. 2012, 45, 885−898.
(16) (a) Li, Z.; Li, C.-J. J. Am. Chem. Soc. 2004, 126, 11810−11811.
(b) Dai, C.; Meschini, F.; Narayanam, J. M. R.; Stephenson, C. R. J. J.
Org. Chem. 2012, 77, 4425−4431.
(17) Wan, Y.; Alterman, M.; Larhed, M.; Hallberg, A. J. Org. Chem.
2002, 67, 6232−6235.
(18) Lei, J.; Wu, X.; Zhu, Q. Org. Lett. 2015, 17, 2322−2325.
(19) Journet, M.; Cai, D.; DiMichele, L. M.; Larsen, R. D.
Tetrahedron Lett. 1998, 39, 6427−6428.
(20) Wendlandt, A. E.; Suess, A. M.; Stahl, S. S. Angew. Chem., Int. Ed.
2011, 50, 11062−11087.
(21) Cheng, G.-J.; Song, L.-J.; Yang, Y.-F.; Zhang, X.; Wiest, O.; Wu,
Y.-D. ChemPlusChem 2013, 78, 943−951.
1267
DOI: 10.1021/acscatal.5b02881
ACS Catal. 2016, 6, 1263−1267