Metal-to-Ligand Ratio-Dependent Chemodivergent Asymmetric Synthesis

Article abstract of DOI:10.1002/anie.202108617

Chemodivergent asymmetric synthesis was achieved by tuning the metal-to-ligand ratio in an organometallic catalytic system. Using N-(aroyloxy)phthalimide as the precursor of either an oxygen-centered aroyloxy radical or a nitrogen-centered phthalimidyl radical, enantioselective oxocyanation or aminocyanation of alkenes was achieved separately through a dual photoredox and copper catalysis. The metal-to-ligand ratio can exert chemoselective control while retaining the high enantiopurity of divergent products. Both reactions proceed efficiently with catalyst loading as low as 0.2 mol % and can be performed on a gram scale without loss of chemoselectivity or enantioselectivity. Chemodivergent asymmetric 1,5-aminocyanation or 1,5-oxocyanation of vinylcyclopropane can also be realized by this protocol. Mechanistic investigations involving electron paramagnetic resonance (EPR) experiments were performed to shed light on the stereochemical and chemodivergent results.

Full text of DOI:10.1002/anie.202108617

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Metal-to-Ligand Ratio-dependent Chemodivergent Asymmetric
Synthesis
Authors: Min Zheng, Ke Gao, Haitao Qin, Guigen Li, and Hongjian Lu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202108617
Link to VoR: https://doi.org/10.1002/anie.202108617

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
Metal-to-Ligand Ratio-Dependent Chemodivergent Asymmetric
Synthesis
Min Zheng, [a] Ke Gao, [a] Haitao Qin,[a] Guigen Li, [a, b] Hongjian Lu*[a]
[a]
M. Zheng, K. Gao, H. Qin, Prof. G. Li and Prof. H. Lu
Institute of Chemistry and BioMedical Sciences, Jiangsu Key Laboratory of Advanced Organic Materials,
School of Chemistry and Chemical Engineering
Nanjing University, Nanjing, 210093, China.
E-mail: hongjianlu@nju.edu.cn.
[b]
Prof. G. Li
Department of Chemistry and Biochemistry
Texas Tech University, Lubbock, Texas 79409-1061, United States.
Supporting information for this article is given via a link at the end of the document.
Abstract: The metal-to-ligand ratio is an important variable in the field
of asymmetric organometallic catalysis. An unsuitable metal-to-ligand
ratio can reduce the catalytic activity and/or the asymmetric induction.
Herein, we report a chemodivergent asymmetric synthesis achieved
by tuning the metal-to-ligand ratio in an organometallic catalytic
system. Using N-(aroyloxy)phthalimide as the precursor of either an
oxygen-centered aroyloxy radical or a nitrogen-centered phthalimidyl
radical, enantioselective oxocyanation or aminocyanation of alkenes
was achieved separately through a dual photoredox and copper
catalysis. It was found that the metal-to-ligand ratio can exert
chemoselective control while retaining the high enantiopurity of
divergent products. In addition to tolerating many functional groups
and working well with a wide range of alkenes, both reactions proceed
efficiently with catalyst loading as low as 0.2 mol% and can be
performed on a gram scale without loss of chemoselectivity or
enantioselectivity. Chemodivergent asymmetric 1,5-aminocyanation
or 1,5-oxocyanation of vinylcyclopropane can also be realized by this
protocol. Mechanistic investigations involving electron paramagnetic
resonance (EPR) experiments were performed to shed light on the
stereochemical and chemodivergent results. The reactions deliver a
diverse range of optically active compounds and also provide a new
scenario which allows optimization of the reaction conditions in
asymmetric organometallic catalysis.
the development of novel synthetic methods and strategies
involving the construction of C−N and C−O bonds is the focus of
intense research efforts. Radical reactions have been going
through a renaissance due to their distinctive selectivity, high
reactivity and functional group tolerance,[4] and are being used
increasingly as conventional synthetic tools, complementary to
ionic reactions.[5] In this context, N-centered radicals are often
considered to be irreplaceable intermediates in the construction
of C−N bonds.[6] Progress involving N-centered radicals however
is slow when compared with carbon-centered radicals because of
a dearth of convenient access to them and a lack of awareness
of their reactivity.[7] Compared with N-centered radicals whose
reactivity can be modified by two substituent groups, O-centered
radicals, such as alkyloxy[8] and acroyloxy[9] radicals are more
electrophilic and reactive, and this leads to difficulties in
controlling their reaction selectivity.[10] The construction of C–O
bonds based on O-centered radicals has been less explored and
the development of
a general method for intermolecular
asymmetric reactions involved O-centered radicals remains an
major challenge.
Introduction
Organometallic catalysts are important in both academic and
industrial chemistry[1] and can improve reaction efficiency, tune
reaction selectivity and facilitate new reaction pathways with
fundamentally different mechanisms. The metal itself is often the
focus of a catalytic reaction, but the ligand is also a key factor
which controls the reactivity and selectivity, especially in
asymmetric catalysis.[2] The metal-to-ligand ratio in such
situations is an important variable and its optimization is normally
pursued in the field of asymmetric catalysis.[3] Excess ligand
occupies the reactive site of the metal center and can reduce the
catalytic activity. On the other hand extra metal is generally
avoided since the background reaction and side reactions can be
induced by a ligand-free metal, leading to low enantioselectivity
and yield. Consequently, a slight excess of ligand is generally
used to facilitate the reactivity and enantioselectivity in
asymmetric catalysis.[2]
Scheme 1. Metal-to-ligand ratio-dependent chemodivergent asymmetric
synthesis (this work).
The difunctionalization of alkenes involving heteroatom-
centered radicals is one of most important methods to construct
heteroatom-containing compounds.[11] The regioselectivity of
these reactions generally follows the anti‐Markovnikov rule,[12]
and complements the ion chemistry which follows the
Markovnikov rule. In this report, we disclose a metal-to-ligand
ratio-dependent generation of either an O-centered aroyloxy
radical or an N-centered phthalimidyl radical from N-(aroyloxy)-
phthalimides (ArCO2NPhth) via merged copper and photoredox
catalysis,[13,14] leading to enantioselective oxocyanation and
aminocyanation of alkenes respectively (Scheme 1). The most
important factor controlling the chemoselectivity of the reaction is
Because compounds containing oxygen and/or nitrogen are
common in biologically active molecules and functional materials,
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry
x (mol%)/y (mol%)
CH3CN (mL)
yield (1c + 1d)
1c/1d
1
1/1
2
2/1
3
5/1
4
7/1
5a,b
5/1
6
1/2
7
1/5
8
1/7.5
7
9b
1/7.5
7
7
7
7
7
2
7
7
68%
1/2.1
95%|83%
77%
1.1/1
95%|85%
84%
3.9/1
95%|84%
83%
5.4/1
95%|85%
83%c
> 14/1
96%|--
62%
1/5.0
97%|87%
62%
1/6.9
97%|87%
62%
1/20
--|90%
59%d
< 1/20
--|91%
ee (1c)|ee (1d)
Reaction conditions: 1a (0.2 mmol), 1b (0.5 mmol), TMSCN (0.3 mmol), Cu(CH3CN)4BF4 (x mol%), L (y mol%), PC (3DPA2FBN, 2,4,6-Tris(diphenylamino)-3,5-
o
difluorobenzonitrile, 1.0 mol%), CH3CN (7 mL), blue LEDs (24 W), Argon, at 5 C, 12 h. Yield determined by crude NMR with CH2Br2 as an internal standard. Ee
values were determined by HPLC on a chiral stationary phase. a CH3CN (2 mL). b PC (0.5 mol%). c Isolated yield of 1c. d Isolated yield of 1d. e Ratio of 1c/1d.
Scheme 2. Reaction discovery and optimizing conditions
indeed the metal-to-ligand ratio which, if tuned, can provide a
straightforward chemodivergent asymmetric synthetic route.[15]
designed NHP esters. In the other hand, Liu et al.[19] and other
groups[20] recently developed a series of elegant copper-catalyzed
asymmetric radical cyanation reactions.[21] The optically pure
organonitriles that are obtained are present in many bioactive
compounds and therapeutic agents[22] but can also be easily
transformed to other useful functional groups.[23] The aromatic
NHP ester (N-(aroyloxy)-phthalimide, ArCO2NPhth) 1a derived
from pentafluorophenyl carboxylic acid was initially selected as an
amino radical source for the development of radical-involved
asymmetric 1,2-aminocyanation of alkenes shown in Scheme 2
(for details of the optimization, see Table S1 in Supporting
Information-SI).[24] The reaction was first performed with a copper
to ligand L ratio of 1/1 and photoredox catalysis (Scheme 2A,
entry 1). The oxocyanation product 1c was obtained together with
the desired aminocyanation product 1d, suggesting that both the
aroyloxy radical and the phthalimidyl radical may be present
under the reaction conditions. This result conflicts with the report
that under visible-light photocatalysis, 1a generates a phthalimidyl
radical exclusively.[17e] Interestingly, upon increasing the loading
of copper from 1 mol% to 7 mol% and maintaining the loading of
L at 1 mol%, the ratio of 1c/1d was increased to 5.4/1 (entries 2-
4). Upon reducing the volume of the solvent from 7 mL to 2 mL
and loading of the photocatalyst (PC) from 1 mol% to 0.5 mol%,
the 1c/1d ratio reached 14/1 and 1c could be isolated in 83% yield
Results and Discussion
The limited number of easily accessible, suitable reactive and
reliable radicals is a major issue which has limited the
development of radical chemistry.[4,5] Recently, because of the
mild reaction conditions involved, radicals generated from visible-
light induced single electron reduction of redox activated esters
such as N-hydroxyphthalimide esters (NHP esters, N-
(acyloxy)phthalimides) have received more attention.[16] The
reduction of aliphatic NHP esters generates alkoyloxy radical
intermediates which decompose to alkyl radicals through an
intrinsic decarboxylation process, leading to a series of alkylation
reactions.[17] Using NHP esters derived from pentafluorophenyl or
trifluoromethyl carboxylic acids, the reduction reactions result in a
phthalimidyl radical, which can lead to radical amination
processes.[18] The generation of acroyloxy or phthalimidyl radicals
is determined by the relative electronegativity of the acroyloxy
segment and the phthalimidyl fragment, and this limits the
production of target radicals or requires the preparations of well-
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
a Conditions A: 1a (0.2 mmol), alkene 1-38b (0.5 mmol), TMSCN (0.3 mmol), CH3CN (7 mL), PC (3DPA2FBN, 0.5 mol%), Cu(CH3CN)4BF4 (1 mol%) and L (7.5
mol%), irradiated by blue LEDs (24 W) at 5 oC for 12 h. Conditions B: 1e (0.2 mmol), alkene 1-38b (0.5 mmol), TMSCN (0.3 mmol), CH3CN (2 mL), PC (3DPA2FBN,
0.5 mol%), Cu(CH3CN)4BF4 (5 mol%) and L (1 mol%), irradiated by blue LEDs (24 W) at 5 oC for 12 h. The isolated yield of the single isomer product, and ee and de
values were determined by HPLC on a chiral stationary phase. See SI for experimental details. b The 1f/1d ratio is 29/1, which is determined by 1H NMR of the reaction
mixture. c The reaction was conducted in 2.0 mmol scale of 1a or 1e. d With 1.5 equiv of olefin b. e Ee or de were not determined. f CH3CN (5 mL).
Scheme 3. Scope of oxocyanation and aminocyanation reactionsa
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
with 96% ee (entry 5). The 1c/1d ratio was decreased when
increasing the loading of ligand L (entries 6-8). When L was used
at 7.5 mol% (entries 8 and 9), the oxocyanation product 1c was
essentially absent and the aminocyanation product 1d was
isolated with 59% yield and 91% ee (entry 9). Excess ligand or
excess copper failed to influence the high enantioselectivity but
controlled the chemoselectivity significantly. Several control
experiments were performed (Table S1 in SI). In absence of light,
photocatalyst or copper, no desired products were achieved from
either of the aminocyanation and oxocyanation reactions.
Without the assistance of a ligand, only the oxocyanation product
1c could be obtained, in 25% yield. In the optimization of reaction
conditions, aryl C–H amination products N-(4-vinylphenyl)-
phthalimide and N-(2-vinylphenyl)-phthalimide were observed as
major side products in the aminocyanation reaction (Table S2 in
SI). Different NHP esters (ArCO2NPhth) derived from 3,5-
ditrifluoromethylphenyl, 4-fluorophenyl, 2,6-dichlorophenyl, 4-
trifluoromethylphenyl, and phenyl carboxylic acid were examined
(Table S3 in SI). The chemoselective oxocyanation products
were achieved exclusively with good to high yield (48-89%)
under conditions B, and the chemoselectivities of the reactions
under conditions A varied, depending on the aryl structure.
Several Box ligands were investigated, giving the selected
results in Scheme 2B (for details of screening ligands, see Table
S4 in SI) and it was found that the new serine-derived
bisoxazoline ligand L whose design was based on the Song’s
research[25] is the best ligand in both the aminocyanation and
oxocyanation reactions. The ester group in the serine-derived
BOX scaffold can further stabilize the copper species and
increase the rigidity of the transition state,[25] which may improve
the efficiency and enantioselectivity in both the aminocyanation
and oxocyanation reactions (Table S4 in SI).
yield; 23d, yield undetermined) were obtained. In addition to the
styrene derivatives, heteroaryl alkenes, such as 2-vinylthiophene
and 2-chloro-3-vinylpyridine were suitable substrates for both the
aminocyanation (24d, 33% yield, 93% ee; 25d, 56% yield, 40%
ee) and the oxocyanation reactions (24f, 41% yield, 91% ee; 25f,
44% yield, 85% ee). Various functional groups, such as alkyl (2b,
3b, 4b, 14b, 18b, 22b, 23b), chloromethyl (5b), phenyl (6b),
fluoro (7b, 15b, 19b), chloro (8b, 16b, 20b), bromo (9b, 17b,
21b), alkoxy (10b), aryloxyl (11b, 12b) and trifluoromethyl (13b)
are tolerated. For substrates containing additional olefinic groups
such as acryl (26b) or alkyl alkenyl (27b) groups, both
aminocyanation (26d, 40% yield, 75% ee; 27d 44% yield, 84%
ee) and oxocyanation (26f, 74% yield, 96% ee; 27f, 57% yield,
93% ee) of aryl olefins could be achieved selectively. The
vinylbenzenes derived from natural products such as an α-amino
acid (28b), camphorsulfonic acid (29b), estrone (30b) or
dehydrocholic acid (31b) went through the chemodivergent
asymmetric process smoothly, and both the aminocyanation (28-
31d, 38-61% yield) and oxocyanation products (28-31f, 64-78%
yield) were isolated with high diastereoselectivity, indicating that
this protocol may have significant potential in late stage
functional group transformations. In the reaction of β-methyl
styrene (32b), a single diastereomeric product (32d, 37% yield,
84% ee; 32f, 43% yield, 84% ee) could be isolated with moderate
yields and enantioselectivities. When cyclic 1,2-disubstituted
alkenes, such as indene (33b) and 1,2-dihydronaphthalene
(34b) were used as substrates, both the aminocyanation (33d,
44% yield; 34d, 30% yield) and oxocyanation products (33f, 78%
yield; 34f, 64% yield) were isolated but with little or no
enantioselectivity. Other type of alkenes, such as vinyl ethers
(35-36b) and unactivated alkyl alkenes (37b) were also
adequate substrates, providing the desired aminocyanation (35-
37d, 66-74% yield) and oxocyanation (35-37f, 56-85% yield)
products in reasonable yields although with no or low
enantioselectivity. Asymmetric 1,5-bifunctionalization still
With the optimal reaction conditions in hand, the scope of
alkenes was explored under conditions A with excess ligand, or
conditions B with excess copper (Scheme 3). Because the
pentafluorophenyl carboxylic acid derived product 1c
decomposed easily to 2-phenylacrylonitrile during the
purification, the 2,6-difluorophenyl carboxylic acid derived NHP
ester 1e was used in place of 1a when exploring the substrate
scope of the oxocyanation reactions. The reaction of 1e with
remains
a
challenge.[26] Recently, Wang et al. reported
enantioselective 1,5-cyanotrifluoromethylation of vinylcyclo-
propanes, which afforded easy access to useful chiral allyl
nitriles.[27] Inspired by these reactions, the radical probe
vinylcyclopropane 38b was used as a substrate. The desired 1,5-
aminocyanation product (38d, 44% yield, 94% ee) and the 1,5-
oxocyanation product (38f, 52% yield, 95% ee) were obtained
with high enantioselectivity under either conditions A or B. The
allyl amine or allyl alcohol substructures are common in bioactive
compounds and are frequently used as synthetic building blocks
in organic synthesis. The chiral center present in the allylic
position increases the utility of these allyl amines and allyl
alcohol. The absolute configurations of typical products were
confirmed by X-ray crystallographic analysis.[28]
Several experiments were performed in an effort to further
understand the reaction mechanism (Scheme 4 and Figures S1-
S2 in SI). Quenching experiments were performed and are
shown in Figure S1. The fluorescence of PC (3DPA2FBN) was
not quenched by either TMSCN, styrene, TMSOTf,
Cu(MeCN)4BF4[29] or a mixture of Cu(MeCN)4BF4 and the ligand
L. In contrast, the fluorescence was quenched by 1a or a mixture
of 1a and TMSOTf with defined Stern–Volmer kinetics. Under
conditions B, the side product TMSNPhth was detected by both
GCMS and HRMS (Figure S2 in SI). Electron paramagnetic
resonance (EPR) spectra were recorded using 5,5-dimethyl-1-
pyrroline-N-oxide (DMPO) for spin trapping (Scheme 4A).
styrene under conditions
B could produce the desired
oxocyanation product 1f with high yield (91%), high
enantioselectivity (95% ee) and high chemoselectivity (1f/1d,
29/1). Neither electron-rich or electron-deficient substituents in
the para- and/or the meta-position of styrene influenced the
reactions. Both aminocyanation products (1-18d, 42-76% yield,
83-94% ee) and those from oxocyanation (1-18f, 44-92% yield,
92-96% ee) were formed in good yields and with high
enantioselectivities. In the reactions of compounds with an ortho-
halogen, such as fluoro, chloro or bromo substituted styrenes,
the aminocyanation products were produced with moderate
enantioselectivities (19-21d, 57-61% yield, 67-85% ee) and
oxocyanation products were formed with high enantio-
selectivities (19-21f, 90-94% ee) and isolated in 64-84% yield.
Since the N-centered radical is a less electrophilic and reactive
species than an O-centered radical, sterically hindered
substrates might be not suitable for the aminocyanation reaction.
With ortho-methyl or ortho-dimethyl styrenes as examples, the
reactions produced the oxocyanation products with good yields
and high enantioselectivities (22f, 85% yield, 93% ee; 23f, 68%
yield, 95% ee) but little or no aminocyanation products (22d, 38%
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
Exp.
Sim.
Exp.
Sim.
3440
3460
3480
3500
3520
3440
3460
3480
B(G)
3500
3520
B(G)
Scheme 4. Mechanistic studies
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
In the reaction of 1a under conditions B, formation of the DMPO-
spin adduct 1g indicates that benzoyloxyl radical was selectively
generated.[30] In the reaction of 1e in the absence of copper and
TMSCN, the resulting spectra showed that the DMPO-spin adduct
1h is consistent with the trapping of the phthalimidyl radical. The
radical trapping reactions of 1a were performed using TEMPO as
a radical scavenger (Scheme 4B). Both two catalytic reactions
were inhibited by TEMPO, no 1c and 1d being observed by TLC
used an excess of L (Scheme 4F), indicating that the progress of
the reaction depends on a metal-to-ligand (1/1 ratio) species.[33]
1
and H NMR spectra. The TEMPO captured products 1i and 1j
were detected by HRMS under the conditions B, and only 1i was
observed under conditions A. This agrees with the observation
that almost no oxocyanation product 1c was produced in the
reaction of the NHP ester 1a with styrene (Scheme 2A, entry 9).
When the reaction was performed under the photoredox
conditions in the presence or absence of TMSCN (Scheme 4C),
the side diamination product 1k (38% and 6% respectively) was
obtained and is believed to have been produced from the cascade
phthalimidyl radical addition to styrene and a Ritter type reaction.
It is known that proton[31] or boron reagents[32] can drive
photoinduced electron transfer between photoexcited catalysts
and aliphatic NHP esters, and thus the cations present in this
reaction system may play a role similar to that of the proton or
boron reagents. Consequently, a catalytic amount of TMSOTf
was added to the photoredox reaction and a 1,2-hydroxylamino
product 1l was isolated but the alternative product 1k was not
observed. Compound 1l was thought to be produced from a Ritter
type reaction involving benzoyloxyl radical. When the reaction
was performed under conditions A with an additional 2.5 mol% of
TMSOTf, both the oxocyanation product 1c (18%) and the
aminocyanation product 1d (43%) were observed. The reaction
(entry 1 in Scheme 2A, metal/ligand = 1/1, total yield 68%, 1c/1d
= 1/2.1) with an additional 10 mol% of potassium phthalate, as a
sacrificial Lewis base to capture TMS+, was performed, no
oxocyanation product 1c was observed and the aminocyanation
product 1d was achieved in 41% yield. When mesitylene was
used in place of styrene as the radical acceptor (Scheme 4D).
Under the photoredox conditions in the presence of copper or
TMSCN, the amination product 1m which is thought to originate
from phthalimidyl radical, was formed. In contrast, the
acetoxylated product 1n was isolated in the presence of a
catalytic amount of TMSOTf while formation of 1m was not
observed. Thus, the evidence for the selective formation of
benzoyloxyl radical and phthalimidyl radical respectively under
different metal-to-ligand ratios appears strong. These results also
indicate that TMS+ plays a key role in the formation of the
benzoyloxyl radical. Only the phthalimidyl radical was formed in
the absence of TMS+. The aminocyanation and oxocyanation
products 33 and 34 from the reactions of cyclic 1,2-disubstituted
alkenes are racemic or minimally enantiomeric (Scheme 3). To
further understand the stereochemical outcomes of the reaction,
the substrate 1o was designed for intramolecular oxocyanation
reactions with different ligands (Scheme 4E, see details in Table
S5 in SI), and led to the desired product 1p which was formed
with moderate yield and almost no enantioselectivity (< 10% ee).
These results suggest that there is almost no asymmetric
induction during the formation of C−O or C−N bonds. The ee
values of the aminocyanation and oxocyanation products were
almost unchanged when altering the metal/ligand ratio from 1/7 to
7.5/1 (Scheme 2A). Furthermore, the ee of the chiral ligand L is
proportional to that of the product 1d under the conditions which
Scheme 5. Synthetic applications
Based on our experiments and earlier reports[20, 34], a proposed
mechanism was developed and is shown in Scheme 4G. Under
the conditions B with excess ligand (right cycle), the direct
oxidative quenching of the excited photocatalyst furnishes a
phthalimidyl radical which adds to the styrene to form a radical
intermediate A. The addition reaction of A with the ligand/CuII(CN)
complex E furnishes the high valent Cu(III) species B. Reductive
elimination of B yields the desired aminocyanation product 1d.
Under conditions A with excess Cu+ (left cycle), the reaction of
excess Cu+ and TMSCN initially generates the TMS+ ion which
may coordinate with the carbonyl group of 1a to form the complex
F, thus supporting an increased positive charge in the phthalimidyl
group. SET reduction of complex F generates an aroyloxy radical
which leads to the final oxocyanation reaction. The TMS+ ion can
be regenerated from the reaction of the ligand/Cu+ complex
species C with TMSCN, completing the Si+ activation cycle.[35]
Since an oxygen-centered radical is a very reactive species which
could abstract a hydrogen atom from the reaction system as a
side reaction pathway,[9] increasing the concentration of the
reaction benefits the desired intermolecular radical oxocyanation
reaction (entry 4 vs 5, Scheme 2A).
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
We studied the synthetic applications of the reactions (Scheme Acknowledgements
5). When gram scale preparations are conducted in the
oxocyanation reactions, the load of catalyst can be further
reduced to 1.25 mol% of copper, 0.25 mol% of ligand L and 0.2
mol% of the photocatalyst 3DPA2FBN (Scheme 5A), providing
the β-aroxyl nitriles 1c (2.62 g, 77% yield, 96% ee, from 1a) and
1f (2.44 g, 85% yield, 93% ee, from 1e) respectively. Similarly, for
the aminocyanation reactions, the catalyst loading of copper
could be reduced to 0.2 mol%, and the β-amino nitrile 1d is
formed in moderate yields with high enantioselectivity (1.55 g,
54% yield, 91% ee from 1a; 1.21 g, 44% yield, 91% ee from 1e).
For the 1,5-oxocyanation reaction on a gram scale (Scheme 5B),
0.5 mol% of L is enough to promote the reaction, providing 5-
aroxyl nitrile 38f in high enantioselectivity (0.88 g, 47% yield, 95%
ee and >20/1 (Z/E)), demonstrating the practicality of the method.
Reduction of the cyano group in 1d efficiently generated an
optically active 1,3-diamine 1q with different protecting groups
(Scheme 5C). Reduction of 1f provided the protected 3‐amino
alcohol 1r in good yield (Scheme 5D). Selective deprotection of
the aroyl group in 1r formed an alcohol 1s, which can couple with
an α-amino acid to generate the ester 1t. Oxidation of 1s provided
the β-amino α-chiral aldehyde 1u with some loss of the original
enantiopurity. This aldehyde 1u was attacked by a phenyl
Grignard reagent, a nucleophile, to provide 1,3‐amino alcohol 1v
with adjacent stereogenic centers. Since the fragments of 1,3‐
amino alcohols and 1,3-diamines are present in natural products
and many useful compounds, and are employed as useful
building blocks in organic synthesis, these chemical
transformations will extend the applications of this reaction.
Financial support for this work was provided by the National
Natural Science Foundation of China (21871131, 22071100).
Keywords: Radical • Asymmetric • Cyanation • Catalytic •
Chemodivergent
[1]
[2]
B. Cornils, W. A. Herrmann, M. Beller, R. Paciello, (Eds) Applied
homogeneous catalysis with organometallic compounds:
a
comprehensive handbook in four volumes. John Wiley & Sons, 2017.
a) E. N. Jacobsen, A. Pfaltz, H. Yamamoto, (Eds) Comprehensive
asymmetric catalysis. Springer Science & Business Media, 2003; b) R.
Noyori, Angew. Chem. Int. Ed. 2002, 41, 2008–2022; c) Sharpless, K. B.
Angew. Chem. Int. Ed. 2002, 41, 2024–2032.
[3]
[4]
a) R. Rasappan, M. Hager, A. Gissibl, O. Reiser, Org. Lett. 2006, 8,
6099–6102; b) M. Hager, S. Wittmann, A. Schätz, F. Pein, P. Kreitmeier,
O. Reiser, Tetrahedron: Asymmetry 2010, 21, 1194–1198; c) J. M.
Alderson, J. R. Corbin, J. M. Schomaker, Acc. Chem. Res. 2017, 50,
2147–2158; d) C. Weatherly, J. M. Alderson, J. F. Berry, J. E. Hein, J. M.
Schomaker, Organometallics 2017, 36, 1649-1661.
a) C. Chatgilialoglu, A. Studer, (Eds) Encyclopedia of Radicals in
Chemistry, Biology and Materials, John Wiley & Sons, 2012; b) D. P.
Curran, N. A. Porter, B. Giese, Stereochemistry of Radical Reactions:
Concepts, Guidelines, and Synthetic Applications. John Wiley & Sons,
2008; c) S. Z. Zard, (Ed) Radical Reactions in Organic Synthesis, Oxford
University Press, 2003; d) P. Renaud, M. Sibi, (Eds) Radicals in Organic
Synthesis, 1st ed., Wiley-VCH: Weinheim, 2001.
[5]
[6]
a) A. Studer, D. P. Curran, Angew. Chem., Int. Ed. 2016, 55, 58–102;
Angew. Chem. 2016, 128, 58–106; b) M. Yan, J. C. Lo, J. T. Edwards, P.
S. Baran, J. Am. Chem. Soc. 2016, 138, 12692–12714; c) D. Leifert, A.
Studer, Angew. Chem., Int. Ed. 2020, 59, 74–108; Angew. Chem., Int.
Ed. 2020, 132, 74–110; d) S. Crespi, M. Fagnoni, Chem. Rev. 2020, 120,
9790–9833.
Conclusion
a) X.-Y. Yu, Q.-Q. Zhao, Z.-J. Chen, W.-J. Xiao, J.-R. Chen, Acc. Chem.
Res. 2020, 53, 1066–1083; b) H. Jiang, A. Studer, CCS Chem. 2019, 1,
38–49; c) P. Xiong, H.-C. Xu, Acc. Chem. Res. 2019, 52, 3339–3350; d)
Y. Zhao, W. Xia, Chem. Soc. Rev. 2018, 47, 2591–2608; e) P. F. Kuijpers,
J. I. van der Vlugt, S. Schneider, B. de Bruin, Chem. - Eur. J. 2017, 23,
13819–13829; f) T. Xiong, Q. Zhang, Chem. Soc. Rev. 2016, 45, 3069–
3087; g) J.-R. Chen, X.-Q. Hu, L.-Q. Lu, W.-J. Xiao, Chem. Soc.
Rev. 2016, 45, 2044–2056; h) M. E. Wolff, Chem. Rev. 1963, 63, 55–64.
S. Z. Zard, Chem. Soc. Rev. 2008, 37, 1603–1618.
In summary, we have developed a novel strategy for
chemodivergent asymmetric synthesis through a dual photoredox
and copper catalysis. The reaction depends on the metal-to-
ligand ratio. Using N-(aroyloxy)phthalimide as the aroyloxy or
phthalimidyl
source,
enantioselective
oxocyanation
or
aminocyanation of alkenes can be independently realized by
tuning the metal-to-ligand ratio. In addition to tolerating many
functional groups and proceeding with a wide range of alkenes,
the reactions are highly efficient. They use a catalyst loading as
low as 0.2 mol% and can be performed at the gram scale without
losing efficiency or enantioselectivity. Mechanistic investigations
including EPR experiments, were performed to illustrate the
possible reaction pathway and the process of stereoinduction.
The organic photocatalyst serves either with excess copper to
generate an O-centered aroyloxy radical or with excess ligand to
produce the N-centered phthalimidyl radical. The copper/ligand
complex functions as an organometallic catalyst, installing a
cyano group in an enantiocontrolled manner. This protocol
provides unprecedented access to a diverse range of optically
active difunctionalized compounds (77 examples, up to 91% yield
and up to 97% ee), which has great potential in pharmaceutical
chemistry and natural product synthesis. The current study
reveals that the metal-to-ligand ratio can contribute substantially
to chemoselective control while retaining the high enantiopurity of
the divergent products, which provides a new scenario when
optimizing metal-to-ligand ratio in asymmetric catalysis.
[7]
[8]
a) E. Tsui, H. Wang, R. R. Knowles, Chem. Sci. 2020, 11, 11124–11141;
b) K.-F. Jia. Y.-Y. Chen. Chem. Commun. 2018, 54, 6105–6112; c) J.
Hartung, T. Gottwald, K. Špehar, Synthesis 2002, 11, 1469–1498.
X.-Q. Hu, Z.-K. Liu, Y.-X. Hou, Y. Gao, iScience 2020, 23, 101266.
[9]
[10] a) K.-F. Jia, F.-Y. Zhang, H.-C. Huang, Y.-Y. Chen, Visible-Light-Induced
Alkoxyl Radical Generation Enables Selective C(sp3)–C(sp3) Bond
Cleavage and Functionalizations. J. Am. Chem. Soc. 2016, 138, 1514–
1517; b) A.-H. Hu, J.-J. Guo, H. Pan, Z.-W. Zuo, Science 2018, 361, 668–
672; c) E. Tsui, A. J. Metrano, Y. Tsuchiya, R. R. Knowles, Angew.
Chem., Int. Ed. 2020, 59, 11845–11849; Angew. Chem. 2020, 132,
11943-11947; d) S. Shirase, S. Tamaki, K. Shinohara, K. Hirosawa, H.
Tsurugi T. Satoh, K. Mashima, J. Am. Chem. Soc. 2020, 142, 5668–5675.
[11] a) Z.-L. Li, G.-C. Fang, Q.-S. Gu, X.-Y. Liu, Chem. Soc. Rev., 2020, 49,
32–48; b) Q.-S. Gu, Z.-L. Li, X.-Y. Liu, Acc. Chem. Res. 2020, 53, 170–
181.
[12] A. L. Barthelemy, B. Tuccio, E. Magnier, G. Dagousset, Angew. Chem.,
Int. Ed. 2018, 57, 13790–13794; Angew. Chem. 2018, 130, 13986–
13990.
[13] a) A. Lipp, S. O. Badir, G. A. Molander, Angew. Chem., Int. Ed. 2021, 60,
1714–1726; Angew. Chem. 2021, 133, 1738–1750; b) A. Hossain, A.
Bhattacharyya, O. Reiser, Science 2019, 364, eaav9713; c) K. L. Skubi,
T. R. Blum, T. P. Yoon, Chem. Rev. 2016, 116, 10035–10074; d) R.
Brimioulle, D. Lenhart, M. M. Maturi, T. Bach, Angew. Chem., Int. Ed.
2015, 54, 3872–3890; Angew. Chem. 2015, 127, 3944–3963; e) C. K.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 2013, 113, 5322–
5363.
[21] a) F. Wang, P.-H. Chen, G.-S. Liu, Acc. Chem. Res. 2018, 51, 2036–
2046; b) W.-B. Wu, J.-S. Yu, J. Zhou, ACS Catal. 2020, 10, 7668–7690.
[22] F. F. Fleming, L.-H. Yao, P. C. Ravikumar, L. Funk, B. C. Shook, J. Med.
Chem. 2010, 53, 7902–7917.
[14] For recent developed photocontrolled cobalt catalysed selective
hydroboration of α,β-unsaturated ketones, see: a) F. Beltran, E.
Bergamaschi, I. Funes-Ardoiz, C. J. Teskey, Angew. Chem. Int. Ed. 2020,
59, 21176-21182; Angew. Chem. 2020, 132, 21362-21368; b) L. N.
Mendelsohn, C. S. MacNeil, L. Tian, Y. Park, G. D. Scholes, P. J. Chirik,
ACS Catal. 2021, 11, 1351-1360.
[23] Z. Rappoport, (Ed) Chemistry of the Cyano Group Wiley, 1970.
[24] a) H.-W. Zhang, W.-Y. Pu, T, Xiong, Y. Li, X. Zhou, K. Sun, Q. Liu, Q.
Zhang, Angew. Chem., Int. Ed. 2013, 52, 2529–2533; Angew. Chem.
2013, 125, 2589–2593; b) Y. Miyazaki, N. Ohta, K. Semba, Y. Nakao, J.
Am. Chem. Soc. 2014, 136, 3732–3735; c) D.-H. Wang, F. Wang, P.-H.
Chen, Z.-Y. Lin, G.-S. Liu, Angew. Chem., Int. Ed. 2017, 56, 2054–2058;
Angew. Chem. 2017, 129, 2086–2090.
[15] a) I. P. Beletskaya, C. Nájera, M. Yus, Chem. Soc. Rev. 2020, 49, 7101–
7166; b) I. P. Beletskaya, C. Nájera, M. Yus, Chem. Rev. 2018, 118,
5080–5200.
[16] a) S. Murarka, Adv. Syn. Catal. 2018, 360, 1735–1753; b) P.-F. Niu, J.
Li, Y.-X. Zhang, C.-D. Huo, Eur. J. Org. Chem. 2020, 36, 5801–5814; c)
S. K. Parida, T.; Mandal, S. Das, S. K. Hota, S. D. Sarkar, S. Murarka,
ACS Catal. 2021, 11, 1640–1683.
[25] N.-K. Fu, L. Song, J.-J. Liu, Y.-F. Shen, J. C. Siu, S. Lin, J. Am. Chem.
Soc. 2019, 141, 14480–14485.
[26] L. Fu, S. Zhou, X.-L. Wan, P.-H. Chen, G.-S. Liu, J. Am. Chem. Soc.
2018, 140, 10965–10969.
[17] a) D.-H. Wang, N. Zhu, P.-H. Chen, Z.-Y. Lin, G.-S. Liu, J. Am. Chem.
Soc. 2017, 139, 15632–15635; b) J.-J. Ma, J.-H. Lin, L.-F. Zhao, K.
Harms, M. Marsch, X.-L. Xie, E. Meggers, Angew. Chem., Int. Ed. 2018,
57, 11193–11197; Angew. Chem. 2018, 130, 11363–11367; c) R. S. J.
Proctor, H. J. Davis, R. J. Phipps, Science 2018, 360, 419–422; d) J. P.
Reid, R. S. J. Proctor, M. S. Sigman, R. J. Phipps, J. Am. Chem. Soc.
2019, 141, 19178–19185; e) F.-D. Lu, L.-Q. Lu, G.-F. He, J.-C. Bai, W.-
J. Xiao, J. Am. Chem. Soc. 2021, 143, 4168-4173.
[27] Z.-Q. Zhang, X.-Y. Meng, J. Sheng, Q. Lan, X.-S. Wang, Org. Lett. 2019,
21, 8256–8260.
[28] CCDC numbers for the X-ray structures: 1c: 2080908; 9d: CCDC,
2080909 and 9f: 2080903.
[29] R. Mao, J. Balon, X. Hu, Angew. Chem. 2018, 13812–13816; Angew.
Chem. Int. Ed. 2018, 57, 13624–13628.
[30] a) B. Vriens, M. Lenz, L. Charlet, M. Berg, L. H. E. Winkel, Nat. Commun.
2014, 5, 1-6; b) B.-H. Shih, R. S. Basha, C. F. Lee, ACS Catal. 2019, 9,
8862–8866.
[18] a) L. J. Allen, P. J. Cabrera, M. Lee, M. S. Sanford, J. Am. Chem. Soc.
2014, 136, 5607–5610; b) X. Zhu, Z. He, Q.-Y. Li, X.-J. Wang, RSC Adv.
2017, 7, 25171–25174.
[31] A. Tlahuext-Aca, R. A. Garza-Sanchez, F. Glorius, Angew. Chem.
2017,129, 3762-3765; Angew. Chem. Int. Ed. 2017, 56, 3708–3711.
[32] L. Gao, G. Wang, J. Cao, D. Yuan, C. Xu, X. Guo, S. Li, Chem. Commun.
2018, 54, 11534–11537.
[19] a) W. Zhang, F. Wang, S. D. McCann, D.-H. Wang, P.-H. Chen, S. S.
Stahl, G.-S. Liu, Science 2016, 353, 1014–1018; b) F. Wang, D.-H. Wang,
X.-L. Wan, L.-Q. Wu, P.-H. Chen, G.-S. Liu, J. Am. Chem. Soc. 2016,
138, 15547–15550; c) G.-Y. Zhang, S. Zhou, L. Fu, P.-H. Chen, Y.-B. Li,
J.-P. Zou, G.-S. Liu, Angew. Chem., Int. Ed. 2020, 59, 20439–20444;
Angew. Chem. 2020,132, 20619–20624.
[33] C. Girard, H. B. Kagan, Angew. Chem., Int. Ed. 1998, 37, 2922–2959;
Angew. Chem. 1998,130, 11363–11367.
[34] a) F.-D. Lu, D. Liu, L. Zhu, L.-Q. Lu, Q. Yang, Q.-Q. Zhou, W. Yi, Y. Lan,
W.-J. Xiao, J. Am. Chem. Soc. 2019, 141, 6167–6172; b) Q. Zhang, T.
Wang, X. Zhang, S. Tong, Y.-D. Wu, M.-X. Wang, J. Am. Chem. Soc.
2019, 141, 18341–18348; c) P. Xu, P. López-Rojas, T. Ritter, J. Am.
Chem. Soc. 2021, 14, 5349–5354.
[20] a) F.-D. Lu, D. Liu, L. Zhu, L.-Q. Lu, Q. Yang, Q.-Q. Zhou, Y. Wei, Y. Lan,
W.-J. Xiao, J. Am. Chem. Soc. 2019, 141, 6167–6172; b) Y.-H. Zeng, M.-
F. Chiou, X.-T. Zhu, J. Cao, D.-Q., Lv, W.-J. Jian, Y.-J. Li, X.-H. Zhang,
H.-L. Bao, J. Am. Chem. Soc. 2020, 142, 18014–18021; c) d) L. Song,
N.-K. Fu, B. G. Ernst, W.-H. Lee, M. O. Frederick, R. A. J. DiStasio, S.
Lin, Nat. Chem. 2020, 12, 747–754; d) Y.-H. Zeng, Y.-J. Li, D.-Q. Lv, H.-
L. Bao, Org. Chem. Front. 2021, 8, 908-914.
[35] a) J. C. L. Walker, H. F. T. Klare, M. Oestreich, Nat. Rev. Chem. 2020,
4, 54–62; b) H. F. T. Klare, L. Albers, L. Süsse, S. Keess, T. Müller, M.
Oestreich, Chem. Rev. 2021, 121, 5889–5985.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202108617
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
A chemodivergent asymmetric synthesis can be achieved by tuning the metal-to-ligand ratio in an organometallic catalytic system.
9
This article is protected by copyright. All rights reserved.