вход по аккаунту



код для вставкиСкачать
Cite This: ACS Catal. 2017, 7, 7957-7961
Nickel/Photoredox-Catalyzed Amidation via Alkylsilicates and
Shuai Zheng, David N. Primer, and Gary A. Molander*
Roy and Diana Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, 231 South 34th Street, Philadelphia,
Pennsylvania 19104-6323, United States
S Supporting Information
ABSTRACT: A nickel/photoredox, dual-catalyzed amidation
reaction between alkylsilicate reagents and alkyl/aryl isocyanates
is reported. In contrast to the previously reported reductive
coupling process, this protocol is characterized by mild reaction
conditions and the absence of a stoichiometric reductant. A
mechanistic hypothesis involving a nickel-isocyanate adduct is
proposed based on literature precedent and further validation by
experimental results.
KEYWORDS: amidation, radicals, nickel/photoredox dual catalysis, alkylsilicates, isocyanates
Scheme 1. Mechanistic Similarities between Reductive and
Photoredox Cross-coupling
eductive coupling has been a powerful tool for the
construction of a variety of structures, including those
accessed via Csp2−Csp3 cross-coupling.1 Compared to crosscoupling reactions utilizing organometallic reagents as
nucleophiles,2 reductive cross-coupling reactions generate
“radical nucleophiles” in situ from the corresponding halides,
significantly expanding the functional group tolerance for both
partners. Despite this clear advantage in substrate scope, the
need for a sacrificial amount of a stoichiometric reductant
(hydride sources, metals, etc.) is a systemic drawback regarding
atom economy and sustainability that limits the widespread
implementation of reductive couplings.
According to the most widely accepted catalytic cycle for
reductive coupling, these transformations proceed through a
putative single-electron transfer (SET) process involving radical
rebound oxidative addition at the metal center.3 Even though
both coupling partners in reductive cross-coupling are electrophiles, the mechanism of these transformations merges with
that of photoredox-mediated cross-coupling reactions, in which
a radical is generated via photoredox catalysis before oxidative
capture by the metal center, generating a Ni(III) intermediate
(I) that is common to both approaches (see Scheme 1).4
Indeed, photoredox catalysis has drawn a significant amount
of attention in recent years. The ability to replace difficult, twoelectron processes with discrete, successive, single-electron
transfer events results in methods with far greater control of
reaction rate.5 Furthermore, these approaches often succeed in
diminishing radical chain processes by obviating the requirement for stoichiometric radical generation. Recently, our group
reported the merging of such photoredox processes with Ni
cross-coupling reactions, utilizing oxidatively generated radical
Given the mechanistic similarities between these two
emerging fields, we pondered whether lessons learned from
the reductive coupling literature could inform the design of new
© XXXX American Chemical Society
chemistry in the photoredox cross-coupling realm. In particular,
it seemed reasonable to explore whether other electrophilic
partners previously employed in reductive coupling processes
could be translationally applied to unexplored photoredox
systems.1c The main advantage here would be that the
transformation would be overall redox-neutral: the reduced
photocatalyst formed upon reductive quenching would turn
over the nickel catalytic cycle, obviating the need for a sacrificial
Several electrophiles have been incorporated within photoredox cross-coupling, including (pseudo)aryl halides,6a,c−e,7
alkenyl halides,8 acyl chlorides,9 acyl imides,10 and activated
carboxylic acids.11 However, unsaturated carbonyl-type electrophiles have not been explored extensively, with most efforts
focusing on CO2.12 Inspired by work from Martin et al. in the
formation of amides from alkyl halides and isocyanates
Received: August 17, 2017
Revised: October 20, 2017
DOI: 10.1021/acscatal.7b02795
ACS Catal. 2017, 7, 7957−7961
ACS Catalysis
Table 1. Optimization of Reaction Conditionsa
(Scheme 2a),13 we decided to explore the ability of isocyanates
to be engaged in a Ni/photoredox dual catalytic cycle to
generate amide groups.
Scheme 2. Previous Research and Present Projecta
Data taken from refs 13 and 16.
yieldb (%)
1a (0.1 mmol, 1.0 equiv), 1b (0.2 mmol, 2.0 equiv), photocatalyst
(1.5 mol %), [M] (2.5 mol %), ligand (2.5 mol %), DMF (1 mL, 0.1
M) at room temperature (rt) under blue LED irradiation. bHPLC
yield. cUsing diisopropylammonium cyclohexylbis(catecholato)silicate
as radical precursor. d4CzIPN = 1,2,3,5-tetrakis(carbazol-9-yl)-4,6dicyanobenzene. e[Ir] = {Ir[dF(CF3)ppy]2[bpy]}[PF6]. fIn darkness.
1.0 equiv of TEMPO was added. hUsing diethyl 4-cyclohexyl-2,6dimethyl-1,4-dihydropyridine-3,5-dicarboxylate as radical precursor.
Using cyclohexyl-BF3K as radical precursors. jdtbpe = 1,2-bis(di-tertbutylphosphino)ethane.
Amides are highly prevalent in biomedically related
molecules. In fact, over 30% of bond-forming reactions in
medicinal chemistry publications are amidations.14 Despite the
fact that amidation methods are well-represented, many are
predicated on sensitive reagents, harsh conditions, or
stoichiometric amounts of activating agents.15 As a representative example, in a contribution from Bode et al. (Scheme 2b),16
alkyl and aryl isocyanates reacted with Grignard reagents to
form sterically hindered amides. However, utilization of
organometallic reagents limits the functional group tolerance
of this approach (i.e., ketones, esters, etc., were not
incorporated within the nucleophile). Because Ni/photoredox
dual catalysis is known for its mild reaction conditions,
processes based on this mechanistic paradigm would be
expected to allow formation of the same bond without the
need to preform an organometallic reagent, thus overcoming
major limitations of traditional organometallic-based approaches.
To begin our investigations, we selected cyclohexyl bis(catecholato)silicate (1a) and phenethyl isocyanate (1b) as
radical partner and electrophile, respectively. Use of [Ru(bpy)3][PF6]2 as a photocatalyst and [(dtbbpy)Ni(H2O)4]Cl2
as the nickel precatalyst afforded the desired amide product in
46% yield (Table 1, entry 1). For organosilicates, the use of iPr2NH2+ as a counterion proved detrimental because of the
tendency of the associated conjugate base to form an N,Ndiisopropyl urea upon reaction with the isocyanates, resulting in
lower yields (Table 1, entry 2). This problem was circumvented
by altering the silicate counterion to Et3NH+ or Me3NH+, the
conjugate bases of which are non-nucleophilic. At this point, a
rigorous optimization effort was carried out, utilizing highthroughput experimentation.17 Unsurprisingly, the choice of
ligand for the Ni catalyst exerted a significant influence in the
reaction outcome. Although bidentate phosphine ligands
afforded poor yields (entries 4 and 16 in Table 1), with a
significant amount of radical homocoupling byproducts,
bispyridyl-type ligands (entries 1−3 in Table 1) gave better
results. It came to our attention that the use of the flexible
dtbbpy ligand afforded modest yields (entries 1 and 2), whereas
the previously employed 6-Mebpy13 gave a yield of only 15%
(entry 3 in Table 1). Given that substitution at the 6-position of
the bpy ligand should cause distortion of the backbone,13 this
suggested that a more rigid system might be beneficial for this
catalysis. In this vein, the use of phenanthroline resulted in
quantitative product formation, while successfully suppressing
the formation of off-cycle products (entry 5 in Table 1). To
improve the operational simplicity, the use of preformed
(Phen)NiCl2 led to similar results (entry 12 in Table 1).
Further optimization of catalyst loading indicated the feasibility
of reducing the catalyst loading to 2.5 mol % of Ni-precatalyst
and 1.5 mol % of the Ru-photocatalyst in DMF. Other radical
precursors, such as 4-alkyl-1,4-dihydropyridines and alkyltrifluoroborates were also attempted under the conditions
developed for the alkylsilicates (entries 14 and 15 in Table
1),6c,e however, significantly lower yields were observed using
these radical precursors.
Encouraged by these results, we explored the substrate scope.
As shown in Table 2, primary (2a), secondary (2b) and tertiary
(2c) isocyanate species all rendered comparable yields,
indicating the low influence of steric factors regarding the
isocyanate counterpart. Notably, even for a sterically demanding 2,6-diisopropylphenyl isocyanate (2h), the yield was not
compromised. When using aromatic isocyanates, only one
equivalent was required to render comparable yields (2n),
likely because of their enhanced stability under the reaction
conditions. Concerning the alkylsilicates, we observed that both
DOI: 10.1021/acscatal.7b02795
ACS Catal. 2017, 7, 7957−7961
ACS Catalysis
Table 2. Isocyanate Scope
Table 3. Silicate Scope
Reaction conditions: alkylsilicate (0.5 mmol, 1.0 equiv), isocyanate
(1.0 mmol, 2.0 equiv for alkyl isocyanates; 0.5 mmol, 1.0 equiv for aryl
isocyanates), DMF (5 mL, 0.1 M). bHNMe3+ was used as a
counterion. cNMR yield. d36 h is needed for a complete conversion.
Reaction conditions: alkylsilicate (0.5 mmol, 1.0 equiv), isocyanate
(1.0 mmol, 2.0 equiv for alkyl isocyanates, 0.5 mmol, 1.0 equiv for aryl
isocyanates), (phen)NiCl2 (0.0125 mmol, 2.5 mol %), [Ru(bpy)3][PF6]2 (0.0075 mmol, 1.5 mol %), DMF (5 mL, 0.1 M). bNMR yield.
Obtained via Tamao−Fleming oxidation of the corresponding
Si(OEt)3 species: 30% H2O2 (1 mL) KF (1.5 mmol, 3.0 equiv)
KHCO3 (1.5 mmol, 3.0 equiv), MeOH (5 mL), THF (5 mL), 0 °C.
Yield of gram-scale reaction.
addition reactions, allyl isocyanate afforded a good yield of 2e
without any evidence of radical addition.
Next, we sought to understand the intricacies of the reaction
pathway. First, control experiments were performed. Unsurprisingly, in the absence of photocatalyst or light, no conversion
was observed (Table 1, entries 10−13). By the addition of
TEMPO in the reaction (Table 1, entry 14), no product was
formed, which supported the radical pathway. The lack of
product formation without a nickel catalyst (Table 1, entry 10)
ruled out the direct radical addition pathway, suggesting a
necessary interaction between the isocyanate and the nickel
complex. Alternative metal catalysts (e.g., Fe, Co) and aprotic
silicate counterions only rendered trace amounts of product,
thus highlighting both the need for a proton source to turn over
the catalytic cycle and the unique role of nickel.17 Furthermore,
preliminary computational studies were carried out to enlighten
the interaction between isocyanate and (Phen)Ni0 complex.24
A barrierless oxidative addition of Ni(0) to the isocyanate
appears to be more favorable than the slightly uphill alkyl
radical addition to a Ni(0) center.4 Based on these results and
our previous understanding of photoredox/nickel dual
catalysis,4 a plausible mechanism was proposed (Scheme 3),
where Ni(II) carbonyl-amido intermediate 4b20,21 is formed
upon oxidative addition of Ni(0) to the isocyanate.
Subsequently, intermediate 4c will be generated upon radical
addition.4,6a The generated Ni(III) complex then undergoes
reductive elimination to generate the new C−C bond, followed
by protonation with the ammonium counterion. The resulting
Ni(I) complex is then reduced by Ru(bpy)3+ to turn over both
the nickel and photoredox catalytic cycles.22
Although attempts to synthesize complex 4b with bipyridine
supporting ligands failed because of the instability of this
complex, by mixing a stable Ni(0) source, Ni(COD)2, in the
presence of phenanthroline and 2,6-diisopropylphenyl iso-
primary and secondary alkylsilicates worked well, with slightly
higher yields using secondary alkylsilicates, probably as a
consequence of the higher stability of the intermediate radical.
A gram-scale reaction generating 2n was also carried out, and a
comparable yield of 69% was observed, indicating that this
reaction is scalable.
A variety of functional groups were well accommodated on
both the silicate and isocyanate coupling partners. According to
results presented in Tables 2 and 3, enolizable ketones (2i),
esters (2d, 3f, and 3i), ethers (3e), fluorides (3c, 3d), nitriles
(2l, 3a), and amides (3g) were accommodated in moderate to
good yields. These functional handles were not tolerated under
previously reported protocols.16 Isothiocyanate species could
also be employed in the reaction (2g), and heteroaromatic
systems were readily accommodated as well. Despite a longer
reaction time, the pyridyl moiety (3k) afforded a reasonable
yield. Notably, although both pyrrole (3h) and thiophene (2m)
ring systems are known for radical polymerization reactivity,18
the desired products were successfully isolated. The functional
group tolerance of this reaction also provides possibilities for
sequential functionalization. For example, an aryl chloride (2j)
was accessed in good yield, providing a handle for further
elaboration via traditional cross-coupling. The (EtO)3Si group
was also compatible, which led to installation of a hydroxy
group (2f) in 68% yield after a one-pot, Tamao−Fleming
oxidation.19 Although alkenyl moieties are vulnerable to radical
DOI: 10.1021/acscatal.7b02795
ACS Catal. 2017, 7, 7957−7961
ACS Catalysis
cyclohexyl radical generated during reaction. Notably, these
CO extrusion-derived side products further support the viability
of 4b and 4c as intermediates along the reaction pathway.
Overall, the proposed mechanism and experiments are
consistent with previous work by Martin, in which a reductive
coupling strategy was used.13,25 In their cases, coupling was
validated to occur via a radical pathway, using superstoichiometric zinc, both to generate the alkyl radical and to
reduce the nickel catalyst.
In conclusion, a nickel-catalyzed photoredox amidation of
alkylsilicates and aryl/alkyl isocyanates has been developed.
Under these mild conditions, this reaction utilizes a similar
pathway to that previously used in reductive couplings, but
eliminates the required stoichiometric reductant by using a
photocatalyst for both radical generation and nickel cycle
turnover. With a series of electrophilic functional groups
tolerated, this reaction is complementary to those afforded by
conventional methods, such as acylation of amines and
isocyanate alkylation via organometallic reagents. Most
excitingly, the recognition regarding the mechanistic similarities
between Ni/photoredox dual cross-coupling and nickel
reductive couplings can serve as a foundation for developing
new chemistry in both fields. As such, a closer examination of
the literature between these two related regimes can help
inform and allow the development of new transformations.
Scheme 3. Plausible Mechanisms of Nickel-Catalyzed
Photoredox Amidation
cyanate, we were able to observe rapid changes by NMR and
IR, corresponding to the disappearance of the isocyanate and
the formation of a new carbonyl peak. These appear to be
indicative of an oxidative addition intermediate. Indeed, the
corresponding product was observed by reaction of the in-situgenerated 5a with cyclohexylsilicate under photoredox
conditions (see Scheme 4a).20
Scheme 4. Mechanistic Investigation
S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.7b02795.
Experimental procedures, HTE data, mechanistic studies,
compound characterization data, and NMR spectra for all
compounds (PDF)
Corresponding Author
Shuai Zheng: 0000-0001-9085-4410
Gary A. Molander: 0000-0002-9114-5584
The authors are grateful for the financial support provided by
NIGMS (No. R01 GM 113878). We thank the NIH (No. S10
OD011980) for supporting the University of Pennsylvania
(UPenn) Merck Center for High Throughput Experimentation,
which funded the equipment used in screening efforts.
The plausibility of this proposal is also consistent with the
work by Hillhouse on the synthesis of related, isolable
(dtbpe)Ni(II)-DippNCO oxidative addition complexes.21,24
As noted in our optimization, the same, sterically bulky,
bisphosphino ligand was viable in the catalysis, affording a
moderate 30% yield of the desired coupled product (Table 1,
entry 16).23
Interestingly, during the reaction between the cyclohexylsilicate and diisopropylphenyl isocyanate, diisopropylphenyl urea (6a) was isolated (Scheme 4b). This off-cycle
product suggests a reaction pathway where CO extrusion from
4b results in the formation of a Ni-imido complex,20 which can
then react with another equivalent of isocyanate to generate the
observed urea.21 In addition, detection of dicyclohexyl ketone
6b in the GC-MS demonstrates the capture of CO by
The authors declare no competing financial interest.
We sincerely thank Dr. Alvaro Gutierrez-Bonet [University of
Pennsylvania (UPenn)] for helpful discussion. We thank
Professor Osvaldo Gutierrez (University of Maryland) for
preliminary computational results concerning the energy of the
proposed reaction intermediates. Dr. Charles W. Ross, III
(UPenn) is acknowledged for obtaining HRMS data.
(1) (a) Everson, D. A.; Weix, D. J. J. Org. Chem. 2014, 79, 4793−
4798. (b) Knappke, C. E. I.; Grupe, S.; Gärtner, D.; Corpet, M.;
Gosmini, C.; Jacobi von Wangelin, A. Chem.Eur. J. 2014, 20, 6828−
DOI: 10.1021/acscatal.7b02795
ACS Catal. 2017, 7, 7957−7961
ACS Catalysis
coordination intermediate. For plausibility of π-coordination complex,
see: Spessard, G. O.; Miessler, G. L. Organometallic Chemistry; Oxford
University Press: New York, 2010; pp 299−313.
(25) Correa, A.; Martin, R. J. Am. Chem. Soc. 2014, 136, 7253−7256.
6842. (c) Moragas, T.; Correa, A.; Martin, R. Chem. - Eur. J. 2014, 20,
8242−8258. (d) Weix, D. J. Acc. Chem. Res. 2015, 48, 1767−1775.
(e) Gu, J.; Wang, X.; Xue, W.; Gong, H. Org. Chem. Front. 2015, 2,
(2) (a) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem. Rev. 2011, 111,
1417−1492. (b) Cherney, A. H.; Kadunce, N. T.; Reisman, S. E. Chem.
Rev. 2015, 115, 9587−9652.
(3) (a) Everson, D. A.; Shrestha, R.; Weix, D. J. J. Am. Chem. Soc.
2010, 132, 920−921. (b) Biswas, S.; Weix, D. J. J. Am. Chem. Soc.
2013, 135, 16192−16197.
(4) Gutierrez, O.; Tellis, J. C.; Primer, D. N.; Molander, G. A.;
Kozlowski, M. C. J. Am. Chem. Soc. 2015, 137, 4896−4899.
(5) (a) Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc. Rev.
2011, 40, 102−113. (b) Prier, C. K.; Rankic, D. A.; MacMillan, D. W.
C. Chem. Rev. 2013, 113, 5322−5363.
(6) (a) Tellis, J. C.; Primer, D. N.; Molander, G. A. Science 2014, 345,
433−436. (b) Zuo, Z.; Ahneman, D. T.; Chu, L.; Terrett, J. A.; Doyle,
A. G.; MacMillan, D. W. C. Science 2014, 345, 437−440. (c) Primer,
D. N.; Karakaya, I.; Tellis, J. C.; Molander, G. A. J. Am. Chem. Soc.
2015, 137, 2195−2198. (d) Jouffroy, M.; Primer, D. N.; Molander, G.
A. J. Am. Chem. Soc. 2016, 138, 475−478. (e) Gutiérrez-Bonet, Á .;
Tellis, J. C.; Matsui, J. K.; Vara, B. A.; Molander, G. A. ACS Catal.
2016, 6, 8004−8008.
(7) Patel, N. R.; Molander, G. A. J. Org. Chem. 2016, 81, 7271−7275.
(8) Patel, N. R.; Kelly, C. B.; Jouffroy, M.; Molander, G. A. Org. Lett.
2016, 18, 764−767.
(9) (a) Amani, J.; Sodagar, E.; Molander, G. A. Org. Lett. 2016, 18,
732−735. (b) Amani, J.; Molander, G. A. J. Org. Chem. 2017, 82,
(10) Amani, J.; Alam, R.; Badir, S.; Molander, G. A. Org. Lett. 2017,
19, 2426−2429.
(11) Amani, J.; Molander, G. A. Org. Lett. 2017, 19, 3612−3615.
(12) Shimomaki, K.; Murata, K.; Martin, R.; Iwasawa, N. J. Am. Chem.
Soc. 2017, 139, 9467−9470.
(13) Serrano, E.; Martin, R. Angew. Chem., Int. Ed. 2016, 55, 11207−
(14) Brown, D. G.; Boström, J. J. Med. Chem. 2016, 59, 4443−4458.
(15) (a) Bariwal, J.; Van der Eycken, E. Chem. Soc. Rev. 2013, 42,
9283−9303. (b) Ruiz-Castillo, P.; Buchwald, S. L. Chem. Rev. 2016,
116, 12564−12649. (c) de Figueiredo, R. M.; Suppo, J.-S.; Campagne,
J.-M. Chem. Rev. 2016, 116, 12029−12122. (d) De Risi, C.; Pollini, G.
P.; Zanirato, V. Chem. Rev. 2016, 116, 3241−3305.
(16) Schäfer, G.; Matthey, C.; Bode, J. W. Angew. Chem., Int. Ed.
2012, 51, 9173−9175.
(17) Schmink, J. R.; Bellomo, A.; Berritt, S. Aldrichimica Acta 2013,
46, 71−80. See the Supporting Information for further details.
(18) (a) Sadki, S.; Schottland, P.; Brodie, N.; Sabouraud, G. Chem.
Soc. Rev. 2000, 29, 283−293. (b) Yagci, Y.; Jockusch, S.; Turro, N. J.
Macromolecules 2007, 40, 4481−4485.
(19) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organometallics
1983, 2, 1694−1696.
(20) See the Supporting Information for further details. The
considerably lower yield was most likely due to the instability of the
nickel-isocyanato complex. See: Braunstein, P.; Nobel, D. Chem. Rev.
1989, 89, 1927−1945.
(21) Mindiola, D. J.; Hillhouse, G. L. Chem. Commun. 2002, 1840−
(22) Alternatively, a migratory-insertion pathway is also plausible. See
the Supporting Information for more details. See: (a) De Sarkar, S.;
Ackermann, L. Chem.Eur. J. 2014, 20, 13932−13936. (b) Beutner,
G. L.; Hsiao, Y.; Razler, T.; Simmons, E. M.; Wertjes, W. Org. Lett.
2017, 19, 1052−1055.
(23) The higher yield observed with the phenanthroline ligand might
also be attributed to its diminished electron-donating capacity. See:
(a) Van Hecke, G. R.; Horrocks, W. D. Inorg. Chem. 1966, 5, 1960−
1968. (b) Pörschke, K.; Pluta, C.; Proft, B.; Lutz, F.; Krüger, C. Z.
Naturforsch., B: J. Chem. Sci. 1993, 48, 608−626.
(24) See the Supporting Information for further details about
preliminary computation result between oxidative-addition and π7961
DOI: 10.1021/acscatal.7b02795
ACS Catal. 2017, 7, 7957−7961
Без категории
Размер файла
1 139 Кб
7b02795, acscatal
Пожаловаться на содержимое документа