A general approach to intermolecular carbonylation of arene C-H bonds to ketones through catalytic aroyl triflate formation

Article abstract of DOI:10.1038/NCHEM.2903

The development of metal-catalysed methods to functionalize inert C-H bonds has become a dominant research theme in the past decade as an approach to efficient synthesis. However, the incorporation of carbon monoxide into such reactions to form valuable ketones has to date proved a challenge, despite its potential as a straightforward and green alternative to Friedel-Crafts reactions. Here we describe a new approach to palladium-catalysed C-H bond functionalization in which carbon monoxide is used to drive the generation of high-energy electrophiles. This offers a method to couple the useful features of metal-catalysed C-H functionalization (stable and available reagents) and electrophilic acylations (broad scope and selectivity), and synthesize ketones simply from aryl iodides, CO and arenes. Notably, the reaction proceeds in an intermolecular fashion, without directing groups and at very low palladium-catalyst loadings. Mechanistic studies show that the reaction proceeds through the catalytic build-up of potent aroyl triflate electrophiles.

Full text of DOI:10.1038/NCHEM.2903

ARTICLES
PUBLISHED ONLINE: 11 DECEMBER 2017 | DOI: 10.1038/NCHEM.2903
A general approach to intermolecular
carbonylation of arene C–H bonds to ketones
through catalytic aroyl triflate formation
R. Garrison Kinney, Jevgenijs Tjutrins, Gerardo M. Torres, Nina Jiabao Liu, Omkar Kulkarni
and Bruce A. Arndtsen
*
The development of metal-catalysed methods to functionalize inert C–H bonds has become a dominant research theme in
the past decade as an approach to efficient synthesis. However, the incorporation of carbon monoxide into such reactions
to form valuable ketones has to date proved a challenge, despite its potential as a straightforward and green alternative to
Friedel–Crafts reactions. Here we describe a new approach to palladium-catalysed C–H bond functionalization in which
carbon monoxide is used to drive the generation of high-energy electrophiles. This offers a method to couple the useful
features of metal-catalysed C–H functionalization (stable and available reagents) and electrophilic acylations (broad scope
and selectivity), and synthesize ketones simply from aryl iodides, CO and arenes. Notably, the reaction proceeds in an
intermolecular fashion, without directing groups and at very low palladium-catalyst loadings. Mechanistic studies show
that the reaction proceeds through the catalytic build-up of potent aroyl triflate electrophiles.
he generation of reactive electrophiles is one of the corner- progress16–20. In contrast, the intermolecular carbonylative coupling
stones of traditional synthetic chemistry. A powerful illus- with simple arenes to generate valuable ketones is not known,
T
tration of this is in Friedel–Crafts reactions, wherein the and has to date required either intramolecular reactions or the
creation of sufficiently high-energy acylating reagents can lead to use of acidic/activated substrates, as shown by the groups of
the functionalization of even the unactivated aromatic C–H bonds Larock, Beller, Skrydstrup, Lei and others21–26. This has been
(Fig. 1a)1,2. The breadth and reliability of Friedel–Crafts acylations attributed, in part, to the common mechanistic pathway for
has made it a mainstay in the synthetic organic repertoire for the palladium-based C–H functionalization, in which the carboxylate
assembly of ketones, and recent advances have expanded the reagents often used in the activation of C–H bonds can react
scope of these reactions, as well as offered novel approaches to with the palladium–acyl intermediate much more rapidly
control product selectivity3–6. Nevertheless, an intrinsic limitation than with arenes (Fig. 1c). Instead, the formation of ketones by
of Friedel–Crafts chemistry is the required initial generation of the carbonylations typically requires coupling with pre-synthesized
electrophilic acylating compounds themselves. This usually entails organometallic agents27.
the use of reactive reagents (for example, SOCl2, PCl3 and strong
In considering these issues, we questioned if metal catalysed carbo-
Lewis acids) that are synthetic, high energy and create significant nylations might offer an alternative approach to arene C–H bond
chemical waste. The reactivity of acylating reagents also leads to functionalization, in which, rather than using transition metals as
significant challenges in their handling and functional group the active bond-functionalizing agent, carbon monoxide might
compatibility. In this regard, transition-metal-catalysed C–H bond instead be employed to drive the build-up of potent organic electro-
functionalization has rapidly evolved over the past decade to have philes capable of reacting with arenes (Fig. 1d). This could provide a
tremendous impact7–12. Relative to stoichiometric Friedel–Crafts method to combine the attractive features of transition-metal catalysis
reactions, metal-catalysed approaches can use benign reagents to (available reagents, functional group compatibility and atom
functionalize traditionally unreactive C–H bonds, generate minimal economy) with that of Friedel–Crafts chemistry (high reactivity,
waste and often display high functional group compatibility.
intermolecular and electrophilic selectivity). The development of
Although metal-catalysed C–H functionalization can be advan- this transformation would require a catalyst that can mediate the cur-
tageous, there remain a number of limitations in this chemistry rently unknown reductive elimination of an electrophile sufficiently
relative to classical electrophilic substitution reactions. For reactive to functionalize simple arenes, in contrast with the highly
example, the catalytic functionalization of simple arenes can often favoured and well-established reverse oxidative addition of acylating
require substrate activation, high catalyst or substrate concen- electrophiles to palladium28,29. We describe below our studies
trations, or the presence of directing groups in order to proceed towards this goal. These have led to the discovery of a palladium-cat-
with high efficiency13,14. In addition, the incorporation of reactive alysed platform toperformthe carbonylative C–H functionalizationof
functionalities into catalytic C–H derivatization platforms can be arenes to form ketones directly from broadly available and stable
plagued by catalyst inhibition. An important example is in the car- reagents (aryl halides, arenes and CO). This transformation occurs
bonylative synthesis of ketones (Fig. 1b). The palladium-catalysed in common solvents, with low catalyst loadings, uses simple
carbonylative C–H functionalization of arenes to form esters, palladium catalysts and illustrates how carbonylation reactions can
amides and carboxylic acids was pioneered by Fujiwara and be employed efficiently to build-up potent electrophilic acylating
co-workers over two decades ago15, and has seen significant recent agents: aroyl triflates.
Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal, Quebec H3A 0B8, Canada. e-mail: bruce.arndtsen@mcgill.ca
*
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
1
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
allow the formation of acid halides (ArCOX, X = Cl, I). However,
no evidence for benzene functionalization was observed, and we
instead simply recovered the starting materials (Fig. 2a). This reac-
tivity mirrors that of aroyl iodides, the presumed intermediate gen-
erated from the carbonylation of aryl iodide, which also cannot
directly react with benzene (Supplementary Fig. 1). In considering
approaches to enhance the electrophilicity, one possibility would
be to employ anions that are even more ionizable, such as triflate.
Aroyl triflates are exceptionally potent electrophiles that are typi-
cally generated in situ for aroylation reactions37. Unfortunately,
their high reactivity would not only make their reductive elimin-
ation challenging, but also lead to their much more rapid re-
addition to Pd(0) than that of the aryl halide reagent (Fig. 2b).
The screening of many triflate salts in this catalytic reaction did
not lead to any reaction (Supplementary Table 1). However, we
were pleased to find that the addition of silver triflate resulted in
the carbonylative functionalization of benzene into ketone 1a in
low yield (12%, Fig. 2c).
a
O
H
O
Friedel–Crafts
+
Ar
Lewis
acid
Ar
Cl
High-energy
electrophiles
b
O
M
Cross-coupling
LnPd
Ar
X
Ar
+
CO
+
Organometallic
reagents
c
H
Ar
O
O
Rapid
Ketones
Ar
LnPd
Nu = O2CR,
ROH, RNH2
Ar
Nu
Unless activated
or intramolecular Ar–H
O2CR
d
This work
O
H
I
Pd cat.
OTf
+
CO
+
Interestingly, further studies showed that a number of phosphine
ligands can be employed in this catalytic reaction to generate ketone
in comparable yields (Fig. 2c, entries 2–5). This contrasts with the
catalytic formation of acid chlorides and iodides, in which the steri-
cally encumbered PtBu3 ligand is critical to drive reductive elimin-
ation as a mechanism to relieve strain38. The origin of the lack of
ligand influence can be seen by monitoring the catalytic reaction
R2
R1
Available feedstock chemicals
R2
R1
O
OTf
1
with PtBu3 by H and 31P NMR analysis, which shows that PtBu3
R1
Electrophile formation
with CO
a
O
2.5% Pd2dba3
10% PtBu3
I
CO
+
+
Figure 1 | Comparison of the classical approaches to ketone synthesis with
the palladium-catalysed carbonylative C–H bond functionalization reaction
described here. a, Typical synthetic approach to ketones by Friedel–Crafts
reactions with synthetic electrophiles. b, Ketone synthesis by carbonylative
cross-coupling reactions with organometallic reagents. Known for
Benzene
4 atm, 24 h, 100 °C
1a
0% yield
O
Steric
strain
b
Ar
X
organometallic reagents (such as with M = R3Sn or R2B), not known for
M = H. c, A challenge in applying palladium-catalysed C–H bond
Ar
X
O
(X = Cl, I)
Pd[PtBu3]2
?
Pd PtBu3
CO
CO X
+
Ar
I
+
functionalizations to carbonylative ketone synthesis is the reaction of the
palladium–acyl intermediate with bases, and typically requires the use of
activated heteroarenes, acidic perfluoroarenes or intramolecular reactions.
d, Concept reported here: the palladium-catalysed formation of high-energy
aroyl electrophiles as an alternative approach to carbonylative arene C–H
functionalization. Ar, aryl group; cat, catalyst; L, ligand; M, metal fragment;
Nu, nucleophile; R, alkyl chains; R1 and R2, substituents; X, halogen.
O
Favoured
reverse
Ar
OTf
c
2.5% [Pd]
10% ligand
AgOTf
O
I
+
+
CO
Benzene
4 atm, 24 h, 100 °C
1a
Results and discussion
The catalytic formation of organic electrophiles sufficiently reactive
to derivatize aromatic C–H bonds, yet doing so from stable and
available substrates, at first glance appears energetically challenging.
However, carbonylations may provide a unique entry to such
systems. Carbon monoxide is an inexpensive feedstock chemical
broadly employed in the assembly of carbonyl-containing
products30,31. In addition, a less-appreciated feature of carbon
monoxide is its energetics, in which the conversion of CO into a
carbonyl-containing product is often highly exergonic (by up to 40
kcal mol–1) (ref. 32). We have recently exploited this feature to gen-
erate aromatic acid chlorides and iodides from aryl halides and CO,
[Pd]
Entry
Ligand
Yield 1a
PtBu3
PPh3
1
2
Pd2dba3
Pd2dba3
12%
31%
9%
Pd2dba3
PoTol3
PCy3
P(C6F6)3
–
3
4
5
6
7
Pd2dba3
21%
Pd2dba3
29%
99%
92%
[Pd(allyl)Cl]2
5% PdCl2
–
0.0075%
[Pd(allyl)Cl]2
8
–
95%
with the latter having the ability to functionalize electron-rich N-het- Figure 2 | Palladium-catalyst development for the intermolecular
erocycles33–35. Morandi and co-workers have also shown that alkyl carbonylative coupling of aryl iodide and benzene into ketones. a, Attempt
and vinyl acid chlorides can be generated by palladium-catalysed at benzene carbonylation by in situ aroyl iodide generation leads to no
functional group exchange36. Based on these results, we questioned product. b, Mechanism of catalytic acid halide formation, and the challenge
if the energetics of CO might be taken even further, and be of generating even more potent aroyl triflate electrophiles. c, Catalyst design
employed to drive the generation of electrophilic acylating agents for arene carbonylation into ketones with AgOTf: p-tolyl iodide (55 mg,
sufficiently reactive to directly derivatize simple arenes.
0.25 mmol); Pd2dba3 (6 mg, 0.0062 mmol); ligand (0.025 mmol); AgOTf
To probe this potential, our initial studies examined the functio- (97 mg, 0.375 mmol); benzene (2 ml); and 4 atm CO. 1H NMR yield. Entry 8
nalization of benzene with the Pd/PtBu3 catalyst system found to performed in DCE solvent. Cy, cyclohexyl; dba, dibenzylideneacetone.
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
a
H
1
1
O
LnPd(0)
O
Ar
Possible mechanistic
pathways
(CO)nPd
Ar
OTf
Ar
CO
I
Path B
AgI
Path C
Path A
O
O
Ar
OTf
Ar
OTf
H
(CO)nPd
(CO)nPd
O
Ag
2
2
Ar
(CO)nPd
AgI
AgI
I
AgOTf
O
AgOTf
b
O4
O1
O3
I
2.5 mol% [Pd(allyl)Cl]2
2 h, 100 °C, DCE
C2
S1
F2
OTf
C8
C5
AgOTf
+
CO
(4 atm)
+
C3
O2
F1
C7
C6
Br
Br
C4
F3
3a
90% yield
Br1
With phosphine
5 min
Without phosphine
pTol
c
d
O
pTol
O
O
pTol
4 atm CO
HOTf
O
O
r.t.
C1
O1
P1
Pd[PtBu3]2
OTf
+
TfO Pd PtBu3
OTf
tBu3P Pd OTf
(CO)nPd OTf
C6D6
30 min
C6D6, r.t.
Pd1
O2
– tBu3PH+OTf
4b
3b
4b
2b
3b
83% yield
O3
S1
C6D6 4 atm
O4
F3
F2
CO
24 h
F1
Slow decomposition
Figure 3 | Mechanistic studies demonstrate this transformation proceeds through the novel catalytic formation of aroyl triflate electrophiles. a, Plausible
mechanisms for ketone formation include the in situ formation of aroyl triflate electrophiles (path A), palladium-based arene functionalization (path B) or the
formation of a silver–aryl intermediates (path C). Ar, aryl group; L, ligand. b, Removal of benzene from the catalytic reaction leads to the formation of aroyl
triflate (ORTEP-style ellipsoids drawn at 50% probability). c, Aroyl triflates undergo rapid oxidative addition to PtBu3-coordinated palladium. d, The
sequestration of PtBu3 with HOTf addition leads to the rapid, near-quantitative elimination of aroyl triflate from palladium. r.t., room temperature.
is rapidly consumed by the strong electrophiles generated in cataly- However, it is in line with common C–H/D isotope effects in
sis to form protonated phosphine and various degradation products Friedel–Craft acylations with aroyl triflates (KIE = 1.1–1.4) (ref. 43).
(Supplementary Fig. 2). As a result, simply removing the phosphine Even greater evidence for the mechanism can be seen by monitoring
ligand from the reaction and employing palladium(II) salts as the catalytic reaction by NMR analysis. As illustrated in Fig. 3b,
catalyst leads to the intermolecular derivatization of benzene in removing benzene from the [Pd(allyl)Cl]2-catalysed reaction of
near-quantitative yield (Fig. 2c, entries 6 and 7).
aryl iodide, CO and silver triflate leads to the high-yield
The unusual ability of simple palladium salts with no added generation of the potent electrophile aroyl triflate 3a as an
ligands to catalyse the high-yield intermolecular functionalization isolable product.
of benzene with CO, aryl iodide and a triflate source raises a
As far as we are aware, aroyl triflates have not been observed
number of mechanistic questions. The transformation presumably before as products of palladium catalysis or carbonylation chem-
proceeds through the formation of a palladium–aroyl intermediate istry. The identity of this product was confirmed by both in situ
from oxidative addition followed by CO insertion (Fig. 3a). One multinuclear NMR analysis and, as shown in Fig. 3b, X-ray crystal-
possibility is that this palladium intermediate may undergo reduc- lography. The subsequent reaction of this aroyl triflate with benzene
tive elimination of an aroyl triflate electrophile (path A). However, occurs in the absence of any catalyst, and with the same KIE as
in the absence of any phosphine to favour elimination, we also con- noted above (kH/kD = 1.27 (Supplementary Fig. 3)), which confirms
sidered the potential that the non-phosphine coordinated Pd–COAr the viability of path A as the mechanism for the catalytic reaction.
intermediate 2 may be sufficiently electrophilic to palladate benzene The actual build-up and isolation of 3a is surprising, as its potent
(path B), or the silver complex could itself be involved in C–H bond electrophilicity would normally be thought to result in its more
activation (path C), similar to recent results of Hartwig39, Larossa40 favoured re-oxidative addition to palladium. For example, the
and Sanford41 and their co-workers. To gain some insight into addition of 3b to Pd(PtBu3)2 leads to the quantitative formation
which pathway is operative, a series of mechanistic studies were of the oxidative addition product 4b within minutes at ambient
performed. The competition kinetic isotope effect (KIE) for the temperature (Fig. 3c). However, such a rapid oxidative addition is
reaction is 1.24 (Supplementary Fig. 3), which is much lower than not necessarily the case with the more electron-deficient, non-
that previously observed for arene palladation or the formation of ligated palladium salt used as catalyst. By itself, the PtBu3-
a silver–aryl intermediate (KIE = ∼4.5 and 2.3, respectively)39,40,42
.
coordinated palladium–aroyl triflate complex 4b does not eliminate
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
3

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
Table 1 | Scope of the carbonylative C–H functionalization of arenes and heteroarenes.
O
O
0.05–2.5 mol% [Pd(allyl)Cl]2
AgOTf
I
X
+
CO
+
or
or
R2
4 atm CO, DCE,
24 h, r.t. to 100 °C
X
R2
O
R2
R1
R2
R1
R1
Aryl iodides
O
O
O
O
Cl
O
Ph
Ph
Ph
Ph
Ph
tBu
O
R
R = H (1e), 92%
1b, 87%
1c, 85%
1d, 87%
1g, 98%
1h, 79%
Br
R = SO3Me (1f), 71%
O
O
O
O
O
O
Ph
Ph
Ph
Ph
Ph
Ph
F
F3C
O2N
O
EtO2C
CO2Me
O
1i, 85%
1j, 82%
1k, 83%
1l, 77%
1m, 92%
1n, 79%
Arenes
R
O
O
O
O
O
pTol
pTol
Ph
pF(C6H4)
pTol
pTol
N
Ts
R
Cl
1s, 72%
(p:o, 3.7:1)
1t, 96%
1u, 97%
81% recovery of arene
R = Me (1o), 88%
R = OMe (1p), 91%
1q, 94%
1r, 85%
O
O
Cl
O
F
O
O
R = SMe (1v), 85%
R = OMe (1w), 91%
R = OSO2Me (1x), 79%
R = OH (1y), 47%
pFC6H4
pFC6H4
pTol
pTol
pTol
Cl
F
F
Cl
R
1z, 95%
1aa, 82%
1bb, 84%
1cc, 69%
(a:b, 3:1)
(p:o, 5.5:1)
Heteroarenes
O
O
O
O
O
O
S
O
S
pTol
pTol
pTol
N
pTol
pTol
pTol
R
N
N
Bn
Br
1dd, 73%
(2-:3- = 2.3:1)
1ee, 54%
R = Ms (1ff), 88%
R = Ts (1gg), 99%
1hh, 81%
1ii, 96%
1jj, 86%
(5-:2- = 2.6:1)
O
O
O
O
O
O
O
S
S
pBrC6H4
pTol
pTol
O
S
N
Ms
O
O
1kk, 95%
1ll, 70%
1mm, 58%
1nn, 55%
1oo, 40%
(a:b, 1:2.7)
(a:b = 3.3:1)
Vinyl iodides
O
O
O
O
O
O
O
EtO
S
S
S
S
S
Ph
Ph
R
O
N
1qq, 86%
1pp, 82%
R = Ph (1rr), 83%
1tt, 53%
1uu, 84%
1vv, 84%
Ms
O
O
R = fluorenyl (1ss), 81%
O
O
O
O
EtO
S
S
S
Ph
Ph
Ph
O
N
1ww, 92%
1xx, 99%
1yy, 87%
1zz, 94%
1aaa, 98%
Ts
Ar–I (1 mmol), arene (2 mmol), [Pd(allyl)Cl]2 (0.2 mg, 5 × 10−4 mmol, from stock solution), AgOTf (385 mg, 1.5 mmol), 4 ml DCE, 4 atm CO, 100 °C, 24 h. 1uu–1ww and 1u formed with 2,6-di-t-butylpyridine
(50 mg, 0.30 mmol). Electron-rich arenes (<80 °C) and vinyl iodides (r.t.) react at lower temperatures; heterocycle added after aroyl triflate formation for 1dd, 1ee, 1hh, 1ll and 1mm (Supplementary Information
gives full details). Excess arene can be recovered (for example, 1u). 1xx–aaa are generated by hydrogenation of the vinyl ketone (0.15 mmol), Pd/C (38 mg, 5% w/w Pd), 3 atm H2, 55 °C, 16 h. X = N, O, S;
Bn = benzyl; Ts = p-toluenesulfonyl; Ms = methanesulfonyl.
aroyl triflate, and slowly decomposes into a range of products in the aroyl triflate in this system is facilitated by the electron-deficient
presence of CO. In contrast, the addition of HOTf to protonate the Pd(II) catalyst generated in the presence of CO that can undergo a
phosphine and generate in situ a presumably CO-bound palladium rapid reductive elimination of even a highly electrophilic product
intermediate 2b leads to the rapid, near-quantitative elimination of such as 3.
aroyl triflate 3b as a product together with protonated PtBu3
In addition to offering a method to generate high-energy aroyl
(Fig. 3d). As such, we postulate that the catalytic generation of triflate electrophiles as the products of carbonylation, a notable
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
HO
O
a
b
2.5 mol%
[Pd(allyl)Cl]2
AgOTf
1) 2.5% [Pd(allyl)Cl]2
AgOTf, 2,6-ditbutylpyridine
24 h, 100 °C, 4 atm
O
O
O
O
H
H
+
I
I
+ CO +
H
+
R1
CO
R1
4 atm, DCE,
60 oC, 24 h
R2
2) LiOH, THF/H2O,
24 h
R2
Ketoprofen (5)
50% yield
5% [Pd(allyl)Cl]2
2,6-ditbutylpyridine
AgX
O
O
O
O
I
O
nBu
O
O
O
O
S
O
O
+
+
CO
30 atm, 65 °C,
24 h, DCE
Ph
Cl
O
Fenofibrate (6)
X = OTf, 57% yield
X = NTf2, 83% yield
Cl
H
7a, 60%
7b, 48%
7c, 57%
c
R
O
2.5 mol% [Pd(allyl)Cl]2
AgNTf2
I
pTol
+
+
CO
4 atm, DCE, 24 h
O
R
pTol
NTf2
O
O
O
O
O
S
pTol
pTol
pTol
pTol
Ph
O
Cl
1ii, 87%
(r.t.)
1a, 97%
(50 °C)
1w, 98% (p:o > 10:1)
1bbb, 97% (p:o > 10:1)
1ccc, 89% (p:o = 8:1)
(r.t.)
(50 °C)
(50 °C)
Figure 4 | Applications of this general methodology to the synthesis of pharmaceutically relevant molecules, double C–H bond functionalization and
low-temperature C–H bond functionalization. a, Synthesis of ketoprofen and fenofibrate by arene carbonylation. b, Palladium-catalysed double C–H
functionalization and indanone synthesis with vinyl iodides by in situ acid-mediated cyclization. c, Generation of the even-more electrophilic aroyl triflimides
allows the carbonylative C–H functionalization of arenes to proceed at ambient temperature and with high regioselectivity (p:o selectivity with AgOTf: 1bbb,
6:1; 1ccc, 5:1).
feature is that the catalytic formation of 3 is more rapid than its sub- pyrroles and indoles, which can react to generate1ddand 1ee, respect-
sequent reaction with benzene (Fig. 3b and Supplementary Fig. 4). ively. Less-activated heterocycles can be similarly functionalized,
The decoupling of the low-concentration palladium catalyst from including furan (1hh), substituted furans (1ll), thiophenes (1ii–1kk)
the bond-functionalization step leads to a number of useful features. or heterocycles with electron-withdrawing substituents (1ff and
For example, it is straightforward to perform benzene C–H functio- 1gg). Heteroaryl iodides are also viable reagents in this reaction to
nalization in common organic solvents (such as 1,2-dichloroethane build-up diheteroaryl ketones (1oo).
(DCE) and dichloromethane), rather than in neat arene solvent,
Vinyl halides were also explored as substrates. Unlike aryl-substi-
because the arene can react with the organic aroyl triflate product tuted acid triflates, vinoyl triflates are often highly labile and difficult
built up during catalysis (Supplementary Table 2). The catalyst load- to generate by traditional methods47. As shown in Table 1, these elec-
ings for arene functionalization can also be significantly lowered, trophiles can be formed in situ through palladium-catalysed
and the reaction proceeds in near-quantitative yields with as little carbonylations and allow the formation of synthetically useful α,β-
as 0.0075 mol% catalyst (or 150 ppm Pd (Fig. 2c, entry 8)). As far unsaturated ketones 1pp–1ww with electron-rich arenes. In the
as we are aware, this represents one of the lowest catalyst loadings case of 1uu–1ww, a bulky 2,6-di-t-butylpyridine base is required to
known for intermolecular arene C–H functionalization44–46
.
remove HOTf. The formation of these products can also be combined
We next examined the generality of this approach to ketone syn- with their subsequent reactivity. As an example, coupling the catalytic
thesis. Examples are shown in Table 1. A wide range of aryl iodides formation of these α,β-unsaturated ketones with hydrogenation can
are compatible with this reaction, including those with electron- provide an overall approach to prepare alkyl-substituted ketones
donating (1a–1d) and electron-withdrawing (1f–1m) substituents. (1xx–1aaa). As such, this offers a general method to perform carbo-
Potentially reactive functionalities can be incorporated, such as nylative arene C–H bond functionalization to generate ketones.
esters (1i and 1l) and nitro substituents (1k), and the chemoselectivity
Ketones are valuable building blocks in synthetic chemistry, as
for aryl iodide oxidative addition makes this chemistry tolerant of well as useful products themselves in many areas of application.
other halogen-containing substituents (1g and 1h). Benzylic C–H The ability to generate these compounds from stable and broadly
bonds, which often exhibit a higher reactivity than arenes in many available aryl iodides, CO and arenes can therefore provide an
C–H functionalization systems, are readily tolerated (1a, 1b and attractive alternative to classical synthetic methods. As an illus-
1d). The reactions can be performed with an equimolar amount of tration, this transformation can be applied to the targeted formation
arene with a slightly lower yield (80% 1a (Supplementary Fig. 5)). It of pharmaceuticals such as ketoprofen, an effective non-steroidal
is also straightforward to tune the arene that is derivatized. anti-inflammatory drug48, from simply benzene, CO and an aryl
Examples include benzene and various electron-rich arenes (1o–1z). iodide in the presence of base to inhibit ester decomposition
With more-elevated temperatures, deactivated arenes can also be (Fig. 4a). Similarly, fenofibrate, a triglyceride and cholesterol regula-
employed, such as chlorobenzenes (1aa and 1bb) and trifluoroben- tor49, can be generated by the carbonylative coupling of 4-chloroio-
zene (1cc), with the selectivity mirroring that of Friedel–Crafts chem- dobenzene and the appropriate arene. Alternatively, the catalytic
istry1,2. The electrophilicity of the aroyl triflate intermediates in this formation of ketones can also be coupled with their subsequent
chemistry can also be applied to functionalize heterocycles, such as reactivity. Performing the carbonylation of arenes with vinyl
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
5

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
6. Liu, Y., Meng, G., Liu, R. & Szostak, M. Sterically-controlled intermolecular
Friedel–Crafts acylation with twisted amides via selective N–C cleavage under
mild conditions. Chem. Commun. 52, 6841–6844 (2016).
7. Hartwig, J. F. Evolution of C–H bond functionalization from methane to
methodology. J. Am. Chem. Soc. 138, 2–24 (2015).
8. Lyons, T. W. & Sanford, M. S. Palladium-catalyzed ligand-directed C−H
functionalization reactions. Chem. Rev. 110, 1147–1169 (2010).
9. Colby, D. A., Bergman, R. G. & Ellman, J. A. Rhodium-catalyzed C−C bond
formation via heteroatom-directed C−H bond activation. Chem. Rev. 110,
624–655 (2010).
10. Chen, X., Engle, K. M., Wang, D.-H. & Yu, J.-Q. Palladium(II)-catalyzed C–H
activation/C–C cross-coupling reactions: versatility and practicality.
Angew. Chem. Int. Ed. 48, 5094–5115 (2009).
11. Liu, C. et al. Oxidative coupling between two hydrocarbons: an update of recent
C–H functionalizations. Chem. Rev. 115, 12138–12204 (2015).
12. Cernak, T., Dykstra, K. D., Tyagarajan, S., Vachal, P. & Krska, S. W. The
medicinal chemist’s toolbox for late stage functionalization of drug-like
molecules. Chem. Soc. Rev. 45, 546–576 (2016).
13. Kuhl, N., Hopkinson, M. N., Wencel-Delord, J. & Glorius, F. Beyond directing
groups: transition-metal-catalyzed C–H activation of simple arenes.
Angew. Chem. Int. Ed. 51, 10236–10254 (2012).
iodides under mild heating in the absence of base leads to the formal
double C–H functionalization of arenes (Fig. 4b). The latter presum-
ably proceeds through subsequent Friedel–Crafts cyclization from
the α,β-unsaturated ketone product under the acidic reaction con-
ditions50, and offers a direct method to form indanones from arenes.
Finally, we have probed if this approach can offer a method to
build-up electrophiles that are even more reactive. Relative even
to triflates, triflimides are exceptional leaving groups. Thus,
performing the catalytic carbonylation of aryl iodides in the pres-
ence of a triflimide salt can allow the intermolecular carbonylative
functionalization of arenes into ketones at ambient temperature
(Fig. 4c). To our knowledge, electrophilic aroyl triflimides have
not been reported. Although these electrophiles are more labile,
their formation through carbonylation can even allow the derivati-
zation of deactivated aromatic substrates at temperatures as low as
50 °C and with high regioselectivity. The latter presumably arises
from a combination of the lower reaction temperatures and larger
triflimide anion that directs this chemistry towards the selective for-
mation of para-derivatized arenes in high yield and under
mild conditions.
14. Hartwig, J. F. & Larsen, M. A. Undirected, homogeneous C–H bond
functionalization: challenges and opportunities. ACS Cent. Sci. 2,
281–292 (2016).
15. Jia, C., Kitamura, T. & Fujiwara, Y. Catalytic functionalization of arenes and
alkanes via C−H bond activation. Acc. Chem. Res. 34, 633–639 (2001).
16. Liu, Q., Zhang, H. & Lei, A. Oxidative carbonylation reactions: organometallic
compounds (R–M) or hydrocarbons (R–H) as nucleophiles. Angew. Chem. Int.
Ed. 50, 10788–10799 (2011).
17. Bai, Y., Davis, D. C. & Dai, M. Natural product synthesis via palladium-catalyzed
carbonylation. J. Org. Chem. 82, 2319–2328 (2017).
18. Yang, L. & Huang, H. Transition-metal-catalyzed direct addition of
unactivated C–H bonds to polar unsaturated bonds. Chem. Rev. 115,
3468–3517 (2015).
19. Willcox, D. et al. A general catalytic β-C–H carbonylation of aliphatic amines to
β-lactams. Science 354, 851–857 (2016).
20. Li, S., Chen, G., Feng, C.-G., Gong, W. & Yu, J.-Q. Ligand-enabled γ-C–H
olefination and carbonylation: construction of β-quaternary carbon centers.
J. Am. Chem. Soc. 136, 5267–5270 (2014).
21. Campo, M. A. & Larock, R. C. Synthesis of fluoren-9-ones by the
palladium-catalyzed cyclocarbonylation of o-halobiaryls. J. Org. Chem. 67,
5616–5620 (2002).
Conclusions
In summary, we have developed a conceptually new approach to
intermolecular arene C–H functionalization to form ketones. The
reaction proceeds through the unprecedented palladium-catalysed
formation of aroyl triflates, and demonstrates how carbonylations
can be employed to drive the formation of exceptionally reactive elec-
trophiles from available reagents. This opens a method to couple
metal-catalysed arene C–H bond functionalization with electrophilic
substitution chemistry to generate ketones in high yield with low pal-
ladium loadings, and from aryl iodides, CO and arenes. Considering
the diverse reactivity of the high-energy acylating agents, we anti-
cipate this approach will find utility as a general platform to
functionalize unreactive bonds by carbonylation chemistry.
22. Wu, X.-F., Anbarasan, P., Neumann, H. & Beller, M. Palladium-catalyzed
carbonylative C–H activation of heteroarenes. Angew. Chem. Int. Ed. 49,
7316–7319 (2010).
23. Lian, Z., Friis, S. D. & Skrydstrup, T. C–H activation dependent Pd-catalyzed
carbonylative coupling of (hetero)aryl bromides and polyfluoroarenes.
Chem. Commun. 51, 1870–1873 (2015).
24. Zhang, H. et al. Palladium-catalyzed oxidative double C–H functionalization/
carbonylation for the synthesis of xanthones. Angew. Chem. Int. Ed. 51,
5204–5207 (2012).
25. Cernak, T. A. & Lambert, T. H. Multicatalytic synthesis of α-pyrrolidinyl ketones
via a tandem palladium(II)/indium(III)-catalyzed aminochlorocarbonylation/
Friedel−Crafts acylation reaction. J. Am. Chem. Soc. 131, 3124–3125 (2009).
26. Zeng, F. & Alper, H. Pd-catalyzed direct coupling of indoles with carbon
monoxide and alkynes: selective synthesis of linear α,β-unsaturated ketones.
Org. Lett. 15, 2034–2037 (2013).
Methods
General procedure. Under an inert nitrogen atmosphere, silver triflate (386 mg,
1.5 mmol) was transferred to a Teflon sealed thick-walled 50 ml glass reaction vessel
equipped with a stir bar, followed by aryl iodide (1.0 mmol), arene (2.0 mmol),
DCE (4 ml) and then a freshly prepared stock solution of [Pd(allyl)Cl]2 (0.2 mg,
5 × 10−4 mmol). The vessel was closed, removed from the glovebox, evacuated and
backfilled with carbon monoxide three times, and finally pressurized with 4 atm
carbon monoxide. After heating at 100 °C for 24 h with stirring, the reaction was
cooled to room temperature and carbon monoxide was released. The reaction
mixture was filtered through Celite, eluting with dichloromethane. Saturated
NaHCO3 was added and the aqueous layer was extracted with dichloromethane.
The combined organic layers were concentrated in vacuo and the residue was
purified by column chromatography (silica gel, gradient hexane/ethyl acetate 0 to
20%) to afford the pure ketone product.
27. Wu, X.-F., Neumann, H. & Beller, M. Palladium-catalyzed carbonylative
coupling reactions between Ar–X and carbon nucleophiles. Chem. Soc. Rev. 40,
4986–5009 (2011).
28. Blangetti, M., Rosso, H., Prandi, C., Deagostino, A. & Venturello, P.
Suzuki–Miyaura cross-coupling in acylation reactions, scope and recent
developments. Molecules 18, 1188–1213 (2013).
29. Kuniyasu, H. et al. σ-Bond metathesis between M–X and RC(O)X′ (M = Pt,
Pd; X, X′ = Cl, Br, I): facile determination of the relative ΔG values of the
oxidative additions of RC(O)X to an M(0) complex, evidence by density
functional theory calculations, and synthetic applications. Organometallics 32,
2026–2032 (2013).
Data availability. All data generated and analysed during this study are included in
this article and its Supplementary Information files, and are also available from the
authors on reasonable request. Crystallographic data have been deposited at the
Cambridge Crystallographic Data Centre (CCDC) as CCDC 1513679 (3a) and
1554343 (4b) and can be obtained free of charge from the CCDC via
http://www.ccdc.cam.ac.uk/getstructures.
Received 14 July 2017; accepted 30 October 2017;
published online 11 December 2017
30. Beller, M. & Wu, X. F. Transition Metal Catalyzed Carbonylations –
Carbonylative Activation of C–X bonds (Springer, 2013).
31. Grigg, R. & Mutton, S. P. Pd-catalysed carbonylations: versatile technology for
discovery and process chemists. Tetrahedron 66, 5515–5548 (2010).
32. Luo, Y.-R. Handbook of Bond Dissociation Energies in Organic Compounds
(CRC, 2003).
33. Quesnel, J. S. & Arndtsen, B. A. A palladium-catalyzed carbonylation approach
to acid chloride synthesis. J. Am. Chem. Soc. 135, 16841–16844 (2013).
34. Quesnel, J. S., Kayser, L. V., Fabrikant, A. & Arndtsen, B. A. Acid chloride
synthesis by the palladium-catalyzed chlorocarbonylation of aryl bromides.
Chem. Eur. J. 21, 9550–9555 (2015).
References
1. Olah, G. A. Friedel–Crafts and Related Reactions (Interscience, 1963).
2. Sartori, G. & Maggi, R. Advances in Friedel–Crafts Acylation Reactions: Catalytic
and Green Processes (CRC, 2010).
3. Boursalian, G. B., Ham, W. S., Mazzotti, A. R. & Ritter, T. Charge-transfer-
directed radical substitution enables para-selective C–H functionalization.
Nat. Chem. 8, 810–815 (2016).
4. Bähr, S. &Oestreich, M. Electrophilicaromaticsubstitutionwithsiliconelectrophiles:
catalytic Friedel–Crafts C−H silylation. Angew. Chem. Int. Ed. 56, 52–59 (2017).
5. Vekariya, R. H. & Aubé, J. Hexafluoro-2-propanol-promoted intermolecular
Friedel–Crafts acylation reaction. Org. Lett. 18, 3534–3537 (2016).
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2903
ARTICLES
35. Tjutrins, J. & Arndtsen, B. A. An electrophilic approach to the palladium-
catalyzed carbonylative C–H functionalization of heterocycles. J. Am. Chem. Soc.
137, 12050–12054 (2015).
36. Fang, X., Cacherat, B. & Morandi, B. CO- and HCl-free synthesis of acid
chlorides from unsaturated hydrocarbons via shuttle catalysis. Nat. Chem. 9,
1105–1109 (2017).
47. Effenberger, F., Epple, G., Eberhard, J. K., Buehler, U. & Sohn, E. Electrophilic
aromatic substitution. 24. Carboxylic trifluoromethanesulfonic and
methanesulfonic anhydrides synthesis and dissociation tendency. Chem. Ber.
116, 1183–1194 (1983).
48. Kantor, T. G. Ketoprofen: a review of its pharmacologic and clinical properties.
Pharmacotherapy 6, 93–102 (1986).
49. Keating, G. M. Fenofibrate: a review of its lipid-modifying effects in dyslipidemia
and its vascular effects in type 2 diabetes mellitus. Am. J. Cardiovasc. Drugs 11,
227–247 (2011).
37. Effenberger, F., Eberhard, J. K. & Maier, A. H. The first unequivocal evidence of
the reacting electrophile in aromatic acylation reactions. J. Am. Chem. Soc. 118,
12572–12579 (1996).
50. Yin, W., Ma, Y., Xu, J. & Zhao, Y. Microwave-assisted one-pot synthesis of
1-indanones from arenes and α,β-unsaturated acyl chlorides. J. Org. Chem. 71,
4312–4315 (2006).
38. Quesnel, J. S. et al. Computational study of the palladium-catalyzed
carbonylative synthesis of aromatic acid chlorides: the synergistic effect of PtBu3
and CO on reductive elimination. Chem. Eur. J. 22, 15107–15118 (2016).
39. Lee, S. Y. & Hartwig, J. F. Palladium-catalyzed, site-selective direct allylation of
aryl C–H bonds by silver-mediated C–H activation: a synthetic and mechanistic
investigation. J. Am. Chem. Soc. 138, 15278–15284 (2016).
40. Whitaker, D., Burés, J. & Larrosa, I. Ag(I)-catalyzed C–H activation: the role of
the Ag(I) salt in Pd/Ag-mediated C–H arylation of electron-deficient arenes.
J. Am. Chem. Soc. 138, 8384–8387 (2016).
41. Lotz, M. D., Camasso, N. M., Canty, A. J. & Sanford, M. S. Role of silver salts in
palladium-catalyzed arene and heteroarene C–H functionalization reactions.
Organometallics 36, 165–171 (2017).
42. Lapointe, D. & Fagnou, K. Overview of the mechanistic work on the concerted
metallation–deprotonation pathway. Chem. Lett. 39, 1118–1126 (2010).
43. Effenberger, F. & Maier, A. H. Changing the ortho/para ratio in aromatic
acylation reactions by changing reaction conditions: a mechanistic explanation
from kinetic measurements. J. Am. Chem. Soc. 123, 3429–3433 (2001).
Acknowledgements
We thank Natural Sciences and Engineering Research Council of Canada, the Canadian
Foundation for Innovation and the Centre for Green Chemistry and Catalysis
(supported by Fonds de recherche du Québec – Nature et Technologies) for funding
this research. We thank L. Kayser for the X-ray crystal structure of 3a and to S. Kelley for
the X-ray crystal structure of 4b.
Author contributions
B.A.A. and R.G.K. conceived the project. R.G.K. conducted the majority of the experiments.
J.T. assisted in the early experiments. G.M.T. conducted the mechanistic experiments
with 4b and assisted with Table 1 and Fig. 4b. N.I.L. and O.K. assisted with Table 1. B.A.A.
and R.G.K. prepared the manuscript with feedback from all of the authors.
44. Boller, T. M. et al. Mechanism of the mild functionalization of arenes by diboron Additional information
reagents catalyzed by iridium complexes. intermediacy and chemistry of
bipyridine-ligated iridium trisboryl complexes. J. Am. Chem. Soc. 127,
14263–14278 (2005).
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online at
www.nature.com/reprints. Publisher’s note: Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations. Correspondence and
requests for materials should be addressed to B.A.A.
45. Press, L. P., Kosanovich, A. J., McCulloch, B. J. & Ozerov, O. V. High-turnover
aromatic C–H borylation catalyzed by POCOP-type pincer complexes of
iridium. J. Am. Chem. Soc. 138, 9487–9497 (2016).
46. Obligacion, J. V., Semproni, S. P. & Chirik, P. J. Cobalt-catalyzed C–H
borylation. J. Am. Chem. Soc. 136, 4133–4136 (2014).
Competing financial interests
The authors declare no competing financial interests.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.