news & views

PHOTOCATALYSIS

Three catalysts tango with olefins

Predictably activating C–H and C–C bonds for the synthesis of new materials remains a challenge within the
synthetic community. A ternary catalytic dance between radicals, metals and light may unlock this puzzle.

Thiago Reschützegger, Thomas Syphan and Gabriel dos Passos Gomes

T
he ability to forge C–C and C–X radical cycles; HAT catalysis, Ni catalysis The first two questions were tackled in
bonds lies at the heart of synthetic and photocatalysis (Fig. 1a). In preparation tandem by surveying the thermodynamics
organic chemistry. Chemists strive for the first step, the radical form of a CPA of four possible catalytic cycles involving
to unlock new transformations to make (chiral phosphoric acid) HAT catalyst is different Ni-species with special attention
molecules ranging from pharmaceuticals generated in the photocatalytic cycle. The to the oxidative addition step. In elucidating
to industrial catalysts and energy- reaction starts with the formation of an the potential cycles involving different
storage materials. However, selective allyl radical via H-atom abstraction from Ni species, a mechanistic picture arises
functionalization of the hydrocarbon the olefin substrate to the CPA radical with only one exergonic path with a
backbone of readily available chemical (Fig. 1a). In parallel, the Ni-catalytic cycle surmountable activation barrier at room
feedstocks remains elusive due to the starts by oxidative addition of an aryl temperature. Moreover, density functional
stability and similarity of C–H bonds. bromide to a Ni complex. The high-valence theory analyses showed that oxidative
Over the past two decades, many Ni-intermediate from the oxidative addition addition is the rate-limiting step in reference
strategies for directly functionalizing step traps the previously formed allyl to the Ni-catalytic turnover. Reductive
C–H bonds have been developed. For radical, with reductive elimination resulting elimination (the final step in the Ni-catalytic
instance, transition-metal catalysis has in the selective coupling of the aryl and allyl cycle) is key to answering the third question
been used to functionalize alkyl and allylic groups. Lastly, all three cycles are closed regarding the origins of regioselectivity,
C–H bonds, often with the assistance with the reduced organophotocatalyst, and with this aim the team compared
of directing groups1,2. However, these transferring an electron to an oxidized the geometry and energies of alternative
strategies are typically limited by the Ni-intermediate so that both return to their transition states for reductive elimination
necessity of using pre-activated starting respective initial state. leading to either terminal or internal olefin
materials. Previous efforts to solve these Such a complex mechanistic picture arylated product. Analysis reveals that
limitations have resorted to variations of demands a wealth of experimental the source of the regioselectivity of the
direct allylic arylation of olefins through evidence, which is reported by the team. reaction is rooted in a steric clash between
C(sp³)–H bond functionalization using iron First, radical trapping with TEMPO the olefin and the Ni-bipyridine ligand in
catalysis, ionic intermediates and harsh (2,2,6,6-tetramethylpiperidine-1-oxyl, the transition state for the internal product
reaction conditions3. A breakthrough came a radical quenching agent) completely (∆∆G‡ = +8.2 kcal mol–1; Fig. 1c).
with the shift from ionic mechanisms to inhibits the reaction, establishing that The approach of Glorius, Houk
radical couplings via the combination of the reaction is likely to proceed via a and co-workers is a mild, efficient
hydrogen atom transfer (HAT) reactions radical rather than an ionic intermediate. and straightforward way to modify
with photocatalysis4. Now, writing in Nature Kinetic isotope effect studies suggest that olefins and produce a range of value-
Synthesis, Glorius, Houk and co-workers the formation of the allyl radical using added molecules. In addition, a diverse
describe the combination of three catalytic HAT catalysis is the rate-determining functional group tolerance for both
cycles in one concerted reaction system, step of the overall reaction. Interestingly, coupling partners and precise regiocontrol
capable of synthesizing a variety of new photocatalysed Z→E olefin isomerization are achieved. In short, combining HAT-,
compounds by cross-coupling of allyl is shown to occur, but positional photoredox- and Ni-catalysis is an
radicals to aryl halides using a wide array of isomerization is unlikely, further excluding innovative solution to the age-old C–H
starting materials5. potential HAT events on the product. This bond functionalization problem.
First, optimization studies using 1-hexene observation illustrates that the reaction, This work overcame many challenges
and ethyl-4-bromobenzoate identified the at least with regard to the olefin system, faced by the concerted use of multiple
best conditions for the catalytic triad (Fig. avoids uncontrolled reactive intermediates catalysts6. Another ‘one-pot system’ where
1a). These optimal conditions were then in a radical cascade and thus prevents the such limitations are elegantly overcome is a
applied to a range of substrates tackling the unwanted proliferation of side products. living cell. Innumerable enzymatic catalysts
allylic moiety and the hetero(aryl) bromide, To further illuminate the mechanistic work in concert to facilitate the complex
showing a wide structural variety for intricacies, the team explored the reactions needed to maintain life. Taking
both coupling partners while maintaining mechanism for the Ni catalytic cycle using lessons from the cell — and perhaps using
good to high yields. Furthermore, the computational techniques. Three main its tools — to guide cutting-edge research
robustness of the approach is exemplified questions are probed: (1) Is Ni(0) or Ni(I) in the field of multi-catalysis seems to be an
by its application to the functionalization of the active coupling catalyst? (2) What is the informative and valuable endeavour7.
natural products and therapeutics (Fig. 1b). rate-determining step in the Ni-catalytic Many questions arise from this and other
This ternary catalytic process is proposed cycle? (3) What is the origin of the studies in the field of cooperative multi-
to proceed through three compatible regioselectivity observed experimentally? catalysis. For example, can a multi-catalytic
Nature Synthesis | VOL 1 | January 2022 | 13–15 | www.nature.com/natsynth 13
news & views

a
H
Hydrogen
atom
R1 transfer
• Toleration of different
Photo- functional groups
+
catalyst R1 • No persistent radicals
R2
Br • Requires mild conditions
Ni A • Consistent regioselectivity
R2 catalyst A = CH, N • > 50 examples
A

Ternary catalysis
Me
t
Bu
Me
N Br
O NTf Ni
P = CPA-H Me
O SH N Br
t Bu t Bu
N
t Bu
Mes
BF4

HAT catalytic cycle Ni catalytic cycle Photoredox catalytic cycle

Ni(III) Single
Oxidative electron
Ni addition
CPA H R1 transfer
*PC PC
CPA H
PC HAT Ni(I) Ni(II) Ni(n – 1)

CPA• Reductive Radical CPA•

R1
elimination addition PCr Ni(n)
Ni(III)

b c
Hetero(aryl) bromides Allylic substrates ΔΔG‡ = 0 kcal mol–1 ΔΔG‡ = 8.2 kcal mol–1
EtO2C R2 R2
General motif: Me General motif:
t Bu t Bu

7O
Me Me N Br N Br
76% O R1
42% Ni R1 Ni
COPh O H
H
7O N H N H
O O N O t Bu
H t
Bu H H
H
66%
Me 52% Steric clash
N 60%
52%

Fig. 1 | Design of a broadly applicable reaction for direct allylic C(sp³)–H arylation. a, General radical mechanism and optimal catalysts for the reaction.
First, interaction with an excited photocatalyst prepares the chiral phosphoric acid–hydrogen atom transfer (CPA–HAT) catalyst to generate the allyl radical
via the proposed radical mechanism. Subsequent radical trapping is accomplished by coordination with the Ni catalyst. Oxidative addition to the Ni complex
is followed by trapping of the allyl-radical intermediate. Thus, this catalyst undergoes the demonstrated transformations in oxidation state throughout the
catalytic cycle. Finally, a photoredox catalytic cycle is used to turnover both HAT and Ni catalysts. PC, photocatalyst; PC*, excited-state photocatalyst; PCr,
reduced-state photocatalyst; Tf, trifluoromethanesulfonyl; Mes, mesityl. b, The product variation shows the excellent synthetic scope of the reaction regarding
both the (hetero)aryl bromide and allylic substrates. c, DFT-computed Gibbs free energies for reductive elimination transition states leading to terminal or
internal aryl-olefin product. The experimentally observed regioselectivity for the terminal olefin is explained by the steric clash between the olefin and the
Ni-bipyridine ligand in the transition state for the internal product (∆∆G‡ = +8.2 kcal mol–1).

system be designed to target a stronger, Thiago Reschützegger   1, Thomas Syphan2 Twitter: @thiagoreschutz; @gabepgomes
✉e-mail: gabegomes@cmu.edu
vinylic C–H bond in the presence of its and Gabriel dos Passos Gomes   3,4 ✉
weaker, allylic counterpart without relying 1
Department of Chemical Engineering,
on olefin activation? That remains to be Federal University of Santa Maria, Published online: 12 January 2022
seen. But, for now, the Houk and Glorius Santa Maria, Brazil. 2Department of Chemistry, https://doi.org/10.1038/s44160-021-00018-9
team successfully capitalized on a broad Saint Vincent College, Latrobe, PA, USA.
foundation of research exploring the use of 3
Department of Chemistry, Carnegie Mellon References
1. Giri, R., Shi, B.-F., Engle, K. M., Maugel, N. & Yu, J.-Q. Chem. Soc.
multiple catalysts in a one-pot reaction in University, Pittsburgh, PA, USA. 4Department of
Rev. 38, 3242–3272 (2009).
what may become a powerful tool for the Chemical Engineering, Carnegie Mellon University, 2. Yang, K., Song, M., Liu, H. & Ge, H. Chem. Sci. 11,
allylic arylation of C(sp³)–H bonds. ❐ Pittsburgh, PA, USA. 12616–12632 (2020).

14 Nature Synthesis | VOL 1 | January 2022 | 13–15 | www.nature.com/natsynth

 news & views

3. Sekine, M., Ilies, L. & Nakamura, E. Org. Lett. 15, 5. Huang, H.-M., Bellotti, P., Chen, P.-P., Houk, K. N. & Glorius, F. 7. DeHovitz, J. S. et al. Science 369, 1113–1118 (2020).
714–717 (2013). Nat. Synth. https://doi.org/10.1038/s44160-021-00006-z (2022).
4. Cuthbertson, J. D. & MacMillan, D. W. C. Nature 519, 6. Sancheti, S. P., Urvashi, Shah, M. P. & Patil, N. T. ACS Catal. 10, Competing interests
74–77 (2015). 3462–3489 (2020). The authors declare no competing interests.

Nature Synthesis | VOL 1 | January 2022 | 13–15 | www.nature.com/natsynth 15