Positional Selectivity in C-H Functionalizations of 2-Benzylfurans with Bimetallic Catalysts

Article abstract of DOI:10.1021/jacs.6b01578

Metal-catalyzed carbon-carbon bond-forming reactions are a mainstay in the synthesis of pharmaceutical agents. A long-standing problem plaguing the field of transition metal catalyzed C-H functionalization chemistry is control of selectivity among inequivalent C-H bonds in organic reactants. Herein we advance an approach to direct site selectivity in the arylation of 2-benzylfurans founded on the idea that modulation of cooperativity in bimetallic catalysts can enable navigation of selectivity. The bimetallic catalysts introduced herein exert a high degree of control, leading to divergent site-selective arylation reactions of both sp² and sp³ C-H bonds of 2-benzylfurans. It is proposed that the selectivity is governed by cation-π interactions, which can be modulated by choice of base and accompanying additives [MN(SiMe₃)₂, M = K or Li·12-crown-4].

Full text of DOI:10.1021/jacs.6b01578

Subscriber access provided by GAZI UNIV
Article
Positional Selectivity in C–H Functionalizations
of 2-Benzyl Furans with Bimetallic Catalysts
Jiadi Zhang, Sheng-Chun Sha, Ana Bellomo, Nisalak
Trongsiriwat, Feng Gao, Neil C. Tomson, and Patrick J Walsh
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b01578 • Publication Date (Web): 03 Mar 2016
Downloaded from http://pubs.acs.org on March 4, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
Positional Selectivity in C–H Functionalizations of 2-Benzyl Furans
1
2
3
4
5
6
7
8
with Bimetallic Catalysts
Jiadi Zhang, Sheng-Chun Sha, Ana Bellomo, Nisalak Trongsiriwat, Feng Gao, Neil C. Tomson,*
and Patrick J. Walsh*
Roy and Diana Vagelos Laboratories, Penn/Merck Laboratory for High-Throughput Experimentation, Department
of Chemistry, University of Pennsylvania, 231 South 34th Street, Philadelphia, Pennsylvania 19104-6323, United
States
9
Supporting Information Placeholder
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABSTRACT: Metal-catalyzed carbon-carbon bond-forming reactions are a mainstay in the synthesis of pharmaceutical
agents. A long-standing problem plaguing the field of transition metal catalyzed C–H functionalization chemistry is con-
trol of selectivity among inequivalent C–H bonds in organic reactants. Herein we advance an approach to direct site selec-
tivity in the arylation of 2-benzyl furans founded on the idea that modulation of cooperativity in bimetallic catalysts can
enable navigation of selectivity. The bimetallic catalysts introduced herein exert a high degree of control, leading to di-
vergent site-selective arylation reactions of both sp2 and sp3 C–H bonds of 2-benzyl furans. It is proposed that the selectiv-
ity is govern by cation-π interactions, which can be modulated by choice of base and accompanying additives
[MN(SiMe3)2, M = K or Li•12-crown-4].
identity of the auxiliary metal could also impart divergent
reaction pathways, as depicted in Figure 2.
1. INTRODUCTION
Recent years have witnessed spectacular advanc-
es in transition metal catalyzed C–H functionalization
reactions,1-7 which have numerous potential applications
in the synthesis of biologically active compounds and
pharmaceuticals.8,9 Despite remarkable progress, a cen-
tral challenge remains achievement of positional selectivi-
ty in the presence of inequivalent C–H bonds.10 One
strategy to approach this problem is the installation of
directing groups that position the catalyst to react selec-
tively with the C–H bond of interest,11-14 through either
oxidative addition or other means of activation. This
strategy is particularly useful when the directing group is
part of the target molecule, or is readily removed.
Figure 1. Oxidative addition of aryl chlorides by deproto-
nated (NIXANTPHOS)Pd(0).
An alternative approach is to exert control over
the reductive elimination step of a C–H functionalization
reaction. Doing so would require a directing group to
influence the equilibrium between various organometallic
isomers. We hypothesized that use of bimetallic catalysts
could provide unique opportunities to enhance catalyst
selectivity beyond that obtainable by the mononuclear
catalysts that are the state-of-the-art in coupling reac-
tions.
Inspiration for this mode of regiochemical con-
trol came from the synergistic reactivity we observed of a
heterobimetallic Pd/M catalyst (M=Li, Na, K)15 based on
the NIXANTPHOS16 scaffolding (Figure 1). This catalyst
exhibits much greater reactivity in the oxidative addition
of aryl chlorides than the most reactive bidentate phos-
phine complexes of palladium known to catalyze cross-
coupling reactions. Importantly, the (NIXANTPHOS)Pd
complex’s N–H must be deprotonated to generate the
active bimetallic catalyst that achieves the room tempera-
ture oxidative addition of aryl chlorides.15 For the purpos-
es of directing C–H functionalization reactions at the re-
ductive elimination step, we envision using the coopera-
tivity between proximate metallic sites in bimetallic cata-
lysts.17,18 Not only could the presence of a second metal
ion engender the types of novel reactivity observed during
Ar–Cl activation, but we thought that variation of the
Figure 2. Cooperativity in heterobimetallic catalysts via
directing group to control regioselectivity in reductive
elimination.
Herein, we advance room temperature site-
selective arylation reactions at both sp2 and sp3 hybridized
carbons with bimetallic catalysts employing 2-benzyl fu-
ran derivatives (Figure 3). Nearly complete selectivity is
reached in the reductive elimination between the two
reaction pathways under conditions that differ only in the
nature of the auxiliary metal and its accompanying ligands.
These divergent routes provide access to differentially
substituted furans, a class of heterocycles ubiquitous in
naturally occurring and biologically active compounds.19,20
Experimental and computational studies support cation-π
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
interactions as the controlling feature enabling the regi- We next explored the impact of different bases
oselective C-3 arylation of 2-benzyl furans.
1
on the selectivity of the heterobimetallic catalyst. Substi-
tution of NaN(SiMe3)2 for KN(SiMe3)2 under otherwise
identical conditions resulted in a remarkable change in
the course of the reaction, generating benzylic arylation
(5ab) with high selectivity, albeit in 29% AY (entry 2).
Based on our prior results,22 we were not surprised that
LiN(SiMe3)2 was not sufficiently reactive to deprotonate
the weakly acidic substrate, and no reaction was observed
(entry 3). We previously demonstrated that additives that
bind to M of MN(SiMe3)2 (M = Li, Na) accelerate the
deprotonation step of deprotonative cross-coupling pro-
cesses (DCCP), presumably by reducing the aggregation
state of the base.22 Crown ethers were found to exhibit
the greatest acceleration. Thus, we explored the use of 2
equiv of the corresponding crown ethers relative to base
(entries 4–6). Interestingly, addition of 12-crown-4 (5
equiv) to the reaction with LiN(SiMe3)2 afforded the ben-
zylic arylation product 5ab exclusively in 94% AY and 91%
isolated yield. The impact of 15-crown-5 on the reaction
with NaN(SiMe3)2 was to increase the yield of the benzylic
arylation product (entry 5). Importantly, and in sharp
contrast to the use of KN(SiMe3)2 alone, the potassium
cation caged with 18-crown-6 (1:2) afforded the benzylic
arylation product 5ab as the major product without for-
mation of the C-3 arylation product 3ab (entry 1 vs. 6). It
is noteworthy that no reaction was observed in the ab-
sence of NIXANTPHOS. The implications of the results in
Table 1 and investigation of the factors controlling selec-
tivity are discussed in the next sections.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Cation-controlled site-selective sp2 C-3–H and
sp3 benzylic C–H arylation of furans
2. RESULTS AND DISCUSSION
2.1. Initial Studies with 2-Benzylfuran.
In our search for substrates to probe cooperativi-
ty in reductive elimination from bimetallic catalysts, we
discovered that 2-benzyl furans gave mixtures of products
derived from C-3 and benzylic arylation reactions (Figure
3). We, therefore, initiated an exploration of site-
selectivity in the arylation of 2-benzylfuran (1a) using
NIXANTPHOS (structure in Figure 1), Pd(OAc)2, base,
and 4-tert-butyl bromobenzene (2b). Given the high pKa
of the benzylic C–H’s of 2-benzylfuran (30.2 in DMSO),21
we chose KN(SiMe3)2 as a starting point. Selected results
are shown in Table 1 (with details in the Supporting In-
formation).
To favor the unusual C-3 arylation product (3ab),
4 ethereal solvents [CPME (cyclopentyl methyl ether),
dioxane, THF, and 2-MeTHF] were tested under other-
wise identical conditions (see SI for details). Among the
four solvents, dioxane proved to be the best. Using 2.5
equivalents KN(SiMe3)2 resulted in 93% assay yield (AY by
1H NMR) of C-3 arylation with a >14:1 ratio of 3ab : the
diarylation product 4ab (entry 1). The C-3 arylation
product was ultimately isolated in 93% yield after flash
chromatography.
2.2. Probing the Divergent Reaction Pathways
Leading to C-3 and Benzylic Arylation.
To rationalize the formation of the C-3 arylation
product, we propose the reaction pathway in Figure 4.23,24
Reversible deprotonation of the substrate by MN(SiMe3)2
(M = Li, Na, K) forms the carbanion (I), which undergoes
transmetallation to the Ar–Br oxidative addition product
II to give benzyl adduct III. Reductive elimination at this
point would generate the CBn-arylation product VIII, but
intermediate III is in equilibrium with the C-3 adduct IV
via a σ−π−σ rearrangement. Reductive elimination of the
C-3 adduct IV and subsequent tautomerization of V gen-
erates the C-3 arylation product (VI). As mentioned
above, by changing MN(SiMe3)2 bases and thereby also
changing the auxiliary metal (Figure 4), we proposed that
we could impact the bifurcation between the C-3 and
benzylic arylation reaction manifolds at the reductive
elimination step (as detailed below).
Table 1. Optimization of Pd-catalyzed C-3 and benzylic
arylation of 1aa
The results in Table 1 provide critical insight into
the function of our bimetallic catalysts and lead us to hy-
pothesize that when the NIXANTPHOS-bound potassium
has available coordination sites, it binds the phenyl group
of the substrate forcing palladium to bind at C-3. In con-
trast, when the potassium is caged by the crown ether it is
coordinatively saturated and unable to bind the phenyl of
the substrate. As a result, palladium binds and arylates
the benzylic site. Although cation-π interactions have
been studied in water,25 less is known about the strength
1
aYield determined by H NMR spectroscopy of the crude
reaction mixture. bIsolated yield after chromatographic
purification. cCrown ether (2 equiv relative to base) em-
ployed.
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
of such interactions relative to solvent binding in differ-
C–H functionalization reactions that occur in the second
coordination sphere are rare,14 let alone those that involve
1
2
3
4
ent organic solvents. Thus, the key to controlling C-3 vs.
benzylic selectivity in this bimetallic catalyst is the ability
to switch on and off cooperativity between the auxiliary
metal and the palladium. It is noteworthy that directed
cation-π interactions.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Proposed reaction pathways for benzylic and C-3 arylations
ly investigated the explicit solvation sphere of the com-
plex by comparing the energies of various geometries of
K[III]•(diox)4, K[IV]•(diox)4, and {K[IV]•(diox)3 + diox-
ane}. In line with the relative energies of [III]– and [IV]–
given above, K[III]•(diox)4 was found to be more stable
than K[IV]•(diox)4, albeit by only 1.3 kcal/mol (Figure 6).
However, dissociation of a dioxane from K[IV]•(diox)4
provided a substantial entropic gain of 12.4 kcal/mol in
free energy. Thus, K[IV]•(diox)3 (Figure 6, bottom) plus
a free molecule of dioxane was determined to be the
ground state of the system, with a clear energetic separa-
tion between it and the next lowest energy conformation
(K[III]•(diox)4, +11.1 kcal/mol, Figure 6).
2.3 Computational Exploration of Arylation Regi-
oselectivity.
Computational chemistry was used to help un-
derstand the factors that control the bifurcation of the
arylation reaction pathways. To start, density functional
theory geometry optimization calculations (BP86/ZORA-
TZVP, COSMO[dioxane]; see Supplementary Infor-
mation) were performed on model compounds of inter-
mediates III and IV (Figure 4). Comparison of the Pd–CBn
([III]—) and Pd–C-3 ([IV]—) tautomers of the anion
[(NIXANTPHOS)Pd(Ph)(2-benzylfuranyl)]– (no auxiliary
metal; Figure 5, top) revealed a 5.0 kcal/mol preference
for the Pd–CBn bond (Figure 6), in line with the observa-
tion that caging the K auxiliary metal with crown ether
additive favors benzylic arylation products.
The geometries about the metal centers of [III]–
and [IV]– are both pseudo square planar, with pairs of
nearly trans-disposed phosphines (∠(P–Pd–P) = 142°) and
carbons (∠(C–Pd–C) = 160°). Long Pd–O distances of ca.
2.8 Å lie well outside the sum of the covalent radii of Pd
and O (2.05 Å).26 The Pd–C(47) ([III]–) and Pd–C(1) ([IV]–
) bond distances of 2.236 and 2.296 Å, respectively, nota-
bly also lie outside the 2.15 Å sum of the covalent radii of
Pd and C and are considerably longer than the Pd–C(26)
bond lengths of ca. 2.08 Å. We attribute the weakness of
Pd–Csubstrate interactions to the resonance stabilization of
the anions that would result from heterolytic dissociation.
A series of solvated non-ionic K-NIXANTPHOS
complexes were next used to explore the C-3 arylation
product. We hypothesized that a cation–π interaction
was instrumental in determining the observed regioselec-
tivity, but for this to be the case, the K–arene interaction
would have to outcompete solvation by dioxane. To in-
vestigate this computationally, we used the conductor-
like screening model (COSMO) as implemented in ORCA,
with the dielectric constant of dioxane. We then manual-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
Figure 5. DFT geometry optimized structures of
benzylfuranyl)] (Na[IV]•(diox)2) plus free dioxane to lie
[(NIXANTPHOS)Pd(Ph)(6-Pd-2-benzylfuranyl)]– ([III]–,
5.5 kcal/mol lower in energy than the Pd–CBn tri-solvate
Na[III]•(diox)3 and 11.4 kcal/mol lower in energy than the
tri-solvate of the Pd–C(3) tautomer, Na[IV]•(diox)3 (see
Supporting Information). Na[IV]•(diox)2 similarly dis-
plays a cation–π interaction, with short Na–Cπ distances
of 2.815 and 2.948 Å. The trend toward a smaller cation–π
stabilization energy is consistent with the diminished
capacity of the harder metal ion to form cation–π interac-
tions capable of outcompeting solvation.
1
2
3
4
5
6
7
8
top left), [(NIXANTPHOS)Pd(Ph)(3-Pd-2-benzylfuranyl)]–
([IV]–,
top
right)
and
[K(diox)3][(NIXANTPHOS)Pd(Ph)(3-Pd-2-benzylfuranyl)]
(K[IV]•(diox)3, bottom); all hydrogen atoms were re-
moved for clarity. Important bond lengths (Å) and angles
(°) for i) [III]–: Pd–C47 2.236, Pd–C26 2.085, Pd–P1 2.398,
Pd–P2 2.401, C44–C47 1.459, P1–Pd–P2 142.1, C26–Pd–C47
159.6; ii) [IV]–: Pd–C1 2.296, Pd–C26 2.076, Pd–P1 2.417,
Pd–P2 2.358, C44–C47 1.373, P1–Pd–P2 141.8, C26–Pd–C47
160.6; and K[IV]•(diox)3: Pd–C1 2.301, Pd–C26 2.072, K–N
2.828, K–C50 3.430, K–C51 3.384, P1–Pd–P2 145.0, C1–Pd–
C26 158.4.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.4
Generality
of
Aryl
Bromides
in
M~Pd(NIXANTPHOS)-Catalyzed C-3 and Benzylic Aryla-
tions of 1.
We next desired to probe whether the site-
The most notable structural feature of
K[IV]•(diox)3 is the cation–π interaction27,28 between po-
tassium and the substrate arene ring (K–Cπ = 3.384, 3.430
Å; Figure 5, bottom). The greater polarizability of K+
compared to the lighter alkali metals is well known to
facilitate cation-π interactions,27,28 and it is proposed this
interaction may be strong enough to perturb the equilib-
rium between Pd–C(3) and Pd–CBn intermediates III and
IV (Figure 4), guiding reductive elimination in the direc-
tion of the observed C-3 arylation product.
selectivity could be influenced by the steric and/or elec-
tronic properties of the substrates, first using different
aryl bromides. As detailed below, the nature of the cou-
pling partner had little effect on the reaction outcome,
enabling the synthesis of an array of C-3 arylated furans
3aa–3gb from sterically and electronically diverse aryl and
heteroaryl bromides. As outlined in Table 2, employing
KN(SiMe3)2 as base led to excellent C-3 selectivity with no
benzylic arylation products 5 observed (1H NMR spectros-
copy). Good to excellent selectivity for mono- over dia-
rylation was also exhibited (C-3:diarylation from 5:1 to >
20:1). All the C-3 arylation products were easily separable
from the diarylation byproducts by column chromatog-
raphy, delivering 3aa−3gb in 27−99% isolated yields. Ex-
cellent yields were obtained using bromobenzene (2a)
and aryl bromides bearing 4-alkyl (2b), 4-methoxy (2c), 4-
N,N-dimethylamino (2d) as well as 4-fluoro (2e) groups,
all with C-3:diarylation ratios
≥
11:1.
1-Bromo-4-
chlorobenzene (2f) reacted with 1a to produce C-3 aryla-
tion product 3af in 55% isolated yield with the C-
3:diarylation ratio of 13:1. Trifluoromethyl (2g), methoxy
(2h), and acetal (2i) groups at the meta position were all
well-tolerated (65−77% yields) with moderate to good C-
3:diarylation ratios (≥ 5:1). The sterically hindered 1-
bromonaphthalene (2j) and 2-bromotoluene (2k) also
participated in C-3 arylation to produce 3aj and 3ak with
C-3:diarylation ratios > 20:1. Aryl bromides containing an
acetyl (2l) and heterocycles such as benzofuran (2m) and
indole (2n, 2o) exhibited excellent C-3:diarylation ratios
(> 20:1), delivering the desired products in 56−80% yields.
Combined with the analogous results given below for var-
ying the identity of 2 under CBn-arylation conditions, the-
se results suggest that site-selectivity is not being deter-
mined by aryl bromide substrate.
A similar result was found when we examined
the impact of substituents on the aryl group of the 2-
benzylfuran. The parent phenyl (1a) and 4-fluoro phenyl
(1b) substrates were able to deliver the C-3 arylation
products 3ab and 3bb in excellent yields and C-
3:diarylation ratios. Still, the 4-methoxy derivative (1c)
gave only 27% AY of 3cb (1H NMR spectroscopy) with the
C-3:diarylation ratio of 1:1 when 4-tert-butyl bromoben-
zene was employed. This result suggests that half of the
initial C-3 arylation product 3cb underwent a second ary-
Figure 6. Graphical description of the energy differences
between i) [III]– and [IV]– (left) and ii) K[III]•(diox)4 and
K[IV]•(diox)3 + dioxane. The tilde symbol represents the
deprotonated NIXANTPHOS ligand.
Analogous computations on the Na complexes of
[III]–
and
[IV]–
found
the
di-solvate
[Na(dioxane)2][(NIXANTPHOS)Pd(Ph)(3-Pd-2-
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
needed to observe the benzylic-arylation product under
lation to form the diarylation byproduct 4cb, because the
1
2
3
4
5
6
7
8
rates of these arylation processes are similar. We hypoth-
esized that the 4-methoxy benzyl group of 1c decreased
the acidity of the benzylic hydrogens (relative to 1a and
1b), slowing the rate of the C-3 arylation. Once arylated
at C-3, however, the impact of the C-3 aryl is to increase
the acidity of the benzylic hydrogens and, therefore, the
relative rate of arylation of the benzylic position to gener-
ate the diarylated product. To probe this hypothesis, ary-
lation of 1c was performed with 4-bromoanisole and 4-
bromo-N,N-dimethylaniline, respectively. The C-3 aryla-
tion intermediate with the dimethyl amino group pos-
sesses less acidic benzylic hydrogens, and the relative rate
of the C-3 arylation to benzylic arylation is expected to be
larger. Consistent with this hypothesis, use of 4-bromo-
N,N-dimethylaniline resulted in a 57% yield of the C-3
arylation product with a C-3:diarylation ratio of 5:1 (See
Supporting Information for more details). Use of 4-
bromoanisole gave intermediate results, as predicted (3cc,
43% yield).
these conditions. Indeed, it appears that the primary in-
fluence of the Bu groups is to prevent solvation of the
t
cation, leading to a K–arene complex that is considerably
more stable than the higher solvate. These results are
again consistent with the notion that cation-π interac-
tions perturb the equilibrium between intermediates III
and IV, away from the benzylic position that would be
favored under cation-free conditions and toward the
furanyl C3-position.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cation-π interactions will likely be impacted by
the electronic and steric properties of the π-donor. We
hypothesized that the reduced electron density in the aryl
ring of the electron-deficient 3,5-C6H3-(CF3)2 group of 1d
would render cation-π interactions less favorable. Sub-
jecting benzylic furan 1d to the C-3 arylation conditions
(M = K)29 resulted in a switch in the C-3 : benzylic aryla-
tion ratio to 1.0 : 1.4; i.e. the minor product was C-3 aryla-
tion and the major product was benzylic arylation. Com-
putationally, we found that the Pd–CBn tautomer of the
alkali metal-free anion [(NIXANTPHOS)Pd(Ph)(3-Pd-2-
(3,5-(CF3)2-benzyl)furanyl)]– ([IIICF3]–) was computed to
lie lower in energy than its Pd–C-3 tautomer ([IVCF3]–) by
2.9 kcal/mol. However, in contrast to the parent sub-
strate discussed above, the preference for Pd–CBn bonding
for the fluorinated substrate carries over to the K com-
plexes, where the Pd-CBn species K[IIICF3]•(diox)3 was
found to lie 5.9 kcal/mol lower than the combined ener-
gies of K[IVCF3]•(diox)2 and free dioxane. This is con-
sistent with the return to majority benzylic-arylation
product with 1d (C-3:CBn = 1.0:1.4), which tracks with the
diminished nucleophilicity of the trifluoromethylated
arene ring, leading to a product distribution weighted
more heavily by the energy of the Pd–C bond than by the
K–arene interaction.
Figure 7. DFT geometry optimized structure of
K[IVtBu]•diox; all hydrogen atoms were removed for clari-
ty. Important bond lengths (Å) and angles (°): Pd–C1
2.322, Pd–C26 2.067, K–N 2.839, K–C48 3.180, K–C50 3.141,
K–C51 3.285, K–C52 3.398, K–C53 3.430, K–C49 3.298, P1–
Pd–P2 145.0, C1–Pd–C26 159.4.
Finally, we subjected 2-benzyl-5-methylfuran (1f),
bearing a C-5 methyl, to the KN(SiMe3)2 and obtained the
C-3 arylation product 3fb with excellent site selectivity in
70% yield. C-3 arylation of 2-benzhydrylfuran (1g, R1 = 5-
CH3, R2 = R3 = Ph) proceeded smoothly in the presence of
KN(SiMe3)2 to afford 3gb in 99% isolated yield.
Table 2. Scope of aryl bromides in K~Pd(NIXANTPHOS)-
catalyzed C-3 arylation of 1a and derivativesa
Next we attempted to perturb the equilibrium
between III and IV through steric influences. Use of 3,5-
tBu2-benzylfuran as a substrate provided the C3-arylated
product in 70% yield under the C3-arylation conditions,
t
suggesting that the increased steric influence of the Bu
groups has a negligible effect on the C-3/CBn-arylation
regiochemistry. Computation of the proposed intermedi-
ate
[K(diox)][(NIXANTPHOS)Pd(Ph)(3-Pd-2-(3,5-tBu2-
benzyl)furanyl)] (K[IVtBu]•diox; Figure 7) indicates a
nearly symmetric cation–π interaction with short K–Cπ
distances, ranging from 3.14–3.43 Å. Related meta-tBu2-
arene potassium-π interactions are known,30–34 suggesting
t
that the Bu groups are not sterically imposing enough to
prevent the formation of cation-π interactions, as may be
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
Aryl bromides bearing 4-alkyl (2b), electron-donating 4-
1
2
3
4
5
6
7
8
methoxy (2c) and 4-N,N-dimethylamino (2d), 4-fluoro
(2e) as well as sterically more hindered 1-naphthyl (2j)
groups all proved to be good cross-coupling partners for
the benzylic arylation. Furthermore, functional groups
such as acetal (2i), acetyl (2l) and heterocycles such as
benzofuran (2m) and indole (2n, 2o) were well tolerated,
delivering the corresponding products in 60−96% yields.
The impact of the furan substrate in the benzylic
9
arylations was next explored with 4-tert-butyl bromoben-
zene 2b. In the presence of LiN(SiMe3)2 with 12-crown-4
only benzylic arylation products 5 were observed (Table 3,
last row). Varying the substituents on the furanyl phenyl
group from H (1a) to 4-F (1b), 4-OCH3 (1c), and 3,5-(CF3)2
(1d) resulted in excellent isolated yields of the benzylic
arylation products (75–91%), suggesting that these 4 sub-
strates were reversibly deprotonated and entered the
catalytic cycle. Note, however, that with the 3,5-(CF3)2
(1d), decomposition was observed with LiN(SiMe3)2 and
that use of NaOtBu gave benzylic arylation (85% yield).
Subjecting 2-benzyl-5-methylfuran (1f), bearing a C-5 me-
thyl, to the LiN(SiMe3)2 and 12-crown-4 reaction condi-
tions resulted in benzylic arylation product with excellent
site selectivity in 94% yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table
3.
Scope
of
aryl
bromides
in
(crown)Li~Pd(NIXANTPHOS)-catalyzed benzylic aryla-
tion of 1a and derivatives a
aReactions conducted on a 0.1 mmol scale using 1.2 equiv
of 1, 2.5 equiv of KN(SiMe3)2, and 1 equiv of 2 at 0.1 M.
Isolated yield after chromatographic purification. The C-
1
3:diarylation ratios in parentheses were determined by H
NMR spectroscopy of the crude reaction mixtures. b45 °C.
c2 equiv of KN(SiMe3)2. dReaction conducted on a 0.3
f
mmol scale. e3 equiv of KN(SiMe3)2. 10 mol % Pd(OAc)2
g
1
and 15 mol % NIXANTPHOS. Yield determined by H
NMR spectroscopy of the crude reaction mixtures. hKOtBu
i
j
was used as base. 2.5 equiv of 1f. CPME was used as sol-
vent.
aReactions conducted on a 0.1 mmol scale using 1.2 equiv
of 1, 2.5 equiv of LiN(SiMe3)2, 5 equiv of 12-crown-4, and 1
equiv of 2 at 0.1 M. Isolated yield after chromatographic
To reverse the chemoselectivity, we next em-
ployed LiN(SiMe3)2 with 12-crown-4 (1:2) and determined
the generality of aryl bromides in the benzylic arylation of
1a. As shown in Table 3, with the same aryl bromides
employed in the C-3 arylation, a complete reversal in se-
lectivity towards benzylic arylation was observed. Neither
C-3 arylation nor diarylation products were detected, ren-
dering the exclusive formation of 5 in 60−96% yields.
purification.
c1a:LiN(SiMe3)2:12-crown-4:2 = 2:2:4:1. d1a:LiN(SiMe3)2:12-
b1a:LiN(SiMe3)2:12-crown-4:2
= 2:1.5:3:1.
crown-4:2
=
1:2:4:2.
e1a:LiN(SiMe3)2:12-crown-4:2
=
1.2:3:6:1. fNaOtBu was used as base. gCPME was used as
solvent.
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
chromatography using Whatman Partisil K6F 250 μm
3. SUMMARY AND OUTLOOK
1
precoated 60 Å silica gel plates and visualized by short-
wavelength ultraviolet light as well as by treatment with
ceric ammonium molybdate (CAM) stain or iodine. Silica
gel (230–400 mesh, Silicycle) was used for flash chroma-
Nature has long recognized that the placement
of two or more metals in close proximity can elicit unique
reactivity, as demonstrated by its use of bimetallic en-
zyme cofactors in various processes (i.e., hydrogenases,
nitrogenases, CO dehydrogenases, etc.). Likewise, in
transition metal catalyzed reactions, bimetallic catalysts
can exhibit synergistic activity leading to enhanced reac-
tivity not attainable by monometallic catalysts. Herein,
we have introduced a novel strategy to manage selectivity
in C–H functionalization reactions by employing bimetal-
lic catalysts, wherein the cooperativity can be switched on
or off by proper choice of main group metals to enable
excellent selectivity over divergent reaction pathways.
Proof-of-concept lies in the highly selective generation of
either C-3 or benzylic arylation products from 2-
benzylfuran derivatives. A key conceptual advance of this
chemistry is the demonstration that the regioselectivity of
reductive elimination can be controlled within a single
ligand framework simply by employing different auxiliary
metals. Our approach can be compared to traditional
approaches, wherein selectivity is determined in the C–H
bond cleavage step.
2
3
4
5
6
7
8
9
1
tography. The H NMR and 13C{1H} NMR spectra were
obtained using a Brüker AM-500 Fourier-transform NMR
spectrometer at 500 and 125 MHz, respectively. Chemical
shifts are reported in units of parts per million (ppm)
downfield from tetramethylsilane (TMS), and all coupling
constants are reported in hertz. The infrared spectra
were obtained with KBr plates using a Perkin-Elmer Spec-
trum 100 Series FTIR spectrometer. High-resolution mass
spectrometry (HRMS) data were obtained on a Waters
LC–TOF mass spectrometer (model LCT-XE Premier)
using chemical ionization (CI) or electrospray ionization
(ESI) in positive or negative mode, depending on the ana-
lyte. Melting points were determined on a Unimelt
Thomas–Hoover melting point apparatus and are uncor-
rected.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.2. General Procedure A. Pd-Catalyzed C3 Arylation of
Furans. An oven-dried 10 mL reaction vial equipped with
a stir bar was charged with KN(SiMe3)2 (49.9 mg, 0.25
mmol, 2.5 equiv) under a nitrogen atmosphere. A solu-
tion (from a stock solution) of Pd(OAc)2 (1.12 mg, 0.0050
mmol, 5 mol %) and NiXantphos (4.14 mg, 0.0075 mmol,
7.5 mol %) in 1 mL of dry dioxane was taken up by syringe
and added to the reaction vial. After stirring for 5 min at
24 °C, 2-benzylfuran (18.1 μL, 0.12 mmol, 1.2 equiv) was
added to the reaction mixture followed by 1-bromo-4-tert-
butylbenzene (17.3 μL, 0.1 mmol, 1 equiv). Note that the
aryl bromide in a solid form was added to the reaction
vial prior to KN(SiMe3)2. The reaction mixture was stirred
for 12 h at 24 °C, quenched with three drops of H2O, dilut-
ed with 3 mL of ethyl acetate, and filtered over a pad of
MgSO4 and silica. The pad was rinsed with additional
ethyl acetate, and the solution was concentrated in vacuo.
The crude material was loaded onto a silica gel column
and purified by flash chromatography.
We have presented computational and circum-
stantial experimental evidence supporting the idea that
the bimetallic (M~NIXANTPHOS)Pd-based catalyst sys-
tem can impact the equilibrium of palladium-bound ben-
zylic and C-3 isomers (III and IV in Figure 4) through
secondary coordination sphere effects. The situation of
the anionic charge on the NIXANTPHOS backbone places
K+ in an advantageous position for binding the benzylic
arene ring when Pd is coordinated to the C-3 carbon of
the furan (IV). In the absence of a suitable main group
cation to bind the benzylic π-system (Li, Na or K•18-
crown-6), the furanyl ligand preferentially binds to the
palladium through the benzylic carbon (III) and leads to
benzylic arylation products. We anticipate that the fea-
tures of the bimetallic support scaffold of the
NIXANTPHOS ligand will lead to a general design feature
to control reactivity and selectivity and will inspire the
synthesis and application of new classes of bimetallic cat-
alysts.
4.3. General Procedure B. Pd-Catalyzed Benzylic Aryla-
tion of Furans. An oven-dried 10 mL reaction vial
equipped with a stir bar was charged with LiN(SiMe3)2
(41.8 mg, 0.25 mmol, 2.5 equiv) and 12-crown-4 (80.9 μL,
0.5 mmol, 5 equiv) under a nitrogen atmosphere. A solu-
tion (from a stock solution) of Pd(OAc)2 (1.12 mg, 0.0050
mmol, 5 mol %) and NiXantphos (4.14 mg, 0.0075 mmol,
7.5 mol %) in 1 mL of dry dioxane was taken up by syringe
and added to the reaction vial. After stirring for 5 min at
24 °C, 2-benzylfuran (18.1 μL, 0.12 mmol, 1.2 equiv) was
added to the reaction mixture followed by 1-bromo-4-tert-
butylbenzene (17.3 μL, 0.1 mmol, 1 equiv). Note that the
aryl bromide in a solid form was added to the reaction
vial prior to LiN(SiMe3)2. The reaction mixture was
stirred for 12 h at 24 °C, quenched with three drops of
H2O, diluted with 3 mL of ethyl acetate, and filtered over
a pad of MgSO4 and silica. The pad was rinsed with addi-
tional ethyl acetate, and the solution was concentrated in
vacuo. The crude material was loaded onto a silica gel
column and purified by flash chromatography.
4. EXPERIMENTAL SECTION
Representative procedures are described herein. Full ex-
perimental details and characterization of all compounds
are provided in the Supporting Information.
4.1. General Methods. All reactions were performed un-
der nitrogen using oven-dried glassware and standard
Schlenk or vacuum line techniques. Air- and moisture-
sensitive solutions were handled under nitrogen and
transferred via syringe. Anhydrous CPME and dioxane
were purchased from Sigma-Aldrich and used as solvent
without further purification. Unless otherwise stated,
reagents were commercially available and used as pur-
chased without further purification. Chemicals were ob-
tained from Sigma-Aldrich, Acros, TCI America or Alfa
Aesar, and solvents were purchased from Fisher Scientific.
The progress of the reactions was monitored by thin-layer
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
(16) van der Veen, L. A.; Keeven, P. H.; Schoemaker, G. C.;
ASSOCIATED CONTENT
1
2
3
4
Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Lutz,
M.; Spek, A. L. Organometallics 2000, 19, 872.
(17) Shibasaki, M.; Kanai, M.; Matsunaga, S.; Kumagai, N. Acc.
Chem. Res. 2009, 42, 1117.
(18) Buchwalter, P.; Rosé, J.; Braunstein, P. Chem. Rev. 2015,
115, 28.
Supporting Information. Procedures, full characteriza-
tion of new compounds, and DFT geometry optimization
calculations. This material is available free of charge via
the Internet at http://pubs. acs.org.
5
AUTHOR INFORMATION
6
7
8
9
(19) (a) Sperry, J. B.; Wright, D. L. Curr. Opin. Drug Discovery
Dev. 2005, 8, 723. (b) Li, J. J. Heterocyclic Chemistry in Drug
Discovery; John Wiley & Sons, 2013.
(20) Palladium-catalyzed bisarylation of 3-alkylbenzofurans
with aryl iodides: Cho, B. S.; Chung, Y. K. Chem. Commun. 2015,
51, 14543.
(21) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.
(22) Bellomo, A.; Zhang, J.; Trongsiriwat, N.; Walsh, P. J.
Chem. Sci. 2013, 4, 849.
(23) Zhang, J.; Bellomo, A.; Creamer, A. D.; Dreher, S. D.;
Walsh, P. J. J. Am. Chem. Soc. 2012, 134, 13765.
(24) Hussain, N.; Frensch, G.; Zhang, J.; Walsh, P. J. Angew.
Chem., Int. Ed. 2014, 53, 3693.
(25) Kumpf, R. A.; Dougherty, D. A. Science 1993, 261, 1708.
(26) Cordero, B.; Gómez, V.; Platero-Prats, A. E.; Revés, M.;
Echeverría, J.; Cremades, E.; Barragán, F.; Alvarez, S. Dalton
Trans. 2008, 2832.
(27) Dougherty, D. A. Acc. Chem. Res. 2012, 46, 885.
(28) Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew.
Chem., Int. Ed. 2003, 42, 1210.
(29) KOtBu was used as base because KN(SiMe3)2 caused
decomposition of the substrate.
(30) Pan, X.; Yang, X.; Chen, L.; Wu, J.; Tang, N. Inorg. Chem.
Commun. 2010, 13, 919.
(31) Mansell, S. M.; Kaltsoyannis, N.; Arnold, P. L. J. Am.
Chem. Soc. 2011, 133, 9036.
(32) Akagi, F.; Matsuo, T.; Kawaguchi, H. Angew. Chem., Int.
Ed. 2007, 46, 8778.
(33) Hsu, Y.-L.; Liang, L.-C. Organometallics 2010, 29, 6201.
(34) Sarazin, Y.; Roşca, D.; Poirier, V.; Roisnel, T.; Silvestru, A.;
Maron, L.; Carpentier, J.-F. Organometallics 2010, 29, 6569.
Corresponding Author
pwalsh@sas.upenn.edu, tomson@sas.upenn.edu
ACKNOWLEDGMENTS
Financial support for this work was provided by
NIH/NIGMS (GM 104349). NT acknowledges Mahidol
University (Thailand) and Feng Gao thanks the China
Scholarship Counsel for fellowships.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCES
(1) Davies, I. W.; Welch, C. J. Science 2009, 325, 701.
(2) Colby, D. A.; Bergman, R. G.; Ellman, J. A. Chem. Rev. 2010,
110, 624.
(3) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110, 1147.
(4) Yeung, C. S.; Dong, V. M. Chem. Rev. 2011, 111, 1215.
(5) Mkhalid, I. A. I.; Barnard, J. H.; Marder, T. B.; Murphy, J.
M.; Hartwig, J. F. Chem. Rev. 2010, 110, 890.
(6) Daugulis, O.; Do, H.-Q.; Shabashov, D. Acc. Chem. Res.
2009, 42, 1074.
(7) Magano, J.; Dunetz, J. R. Chem. Rev. 2011, 111, 2177.
(8) Bryan, M. C.; Dillon, B.; Hamann, L. G.; Hughes, G. J.;
Kopach, M. E.; Peterson, E.A.; Pourashraf, M.; Raheem, I.;
Richardson, P.; Richter, D.; Sneddon, H. F. J. Med. Chem. 2013,
56, 6007.
(9) Ritchie, T. J.; Macdonald, S. J. F.; Young, R. J.; Pickett, S. D.
Drug Discov. Today 2011, 16, 164.
(10) Liégault, B.; Petrov, I.; Gorelsky, S. I.; Fagnou, K. J. Org.
Chem. 2010, 75, 1047.
(11) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem.
Res. 2011, 45, 788.
(12) Liu, Y.-J.; Xu, H.; Kong, W.-J.; Shang, M.; Dai, H.-X.; Yu, J.-
Q. Nature 2014, 515, 389.
(13) Leow, D.; Li, G.; Mei, T.-S.; Yu, J.-Q. Nature 2012, 486, 518.
(14) Roosen, P. C.; Kallepalli, V. A.; Chattopadhyay, B.;
Singleton, D. A.; Maleczka, R. E., Jr.; Smith, M. R., III. J. Am.
Chem. Soc. 2012, 134, 11350.
(15) Zhang, J.; Bellomo, A.; Trongsiriwat, N.; Jia, T.; Carroll, P.
J.; Dreher, S. D.; Tudge, M. T.; Yin, H.; Robinson, J. R.; Schelter,
E. J.; Walsh, P. J. J. Am. Chem. Soc. 2014, 136, 6276.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
TOC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment