[4+2] or [4+1] Annulation: Changing the Reaction Pathway of a Rhodium-Catalyzed Process by Tuning the Cp Ligand

Article abstract of DOI:10.1002/anie.201612559

A change in reaction pathway was achieved for the first time by tuning the cyclopentadienyl (Cp) ligand used for the rhodium-catalyzed cyclization of benzamides with conjugated enynones. Depending on the Cp ligand, the reaction pathway switched between [4+2] and [4+1] annulation. Electronic effects turned out to be crucial for the product distribution. The dichotomy was attributed to the alteration of the Lewis acidity of the resultant Cp-bound rhodium species.

Full text of DOI:10.1002/anie.201612559

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201612559
German Edition: DOI: 10.1002/ange.201612559
Rhodium Catalysis
[4+2] or [4+1] Annulation: Changing the Reaction Pathway of
a Rhodium-Catalyzed Process by Tuning the Cp Ligand
Seung Youn Hong, Jisu Jeong, and Sukbok Chang*
Abstract: A change in reaction pathway was achieved for the
first time by tuning the cyclopentadienyl (Cp) ligand used for
the rhodium-catalyzed cyclization of benzamides with con-
jugated enynones. Depending on the Cp ligand, the reaction
pathway switched between [4+2] and [4+1] annulation.
Electronic effects turned out to be crucial for the product
distribution. The dichotomy was attributed to the alteration of
the Lewis acidity of the resultant Cp-bound rhodium species.
À
S
eminal studies on the stoichiometric C H bond activation
of hydrocarbons using half-sandwich iridium or rhodium
complexes were independently reported by the groups of
Bergman and Jones a few decades ago.[1] While the recent
À
developments in catalytic C H functionalization have bene-
fitted from the use of other transition-metal species,[2]
Cp*RhIII-catalyzed
(Cp* = pentamethylcyclopentadienyl)
À
C H functionalizations have recently emerged as a powerful
À
synthetic process for the formation of various C X (X = C, N,
O, or halides) bonds (Scheme 1a, left).[3] This success was
attributed to the strong p-donating ability of the Cp* ligand,
which facilitates access to labile intermediates such as cationic
or high-valent metal species.[4] Given that ligands can
dramatically influence reactivity and/or selectivity, we antici-
pated that tuning cyclopentadienyl (Cp) ligands will induce
significant changes in the kinetic and/or thermodynamic
behavior of reactions that employ such ligands (Scheme 1a,
right).
À
Scheme 1. Rhodium(III)-catalyzed C H annulations.
Rovis and co-workers recently demonstrated that Cp
ligands that are sterically more demanding than Cp* greatly
improve the regioselectivity in the RhIII-catalyzed synthesis of
N-heterocycles.[5] Furthermore, the groups of Rovis and
Tanaka independently showed that an electronic variation
of the Cp ligand can result in a significant enhancement of
cyclization reactivity in RhIII catalysis.[6] In addition, the
Cp ligand in a RhIII catalyst system and slightly modifying the
reaction conditions,[8b,d] thus giving rise to completely differ-
ent types of products from the same reactants.
À
Transition-metal-catalyzed C H annulation has emerged
as a facile route to heterocycles.[9] The Cp*RhIII catalyst
system is especially noteworthy as it displays high reactivity
and selectivity in reactions with various alkynes or alkenes to
afford [n+2] annulation products (Scheme 1b, left).[10] In
addition, diazo compounds are known to serve as carbene
precursors to undergo [n+1] cyclizations by Cp*RhIII catalysis
(Scheme 1b, right).[11]
groups of Rovis and Cramer independently introduced chiral
III
À
Cp ligands for Rh -catalyzed asymmetric C H functionali-
zation.[7] Despite these notable advances,[8] to the best of our
best knowledge, there have been only a few reports that have
shown that reaction pathways can be changed by tuning the
We envisaged that our working hypothesis, namely that
tuning the Cp ligands may change the reaction pathway, could
be examined by employing conjugated enynones as model
reactants in a rhodium-catalyzed cyclization (Scheme 1c).
This was based on the anticipation that enynones can serve
either as a two-carbon source (alkyne) or as a one-carbon
component (carbene), thus undergoing [n+2] or [n+1]
annulation, respectively. In fact, enynones have been utilized
as carbene sources in certain metal-catalyzed reactions that
proceed via 5-exo-dig cyclization.[12] Herein, we present a new
example where the mechanistic pathway was switched
between [4+2] and [4+1] annulation in the reaction of
[*] S. Y. Hong, J. Jeong, Prof. Dr. S. Chang
Department of Chemistry
Korea Advanced Institute of Science and Technology (KAIST)
Deajeon 305-701 (Korea)
and
Center for Catalytic Hydrocarbon Functionalizations
Institute for Basic Science (IBS)
Deajeon 305-701 (Korea)
E-mail: sbchang@kaist.ac.kr
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201612559.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
enynones with benzamides by tuning the Cp ligand in the
RhIII catalyst system.[13]
ing the reaction time (entry 7). Finally, product 4a was
obtained in satisfactory yield upon replacing CsOAc with
AgOAc additive and slightly increasing the catalyst loading
(entry 8). Almost equal amounts of the two reactants (1a/2a
1.1:1) were employed under these optimized conditions.
On the other hand, the use of more Lewis basic rhodium
catalysts with Cp*t-Bu or Cp*Cy ligands in the reaction of 1a
with 2a strongly favored the formation of 3a over 4a
(entries 9–12). We assume that the steric hindrance of the
Cp substituents may contribute cooperatively with the
electronic properties although the exact contributions of
these two factors are not clear at the moment. Interestingly,
the formation of 3a was even more selective at lower reaction
temperatures (entry 11). Finally, high efficiency (3a isolated
in 84% yield) and excellent selectivity (3a/4a 13.2:1) were
achieved when the reaction was run at 258C for an extended
period of time (entry 12).
First, N-pivaloyloxy benzamide (1a) was reacted with
conjugated enynone 2a to optimize the reaction conditions
(Table 1). When Cp*RhIII was used as the catalyst in the
presence of CsOAc (20 mol%), a mixture of the tricyclic
isoquinolinone 3a and isoindolinone 4a was formed in a ratio
of 3.3:1 (entry 1). This result suggests that both [4+2] and
[4+1] annulations are operative with the same catalyst
system, as initially envisioned. Although we extensively
varied the reaction parameters to improve the selectivity,
only incremental changes were observed (see the Supporting
Information). We then turned our attention to the modifica-
tion of the Cp ligand based on our hypothesis that more Lewis
acidic rhodium species may prefer a carbene transfer process
via 5-exo-dig cyclization.[12]
The generality of the switch between [4+2] and [4+1]
annulation was investigated next. First, the scope of the [4+2]
cyclization was examined with the Cp*CyRhIII catalytic system
(Scheme 2). Amide substrates bearing electron-donating
Table 1: Screening of Cp ligands.[a]
Entry
CpG
T
[8C]
Conv.
[%][b]
3a
4a
3a/4a[b]
[%][b]
[%][b]
1
2
3
4
5
6
Cp*
Cp*Ph
Cp*H
Cpt
50
50
50
50
50
50
60
60
100
81
85
72
53
40
100
100
73 (69)
28
41
22
11
35
36
30
19
58
73 (71)
3.3:1
2.5:1
1.2:1
1:1.8
1:>99
1:>99
1:>99
1:>99
19
CpE
<1
<1
<1
<1
Cp*CF
CpE
3
7[c]
8[c,d]
CpE
Scheme 2. Scope of the [4+2] annulation. Reaction conditions:
1 (0.15 mmol), 2 (0.10 mmol), [RhCp*CyCl2]2 (4 mol%), CsOAc
(20 mol), CH3CN (0.5 mL). The ratio of 3/4 as determined by 1H NMR
spectroscopy is given in parentheses. [a] At 408C. [b] For 48 h. [c] The
major regioisomer is shown (isomer ratio: 1.5:1). [d] At 508C.
9
Cp*t-Bu
Cp*Cy
Cp*Cy
Cp*Cy
50
50
40
RT
100
100
100
100
59
78
86
8
10
10
7
7.4:1
7.8:1
8.6:1
13.2:1
10
11[e]
12[f]
89 (84)
[a] Reaction conditions: 1a (0.15 mmol), 2a (0.10 mmol), catalyst
(4 mol%), CsOAc (20 mol%), CH3CN (0.5 mL), 12 h. [b] Determined by
1H NMR analysis of the crude reaction mixture. Yields of isolated
products given in parentheses. [c] Catalyst (5 mol%). [d] For 48 h with
AgOAc (20 mol%) instead of CsOAc and 0.11 mmol of 1a. [e] 24 h.
[f] 72 h.
substituents efficiently reacted with enynone 2a to afford
the corresponding tricyclic isoquinolinones (3a–3e) in good
yields with high selectivity. However, the presence of
electron-withdrawing substituents on the benzamide led to
a decrease in selectivity while the combined yields were
almost quantitative (3 f and 3g). A thiophene substrate (3h)
reacted with 2a highly selectively without formation of the
[4+1] product. The present method was successfully applied
to a range of conjugated enynones (3i–3l). Furthermore, an
enynone with a benzoyl (R3 = Ph) instead of the acetyl
substituent (R3 = Me) gave the corresponding product in high
yield with perfect selectivity (3m).
The scope of the [4+1] annulation was then explored by
employing [RhCpECl2]2 as the rhodium catalyst (Scheme 3).
All tested benzamides reacted with 2a to afford the corre-
sponding [4+1] cyclized products (4a–4j) in good yields and
with perfect selectivity. Halide (4e and 4 f), nitro (4g), cyano
While replacing the Cp* ligand with Cp*Ph, Cp*H, or Cpt
resulted in only a slight improvement in the yield of 4a
(entries 2–4), the installation of carboxylate groups (CpE) or
a trifluoromethyl group (Cp*CF ) on the Cp ligand led to the
3
exclusive formation of 4a, albeit with rather lower conversion
at 508C (entries 5 and 6). The reaction efficiency was further
improved by increasing the reaction temperature and extend-
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
tron-donating Cp ligands will more readily induce [4+2]
alkyne insertion into the rhodacycle to give the seven-
membered rhodacycle III-2, which then undergoes reductive
À
elimination with N O bond cleavage to furnish isoquinoli-
none IV-2.[10] Finally, base-mediated cyclization of IV-2
provides tbe [4+2] product.[10f]
We postulate that electron-deficient Cp ligands will
induce a completely different pathway from the same
intermediate II. A more Lewis acidic rhodium center bearing
Cp ligands such as CpE can readily initiate a 5-exo-dig
cyclization to afford the key rhodium carbene intermediate
III-1,[14] which undergoes a migratory insertion to form six-
À
membered rhodacycle IV-1. Subsequent C N bond formation
Scheme 3. Scope of the [4+1] annulation. Reaction conditions:
1 (0.11 mmol), 2 (0.10 mmol), [RhCpECl2]2 (5 mol%), AgOAc
(20 mol%), CH3CN (0.5 mL). The ratio of 3/4 as determined by
1H NMR spectroscopy is given in parentheses. [a] 1 (0.15 mmol).
[b] The major regioisomer is shown (isomer ratio: 7.1:1). [c] CsOAc
(25 mol%) was used instead of AgOAc.
À
and N O bond cleavage of IV-1 will give the [4+1]
product.[11b,c]
The above mechanistic considerations were supported by
a series of stoichiometric reactions of rhodacycles I bearing
a number of modified Cp ligands (Scheme 5). Initially, we
observed the rhodacycle I-CpG by 1H NMR spectroscopy
(4h), and ester (4i) groups were all tolerated under these
reaction conditions. A range of enynones also readily reacted,
and the electronic nature of the aryl moiety (R2, 4k–4n) had
no influence on the reaction outcome. The high efficiency was
maintained when an alkyl acetylenic enynone (R2 = C5H11)
was used (4o). It is noteworthy that the [4+1] products were
obtained exclusively in all reactions.
Scheme 5. Stoichiometric experiments.
The present reaction-path dichotomy determined by the
Cp ligand is depicted in Scheme 4 along with preliminary
experimental evidence. With the Cp*Cy ligand, rhodacycle
formation was rate-limiting as evidenced by the high kinetic
isotope effects (parallel experiments: kH/kD = 3.38; intermo-
lecular: PH/PD = 8.09). In stark contrast, no significant KIE
values were measured when the CpE ligand was employed
(kH/kD = 1.01; PH/PD = 1.63). We propose that the mechanistic
dichotomy is critically associated with the electronic nature of
the postulated enynone-bound rhodacycle species II. Elec-
when benzamide 1a was treated with 0.5 equiv of
[RhCpGCl2]2 in the presence of AgOAc (see the Supporting
Information). One rhodacycle obtained from Cp*RhIII was
characterized by X-ray analysis after stabilization with
a phosphine ligand (I-PPh3), and this species was shown to
catalyze the present annulation reaction (see the Supporting
Information).[16] The rhodacycle I-Cp* reacted with enynone
2a to afford a mixture of 3a and 4a (7.0:1). The formation of
3a was significantly favored when I-Cp*Cy was used (3a/4a
> 17.8:1). On the other hand, the reaction of I-CpE with 2a
was slow to give the [4+1] product 4a at 258C. Importantly, all
product ratios observed in the stoichiometric reactions were
similar to those of the catalytic variants (see Table 1).
The six-membered rhodacyclic intermediate IV-1 was
1
observed by H NMR spectroscopy in the reaction of I-CpE
with 2a (see the Supporting Information). The structure of
one derivative (IV-1-CN) was determined by X-ray analysis
(Figure 1).[16] Importantly, IV-1 smoothly underwent reduc-
tive elimination to the [4+1] product 4a at 608C as monitored
by 1H NMR spectroscopy, implying that this step is rate-
limiting (Figure 2). To the best of our best knowledge, IV-1 is
the first characterized intermediate that was caught in action
proceeding to reductive elimination in a RhIII-catalyzed [n+1]
annulation.
Density functional calculations provided insight into the
origin of this dramatic change in reaction pathway as
a function of the electronic properties of the Cp ligand
(Figure 3). The activation barriers for alkyne insertion into
the II-Cp* and II-Cp*Cy complexes were lower by 1.0 and
2.9 kcalmolÀ1 than those for the conjectured carbene transfer
Scheme 4. Proposed dichotomous reaction pathways.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
ent is present on the Cp ligand of the rhodacycle. Indeed,
carbene insertion with II-CpE was more favorable than alkyne
insertion by 6.8 kcalmolÀ1 (11.2 vs. 18.0 kcalmolÀ1 in red,
Figure 3a).
Additional information on the reaction pathway dichot-
omy was obtained by natural bond orbital (NBO) analysis
(Figure 3b).[15] During alkyne insertion into the rhodacycle II-
Cp*, the NBO charge of the rhodium center increases from
0.170 to 0.277. Along with this, a notable decrease in the NBO
charge on the triple bond of enynone 2a (0.099 to À0.167) is
seen, which implies the flow of electrons from the rhodium
center to the enynone during the alkyne insertion (Figure 3b,
left). However, during carbene transfer at the same rhoda-
cycle II-Cp*, the decrease in the charge on Rh (0.170 to 0.149)
is accompanied by an increase in the charge on the triple bond
(0.099 to 0.365), implying that the Rh center is reduced by
electron transfer from the enynone (Figure 3b, right).
Based on the above interpretations, “arrow-pushing”
schemes can depict how electronic variations of the Cp
ligand will affect the annulation pathway. With electron-rich
Cp ligands, the resultant Lewis basic rhodacycle will favor
alkyne insertion (Figure 3c, left). Indeed, this view was
reinforced by the decreased NBO charge of rhodium in II-
Cp*Cy (0.163) compared with that in II-Cp* (0.170). In
contrast, the corresponding rhodium species bearing the
CpE ligand displays a higher NBO charge at rhodium (0.235 in
II-CpE; Figure 3c, right). This increase in Lewis acidity of
rhodacycle II-CpE will preferably induce a 5-exo-dig cycliza-
tion, leading to formal carbene transfer. On the other hand,
the steric influence of the Cp derivatives on the product
dichotomy was not as obvious as the influence of the
electronic factors (see the Supporting Information).
Figure 1. XRD structure of IV-1-CN. Thermal ellipsoids set at 30%
probability. H atoms omitted for clarity.[16]
Figure 2. Reaction profile of the reductive elimination. Red: IV-1;
black: 4a.
In conclusion, we have achieved a switch between [4+2]
and [4+1] annulation by tuning the Cp ligand in the RhIII-
catalyzed annulation of benzamides with conjugated eny-
nones. While triple bond insertion was favored by electron-
rich Cp ligands to give the [4+2] annulation products, the
presence of electron-withdrawing substituents on the Cp
ligand resulted in a complete switch in reactivity to a formal
carbene transfer, affording the [4+1] cyclization products.
KIE experiments, the characterization of two key intermedi-
ates, and DFT studies indicate that this dichotomy can be
attributed to the change in Lewis acidity of the Cp-bound
rhodium species upon modification of the Cp ligand.
Acknowledgements
This research was supported by the Institute for Basic Science
(IBS-R010-D1).
Conflict of interest
Figure 3. a) Energy profiles with three Cp ligands. b) NBO charge
analysis. c) Distinct annulation paths.
The authors declare no conflict of interest.
(16.0 vs. 17.0 kcalmolÀ1 in black and 12.7 vs. 15.6 kcalmolÀ1 in
blue, Figure 3a), respectively. However, this situation is
completely reversed when an electron-withdrawing substitu-
Keywords: annulations · cyclopentadienyl ligands ·
reaction mechanisms · rhodium catalysis
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[9] a) J.-R. Chen, X.-Q. Hu, L.-Q. Lu, W.-J. Xiao, Chem. Rev. 2015,
115, 5301; b) R.-Y. Zhu, M. E. Farmer, Y.-Q. Chen, J.-Q. Yu,
Angew. Chem. Int. Ed. 2016, 55, 10578; Angew. Chem. 2016, 128,
10734.
[1] a) R. G. Bergman, Science 1984, 223, 902; b) W. D. Jones, F. J.
Feher, Organometallics 1983, 2, 562.
[10] a) N. Guimond, S. I. Gorelsky, K. Fagnou, J. Am. Chem. Soc.
2011, 133, 6449; b) X. Xu, Y. Liu, C.-M. Park, Angew. Chem. Int.
Ed. 2012, 51, 9372; Angew. Chem. 2012, 124, 9506; c) J. M. Neely,
T. Rovis, J. Am. Chem. Soc. 2013, 135, 66; d) N. QuiÇones, A.
Seoane, R. Garcia-Fandino, J. L. Mascarenas, M. Gulias, Chem.
Sci. 2013, 4, 2874; e) Y. Fukui, P. Liu, Q. Liu, Z.-T. He, N.-Y. Wu,
P. Tian, G.-Q. Lin, J. Am. Chem. Soc. 2014, 136, 15607; f) K. K.
Gollapelli, S. Kallepu, N. Govindappa, J. B. Nanubolu, R.
Chegondi, Chem. Sci. 2016, 7, 4748.
[11] a) T. K. Hyster, K. E. Ruhl, T. Rovis, J. Am. Chem. Soc. 2013,
135, 5364; b) H.-W. Lam, K.-Y. Man, W.-W. Chan, Z. Zhou, W.-
Y. Yu, Org. Biomol. Chem. 2014, 12, 4112; c) Y. Zhang, D. Wang,
S. Cui, Org. Lett. 2015, 17, 2494; d) B. Zhou, Z. Chen, Y. Yang,
W. Ai, H. Tang, Y. Wu, W. Zhu, Y. Li, Angew. Chem. Int. Ed.
2015, 54, 12121; Angew. Chem. 2015, 127, 12289.
[2] a) T. Satoh, M. Miura, Chem. Lett. 2007, 36, 200; b) J. F. Hartwig,
Acc. Chem. Res. 2012, 45, 864; c) S. R. Neufeldt, M. S. Sanford,
Acc. Chem. Res. 2012, 45, 936; d) S. H. Cho, J. Y. Kim, J. Kwak, S.
Chang, Chem. Soc. Rev. 2011, 40, 5068; e) T. Gensch, M. N.
Hopkinson, F. Glorius, J. Wencel-Delord, Chem. Soc. Rev. 2016,
45, 2900; f) H. M. L. Davies, D. Morton, J. Org. Chem. 2016, 81,
343.
[3] a) D. A. Colby, R. G. Bergman, J. A. Ellman, Chem. Rev. 2010,
110, 624; b) T. Satoh, M. Miura, Chem. Eur. J. 2010, 16, 11212;
c) S. Chiba, Chem. Lett. 2012, 41, 1554; d) G. Song, F. Wang, X.
Li, Chem. Soc. Rev. 2012, 41, 3651; e) C. Zhu, R. Wang, J. R.
Falck, Chem. Asian J. 2012, 7, 1502; f) N. Kuhl, N. Schrçder, F.
Glorius, Adv. Synth. Catal. 2014, 356, 1443.
[4] a) S. H. Park, J. Kwak, K. Shin, J. Ryu, Y. Park, S. Chang, J. Am.
Chem. Soc. 2014, 136, 2492; b) Y.-F. Yang, K. N. Houk, Y.-D. Wu,
J. Am. Chem. Soc. 2016, 138, 6861.
[12] a) K. Miki, F. Nishino, K. Ohe, S. Uemura, J. Am. Chem. Soc.
2002, 124, 5260; b) Y. Kato, K. Miki, F. Nishino, K. Ohe, S.
Uemura, Org. Lett. 2003, 5, 2619; c) Y. Xia, S. Qu, Q. Xiao, Z. X.
Wang, P. Qu, L. Chen, Z. Liu, L. Tian, Z. Huang, Y. Zhang, J.
Wang, J. Am. Chem. Soc. 2013, 135, 13502.
[5] a) T. K. Hyster, T. Rovis, Chem. Commun. 2011, 47, 11846;
b) T. K. Hyster, T. Rovis, Chem. Sci. 2011, 2, 1606; c) T. K.
Hyster, D. M. Dalton, T. Rovis, Chem. Sci. 2015, 6, 254.
[6] a) J. M. Neely, T. Rovis, J. Am. Chem. Soc. 2014, 136, 2735; b) F.
Romanov-Michailidis, K. F. Sedillo, J. M. Neely, T. Rovis, J. Am.
Chem. Soc. 2015, 137, 8892; c) M. Fukui, Y. Hoshino, T. Satoh,
M. Miura, K. Tanaka, Adv. Synth. Catal. 2014, 356, 1638; d) Y.
Takahama, Y. Shibata, K. Tanaka, Chem. Eur. J. 2015, 21, 9053;
e) Y. Takahama, Y. Shibata, K. Tanaka, Org. Lett. 2016, 18, 2934;
f) M. Fukui, Y. Shibata, Y. Hoshino, H. Sugiyama, K. Teraoka,
H. Uekusa, K. Noguchi, K. Tanaka, Chem. Asian J. 2016, 11,
2260.
[13] We recently reported a switch in reaction pathway for a Cp*IrIII
-
catalyzed process that depended on the directing group in the
substrate; see: J. Kim, S.-W. Park, M.-H. Baik, S. Chang, J. Am.
Chem. Soc. 2015, 137, 13448.
[14] C. Werlꢀ, R. Goddard, P. Philipps, C. Farꢁs, A. Fꢂrstner, J. Am.
Chem. Soc. 2016, 138, 3797.
[15] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899.
[16] CCDC 1524489 (3a), 1524491 (4a), 1524490 (I-PPh3), and
1524492 (IV-1-CN) contain the supplementary crystallographic
data for this paper. These data are provided free of charge by
The Cambridge Crystallographic Data Centre.
[7] a) T. K. Hyster, L. Knçrr, T. R. Ward, T. Rovis, Science 2012, 338,
500; b) B. Ye, N. Cramer, Science 2012, 338, 504.
[8] a) M. D. Wodrich, B. Ye, J. F. Gonthier, C. Corminboeuf, N.
Cramer, Chem. Eur. J. 2014, 20, 15409; b) T. Piou, T. Rovis, J.
Am. Chem. Soc. 2014, 136, 11292; c) T. Piou, T. Rovis, Nature
2015, 527, 86.
Manuscript received: December 28, 2016
Final Article published: && &&, &&&&
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Rhodium Catalysis
S. Y. Hong, J. Jeong,
S. Chang*
&&& — &&&
[4+2] or [4+1] Annulation: Changing the
Reaction Pathway of a Rhodium-
Catalyzed Process by Tuning the
Cp Ligand
The choice is yours: A change in reaction
pathway was achieved by tuning the
cyclopentadienyl (Cp) ligand used for the
rhodium-catalyzed cyclization of benza-
mides with conjugated enynones.
Depending on the Cp ligand, the reaction
pathway switched between [4+2] and
[4+1] annulation. Electronic effects
turned out to be crucial for the product
distribution.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!