Site Selective Chlorination of C(sp3)?H Bonds Suitable for Late-Stage Functionalization

Article abstract of DOI:10.1002/anie.202016548

C(sp³)?Cl bonds are present in numerous biologically active small molecules, and an ideal route for their preparation is by the chlorination of a C(sp³)?H bond. However, most current methods for the chlorination of C(sp³)?H bonds are insufficiently site selective and tolerant of functional groups to be applicable to the late-stage functionalization of complex molecules. We report a method for the highly selective chlorination of tertiary and benzylic C(sp³)?H bonds to produce the corresponding chlorides, generally in high yields. The reaction occurs with a mixture of an azidoiodinane, which generates a selective H-atom abstractor under mild conditions, and a readily-accessible and inexpensive copper(II) chloride complex, which efficiently transfers a chlorine atom. The reaction's exceptional functional group tolerance is demonstrated by the chlorination of >30 diversely functionalized substrates and the late-stage chlorination of a dozen derivatives of natural products and active pharmaceutical ingredients.

Full text of DOI:10.1002/anie.202016548

Angewandte
Research Articles
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 8276–8283
International Edition:
German Edition:
doi.org/10.1002/anie.202016548
doi.org/10.1002/ange.202016548
À
C H Functionalization Very Important Paper
3
À
Site Selective Chlorination of C(sp ) H Bonds Suitable for Late-Stage
Functionalization
Alexander Fawcett, M. Josephine Keller, Zachary Herrera, and John F. Hartwig*
3
À
Abstract: C(sp ) Cl bonds are present in numerous biologi-
an industrially important process for the production of small
chlorinated molecules from light alkanes, but its use on the
benchtop for the chlorination of more complex small
molecules is greatly limited by the toxicity of chlorine gas,
the strongly oxidizing nature of this reagent, and the low site
selectivity with which these chlorinations occur.[9,6d] The site
selectivity of these chlorinations is low because the reaction is
initiated by the homolysis of chlorine gas, giving two chlorine
radicals, and is propagated by the reaction of Cl2 with an alkyl
cally active small molecules, and an ideal route for their
preparation is by the chlorination of a C(sp ) H bond.
However, most current methods for the chlorination of
C(sp ) H bonds are insufficiently site selective and tolerant
of functional groups to be applicable to the late-stage
functionalization of complex molecules. We report a method
for the highly selective chlorination of tertiary and benzylic
3
À
3
À
3
À
C(sp ) H bonds to produce the corresponding chlorides,
generally in high yields. The reaction occurs with a mixture
of an azidoiodinane, which generates a selective H-atom
abstractor under mild conditions, and a readily-accessible and
inexpensive copper(II) chloride complex, which efficiently
transfers a chlorine atom. The reactionꢀs exceptional functional
group tolerance is demonstrated by the chlorination of > 30
diversely functionalized substrates and the late-stage chlorina-
tion of a dozen derivatives of natural products and active
pharmaceutical ingredients.
radical (Scheme 1B); because the chlorine radical undergoes
3
À
unselective abstraction of a C(sp ) H bond, mixtures of
isomeric mono-chloride and poly-chloride products are often
produced.[10]
3
À
Interest in the development of C(sp ) H bond chlorina-
tion reactions has been driven by the high value of alkyl
3
À
chlorides. C(sp ) Cl bonds are present in numerous natural
products[11] and can imbue beneficial properties to biologi-
cally-active molecules,[12] such as increased lipophilicity,
modulation of the electronic properties of nearby functional
groups, and the prevention of metabolic oxidation at the site
Introduction
of chlorination (for some selected examples of biologically
3
3
À
À
C(sp ) H bond functionalization reactions enable the
active molecules containing C(sp ) Cl bonds, see Sche-
introduction of functional groups at previously inert sites,
circumventing the synthesis of pre-functionalized substrates
and streamlining the preparation of new organic molecules.[1]
Such reactions have been heralded as one of many emerging
synthetic strategies that could revolutionize how we prepare
organic molecules[2] for application in medicinal chemistry,[3]
total synthesis,[4] and the late-stage functionalization of
me 1C). Furthermore, alkyl chlorides can be used as a site
for diversification by substitution at the C(sp ) Cl bond.
3
[13]
À
natural products.[5] However, the development of site-selec-
tive functionalizations of C(sp ) H bonds, particularly those
destined to be applied to late-stage functionalization, is
3
À
challenging because of the need to distinguish between
3
À
multiple non-activated C(sp ) H bonds possessing similar
steric and electronic properties.[6] While many advances have
2
À
been made toward the functionalization of C(sp ) H bonds
with high site-selectivity,[7] far fewer methods functionalize
3
[8]
À
C(sp ) H bonds with high site-selectivity, especially those
within complex molecules.[2–5]
3
À
The archetypal C(sp ) H bond functionalization is the
radical chlorination of alkanes initiated by the homolysis of
chlorine gas by ultraviolet light (Scheme 1A). This reaction is
[*] Dr. A. Fawcett, M. J. Keller, Z. Herrera, Prof. J. F. Hartwig
Department of Chemistry, University of California, Berkeley
Berkeley, CA 94720 (USA)
E-mail: jhartwig@berkeley.edu
3
À
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202016548.
Scheme 1. Classical C(sp ) H bond chlorination with chlorine gas and
its reaction mechanism, and selected biologically-active molecules
containing C(sp ) Cl bonds.
3
À
8276
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283

Angewandte
Research Articles
Chemie
Therefore, the development of methods capable of selectively
incorporating a chlorine atom into biologically-active small-
molecules or natural products at a late stage could positively
alter their efficacy and expedite the discovery of pharma-
ceuticals based on these scaffolds.[14]
presence of a photoredox catalyst. Next, Kanai[19] reported
3
À
the chlorination of tertiary and benzylic C(sp ) H bonds by
tert-butyl hypochlorite in the presence of a silver catalyst
(Scheme 2B). Again, the substrate scope and functional
group tolerance was limited to small molecules containing
3
À
These potential benefits and the drawbacks of the classical
few potentially reactive C(sp ) H bonds, in this case because
3
À
C(sp ) H bond chlorination methods have led to the devel-
the promiscuous tert-butyloxy radical, which is generated by
opment of alternative methods for the chlorination of
the reaction of the silver catalyst or an alkyl radical with tert-
3
[15–20]
3
À
À
C(sp ) H bonds,
but many of these methods rely on
butyl hypochlorite, abstracts the C(sp ) H bond. More
common reagents (such as N-chlorosuccinimide, TCCA, or
tert-butyl hypochlorite) with the chlorine atom bound to an
complex molecules, even those derived from simple natural
products such as menthol and citronellol, underwent reaction
to form complex mixtures of products. In contrast to these two
À
electronegative heteroatom. With these reagents, the X Cl
bond is sufficiently weak to undergo homolysis and initiate/
propagate the reaction, but, after such a step, generates
a heteroatom-centered radical that is poorly chemo-selective
studies, Alexanian[20] reported recently the chlorination of
3
À
primary and secondary C(sp ) H bonds with N-chloroamide
reagents, a reaction whose site selectivity complements that of
other radical chlorinations (Scheme 2C). Although this
method was applied to the chlorination of sclareolide with
high selectivity in the total synthesis of chlorolissoclimide,[20a]
the chlorination of other small molecules was less selecti-
ve,[20b] and an excess of substrate was required in some cases.
In these reactions, an N-centered radical, which is generated
and site-selective for H-atom abstraction. As a consequence,
3
À
multiple products form, and C(sp ) H bond chlorination
reactions require a large excess of the substrate in which the
À
C H bond reacts. The exceptions to this trend include
3
À
chlorinations of benzylic C(sp ) H bonds, which can be
conducted often with high site selectivity because this bond is
weaker than others,[16] and chlorination reactions that are
directed intramolecularly to a particular site by a pre-installed
functional group.[17]
À
by the homolytic cleavage of the N Cl bond or by its reaction
with an alkyl radical, abstracts the H-atom and, thus, dictates
the reactionꢀs site selectivity.[21] These prior studies show that
3
À
Three recent publications describe the chlorination of
new methods are needed for the chlorination of C(sp ) H
3
À
C(sp ) H bonds with the substrate as the limiting reagent.
bonds to create a suite of systems that are highly site selective
First, Chen[18] reported that a mixture of Zhdankinꢀs azidoio-
and tolerate a wide range of functional groups.
3
À
dinane (1), lithium chloride, and a ruthenium photocatalyst in
Previously, we reported the azidation of C(sp ) H bonds
HFIP under visible light irradiation chlorinated tertiary
in a range of complex small molecules[22] with 10 mol% of
(L1)Fe(OAc)2 (L1 = iPrPybox, Table 1) and 2.0 equivalents of
Zhdankinꢀs azidoiodinane reagent (1) (Scheme 3).[23] We later
exploited the high selectivity of this reaction to achieve the
late-stage azidation of a wide range of natural products and
active pharmaceutical ingredients.[24] While details of the
3
À
C(sp ) H bonds (Scheme 2A). However, just six substrates
were evaluated, and studies described here reveal the limited
tolerance of this process for auxiliary functional groups,
perhaps due to the high concentration of oxidant and
mechanism of this azidation reaction are under study and will
3
À
be reported elsewhere, abstraction of a C(sp ) H bond by
a benzoyloxy radical generated by the homolysis of the weak
Table 1: Evaluation of reactions conditions for the chlorination of the
3
À
tertiary C(sp ) H bond of isopentyl benzoate.
Entry Equiv 1 Chloride source
2 (3) [%][a]
1
2
3
4
5
6
7
8
2.0
2.0
–
2.0
2.0
2.0
2.0
2.5
FeCl2/L1 (10 mol%)
FeCl2/L1 (1.0 equiv)
FeCl2/L1 (10 mol%), 4 (2.0 equiv)
FeCl3, NiCl2, CoCl2, or MnCl2 (1.0 equiv) <5 (0)
CuCl2·2H2O (1.0 equiv)
CuCl2·2H2O (1.0 equiv), L2 (1.0 equiv)
(L2)CuCl2 (1.0 equiv)
20 (41)
13 (10)
0
16 (0)
69 (0)
54 (0)
92 (0)[b]
(L2)CuCl2 (1.5 equiv)
3
1
À
Scheme 2. Recent progress in the site-selective chlorination of C(sp )
H bonds.
[a] Yields determined by H NMR. [b] Allowed to react for 48 h in DCE
solvent.
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
8277

Angewandte
Research Articles
Chemie
nane 4 (entry 3), resulted in no observable chloride product.
The last result was consistent with the unusual ability of 1 to
3
À
generate a species that abstracts a tertiary C(sp ) H bond.
3
À
We considered that the weak C(sp ) H bonds in L1 (those
a to the oxygen and nitrogen atoms, and the tertiary C(sp ) H
bond of the isopropyl group) could interfere with abstraction
of the tertiary C(sp ) H bond in the substrate, and our results
3
À
3
À
made clear that a species was needed that would transfer
a chlorine atom without reacting with 1 to form a metal azide
complex that would transfer an azide unit to an alkyl radical.
On the basis of this logic, we tested a series of additional
metal chloride salts in the presence of 1 and isopentyl
benzoate under otherwise identical conditions. The reactions
with iron(III), nickel(II), cobalt(II) and manganese(II) chlor-
ides formed the alkyl chloride product 2 in trace quantities
(entry 4). In all of these reactions, the presence of alkyl azide
3 was not observed.
Scheme 3. Our previously reported iron-catalyzed azidation of
3
À
C(sp ) H bonds suitable for the late-stage azidation of complex
molecules.
[25]
À
N I bond of Zhdankinꢀs azidoiodinane leads to an alkyl
radical that is trapped by an iron-azide complex. Because of
the exquisite site selectivity and functional group tolerance of
Reactions with copper(II) chlorides bore more fruit. The
reaction with CuCl2·2H2O, formed 2 in 16% yield, again,
without formation of azide 3 (entry 5). The addition of 1,10-
phenanthroline (phen, L2),[26] a simple and readily-available
this method, we sought to develop new functionalizations of
3
À
C(sp ) H bonds based on the combination of Zhdankinꢀs
À
reagent and a transition metal complex that transfers the new
functional group.
nitrogen ligand lacking weak C H bonds, to this reaction
resulted in the formation of 2 in 69% yield (entry 6), but this
result was not consistently reproducible. Fortunately, con-
ducting the reaction with the preformed, isolated copper
chloride complex (phen)CuCl2 reproducibly formed 2 in 54%
yield (entry 7). This (phen)CuCl2 complex is convenient and
easily prepared in one step on multigram scale by simply
mixing cheap CuCl2·2H2O ($0.08gÀ1, Acros Organics) and
1,10-phenanthroline ($0.60gÀ1, Combi-Blocks) or can be
purchased from common chemical suppliers and is fully
air-stable over extended periods of time.
Here, we report the design and implementation of
a method for the chlorination of tertiary, as well as benzylic,
3
À
C(sp ) H bonds, that occurs with high site-selectivity and
functional group tolerance (Scheme 2D). The design involves
a mechanism that is distinct from those of previous studies
and divorces the H-atom abstraction and chlorine atom
transfer agents: Zhdankinꢀs azidoiodinane reagent[23] is used
to generate the o-iodobenzoyloxy radical, which is highly
selective for H-atom abstraction under mild conditions, and
the cheap and readily-accessible copper(II) complex, (phen)-
CuCl2,[26] is used to mediate the controlled, rapid delivery of
a chlorine atom to an alkyl radical without the generation of
reactive radical byproducts. The mild conditions and high site
selectivity of this system made possible the isolation of
a single product in high yield from the chlorination of a range
small molecules, as well as a number of derivatives of complex
To further increase the yield of 2, we increased the
equivalents of 1, increased the reaction time, and changed the
solvent to 1,2-dichloroethane (DCE), resulting in 92% of 2
(90% isolated yield) (entry 8). While we primarily foresee
this method being valuable for the preparation of function-
alized small molecules on the small scales appropriate for
investigation of biological activity, the reaction also occurred
in equally high yield on gram scale. The chlorination of 1.0 g
of isopentyl benzoate gave 1.08 g of isolated 2, corresponding
to a yield of 91%.
natural products and active pharmaceutical ingredients.
3
À
These studies bring C(sp ) H bond chlorination into the
3
[3–5]
À
lexicon of late-stage C(sp ) H bond functionalizations.
These newly developed conditions led to chlorination
reactions occurring in higher yields than those previously
reported by others[18,19] on similar substrates and, more
important, enabled the chlorination of a much broader scope
of complex molecules with high site selectivity (vide infra).
Furthermore, the reaction assembly and work-up are simple.
All chlorination reactions were conducted on the benchtop in
laboratory-grade solvents, and a brief purging of the reaction
vessel headspace with nitrogen gas was the only precaution
taken that was required to maintain high yields. In most cases,
the (phen)CuCl2 complex and unreacted azidoiodinane
reagent was safely and quantitatively removed by washing
the diluted reaction mixture with an aqueous solution of
sodium hydroxide to deliver crude products with few
contaminants.
Results and Discussion
We sought to identify conditions that would enable the
3
À
chlorination of C(sp ) H bonds using our previously devel-
oped azidation conditions as a starting point. We began our
investigations by studying the reaction of isopentyl benzoate
with 1 and metal-chloride salts (Table 1). The product of
3
À
C(sp ) H bond chlorination was first observed from reactions
with FeCl2 in place of Fe(OAc)2 to form the (L1)FeX2
complex under conditions typical for the azidation reac-
tion.[22,24] In this case, 20% of the tertiary chloride 2 was
observed, alongside 41% of the tertiary azide 3 (Table 1,
entry 1). The yield of 2 from this reaction did not increase
with increased equivalents of FeCl2 (entry 2), and the reaction
with the chloride analog of Zhdankinꢀs reagent, chloroiodi-
As shown in Scheme 4A, the chlorination of a set of small
3
À
organic molecules containing C(sp ) H bonds in a range of
8278
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283

Angewandte
Research Articles
Chemie
chemical environments and possessing a series of common
functional groups occurred under the developed conditions to
give products 5–21 in high yield. The reactions of benzyl (5)
and butyl (6) esters of 4-methylvaleric acid formed the
corresponding tertiary chlorides in good to excellent yields.
3
À
These results demonstrate that the C(sp ) H bonds of linear
3
À
aliphatic chains and the weak benzylic C(sp ) H bonds of the
benzyl ester are not chlorinated under our conditions.
Derivatives of leucine (7) and a Leu-Glu dipeptide (8) also
3
À
underwent chlorination at the tertiary C(sp ) H bond on the
side-chain without epimerization of any of the stereocenters,
indicating the potential utility of this method for the
chlorination of leucine residues in polypeptides. Substrates
containing a free carboxylic acid (9), a primary amide (10),
a phthalimide-protected amine (11), and a cyclohexyl group
(12), all underwent chlorination in excellent yields, further
demonstrating the tolerance of this method for common
functional groups. The reaction of dihydrocitronellyl acetate,
which contains two sterically similar but electronically differ-
3
À
entiated tertiary C(sp ) H bonds formed the 7- and 3-
chlorinated products in a 4.0:1 mixture (only major isomer
13 shown). These ratios are similar to that observed in our
previously reported iron-catalyzed azidation reaction of the
same substrate,[22] indicating that the site selectivity of these
reactions does not depend on the functional group transfer
reagent; rather, it depends on the properties of the common
azidoiodinane reagent, which forms the species that abstracts
the H-atom. Finally, adamantyl pinacol boronic ester, which
3
À
contains 12 weak secondary C(sp ) H bonds and a potentially
labile boronic ester, underwent chlorination with selectivity
for the tertiary C(sp ) H bond (14). This site selectivity can
be attributed to steric repulsion between the bulky putative
benzoyloxy radical and the axial protons, which favors
abstraction from the stronger[27] but more sterically accessible
bridgehead tertiary C(sp ) H bonds.
Benzylic C(sp ) H bonds also underwent chlorination
with excellent site selectivity. A secondary benzylic C(sp ) H
3
À
3
[28]
À
3
À
3
À
bond in a series of functionalized 3-phenylpropanes contain-
ing a benzoate (15), bromide (16) and nitro group (17) were
chlorinated in good to excellent yields. The tertiary benzylic
3
À
C(sp ) H bond of a derivative of a precursor to the
antiparasitic drug atovaquone[29] (18) also was successfully
chlorinated, a functionalization that we found did not occur
under published chlorination conditions[18,19] and, in general,
is a poorly precedented transformation.
3
À
Scheme 4. Scope of small molecules that undergo C(sp ) H bond
chlorination. Reaction conditions: 1 (2.5 equiv), (phen)CuCl2
(1.5 equiv) and 0.2 mmol of the substrate in DCE (1.5 mL, 0.13 m) at
508C. Isolated yields of chloride products are reported. [a] Isolated as
The precursors to alkyl chlorides 5–18 all possess electron-
withdrawing functional groups that inductively deactivate
3
À
nearby C(sp ) H bonds and could be responsible for enabling
the high site-selectivity observed. To test this hypothesis, n-
pentyl benzoate, which features an electron-withdrawing
benzoate functional group, was subjected to our chlorination
conditions. This substrate did not react, demonstrating the
high sensitivity of our system to electronic effects. In contrast,
the same substrate underwent chlorination at the g and d sites
to give a mixture of products (19) in a 1:1.7 ratio under
Kanaiꢀs conditions,[19,30] and underwent chlorination at all
positions except that a to the electron-withdrawing functional
group under Alexanianꢀs conditions with the corresponding
acetate substrate.[20b] These results clearly demonstrate that
1
the methyl ester. [b] The ratio of isomers was determined by H NMR
analysis of the crude reaction mixture. [c] Determined by the ratio of
the obtained masses of the two constitutional isomers following
column chromatography. [d] [Ag]=Ag(phen)2OTf (0.2 mol%), tBuOCl
(2.0–3.0 equiv), MeCN, r.t., 48 h.[19] [e] Reaction was conducted at
ambient temperature in MeCN. [f] Yield determined by 1H NMR
analysis of the crude reaction mixture. [g] hn=Ru(bpy)3Cl2
(0.1 mol%), LiCl (4.0 equiv), 1 (2.0 equiv), visible light, HFIP, r.t.,
24 h.[18] [h] Yield determined by GC-FID analysis of the crude reaction
mixture.
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 8279

Angewandte
Research Articles
Chemie
the intermediate that abstracts the H-atom in our system,
a benzoyloxy radical, is much more selective than those in
polycyclic arene (33) and electron-poor heteroarenes, such as
pyridine (34), quinoline (35), and pyrazine (36), were well
tolerated and did not undergo alkylation by a Minisci-type
Kanaiꢀs (a tert-butyloxy radical)[19] and Alexanianꢀs (an N-
3
3
À
À
centered radical) toward tertiary and benzylic C(sp ) H
reaction. The tertiary C(sp ) H bond of the 4-methylpentyl
bonds that are deactivated only weakly by a nearby electron-
withdrawing functional group.
group of a thienyl ester was selectively chlorinated in the
presence of three weak primary benzylic C(sp ) H bonds of
the methyl group on the thienyl ring; 73% of the tertiary
chloride was isolated (40), with only a trace of the bis-
chlorinated product observed. The product of mono-chlori-
nation at the benzylic position was not detected.
3
À
To assess whether the presence of electron-withdrawing
functional groups was required to achieve our high site-
selectivity, we conducted the chlorination of the unfunction-
alized alkane, 3-methylpentane. The product of tertiary
chlorination, compound 20, formed in 91% yield, as deter-
By contrast, electron-rich arenes and polycyclic arenes,
aryl iodides, boronic esters, and heteroarenes gave products
from arene halogenation or complex mixtures of products by
halogenation procedures published previously,[18,19] highlight-
ing the higher functional group tolerance of our chlorination
method.
While the tolerance for functional groups was high, some
common functional groups, such as primary and secondary
alcohols, alkenes, and alkynes, were not tolerated under any
mined by GC-FID; the chlorides produced by functionaliza-
3
À
tion of the secondary and primary C(sp ) H bonds were not
observed. These results contrast sharply with a classical
photochemical chlorination of the same substrate with carbon
tetrachloride,[31] which formed products from chlorination of
3
À
primary and secondary C(sp ) H bonds of 3-methylpentane
in 82% yield, with the tertiary chloride detected as a minor
reaction product in only 18% yield. Furthermore, under our
conditions, chlorocyclohexane (21) was produced by the
chlorination of cyclohexane in 69% yield; higher molecular-
weight products, such as dichlorocyclohexanes, were not
observed, and this result contrasts strongly with the outcomes
obtained by other methods involving more reactive chlori-
nating reagents.[20a]
of these conditions for chlorination (see Supporting Informa-
3
À
tion for details of substrates containing tertiary C(sp ) H
bonds that did not react to give an isolable tertiary chloride
product). Such limitations are consistent with those of other
3
À
C(sp ) H bond functionalization reactions that proceed by
H-atom abstraction to form radical intermediates.
The tolerance of the reaction toward functional groups
was further investigated by conducting the chlorination of
a series of 4-methylpentyl aryl esters containing a variety of
functional groups on the aryl ring (Scheme 4B). Good to
excellent yields of the tertiary chloride products 22–40 were
obtained from the corresponding substrates, showing that this
method is tolerant of numerous functional groups and
pharmaceutically important heteroarenes.[32] For example,
substrates bearing aryl halides (28–31) and pinacol boronic
esters (32) underwent chlorination in good yield without
modification of the functional groups.[33] Furthermore, a fused
To determine whether this reaction is amenable to the
late-stage chlorination of molecules that already are densely
functionalized, we conducted the chlorination of a series of
natural products and active pharmaceutical ingredients con-
3
À
taining tertiary C(sp ) H bonds or, in one case, benzylic
3
À
C(sp ) H bonds. These results highlight the value of this new,
mild, method for achieving chlorination of such C(sp ) H
bonds in complex substrates.
3
À
Complex molecules that underwent chlorination are
shown in Figure 1. The monoterpenoid (À)-menthone, a fla-
voring and fragrance molecule that is responsible for the
3
À
Figure 1. Scope of natural product and drug derivatives that undergo C(sp ) H bond chlorination. Reaction conditions: 1 (2.5 equiv), (phen)CuCl2
(1.5 equiv) and the substrate in DCE (0.13 m) at 508C. Isolated yields of chloride products are reported. [a] Reaction was conducted at ambient
temperature in MeCN. [b] [Ag]=Ag(phen)2OTf (0.2 mol%), tBuOCl (2.0–3.0 equiv), MeCN, r.t., 48 h.[19] [c] Yield determined by 1H NMR analysis
of the crude reaction mixture. [d] hn=Ru(bpy)3Cl2 (0.1 mol%), LiCl (4.0 equiv), 1 (2.0 equiv), visible light, HFIP, r.t., 24 h.[18]
8280 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283

Angewandte
Research Articles
Chemie
characteristic taste and aroma of peppermint, was selectively
tioned toward the aromatic system to minimize steric
3
À
chlorinated at the tertiary C(sp ) H bond of its pendent
isopropyl group (> 20:1 regioselectivity) to give chloride 41.
This product was isolated in 48% isolated yield, albeit with
repulsion between these two groups, and the benzylic one is
likewise sterically inaccessible due to being almost co-planar
with the aromatic system. Finally, the chlorination of a pre-
cursor to the anti-malarial drug artemisinin gave a mixture of
small amounts of impurities that we assumed to be minor
3
À
constitutional isomers. Likewise, the tertiary C(sp ) H bond
chloride products. The two main products were the result of
3
À
of the pendent isopropyl group of the bicyclic monoterpenoid
1,4-cineole, which is an isomer of eucalyptol and a component
of the flavor components of lime juice, was chlorinated to give
42 in 93% yield. The fully hydrogenated derivative of the
chlorination of the C(sp ) H bonds at C2 and C5, giving
tertiary chlorides 53 and 54, respectively, in modest yields.
The most electron-rich tertiary C(sp ) H bond at C6, which is
farthest from the electron-withdrawing ester, did not undergo
chlorination because it is on the concave face of the molecule
and thus sterically inaccessible.
3
À
ubiquitous fragrance molecule linalool, tetrahydrolinalool,
3
À
also was selectively chlorinated at its tertiary C(sp ) H bond
to give 43 in 79% yield. This result contrasts with the
chlorination of tetrahydrolinalool acetate with Alexanianꢀs
chlorination system (Scheme 2C), which produced a mixture
To compare the value of other reported methods for
3
À
C(sp ) H bond chlorination to our procedure closely, we
tested Chen[18] and Kanaiꢀs[19] systems (Scheme 2A and B) for
the chlorination of several complex molecules. In all cases, the
yields of these chlorination reactions were much lower than
those obtained under our conditions. Memantine was the
most complex molecule reported to undergo chlorination
under Chenꢀs[18] conditions (63% yield of 45 vs. 97% under
the newly developed conditions), and no other substrates
gave yields higher than 7% under these conditions. Likewise,
the highest yield for any example under Kanaiꢀs silver-
catalyzed conditions was the chlorination of the N-alkyl
thalidomide substrate, which gave just 28% yield, a reaction
that gave a yield of 98% under our conditions.
of primary and secondary chlorinated products.[20b] The
3
À
secondary benzylic C(sp ) H bond of the indane musk
celestolide underwent chlorination to give the benzylic
chloride 44 in 80% yield, and a derivative of the adaman-
tane-based Alzheimerꢀs drug Memantine was selectively
chlorinated at the tertiary bridgehead position to give
adamantyl chloride 45 in quantitative yield.
The 4-methylpentyl esters of sulbactam, a b-lactamase
inhibitor used in combination with the antibiotic ampicillin
for the treatment of bacterial infections resistant to b-lactam
antibiotics, niflumic acid, a drug used for the treatment of
joint and muscular pain, and the acid derived from fenofi-
brate, a drug used to treat abnormal blood lipid levels, all
3
À
underwent chlorination selectively at the tertiary C(sp ) H Conclusion
bonds of the 4-methylpentyl moiety to give chlorides 46, 47
and 48, respectively, in excellent yields. An N-2-methylbutyl
derivative of thalidomide, a multipurpose drug which is used
for the treatment of multiple myeloma, was also chlorinated
In summary, we have developed a method for the
3
À
chlorination of tertiary C(sp ) H bonds embedded within
a diverse set of small organic molecules featuring a broad
array of functional groups to form the corresponding tertiary
chlorides in good to excellent yields and with excellent site
3
À
with perfect site selectivity for the tertiary C(sp ) H bond of
the pendent 2-methylbutyl group to give chloride 49 in
excellent yield. In these examples, the complex biologically-
active cores remained unfunctionalized and intact, highlight-
ing both the mild conditions and functional-group tolerance
of our chlorination process.
selectivity. This method also was suitable for the transforma-
3
À
tion of secondary and tertiary benzylic C(sp ) H bonds into
the corresponding alkyl chlorides. Furthermore, we demon-
strated that the site selectivity, yield, and functional-group
The site-selective chlorination of natural product and drug
derivatives with more topologically complex structures con-
tolerance of this method are far higher than those of
3
À
previously reported chlorination reactions of C(sp ) H
3
À
taining multiple tertiary C(sp ) H bonds was also investigat-
bonds, especially when applied to the late-stage chlorination
of complex molecules. The process we report is highly
practical and occurs on the benchtop under mild reaction
conditions with readily-available materials to deliver high
purity alkyl chlorides. Additional studies on the mechanism
and origin of site-selectivity of this chlorination reaction and
the related azidation reaction are ongoing in our laboratory,
as is the expansion of this reaction manifold to achieve the
introduction of other valuable functional groups into complex
small molecules, natural products, and active pharmaceutical
ingredients.
ed. A derivative of betulin, a naturally-occurring pentacyclic
triterpene that exhibits a broad spectrum of biological effects,
including anti-cancer activity,[34] and that contains seven
3
À
electron-rich tertiary C(sp ) H bonds, was chlorinated in
3
À
62% yield with high selectivity for a single tertiary C(sp ) H
bond (50) (no other isomeric chloride products could be
isolated or identified). A derivative of the plant and fungi
growth hormone gibberellic acid, a pentacyclic diterpene, was
also chlorinated in good yield and with high selectivity for
3
À
a single tertiary C(sp ) H bond out of four others, giving
chloride 51. A pyrano[3,2-a]carbazole, a common precursor
to the murrayamine natural product family developed
recently by Sarpong,[35] was selectively chlorinated at the
Acknowledgements
3
À
non-benzylic tertiary C(sp ) H bond on the cyclohexane ring
to give tertiary chloride 52 in good yield. The two other
We thank the NIH R35-GM130387 for support of this work
and Dr Siying Zhong for assistance with the structural
assignment of compounds 52–54. We also thank the College
3
À
tertiary C(sp ) H bonds did not react. The one in the pendent
isopropyl group is sterically inaccessible because it is posi-
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
8281

Angewandte
Research Articles
Chemie
M. M. Gilbert, D. J. Weix, J. Am. Chem. Soc. 2020, 142, 9902 –
9907; e) H. A. Sakai, W. Liu, C. Le, D. W. MacMillan, J. Am.
Chem. Soc. 2020, 142, 11691 – 11697; f) M. Bçrjesson, T.
Moragas, R. Martin, J. Am. Chem. Soc. 2016, 138, 7504 – 7507.
[14] a) D. J. Newman, G. M. Cragg, J. Nat. Prod. 2012, 75, 311 – 335;
b) G. M. Cragg, D. J. Newman, Biochem. Biophys. Acta Gen.
Subj. 2013, 1830, 3670 – 3695; c) D. J. Newman, G. M. Cragg, J.
Nat. Prod. 2016, 79, 629 – 661; d) Bioactive Compounds from
Natural Sources: Natural Products as Lead Compounds in Drug
Discovery (Ed.: C. Tringali), CRC Press, Boca Raton, 2011.
[15] a) W. Liu, J. T. Groves, J. Am. Chem. Soc. 2010, 132, 12847 –
12849; b) G. Li, A. K. Dilger, P. T. Cheng, W. R. Ewing, J. T.
Groves, Angew. Chem. Int. Ed. 2018, 57, 1251 – 1255; Angew.
Chem. 2018, 130, 1265 – 1269; c) K. Tanemura, T. Suzuki, Y.
Nishida, T. Horaguchi, Tetrahedron 2010, 66, 2881 – 2888; d) L.
Han, J.-B. Xia, L. You, C. Chen, Tetrahedron 2017, 73, 3696 –
3701.
[16] a) S. H. Combe, A. Hosseini, A. Parra, P. R. Schreiner, J. Org.
Chem. 2017, 82, 2407 – 2413; b) M. Xiang, C. Zhou, X.-L. Yang,
B. Chen, C.-H. Tung, L.-Z. Wu, J. Org. Chem. 2020, 85, 9080 –
9087; c) Z.-H. Li, B. Fiser, B.-L. Jiang, J.-W. Li, B.-H. Xu, S.-J.
Zhang, Org. Biomol. Chem. 2019, 17, 3403 – 3408; d) S. Sasmal,
S. Rana, G. K. Lahiri, D. Maiti, J. Chem. Sci. 2018, 130, 88.
[17] a) M. A. Short, M. F. Shehata, M. A. Sanders, J. L. Roizen,
Chem. Sci. 2020, 11, 217 – 223; b) A. N. Herron, D. Liu, G. Xia, J.-
Q. Yu, J. Am. Chem. Soc. 2020, 142, 2766 – 2770; c) E. A.
Wappes, A. Vanitcha, D. A. Nagib, Chem. Sci. 2018, 9, 4500 –
4504; d) S. P. Morcillo, E. M. Dauncey, J. H. Kim, J. J. Douglas,
N. S. Sheikh, D. Leonori, Angew. Chem. Int. Ed. 2018, 57, 12945 –
12949; Angew. Chem. 2018, 130, 13127 – 13131; e) H. Guan, S.
Sun, Y. Mao, L. Chen, R. Lu, J. Huang, L. Liu, Angew. Chem. Int.
Ed. 2018, 57, 11413 – 11417; Angew. Chem. 2018, 130, 11583 –
11587; f) E. Del Castillo, M. D. Martꢂnez, A. E. Bosnidou, T.
Duhamel, C. Q. OꢀBroin, H. Zhang, E. C. Escudero-Adꢁn, M.
Martꢂnez-Belmonte, K. MuÇiz, Chem. Eur. J. 2018, 24, 17225 –
17229; g) M. A. Short, J. M. Blackburn, J. L. Roizen, Angew.
Chem. Int. Ed. 2018, 57, 296 – 299; Angew. Chem. 2018, 130, 302 –
305; h) Q. Qin, S. Yu, Org. Lett. 2015, 17, 1894 – 1897; i) R.
Breslow, D. Heyer, J. Am. Chem. Soc. 1982, 104, 2045 – 2046;
j) R. Breslow, R. J. Corcoran, B. B. Snider, R. J. Doll, P. L.
Khanna, R. Kaleya, J. Am. Chem. Soc. 1977, 99, 905 – 915.
[18] Y. Wang, G.-X. Li, G. Yang, G. He, G. A. Chen, Chem. Sci. 2016,
7, 2679 – 2683.
of Chemistryꢀs NMR facility for resources provided and Dr.
Hasan Celik for assistance. Instruments in the CoC-NMR are
supported in part by NIH S10OD024998.
Conflict of interest
The authors declare no conflict of interest.
À
Keywords: C H functionalization · chlorination · halogenation ·
late-stage functionalization · radical reactions
[1] a) R. G. Bergman, Nature 2007, 446, 391 – 393; b) M. C. White,
Science 2012, 335, 807 – 809.
[2] J. F. Hartwig, Acc. Chem. Res. 2017, 50, 549 – 555.
[3] a) M. Moir, J. J. Danon, T. A. Reekie, M. Kassiou, Expert Opin.
Drug Discovery 2019, 14, 1137 – 1149; b) T. Cernak, K. D.
Dykstra, S. Tyagarajan, P. Vachal, S. W. Krska, Chem. Soc. Rev.
2016, 45, 546 – 576.
[4] a) W. R. Gutekunst, P. S. Baran, Chem. Soc. Rev. 2011, 40, 1976 –
1991; b) D. J. Abrams, P. A. Provencher, E. J. Sorensen, Chem.
Soc. Rev. 2018, 47, 8925 – 8967.
[5] a) R. R. Karimov, J. F. Hartwig, Angew. Chem. Int. Ed. 2018, 57,
4234 – 4241; Angew. Chem. 2018, 130, 4309 – 4317; b) B. Hong, T.
Luo, X. Lei, ACS Catal. 2020, 6, 622 – 635.
[6] a) J. F. Hartwig, M. A. Larsen, ACS Cent. Sci. 2016, 2, 281 – 292;
b) T. Newhouse, P. S. Baran, Angew. Chem. Int. Ed. 2011, 50,
3362 – 3374; Angew. Chem. 2011, 123, 3422 – 3435; c) A. Fawcett,
Pure Appl. Chem. 2020, 92, 1987 – 2003; d) A. A. Fokin, P. R.
Schreiner, Adv. Synth. Catal. 2003, 345, 1035 – 1052.
[7] P. Wedi, M. van Gemmeren, Angew. Chem. Int. Ed. 2018, 57,
13016 – 13027; Angew. Chem. 2018, 130, 13198 – 13209.
3
À
[8] For selected examples of site-selective aliphatic C(sp ) H bond
functionalizations, see: a) M. S. Chen, M. C. White, Science 2007,
318, 783 – 787; b) M. S. Chen, M. C. White, Science 2010, 327,
566 – 571; c) P. E. Gormisky, M. C. White, J. Am. Chem. Soc.
2013, 135, 14052 – 14055; d) M. C. White, J. Zhao, J. Am. Chem.
Soc. 2018, 140, 13988 – 14009; e) K. Liao, T. C. Pickel, V.
Boyarskikh, J. Bacsa, D. G. Musaev, H. M. L. Davies, Nature
2017, 551, 609 – 613; f) J. L. Roizen, D. N. Zalatan, J. Du Bois,
Angew. Chem. Int. Ed. 2013, 52, 11343 – 11346; Angew. Chem.
2013, 125, 11553 – 11556; g) N. D. Chiappini, J. B. C. Mack, J.
Du Bois, Angew. Chem. Int. Ed. 2018, 57, 4956 – 4959; Angew.
Chem. 2018, 130, 5050 – 5053; h) E. McNeill, J. Du Bois, J. Am.
Chem. Soc. 2010, 132, 10202 – 10204; i) S. Mukherjee, B. Maji, A.
Tlahuext-Aca, F. Glorius, J. Am. Chem. Soc. 2016, 138, 16200 –
16203; j) B. L. Tran, B. Li, M. Driess, J. F. Hartwig, J. Am. Chem.
Soc. 2014, 136, 2555 – 2563; k) I. B. Perry, T. F. Brewer, P. J.
Sarver, D. M. Schultz, D. A. DiRocco, D. W. C. MacMillan,
Nature 2018, 560, 70 – 75; l) P. J. Sarver, V. Bacauanu, D. M.
Schultz, D. A. DiRocco, Y.-h. Lam, E. C. Sherer, D. W. C.
MacMillan, Nat. Chem. 2020, 12, 459 – 467.
[19] J. Ozawa, M. Kanai, Org. Lett. 2017, 19, 1430 – 1433.
[20] a) R. K. Quinn, Z. A. Kçnst, S. E. Michalak, Y. Schmidt, A. R.
Szklarski, A. R. Flores, S. Nam, D. A. Horne, C. D. Vanderwal,
E. J. Alexanian, J. Am. Chem. Soc. 2016, 138, 696 – 702; b) A. M.
Carestia, D. Ravelli, E. J. Alexanian, Chem. Sci. 2018, 9, 5360 –
5365.
[21] M. M. Tierney, S. Crespi, D. Ravelli, E. J. Alexanian, J. Org.
Chem. 2019, 84, 12983 – 12991.
[22] A. Sharma, J. F. Hartwig, Nature 2015, 517, 600 – 604.
[23] a) V. V. Zhdankin, C. J. Kuehl, A. P. Krasutsky, M. S. Forma-
neck, J. T. Bolz, Tetrahedron Lett. 1994, 35, 9677 – 9680; b) V. V.
Zhdankin, A. P. Krasutsky, C. J. Kuehl, A. J. Simonsen, J. K.
Woodward, B. Mismash, J. T. Bolz, J. Am. Chem. Soc. 1996, 118,
5192 – 5197.
[24] R. R. Karimov, A. Sharma, J. F. Hartwig, ACS Cent. Sci. 2016, 2,
715 – 724.
[25] J.-D. Yang, M. Li, X.-S. Xue, Chin. J. Chem. 2019, 37, 359 – 363.
[26] J. D. Griffin, C. L. Cavanaugh, D. A. Nicewicz, Angew. Chem.
Int. Ed. 2017, 56, 2097 – 2100; Angew. Chem. 2017, 129, 2129 –
2132.
[9] S. M. Aschmann, R. Atkinson, Int. J. Chem. Kinet. 1995, 27, 613 –
622.
[10] P. Skell, H. N. Baxter, J. Am. Chem. Soc. 1985, 107, 2823 – 2824.
[11] a) J. Zeng, J. Zhan, Isr. J. Chem. 2019, 59, 387 – 402; b) G. W.
Gribble in Naturally Occurring Organohalogen Compounds—A
Comprehensive Update, Springer, Wien, 2010.
[12] a) K. Naumann, Pest Manage. Sci. 2000, 56, 3 – 21; b) B. Gꢁl, C.
Bucher, N. Z. Burns, Mar. Drugs 2016, 14, 206.
[13] a) X. Wu, W. Hao, K.-Y. Ye, B. Jiang, G. Pombar, Z. Song, S. Lin,
J. Am. Chem. Soc. 2018, 140, 14836 – 14843; b) T. C. Atack,
R. M. Lecker, S. P. Cook, J. Am. Chem. Soc. 2014, 136, 9521 –
9523; c) M. Giedyk, R. Narobe, S. Weib, D. Touraud, W. Kunz, B.
Kçnig, Nat. Catal. 2020, 3, 40 – 47; d) S. Kim, M. J. Goldfogel,
[27] G. H. Kruppa, J. L. Beauchamp, J. Am. Chem. Soc. 1986, 108,
2162 – 2169.
[28] I. Tabushi, S. Kojo, K. Fukunishi, J. Org. Chem. 1978, 43, 2370 –
2374.
8282 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283

Angewandte
Research Articles
Chemie
[29] S. Y. Dike, D. Singh, B. N. Thankachen, B. Sharma, P. K. Mathur,
[34] A. Hordyjewska, A. Ostapiuk, A. Horecka, J. Kurzepa, Phyto-
S. Kore, A. Kumar, Org. Process Res. Dev. 2014, 18, 618 – 625.
[30] Kanaiꢀs method is also not selective in its chlorination of
dihydrocitronellyl acetate (precursor to compound 13), a reac-
tion which produced a “complex mixture” of products. See
ref. [18].
chem. Rev. 2019, 18, 929 – 951.
[35] K. Norseeda, V. Gasser, R. Sarpong, J. Org. Chem. 2019, 84,
5965 – 5973.
[31] D. Estel, K. Mateew, W. Pritzkow, W. Schmidt-Renner, J. Prakt.
Chem. 1981, 323, 262 – 268.
[32] R. D. Taylor, M. MacCoss, A. D. G. Lawson, J. Med. Chem. 2014,
57, 5845 – 5859.
[33] a) G. Evano, A. Nitelet, P. Thilmany, D. F. Dewez, Front. Chem.
2018, 6, 114; b) H. Wu, J. Hynes, Jr., Org. Lett. 2010, 12, 1192 –
1195.
Manuscript received: December 13, 2020
Revised manuscript received: January 3, 2021
Accepted manuscript online: January 21, 2021
Version of record online: March 4, 2021
Angew. Chem. Int. Ed. 2021, 60, 8276 – 8283
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
8283