Isolation of Key Organometallic Aryl-Co(III) Intermediates in Cobalt-Catalyzed C(sp2)-H Functionalizations and New Insights into Alkyne Annulation Reaction Mechanisms

Article abstract of DOI:10.1021/jacs.6b08593

The selective annulation reaction of alkynes with substrates containing inert C-H bonds using cobalt as catalyst is currently a topic attracting significant interest. Unfortunately, the mechanism of this transformation is still relatively poorly understoo

Full text of DOI:10.1021/jacs.6b08593

Subscriber access provided by RYERSON UNIV
Article
Isolation of Key Organometallic Aryl-Co(III) Intermediates
in Co-Catalyzed C(sp2)-H Functionalizations and New
Insights into Alkyne Annulation Reaction Mechanisms
Oriol Planas, Christopher J Whiteoak, Vlad Martin-Diaconescu, Ilaria
Gamba, Josep M. Luis, Teodor Parella, Anna Company, and Xavi Ribas
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b08593 • Publication Date (Web): 10 Oct 2016
Downloaded from http://pubs.acs.org on October 10, 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 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Isolation of Key Organometallic Aryl-Co(III) Intermediates in Co-
Catalyzed C(sp2)-H Functionalizations and New Insights into Alkyne
Annulation Reaction Mechanisms
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
Oriol Planas,1 Christopher J. Whiteoak,1,2 Vlad MartinꢀDiaconescu,1 Ilaria Gamba,1 Josep M. Luis,1
Teodor Parella,3 Anna Company,1 Xavi Ribas1,*
1Institut de Química Computacional i Catàlisi (IQCC) and Departament de Química, Universitat de Girona, Campus Montiliꢀ
vi, Girona, Eꢀ17071, Catalonia, Spain.
2Biomolecular Sciences Research Centre, Faculty of Health and Wellbeing, City Campus, Sheffield Hallam University, Shefꢀ
field, S1 1WB, England.
3Servei de RMN, Facultat de Ciències, Universitat Autònoma de Barcelona, Campus UAB, Bellaterra, Eꢀ08193, Catalonia,
Spain
ABSTRACT: The selective annulation reaction of alkynes with substrates containing inert CꢀH bonds using Co as catalyst is curꢀ
rently a topic attracting significant interest. Unfortunately, the mechanism of this transformation is still relatively poorly underꢀ
stood, with little experimental evidence for intermediates, although an organometallic Co(III) species is generally implicated. Hereꢀ
in, we describe a rare example of the preparation and characterization of benchꢀtop stable organometallic arylꢀCo(III) compounds
(NMR, HRMS, XAS and XRD) prepared through a C(sp2)ꢀH activation, using a model macrocyclic arene substrate. Furthermore,
we provide crystallographic evidence of an organometallic arylꢀCo(III) intermediate proposed in 8ꢀaminoquinoline directed Coꢀ
catalyzed CꢀH activation processes. Subsequent insights obtained from the application of our new organometallic arylꢀCo(III) comꢀ
pounds in alkyne annulation reactions are also disclosed. Evidence obtained from the resulting regioselectivity of the annulation
reactions and DFT studies indicates that a mechanism involving an organometallic arylꢀCo(III)ꢀalkynyl intermediate species is
preferred for terminal alkynes, in contrast to the generally accepted migratory insertion pathway.
through a mild and facile Coꢀcatalyzed alkyne annulation
reaction.9 Since this report, an increasing number of Coꢀ
catalyzed annulation protocols have been reported using a
range of coupling partners,10 including several examples highꢀ
lighting the potential of alkynes.11 Unfortunately, the mechaꢀ
nisms of these reactions are still relatively poorly understood,
although organometallic Co(III) intermediates are generally
implicated.5 Stable organometallic arylꢀCo(III) compounds
have been prepared via transmetalation reactions in the
past,11g,12 although examples of the preparation of stable orꢀ
ganometallic Co(III) compounds through direct CꢀH activation
are still extremely rare which has limited the understanding of
this field. Organometallic arylꢀCo(III) compounds derived
from Nꢀconfused porphyrins and mꢀbenziphthalocyanines
have been metalated using Co(II) salts furnishing the Co(III)
CꢀH activated products,13 although further reactivity studies
were somewhat limited as result of their stability.14
INTRODUCTION
Selective functionalization of inert CꢀH bonds is currently
attracting significant interest as the desire to develop new
simplified protocols for the synthesis of complex pharmaceuꢀ
ticals, agrochemicals and natural products gathers pace.1 Hisꢀ
torically, most CꢀH activation and coupling protocols have
been based on expensive, precious secondꢀ and thirdꢀrow
transition metals, most notably Pd.2 More recently, the develꢀ
opment of catalytic CꢀH functionalization protocols based on
naturally more abundant and cheaper firstꢀrow transition metꢀ
als, including Fe,3 Cu,4 Co5 and Ni,6 has seen an explosion of
activity. Unfortunately, mechanistic understanding concerning
how these new firstꢀrow transition metal based protocols operꢀ
ate is significantly less advanced than their secondꢀ and thirdꢀ
row analogues. This is likely a consequence of the elusiveness
and instability of key reaction intermediates, although there is
an increasing interest in their detection and isolation.7 One
reaction which is receiving particular interest at the current
time is the annulation reaction of alkynes to substrates with
inert CꢀH bonds, providing a facile route towards a variety of
biologically and pharmaceutically relevant heterocyclic comꢀ
pounds. To date, Rhꢀcatalysis has provided significant success
in this field,8 although the replacement of expensive Rh with a
cheaper firstꢀrow transition metal has remained a challenge
until recently. In this context, in 2013 Yoshikai reported the
first synthesis of polysubstituted dihydropyridine derivatives,
In the field of Coꢀcatalyzed alkyne annulation protocols
through CꢀH activation, the most solid experimental evidences
so far of reaction intermediates has been provided by succinct
NMR, MALDIꢀTOF and most recently XꢀRay diffraction
(XRD). In 2014, Daugulis and coꢀworkers reported the prepaꢀ
ration and characterization (1D NMR) of an organometallic
arylꢀCo(III) compound, obtained through C(sp2)ꢀH activation,
which was implicated as the key intermediate in this Coꢀ
catalyzed annulation protocol (Scheme 1A).11b In 2015, Zhang
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
R3
H
B
A
1
2
3
4
Co(OAc)2.4H2O (20 mol%)
Ag2CO3 (3.0 equiv.)
TBAI (4.0 equiv.)
Na2CO3 (3.0 equiv.)
Py (2.0 equiv.)
R1
R2
Co(OAc)2.4H2O (10 mol%)
NaOPiv.H2O (2.0 equiv.)
Mn(OAc)2 (1.0 equiv.)
O
O
O
O
N
R1
R2
N
R1
R2
N
N
N
H
CF3CH2OH, 80oC, air, 6h
H
N
R2
N
C6H5CF3, 150oC, N2, 22h
5
N
H
H
R1
R3
6
7
8
9
O
N
O
R1
via
N
via
R2
N
N
Co
Co
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
Py
X
Py
CF3
AcO
Daugulis, 2014
- Isolated and characterised by 1D NMR
Zhang, 2015
- Observed in MALDI-TOF-MS
Maiti, 2016
- Isolated and characterized by XRD
This work:
C
-All intermediates and products isolated and characterised
R1
R1
N
OAc
N
Co(OAc)2
O
R2
R3
N
N
O
O
or
R1
N
H
N
R1
N
NH
N
NH
Co
Co
R3
CF3CH2OH
air, 100oC, 16h
CF3CH2OH
(when R1 = H)
CF3CH2OH
air, 25oC, 16h
H
N
O
O
R2
N
R2
O
N
N
N
R1
R1
(R3 = H)
2x-H
3x-OAc
(3b-OAc, R1 = H)
4x-OAc
4b-OAc, R1 = H)
Different products depending upon:
temperature / variation of R2 and R3
(
(
(
2b-H, R1 = H)
4c-OAc, R1 = Me)
(
2c-H, R1 = Me)
(
3c-OAc, R1 = Me)
Co(OAc)2
CF3CH2OH
air, 100oC, 36h
Scheme 1. Characterized organometallic Co(III) intermediates in Co-catalyzed alkyne annulation protocols. (A) C(sp2)ꢀH activation
and the organometallic arylꢀCo(III) intermediate isolated by Daugulis and coꢀworkers and Maiti and coꢀworkers. (B) C(sp3)ꢀH activation
and organometallic alkylꢀCo(III) intermediate reported by Zhang and coꢀworkers. (C) Overview of this work; Co(II) and Co(III) intermeꢀ
diates and annulation products characterized by 1D and 2D NMR, HRMS, XRD and XAS.
and coꢀworkers observed, by MALDIꢀTOF analysis of a crude
reaction mixture during a Co(II)ꢀcatalyzed annulation reaction,
evidence of an organometallic C(sp3)ꢀCo(III) intermediate,
resulting from direct activation of C(sp3)ꢀH bonds (Scheme
1B).11a Very recently, in a related study, Maiti and coꢀworkers
have successfully provided XRD evidence of a species inꢀ
voked as an intermediate in Coꢀcatalyzed alkene coupling
reactions.15
from the application of these isolable organometallic arylꢀ
Co(III) intermediates in alkyne annulation reactions, in which
they are proposed as key reaction intermediate (Scheme 1C)
will also be disclosed, including evidence for an unprecedentꢀ
ed ‘acetylide pathway’ for terminal alkynes in annulation
reactions. Putting this work into context, we also report crysꢀ
tallographic evidence of the soughtꢀafter organometallic arylꢀ
Co(III) intermediate proposed in 8ꢀaminoquinoline directed
Coꢀcatalyzed CꢀH activation processes.
Taking into account the increasing interest in Coꢀcatalyzed
transformations and the absence of benchꢀtop stable organoꢀ
metallic arylꢀCo(III) compounds prepared through direct CꢀH
activation, we became interested in the possibility of isolating
organometallic arylꢀCo(III) compounds, in order to understand
their synthesis, features and reactivity. Our approach was
based on expertise within our laboratory using model macroꢀ
cyclic arene substrates to successfully stabilize organometallic
arylꢀmetal intermediate species, including Ni(II)16 and
Cu(III).16,17 These isolable compounds have now been impliꢀ
cated as the intermediate species in a wide variety of crossꢀ
coupling transformations.
RESULTS
Synthesis and characterization of organometallic aryl-
Co(III) compounds. Macrocyclic model arene substrates 2b-
H and 2c-H were synthesized through cyclization of the tosylꢀ
protected amine 1 with 2,6ꢀbis(bromothethyl)pyridine, folꢀ
lowed by a deprotection of tosylated macrocyclic compound
2a-H, furnishing 2b-H, which could be further methylated at
the secondary amines to obtain 2c-H (see Scheme S1). Thereꢀ
after, Co(II) coordination compounds were prepared by reacꢀ
tion of Co(OAc)2 with 2b-H and 2c-H (Scheme 1C), at 25ºC,
using 2,2,2ꢀtrifluoroethanol (TFE) as solvent ([(2b-
H)Co(II)(OAc)2] (3b-OAc) and [(2c-H)Co(II)(OAc)2] (3c-
OAc)). The structure of 3b-OAc was initially confirmed by
HRMS ([M ꢀ OAc]+; calcd. m/z = 357.0882; found: 357.0874)
and XRD analysis (see Figure S15). The solid state molecular
structure of 3b-OAc indicated that the compound has two
acetates coordinated in a bidentate fashion and that the potenꢀ
tially tridentate macrocycle acts as bidentate ligand (coordiꢀ
Herein, we will describe the synthesis and characterization of
benchꢀstable stable organometallic arylꢀCo(III) compounds
obtained through C(sp2)ꢀH activation, in a clear stepꢀwise
manner, using a 12ꢀmembered macrocyclic model substrate,
which have been fully crystallographically and spectroscopiꢀ
cally characterized. A similar macrocyclic substrate has previꢀ
ously been utilized to investigate the reactivity of organomeꢀ
tallic Ni(III) intermediates.18 Subsequent insights obtained
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
nated through only the pyridine and one amine). Upon perꢀ
forming the same reaction with CoBr2 and 2c-H, complex 3c-
1
2
3
4
5
6
7
Br was obtained, which in contrast, had both amines of the
macrocyclic ligand coordinated (XRD, see Figure S16). Careꢀ
ful analysis of these crystal structures shows that the aryl CꢀH
bond is out of plane (torsion angles of 14º for 3b-OAc and 21º
for 3c-Br), indicating an incipient interaction between the
Co(II) and hydrogen of the arene.
Upon reacting Co(OAc)2 with 2b-H and 2c-H in TFE at 100
oC (or heating of the preformed 3b-OAc and 3c-OAc), comꢀ
plete conversion to the target organometallic arylꢀCo(III)
complexes, [(2b)Co(III)(η2ꢀOAc)](OAc) (4b-OAc) and
[(2c)Co(III)(η2ꢀOAc)](OAc) (4c-OAc), was achieved (Scheme
1C). Due to the stability of the isolated organometallic arylꢀ
Co(III) compounds, it was possible to characterize them by
NMR and HRMS (Figure 1 and Figures S36ꢀS45), providing
spectra consistent with a low spin diamagnetic Co(III) metal
centre. Two nonꢀequivalent acetate signals were observed,
indicating that only one acetate is likely to be coordinated to
the Co complex with the second acetate displaced as an outerꢀ
sphere counter ion. The proposed structure was further supꢀ
ported by HRMS studies of 4b-OAc and 4c-OAc, which afꢀ
forded clean spectra with major signals corresponding to the
[(2)Co(III)(OAc)]+ ions (4b-OAc, [M – OAc]+; calcd. m/z =
356.0804; found: 356.0802; 4c-OAc, [M – OAc]+; calcd. m/z
= 384.1117; found: 384.1122). An analogous organometallic
Rh(III) complex (5c-Cl) was also prepared following a similar
protocol (see Scheme S15 and Figure S17).
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
-CH3
Me
Figure 2. Characterization of arylꢀCo(III) compounds through
EXAFS spectroscopy. (A) XAS spectra at the Co Kꢀedge highꢀ
lighting the XANES region of the spectrum, and the 1s → 3d preꢀ
edge transitions (inset). (B) EXAFS analysis of 4b-OAc. Shown
are Fourierꢀtransformed EXAFS spectra (no phase correction, kꢀ
window = 2 ꢀ12.5 Åꢀ1) as well as the k3ꢀweighted unfiltered
EXAFS spectra (inset) and comparison of selected EXAFS deꢀ
rived and theoretical bond distances (bottom table). (C) EXAFS
analysis of 4c-OAc. Shown are Fourierꢀtransformed EXAFS
spectra (no phase correction, kꢀwindow = 2 ꢀ12.5 Åꢀ1) as well as
the k3ꢀweighted unfiltered EXAFS spectra (inset) and comparison
of selected EXAFS derived and theoretical bond distances (botꢀ
tom table). (D) Geometry optimized structure for 4b-OAc. Selectꢀ
ed bond distances [Å] and angles [º]: CoꢀC(1) 1.85, CoꢀN(1) 1.85,
CoꢀN(2) 2.00, CoꢀN(3) 2.00, CoꢀO(1) 1.96, CoꢀO(2) 2.11; C(1)ꢀ
CoꢀN(1) 89.8, C(1)ꢀCoꢀN(2) 84.5, C(1)ꢀCoꢀN(3) 84.4, C(1)ꢀCoꢀ
O(1) 99.6, C(1)ꢀCoꢀO(2) 164.9. (E) Geometry optimized structure
for 4c-OAc. Selected bond distances [Å] and angles [º]: CoꢀC(1)
1.85, CoꢀN(1) 1.85, CoꢀN(2) 2.05, CoꢀN(3) 2.05, CoꢀO(1) 1.96,
CoꢀO(2) 2.11; C(1)ꢀCoꢀN(1) 86.5, C(1)ꢀCoꢀN(2) 84.6, C(1)ꢀCoꢀ
N(3) 84.6, C(1)ꢀCoꢀO(1) 101.0, C(1)ꢀCoꢀO(2) 166.3.
OAc
N
N
O
Co
O
-OAc
N
Me
-OAc
III
MeLCo (OAc2
4c-OAc)
HAr
-CH2-
-CH2-
HPh
Hpyr
Figure 1. 1H NMR spectrum with assignments for 4c-OAc in
CDCl3 at 298 K.
In our initial studies, XRD characterization of 4b-OAc and 4c-
OAc remained elusive and as a result we turned to Xꢀray Abꢀ
sorption Spectroscopy (XAS) at the Co Kꢀedge, in order to
probe the electronic structure (XANES) and coordination
environment (EXAFS) (Figure 2AꢀC and Figures S19ꢀS22).
Figure 2A shows the rising edge spectra for the starting mateꢀ
rial and the organometallic arylꢀCo(III) 4b-OAc and 4c-OAc
analogues. 3c-OAc shows a weak preꢀedge feature due to
1sꢀ3d transitions at ~7709.8 eV consistent with a Co(II)
centre having a centrosymmetric local environment.19 EXAFS
analysis of 3c-OAc shows a six coordinate complex with three
pairs of CoꢀN/O scattering shells at 2.04 Å, 2.22 Å and 2.35
Å, suggesting a distorted octahedral coordination environment
at the metal centre (Figure S20 and Table S15). XAS data for
3c-Br on the other hand is consistent with the proposed 5ꢀ
coordinate crystal structure (Figure S19 and Table
S14). Upon CꢀH activation, a 0.5 eV shift in the rising edge at
half height is observed for both 3c-OAc and the 3b-OAc anaꢀ
logue, concomitant with a ~1.6 eV shift to higher energy of the
preꢀedge, indicating oxidation of the metal center to Co(III).19
Additionally, the increase in intensity of the preꢀedge feature
observed, not only accounts for the increase in oxidation state
and the extra electron hole, but also suggests a coordination
geometry more favorable to pꢀd mixing. EXAFS analysis of
4b-OAc and 4c-OAc (Figures 2B and 2C) intermediates indiꢀ
cates the presence of two short N/O/C scattering atoms at
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
Thereafter, formation of the organometallic arylꢀCo(III) comꢀ
1
2
3
4
5
6
7
8
pounds was further studied (see Scheme 2, and Schemes S17ꢀ
S19). It was found that the addition of 2.0 equivalents of
KOAc to a mixture of 2c-H and CoBr2 in MeCN resulted in
complete conversion to 4c-OAc, whereas in the absence of
base a mixture of Co(II) complex 3c-Br and Co(III) organoꢀ
metallic 4c-Br was obtained, suggesting that both solvent and
base have an important role in the CꢀH activation step. In
particular, fluorinated alcohols appear be crucial (Table S2), as
has previously been identified for a number of other metalꢀ
catalyzed transformations.20 Preparation of the organometallic
arylꢀCo(III) compounds was also attempted under an inert
atmosphere (N2), but the reaction did not proceed when startꢀ
ing from Co(II) salts and 2c-H. However, if the reaction was
performed under an inert atmosphere in the presence of oxiꢀ
dants, such as silver salts or TEMPO, organometallic arylꢀ
Co(III) intermediates could be detected using HRMS (Scheme
S19, Figures S5ꢀS6). When Co(acac)3 was used as Co source,
under an inert atmosphere, the corresponding organometallic
arylꢀCo(III) species (4c-acac) was formed as the major prodꢀ
uct (Figure S4). These observations indicate that Co(II) is
likely first oxidized before the CꢀH activation step, forming a
highly electrophilic pentaꢀcoordinated Co(III) 16eꢀ complex
(Scheme 2, 3x’-OAc). Interaction of Co(III) with the correꢀ
sponding CꢀH bond promotes its deprotonation by the base
(acetate, bromide or perchlorate in this study). These experiꢀ
mental observations lead us to propose that CꢀH bond cobaltaꢀ
tion is likely to proceed via a Concerted Metalationꢀ
Deprotonation (CMD) pathway.
B
A
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
C1-Co = 1.876(7) Å
C1-Co = 1.865(2)Å
Figure 3. Solid state structures of organometallic aryl-Co(III)
compounds. Hydrogen atoms, perchlorate anions and solvent
molecules have been omitted for clarity; ellipsoids are set at 50%
probability. (A) Crystal data for 4b-CH3CN. Selected bond disꢀ
tances [Å] and angles [º]: CoꢀC(1) 1.876(7), CoꢀN(1) 1.853(4),
CoꢀN(2) 1.997(6), CoꢀN(3) 2.008(5), CoꢀN(4) 2.025(6), CoꢀN(5)
1.897(4); C(1)ꢀCoꢀN(1) 90.55(2), C(1)ꢀCoꢀN(2) 83.68(2), C(1)ꢀ
CoꢀN(3) 83.66(2), C(1)ꢀCoꢀN(4) 177.87 (2), C(1)ꢀCoꢀN(5)
88.40(2). (B) Crystal data for 4c-CH3CN. Selected bond distancꢀ
es [Å] and angles [º]: CoꢀC(1) 1.865(2), CoꢀN(1) 1.856(2), Coꢀ
N(2) 2.029(2), CoꢀN(3) 2.027(2), CoꢀN(4) 1.999(2), CoꢀN(5)
1.927(2); C(1)ꢀCoꢀN(1) 88.16(9), C(1)ꢀCoꢀN(2) 84.13(9), C(1)ꢀ
CoꢀN(3) 84.26(9), C(1)ꢀCoꢀN(4) 178.30(9), C(1)ꢀCoꢀN(5)
91.30(9).
~1.89 Å and four longer N/O/C scattering paths with an averꢀ
age effective path length spanning ~2.04 ꢀ 2.09 Å. This is
consistent with the proposed theoretical models that predict
two short N/O/C pyridine nitrogen and CꢀH activated carbon
(Cphenyl) and four longer scattering paths (~2.0 Å) coming from
the tertiary amines and an acetate, resulting in a trigonal biꢀ
pyramidal geometry at the meal center. The proposed strucꢀ
tures for the 4b-OAc and 4c-OAc complexes are in agreement
with the NMR and HRMS data (vide supra) and help to exꢀ
plain the intense preꢀedge features from XAS, whereby a
trigonal bipyramidal local metal geometry facilitates pꢀmixing
into the dꢀmanifold resulting in a more intense preꢀedge.
As mentioned, it was not possible to obtain suitable crystals of
4x-OAc for XRD studies. As reaction using silver salts as
oxidant was also satisfactory, another strategy to grow crystals
was followed; 3x-Br was mixed with AgClO4 (3.0 equiv.) in
an CH3CN/TFE (10:1) mixture at 100ºC, whereby the correꢀ
sponding organometallic complex 4x-CH3CN was obtained
(see Scheme S14 and Fig S13ꢀS14). Crystals of 4b-CH3CN, as
well as 4c-CH3CN, suitable for XRD studies were this time
obtained and fully characterized using XRD and other techꢀ
niques (Figure 3A and 3B, Figures S46ꢀ55). The experimentalꢀ
ly determined EXAFS CoꢀCphenyl bond distances of 1.88 Å for
4b-OAc and 1.90 Å for 4c-OAc (Figure 2Bꢀ2E) are consistent
with the Co(III)ꢀCphenyl distance of 1.86 Å determined for both
4b-CH3CN and 4c-CH3CN from their crystal structures (Figꢀ
ure 3A and 3B). In order to put our new organometallic arylꢀ
Co(III) compounds into context, we have also been able to
obtain XRD evidence of an arylꢀCo(III) intermediate furnished
through assistance of the 8ꢀaminoquinoline directing group.21
Using a nitroꢀderivative of the aminoquinoline scaffold (6),22
together with the addition of 2,2’ꢀbipyridine (bipy), we were
able to successfully crystallize this soughtꢀafter intermediate,
[(6)Co(III)(bipy)(OAc)] (8) (Scheme S1 and Figure S18). This
result sets definitive proof of the structure of this intermediate
organometallic species and highlights the relevance of our new
isolable organometallic arylꢀCo(III) compounds for studying
highꢀvalent Coꢀcatalyzed C(sp2)ꢀH annulation reactions.
R1
N
Co Y2
N
Co (acac)3
Y
Co
R1
N
H
N R1
R1
N
H
N R1
solvent
Y
CF3CH2OH
N2, 100oC, 16h
N
N
oxidant, 100oC, 16h
N
R1
2x-H
4x-Y
(2b-H, R1 = H)
(2c-H, R1 = Me)
(4b-Y, R1 = H)
(4c-Y, R1 = Me)
Co Y2
rt, solvent
CMD
base
R1
N
R1
N
oxidant
(O2, Ag+, TEMPO)
Proposed electrophilic CoIII
intermediate that promotes
a fast C-H activation step
through a Concerted
Metalation Deprotonation
pathway
N
Y
Co
N
Y
H
Co
Y
Y
100ºC, 16h
H
N
N
R1
R1
Study of isolated organometallic aryl-Co(III) compounds
in stoichiometric alkyne annulation reactions. With the
isolated organometallic arylꢀCo(III) compounds in hand, we
then turned our attention towards the reactivity of 4b-OAc
using alkynes as coupling partners to furnish annulated prodꢀ
ucts. We envisioned that this compound may provide similar
products to the previously reported Coꢀcatalyzed annulation
reactions with alkynes,11 which are believed to proceed via βꢀ
migratory insertion of the alkyne into the organometallic arylꢀ
3x-Y
3x'-OAc
(3b-Y, R1 = H)
(3c-Y, R1 = Me)
16e- Co(III) intermediate
Scheme 2. Formation of aryl-Co(III) intermediates. Optiꢀ
mization and evaluation of conditions using different solvents
(TFE, CH3CN) and oxidants (air, silver salts, TEMPO) to
furnish organometallic arylꢀCo(III) intermediates through a Cꢀ
H activation approach.
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
Co(III) bond, furnishing a dihydroisoquinoline product (6ꢀ 11a and 11aa (Table 2, entries 2 and 3), suggesting again
1
2
3
4
5
6
7
8
membered ring, 9a and 10a) upon reductive elimination of the
Co with our macrocyclic ligand. However, reaction of 4b-OAc
with 1ꢀethynylꢀ4ꢀnitrobenzene at 100ºC in TFE unexpectedly
resulted in the regioselective formation of the dihydroisoindoꢀ
line product 9b (5ꢀmembered ring) in good yields (Table 1,
entry 1). The unexpected formation of 9b led us to explore this
reactivity further, changing the reaction conditions, such as
temperature and electronic properties of the alkyne coupling
partner. Thus, reactions with 1ꢀethynylꢀ4ꢀnitrobenzene were
carried out at lower temperatures (Table 1, entries 2 and 3) and
interestingly, mixtures of 9a and 9b were obtained, with inꢀ
creasing ratio of 9a with decreasing temperature.
kinetic and thermodynamic products.
Table 2. Reactivity of organometallic aryl-Co(III) complex
4b-OAc with 1-phenyl-1-propyne as coupling partner.
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
Entrya
atm. temp (ºC) Yieldb (%) Ratioc (11a:11aa)
air
air
air
N2
100
60
rt
72
70
54
65
99:1
4:1
2:1
4:1
1
2
3
4
Table 1. Reactivity of organometallic aryl-Co(III) complex
4b-OAc with terminal alkynes as coupling partners.
60
a
Reaction conditions: 4b-OAc (0.077 mmol) and 1ꢀphenylꢀ1ꢀ
propyne (1.0 equiv.) were mixed in TFE and stirred at different
temperatures and atmospheres over 16h. b Products were isolated
using silicaꢀgel chromatography. c Ratios were determined by H
1
NMR.
Entrya
R
atm. temp (ºC) Yieldb (%) Ratio (a:b)c
Study of isolated aryl-Co(III) compounds in catalytic al-
kyne annulation reactions. The annulation reaction was also
studied in a catalytic fashion using 2b-H and terminal alkynes.
Initial tests were performed using catalytic quantities of
Co(OAc)2, 3b-OAc and 4b-OAc, together with 1ꢀethynylꢀ4ꢀ
nitrobenzene and 2b-H (Table 3). At room temperature, startꢀ
ing from Co(OAc)2 the reactions did not proceed, in contrast
to the stoichiometric experiments described above (Table 1,
entries 3 and 7). This result indicates that CꢀH activation is the
rate determining step for these transformations, in agreement
with what is suggested for most studies in this field. 1ꢀethynylꢀ
4ꢀnitrobenzene was successfully coupled to 2b-H at 80ºC
using TFE as solvent, under an oxygenated atmosphere, with
20 mol% of Co(OAc)2, 3b-OAc or 4b-OAc (Table 3, entries
1, 3 and 7).
1
2
3
4
5
6
7
8
4ꢀNO2Ph air
4ꢀNO2Ph air
4ꢀNO2Ph air
4ꢀNO2Ph N2
100
60
rt
60
100
60
rt
54
66
63
60
30
36
40
34
1:99 (9a:9b)
1:12
1:1.5
1:12
Ph
Ph
Ph
Ph
air
air
air
N2
99:1 (10a:10b)
99:1
4:1
99:1
60
a Reaction conditions: 4b-OAc (40mg, 0.077 mmol) and alkyne
(2.0 equiv.) were mixed in TFE and stirred at different temperaꢀ
b
tures and atmospheres over 16h. Products were isolated using
c
silicaꢀgel chromatography. Ratios were determined by NMR
spectrometry.
Table 3. Study of off-cycle intermediates in catalytic reac-
tions starting from 2b-H and terminal alkynes.
This result suggests that 9b is the thermodynamic product,
whilst 9a is the kinetic product. The structure of 9a and 9b
were unambiguously determined using NMR and MS techꢀ
niques (see Supporting Information for more details). Interestꢀ
ingly, under an inert atmosphere it was also possible to obtain
products (Table 1, entry 4), indicating that no oxidant is reꢀ
quired in the annulation reaction. Thereafter, reaction of 4b-
OAc with phenylacetylene was tested at 100ºC in TFE, which
resulted in the selective formation of the 6ꢀmembered ring
product 10a (Table 1, entry 5). Surprisingly, other 6ꢀ
membered ring isomers were never observed, showing that the
reaction is regioselective towards the dihydroisoquinoline
product 10a. The same results were obtained when reactions
were carried out at 60ºC in TFE, under air or N2 atmospheres.
However, when reactions were carried out at room temperaꢀ
ture, a mixture of 10a and 10b was obtained (Table 1, entries 6
and 7). Therefore, contrary to the reaction with 1ꢀethynylꢀ4ꢀ
nitrobenzene, for the reaction with phenylacetylene the therꢀ
modynamic product is 10a, while 10b is the kinetic product.
Finally, the reactivity of the arylꢀCo(III) compounds with an
internal alkyne, 1ꢀphenylꢀ1ꢀpropyne, was also evaluated. At
100ºC in TFE the reaction furnished 1,2ꢀdihydroisoquinoline
product (11a) in good yield (Table 2, entry 1). Variation of
temperature resulted in the formation of both regioisomers,
Entrya [Co] (x mol%) atm. Yieldb (%)
Ratioc (9a:9b)
Co(OAc)2 (20) air
Co(OAc)2 (20) N2
4b-OAc (20) air
4b-OAc (20) N2
4b-OAc (10) air
4b-OAc (10) N2
3b-OAc (20) air
3b-OAc (20) N2
66
5d
1:99
n.d.
1
2
3
4
5
6
7
8
69
1:99
1:99
1:99
n.d.
13d
43
traced
58
1:99
n.d.
7d
a
Reaction conditions: 2b-H (0.077 mmol) and 1ꢀphenylꢀ1ꢀ
propyne (2.0 equiv.) were mixed in TFE with 10ꢀ20mol% of 4b-
OAc and stirred at 100ºC under different atmospheres over 16h. b
c
Products were isolated using silicaꢀgel chromatography. Ratios
were determined by H NMR. Yield determined using 1,3,5ꢀ
trimethoxybenzene as internal standard.
1
d
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
Under an inert atmosphere the reaction was significantly reꢀ In the ‘acetylide pathway’ (Figure 5), acetateꢀassisted deproꢀ
1
2
3
4
5
6
7
8
tarded when starting from both Co(II) precursors or the offꢀ
cycle intermediate 4b-OAc (Table 3, entries 2, 4 and 8), sugꢀ
gesting that an external oxidant is needed to regenerate the
catalyst.
tonation of the alkyne occurs via 4b-alkyne-2OAc (ꢀG = 17.5
kcal molꢀ1), which presents a hydrogen bond between the
alkyne and one acetate anion. Subsequent proton transfer from
the alkyne to the acetate anion and loss of acetic acid results in
a low spin Co(III) acetylide intermediate, 4b-alkynyl-OAc
(ꢀG = 3.7 kcal/mol), through a 20.2 kcal molꢀ1 transition state.
Then, reductive elimination from 4b-alkynyl-OAc produces
[(2bꢀCCPh)CoI](OAc) (14), which is computed to be most
stable as a triplet state by 19.3 kcal molꢀ1. The most accessible
CꢀC coupling pathway then proceeds from the singlet form of
4b-alkynyl-OAc, which after a spinꢀcrossing to the S=1 state
overcomes an overall barrier of 19.1 kcal molꢀ1. Thereafter, a
Coꢀassisted intramolecular cyclization furnishes
Next, the annulation reaction with internal alkynes was tested
in a catalytic fashion. At 80ºC, 1ꢀphenylꢀ1ꢀpropyne could be
coupled to 2b-H in good yields under air using 20 mol% of
Co(OAc)2 or the proposed offꢀcycle intermediates 3b-OAc
and 4b-OAc, obtaining similar yields and regioselectivities
(Table 4, entries 1, 3 and 5). As with the terminal alkynes,
under an inert atmosphere the reaction did not proceed starting
from 3b-OAc or Co(OAc)2 (Table 4, entries 2 and 6) and a
14% yield was obtained starting from 4b-OAc (Table 4, entry
4), highlighting the necessity for air to regenerate the required
+3 oxidation state of Co for catalytic turnꢀover.
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
Co-C
Co-C
2.07 Å
2.39 Å
Table 4. Study of off-cycle intermediates in catalytic reac-
tions starting from 2b-H and 1-phenyl-1-propyne.
20.6
19.1
18.5
B
4b-alkyne-OAc
A
Entrya [Co] (x mol%) atm. Yieldb (%) Ratioc (11a:11aa)
Co(OAc)2 (20) air
Co(OAc)2 (20) N2
67
traced
64
5:1
n.d.
5:1
5:1
5:1
n.d.
1
2
3
4
5
6
4b-OAc (20)
4b-OAc (20)
3b-OAc (20)
3b-OAc (20)
air
N2
air
N2
0.0
14d
4b-OAc
66
traced
Figure 4. Gibbs energy profile of the formation of 4b-alkyne-
OAc. Relative Gibbs energy values are given in kcal·mol−1 and
selected bond distances in Å. Hydrogens not involved in reactivity
are omitted for clarity (N:blue, O:red, Co:pink, C:grey, H:white).
aReaction conditions: 2b-H (0.077 mmol) and internal alkyne (2.0
equiv.) were mixed in TFE with 10ꢀ20mol% of 4b-OAc and
b
stirred at 100ºC under different atmospheres over 16h. Products
c
were isolated using silicaꢀgel chromatography. Ratios were deꢀ
29.9
termined by 1H NMR. dYield determined using 1,3,5ꢀ
trimethoxybenzene as internal standard.
O--H
TS2
1.39
Å
Co-C
Co-C
2.07 Å
2.03
Å
C-C
1.96 Å
Mechanistic insights. From the experimental studies we can
conclude that initially Co(OAc)2 coordinates to 2b-H to furꢀ
nish 3b-OAc. The new ligand environment permits facile
oxidation to Co(III) followed by CꢀH cobaltation, which we
propose proceeds through a Concerted Metalation Deprotonaꢀ
tion (CMD) process, furnishing 4b-OAc. Starting from the key
arylꢀCo(III) intermediates, PB86/B3LYP calculations (Figures
4ꢀ6) revealed that reaction with terminal alkynes proceeds
following detachment of the η2ꢀOAc resulting in a monocoorꢀ
dinated acetate (A, ꢀG = 18.5 kcal molꢀ1), followed by a forꢀ
mation of an adduct with the incoming terminal alkyne with
the metal complex (B, ꢀG = 20.6 kcal molꢀ1) and finally coorꢀ
dination of the alkyne to furnish a cationic πꢀcomplex (Figure
4, 4b-alkyne-OAc, ꢀG = 19.1 kcal molꢀ1). At this point the
mechanism can diverge to two different pathways, i.e. the
‘acetylide pathway’ (Scheme 3, upper green highlighted route)
and the ‘migratory insertion pathway’ (Scheme 3, lower grey
highlighted route). In order to fully understand the implication
of our experimental results, DFT studies were applied to disꢀ
tinguish whether the 6ꢀmemberd ring products 9a and 10a
20.2
19.1
19.1
TS1
18.2
TS2
S=0
4b-alkyne-OAc
17.5
4b-alkyne-2OAc
14.5
O--H
S=1
1.77 Å
Co-C
2.12 Å
8.5
4b-alkynyl-2OAc
Co-C
1.85 Å
3.7
S=0
4b-alkynyl-OAc
C-C
2.56 Å
-1.1
S=1
14
Figure 5. Gibbs energy profile of the formation of the acetylide
pathway. Relative Gibbs energy values are given in kcal·mol−1
and selected bond distances in Å. Hydrogens not involved in
reactivity are omitted for clarity (N:blue, O:red, Co:pink, C:grey,
H:white).
were formed via the ‘acetylide pathway’ or the prototypical βꢀ
migratory insertion when using phenylacetylene as alkyne
coupling partner.
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
Acetylide pathway
1
H
N
H
N
O
H
N
2
3
4
5
O
O
O
O
H O
N
N
N
O
O
Co
Co
Co
6
N
N 5
NH
Ph
H NH
Reductive
elimination
-AcOH
N
H
N
H
N
H
Ph
N
N
OAc
6
7
14
(2b-CCPh)CoI(OAc)
4b-alkynyl-OAc
(
4b-alkyne-2OAc
[2b-CoIII(η1-OAc)(HCCPh)](OAc)
10a
10b
[2b-CoIII
η
1-OAc)(CCPh)]
8
9
Cobalt-assisted
Intramolecular
Cyclization
ꢀ
G‡ = 20.2 kcal mol-1
10
11
12
13
14
15
16
H
H
N
H
N
N
OAc
[O]
Oxidation
and
C-Hcobaltation
[O], B-
O
N
O
H
O
N
H Co
H
Co
N
O
O
Co
O
2b-H
O
Co(OAc)2
[O]
O
- BH
N
H
N
H
N
H
B = base
3b-OAc
4b-OAc
(
4b-alkyne-OAc
[2b-CoIII(η2-OAc)(π−HCCPh)]+
(2b-H)CoII(
η
2-OAc)2
[2b-CoIII
η
2-OAc)](OAc)
β
-migratory insertion pathway
17
18
19
20
21
22
23
24
25
26
27
ꢀ
G‡ = 29.4 kcal mol-1
H
OAc
Reductiveelimination
N
β-MigratoryInsertion
O
6
N
NH
N
H
O
Co
Ph
N
Ph
N
H
H
10a
15a
[2b-(HCCPh)-CoIII(η1-OAc)](OAc)
Scheme 3. Proposed mechanistic cycle of Coꢀcatalyzed phenylacetylene annulation through arylꢀCo(III) intermediates starting from
Co(OAc)2 and macrocyclic ligand 2b-H. Highest energy barrier for each pathway indicated. Color code for Co oxidation state: green (I),
pink (II) and orange (III).
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
either the dihydroisoquinoline (6ꢀmembered ring, 10a) or
dihydroisoindoline (5ꢀmembered ring, 10b) product dependꢀ
ing on both the reaction temperature and the electronic propꢀ
C-C
1.98 Å
erties of the alkyne.
In the ‘βꢀmigratory insertion pathway’ (Figure 6), no deproꢀ
29.4
tonation occurs and 4b-alkyne-OAc undergoes βꢀmigratory
TS3
insertion towards the commonly proposed 7ꢀmembered
cyclic intermediate (15a, Scheme 3), with a barrier of (ꢀG‡ =
29.4 kcal molꢀ1). Thereafter, reductive elimination from 15a
furnishes the 6ꢀmembered ring 10a, whereas the 5ꢀ
membered product 10b cannot be obtained through this
19.1
mechanism.
4b-alkyne-OAc
Our experimental observations of the formation of 10a and
10b when 4b-OAc is reacted with phenylacetylene and the
higher energy barrier to form 15a (ꢀꢀG‡ = 9.2 kcal molꢀ1)
indicate that the key intermediate is likely to be an organoꢀ
metallic arylꢀCo(III)ꢀalkynyl species (4b-alkynyl-OAc,
Figure 5 and Scheme 2). Thus, our proposal involves the
deprotonation of the terminal acetylene by an acetate anion
7.1
15a
and the subsequent formation of an organometallic arylꢀ
Co(III)ꢀalkynyl intermediate, 4b-alkynyl-OAc, which furꢀ
nishes the linear intermediate 14 after a reductive elimination
(upper green highlighted route in Scheme 3). Then, it is
proposed that 6ꢀmembered ring (9a and 10a) or 5ꢀmembered
ring (9b and 10b) products are obtained through a nucleoꢀ
philic attack of one of the lateral amines through a Coꢀ
assisted intramolecular cyclization. Taking into account the
DFT calculations, we propose that the mechanism exclusiveꢀ
ly operates through an ‘acetylide pathway’, whereas the
commonly proposed ‘migratory insertion pathway’ is exꢀ
cluded for terminal alkynes.
Figure 6. Gibbs energy profile of the formation of the βꢀ
migratory insertion pathway. Relative Gibbs energy values are
given in kcal·mol−1 and selected bond distances in Å. Hydroꢀ
gens not involved in reactivity are omitted for clarity (N:blue,
O:red, Co:pink, C:grey, H:white).
Furthermore, the fact that no other 6ꢀmembered ring regioiꢀ
somers are detected under any experimental conditions furꢀ
ther supports the ‘acetylide pathway’ operating for terminal
acetylenes.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
Our proposed mechanism also provides potential reasoning Light Source (3.0GeV, 300mA storage ring) beamline B18.
for experimental observations reported by Song,11f, 11k Gloriꢀ
us11h and Ackermann11j for other Co(III)ꢀcatalyzed annulation
protocols displaying different selectivity, but starting from
identical substrates. Furthermore, formation of 4b-alkynyl-
OAc is consistent with both the previously described organꢀ
ometallic Co(III) acetylide complexes23 and the organometalꢀ
lic arylꢀCo(III)ꢀalkynyl species proposed by Zhang,11a
Song11f, 11k and Balaraman24 in alkyne annulation reactions.
Additionally, formation of two regioisomers (10a and 10b)
from a common starting material (14, Scheme 3, upper green
highlighted route) can be explained by a nucleophilic attack
of a lateral amine to one of the C(sp) centres. The kinetically
and thermodynamically favored products depend on the
substituents attached to the phenyl group.25 On the other
hand, when using internal alkynes (eg. 1ꢀphenylꢀ1ꢀpropyne)
the organometallic arylꢀCo(III)ꢀalkynyl intermediate species
are not accessible. However, DFT calculations show that the
migratory insertion pathway is a plausible mechanism (See
Figure S23). Therefore, we propose formation of 11a and
11aa via the βꢀmigratory insertion and reductive elimination
from a 7ꢀmembered ring intermediate analogous to 15a.
Recently, Zhang and coꢀworkers have provided MALDIꢀ
TOF evidence of the formation of these intermediates.26
Lastly data on the 4b-OAc species was collected at SOLEIL
synchrotron (2.75 GeV, 400mA storage ring) at 20 K using a
liquid helium cryostat and Si(220) double crystal monoꢀ
chromator. Data calibration and normalization was carried
out using the Athena software package. Energies were caliꢀ
brated to the first inflection point of Co foil spectra set at
7709.5 eV. A linear preꢀedge function and a quadratic polyꢀ
nomial for the postꢀedge were used for background subtracꢀ
tion and normalization of the edge jump. EXAFS data was
extracted using the AUTOBK algorithm with a spline beꢀ
tween a k of 1 and 15 Åꢀ1 having a Rbkg value of 1.0 Å.
EXAFS analysis was carried out with the Artemis software
program running the IFEFFIT engine and the FEFF6 code.27
Unless otherwise specified k3ꢀweighted data was fit in rꢀ
space using a Hannings window (dk=2) over a kꢀrange of 2
to 12.5 Åꢀ1, and an rꢀrange of 1 to 3 Å. The S02 value was set
to 0.9, and a global ꢀE0 was employed with the initial E0
value set to the inflection point of the rising edge. Single and
multiple scattering paths were fit in terms of a ꢀreff and σ2 as
previously described.28 To assess the goodness of the fits the
Rfactor (%R) was minimized. Overꢀfitting the data was conꢀ
trolled by minimizing the number of adjustable parameters
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
and ensuring that the reduced χ2 (χ2 ) decreases with inꢀ
v
creasing number of adjustable parameters.
CONCLUSIONS
Density Functional Theoretical Calculations. Theoretical
calculations were carried out using the ORCA package.29
Geometry optimizations were carried out using the spinꢀ
unrestricted KohnꢀSham formalism employing a BP86 funcꢀ
tional30 with a def2ꢀTZVP(ꢀf) basis set on the metal, nitroꢀ
gens and oxygens, a SVP basis set on the hydrogens and
def2ꢀSVP on carbons, as well as a def2ꢀTZVP/J auxiliary
basis set on all atoms.31 A dense integration grid (ORCA
Grid 5 = Lebedev 434 points) was used for all atoms. Furꢀ
thermore, dispersion corrections were included using the
Grimme and coworkers DFTꢀD3BJ approach32 and solvent
effects were incorporated using a conductor like screening
model (COSMO) using 2,2,2 trifluoroethanol as solvent (ε =
27). Subsequent frequency calculations were done to evaluꢀ
ate enthalpy and entropy corrections at 298.15 K (Gcorr.) and
ensured that all local minima had only real frequencies while
a single imaginary frequency confirmed the presence of
transition states. Single point calculations on the equilibrium
geometries, including the solvent effects and D3BJ disperꢀ
sion corrections were computed using the B3LYP functional
and def2ꢀTZVPP basis set for all the atoms (EB3LYP). The
density fitting and chain of spheres (RIJCOSX) approximaꢀ
tions33 were employed together with the def2ꢀQZVP/JK
auxiliary basis set.31 The free energy change associated with
moving from a standardꢀstate gas phase pressure of 1 atm to
a standard state gas phase concentration of 0.046 M for
solutes (∆ꢀꢁ/∗) was also included in the final free energies.
The value of ∆ꢀꢁ/∗ at 298.15 K is 0.079 kcal·molꢀ1 for 0.046
M standard state solutes. Then, the final total Gibbs free
energy (G) was given by:
In summary, we have prepared a number of organometallic
arylꢀCo(III) complexes using model macrocyclic arene subꢀ
strates which enable the isolation and complete spectroscopic
characterization of these key intermediates in Coꢀcatalyzed
CꢀH activation protocols. The design of the model arene
substrates 2b-H and 2c-H is key for the stabilization of the
organometallic arylꢀCo(III) in a preferredꢀoctahedral enviꢀ
ronment. Additionally, we provide definitive crystallographꢀ
ic proof of a soughtꢀafter organometallic arylꢀCo(III) interꢀ
mediate proposed in 8ꢀaminoquinoline directed Coꢀcatalyzed
CꢀH activation processes. Reaction of organometallic arylꢀ
Co(III) intermediates has been proven to follow different
reaction pathways depending on the nature of the alkyne
coupling partners. Terminal alkynes furnished 5ꢀ and 6ꢀ
membered ring products depending on both the electronic
properties of the alkynes and the reaction temperature. Clear
evidence arising from the regioselectivity of the annulation
reactions, in combination with a DFT study, indicates that
the ‘acetylide pathway’ is preferred for terminal alkynes, in
contrast to the usually proposed ‘βꢀmigratory insertion pathꢀ
way’. These findings strongly suggest that the ‘acetylide
pathway’ is general for terminal alkynes, representing the
first solid experimental evidence of an organometallic arylꢀ
Co(III)ꢀalkynyl intermediate in alkyne CꢀH annulation reacꢀ
tions. This work constitutes a unique fundamental basis for
the understanding of Coꢀcatalyzed CꢀH activation protocols.
METHODS
XAS Data Acquisition and processing. Samples were
prepared as solids diluted in boron nitride, loaded into holdꢀ
ers with Kapton windows and stored at liquid nitrogen temꢀ
peratures until run. All data was collected in transmission
mode. Complexes 3c-Br and 3c-OAc were run under vacuꢀ
um at 77 K using a liquid nitrogen finger dewar available
from the XAFS beamline at Elettra Sincrotrone Trieste (2.0
GeV, 300 mA storage ring) equipped with a Si(111) double
crystal monochromator. Data on the 4c-OAc complex was
collected at 77K using a liquid nitrogen cryostat and a
Si(111) double crystal monochromator available at Diamond
G = EB3LYP + Gcorr. + ∆ꢀꢁ/∗ (Equation 1)
Representative C-H cobaltation of 2b-H and 2c-H. In a 10
mL vial, 2b-H or 2c-H (0.41 mmol) and Co(OAc)2 (0.41
mmol) were mixed in TFE (2.5 mL). The vial was then
o
sealed with a septum and the mixture was heated at 100 C
for 36h. Thereafter, the solvent was removed, and the crude
was dissolved in CHCl3 and layered with pentane. After 24h
at 4oC, the resulting oil was dried under vacuum for 6h,
giving the product as a greyꢀred foam (58ꢀ70%).
ACS Paragon Plus Environment
8

Page 9 of 11
Journal of the American Chemical Society
with experimental setup. XRD (Xavi Fontrodona) and HRMS
(Dr. Laura Gómez) data was collected at STRꢀUdG.
Representative stoichiometric reaction of 4b-OAc with
1
2
3
4
5
6
7
8
alkynes. In a 2 mL vial, 4b-OAc (40mg, 0.077 mmol) and
alkyne (0.077 mmol) were mixed in TFE (1.5 mL). The vial
was then sealed with a septum and the mixture was stirred
for 16h at different temperatures. Then, after removal of
TFE, NH4OH (2 mL) was added and the solution was exꢀ
tracted using CH2Cl2 (2x5mL). Products were purified by
column chromatography on silica gel (CH2Cl2, then
CH2Cl2/MeOH 95:5) and characterized by NMR techniques
and HRMS (see Supporting Information for full details).
REFERENCES
(1) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K. Angew. Chem. Int.
Ed. 2012, 51, 8960.
(2) Chen, X.; Engle, K. M.; Wang, D.ꢀH.; Yu, J.ꢀQ. Angew. Chem.
Int. Ed. 2009, 48, 5094.
(3) (a) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. Rev. 2004,
104, 6217. (b) Ambhaikar, N. In C-H Bond Activation in Organic
Synthesis; CRC Press: Boca Raton, FL, 2015, p 145.
9
Representative catalytic reaction of 4b-OAc with alkynes.
In a 2 mL vial, 2b-H (0.077 mmol) and alkyne (0.154
mmol) were mixed in TFE (1.5 mL) with 10ꢀ20 mol% of 4b-
OAc. The vial was then sealed with a septum and the mixꢀ
ture was stirred for 16h at 80ºC. After removal of the solꢀ
vent, NH4OH (2 mL) was added and the solution was exꢀ
tracted using CH2Cl2 (2x5mL). Then, products were purified
by column chromatography on silica gel (CH2Cl2, then
CH2Cl2/MeOH 95:5) and characterized by NMR techniques.
1,3,5ꢀtrimethoxybenzene was used as internal standard in
selected cases (see Supporting Information for full details).
(4) (a) Guo, X.ꢀX.; Gu, D.ꢀW.; Wu, Z.; Zhang, W. Chem. Rev. 2015,
115, 1622. (b) Ahmad, N. In C-H Bond Activation in Organic
Synthesis; CRC Press: Boca Raton, FL, 2015, p 175.
(5) (a) Moselage, M.; Li, J.; Ackermann, L. ACS Catal. 2016, 6,
498. (b) Wei, D.; Zhu, X.; Niu, J.ꢀL.; Song, M.ꢀP. ChemCatChem
2016, 8, 1242. (c) Gao, K.; Yoshikai, N. Acc. Chem.
Res. 2014, 47, 1208.
(6) (a) Tasker, S. Z.; Standley, E. A.; Jamison, T. F. Nature 2014,
509, 299. (b) Williams, A. In C-H Bond Activation in Organic
Synthesis; CRC Press: Boca Raton, FL, 2015, p 113.
(7) Gensch, T.; Klauck, F. J. R.; Glorius, F. Angew. Chem. Int. Ed.
2016, 55, 11287.
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
(8) (a) Colby, D. A.; Tsai, A. S.; Bergman, R. G.; Ellman, J. A. Acc.
Chem. Res. 2012, 45, 814. (b) Song, G.; Wang, F.; Li, X. Chem.
Soc. Rev. 2012, 41, 3651. (c) Satoh, T.; Miura, M. Chem. Eur. J.
2010, 16, 11212.
ASSOCIATED CONTENT
Supporting Information
(9) Yamakawa, T.; Yoshikai, N. Org. Lett. 2013, 15, 196.
(10) (a) Wu, X.; Yang, K.; Zhao, Y.; Sun, H.; Li, G.; Ge, H. Nat.
Commun. 2015, 6, 6462. (b) Grigorjeva, L.; Daugulis, O. Org. Lett.
2014, 16, 4684. (c) Grigorjeva, L.; Daugulis, O. Org. Lett. 2014, 16,
4688. (d) Hummel, J. R.; Ellman, J. A. J. Am. Chem. Soc. 2015,
137, 490. (e) Zhao, D.; Kim, J. H.; Stegemann, L.; Strassert, C. A.;
Glorius, F. Angew. Chem. Int. Ed. 2015, 54, 4508. (f) Gandeepan,
P.; Rajamalli, P.; Cheng, C.ꢀH. Angew. Chem. Int. Ed. 2016, 13,
4308. (g) Li, J.; Tang, M.; Zang, L.; Zhang, X.; Zhang, Z.;
Ackermann, L. Org. Lett. 2016, 18, 2742. (h) Li, L.; Wang, H.; Yu,
S.; Yang, X.; Li, X. Org. Lett. 2016, 18, 3662. (i) Ma, W.;
Ackermann, L. ACS Catal. 2015, 5, 2822. (j) Wang, H.; Lorion, M.
M.; Ackermann, L. Angew. Chem. Int. Ed. 2016, 55, 10386.
(11) (a) Zhang, J.; Chen, H.; Lin, C.; Liu, Z.; Wang, C.; Zhang, Y.
J. Am. Chem. Soc. 2015, 137, 12990. (b) Grigorjeva, L.; Daugulis,
O. Angew. Chem. Int. Ed. 2014, 53, 10209. (c) Nguyen, T. T.;
Grigorjeva, L.; Daugulis, O. ACS Catal. 2016, 6, 551. (d) Planas,
O.; Whiteoak, C. J.; Company, A.; Ribas, X. Adv. Synth. Catal.
2015, 357, 4003. (e) Kalsi, D.; Sundararaju, B. Org. Lett. 2015, 17,
6118. (f) Zhang, L.ꢀB.; Hao, X.ꢀQ.; Liu, Z.ꢀJ.; Zheng, X.ꢀX.; Zhang,
S.ꢀK.; Niu, J.ꢀL.; Song, M.ꢀP. Angew. Chem. Int. Ed. 2015, 54,
10012. (g) Ikemoto, H.; Yoshino, T.; Sakata, K.; Matsunaga, S.;
Kanai, M. J. Am. Chem. Soc. 2014, 136, 5424. (h) Lu, Q.; Vásquezꢀ
Céspedes, S.; Gensch, T.; Glorius, F. ACS Catal. 2016, 6, 2352. (i)
Kim, J. H.; Greßies, S.; Glorius, F. Angew. Chem. Int. Ed. 2016, 55,
5577. (j) Mei, R.; Wang, H.; Warratz, S.; Macgregor, S. A.;
Ackermann, L. Chem. Eur. J. 2016, 22, 6759. (k) Hao, X.ꢀQ.; Du,
C.; Zhu, X.; Li, P.ꢀX.; Zhang, J.ꢀH.; Niu, J.ꢀL.; Song, M.ꢀP. Org.
Lett. 2016, 18, 3610.
General considerations, experimental details for the preparation
and characterization of ligands 2b-H and 2c-H, and compounds
4b-OAc, 4c-Oac, 4b-CH3CN, 4c-CH3CN, 5c-Cl, 7 and 8;
optimization details; experimental details of stoichiometric
reactions of 4b-OAc with alkynes; experimental details of cataꢀ
lytic reactions of 4b-OAc with alkynes; XYZ coordinates for
DFT structures; full XAS analysis; full characterization of comꢀ
pounds; XꢀRay parameters and original NMR spectra (PDF).
Crystallographic data for compounds 3b-OAc (CCDCꢀ
1493341), 3c-Br (CCDCꢀ1493342), 4b-CH3CN (CCDCꢀ
1493343), 4c-CH3CN (CCDCꢀ1493344), 5c-Cl (CCDCꢀ
1493345) and arylꢀCo(III) species using nitroꢀaminoquinoline
model substrate 8 (CCDCꢀ1493346) can be obtained free of
charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*xavi.ribas@udg.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
(12) (a) Robitzer, M.; Bouamaïed, I.; Sirlin, C.; Chase, P. A.; van
Koten, G.; Pfeffer, M. Organometallics 2005, 24, 1756.
(b) Meneghetti, Mario R.; Grellier, M.; Pfeffer, M.; Cian, André D.;
Fischer, J. Eur. J. Inorg. Chem. 2000, 2000, 1539.
(13) (a) Harvey, J. D.; Ziegler, C. J. Chem. Commun. 2004, 14,
1666. (b) Çetin, A.; Sripothongnak, S.; Kawa, M.; Durfee, W. S.;
Ziegler, C. J. Chem. Commun. 2007, 41, 4289.
(14) Wang, Y.ꢀC.; Chen, J.ꢀH.; Wang, S.ꢀS.; Tung, J.ꢀY. Inorg.
Chem. 2013, 52, 10711.
(15) Maity, S.; Kancherla, R.; Dhawa, U.; Hoqtungue, E.;
Pimparkar, S.; Maiti, D. ACS Catal. 2016, 6, 5493.
We acknowledge financial support from the European Research
Council for the Starting Grant Project ERCꢀ2011ꢀStGꢀ277801 to
X.R. and from MINECO of Spain for project CTQ2013ꢀ43012ꢀ
P to A.C. and X.R., CTQ2014ꢀ52525ꢀP to J.M.L and STQ2015ꢀ
64436ꢀP to T.P. We also thank Generalitat de Catalunya for
projects 2014SGR862 and 2014SGR931. We thank the MECD
of Spain for a FPU PhD grant to O.P. and MINECO for a
Ramón y Cajal contract to A.C. X.R. also thanks ICREA for an
Academia award. In addition, the research leading to these
results has received funding from the European Community’s
Seventh Framework Programme (FP7/2007ꢀ2013) under grant
agreement nº 312284. XAS data was collected at Elettra and
SOLEIL synchrotron facilities. We thank Dr. Luca Olivi from
Elettra and Dr. Landrot Gautier from SOLEIL for their help
(16) Ribas, X.; Calle, C.; Poater, A.; Casitas, A.; Gómez, L.; Xifra,
R. l.; Parella, T.; BenetꢀBuchholz, J.; Schweiger, A.; Mitrikas, G.;
Solà, M.; Llobet, A.; Stack, T. D. P. J. Am. Chem. Soc. 2010, 132,
12299.
(17) (a) Casitas, A.; King, A. E.; Parella, T.; Costas, M.; Stahl, S. S.;
Ribas, X. Chem. Sci. 2010, 1, 326. (b) Huffman, L. M.; Casitas, A.;
ACS Paragon Plus Environment
9

Journal of the American Chemical Society
Page 10 of 11
Font, M.; Canta, M.; Costas, M.; Ribas, X.; Stahl, S. S. Chem. Eur.
J. 2011, 17, 10643. (c) King, A. E.; Huffman, L. M.; Casitas, A.;
Costas, M.; Ribas, X.; Stahl, S. S. J. Am. Chem. Soc. 2010, 132,
12068.
1
2
3
4
5
6
7
8
(18) (a) Zhou, W.; Schultz, J. W.; Rath, N. P.; Mirica, L. M. J. Am.
Chem. Soc. 2015, 137, 7604. (b) Zhou, W.; Zheng, S.; Schultz, J.
W.; Rath, N. P.; Mirica, L. M. J. Am. Chem. Soc. 2016, 138, 5777.
(c) Zhou, W.; Rath, N. P.; Mirica, L. M. M. Dalton Trans. 2016, 45,
8693. (d) Zhou, W.; Watson, M. B.; Zheng, S.; Rath, N. P.; Mirica,
L. M. M. Dalton Trans. 2016, in press, doi: 10.1039/C6DT02185A.
(19) Goswami, M.; Lyaskovskyy, V.; Domingos, S. R.; Buma, W.
J.; Woutersen, S.; Troeppner, O.; IvanovićꢀBurmazović, I.; Lu, H.;
Cui, X.; Zhang, X. P.; Reijerse, E. J.; DeBeer, S.; van Schooneveld,
M. M.; Pfaff, F. F.; Ray, K.; de Bruin, B. J. Am. Chem. Soc. 2015,
137, 5468.
(20) For examples of reactions where the use of fluorinated alcohols
has proved highly beneficial see: (a) Shuklov, I. A.; Dubrovina, N.
V.; Börner, A. Synthesis 2007, 19, 2925. (b) WencelꢀDelord, J.;
Colobert, F. Org. Chem. Front. 2016, 3, 394.
(21) During the preparation of this manuscript, Maiti and coꢀ
workers reported a related XRD structure, see ref. 15.
(22) Whiteoak, C. J.; Planas, O.; Company, A.; Ribas, X. Adv.
Synth. Catal. 2016, 358, 1679.
(23) Cummins, D.; McKenzie, E. D.; Segnitz, A. In J. Organomet.
Chem. 1975, 87, C19.
(24) Landge, V. G.; Jaiswal, G.; Balaraman, E. Org. Lett. 2016, 18,
812.
(25) Rovira, M.; Font, M.; AcuñaꢀParés, F.; Parella, T.; Luis, J. M.;
LloretꢀFillol, J.; Ribas, X. Chem. Eur. J. 2014, 20, 10005.
(26) Yu, W., Zhang, W., Liu, Z. & Zhang, Y. Chem. Commun.
2016, 52, 6837.
(27) (a) Ravel, B.; Newville, M. J. Synchrotron Rad. 2005, 12, 537.
(b) Newville, M. J. Synchrotron Rad. 2001, 8, 96. (c) Rehr, J. J.;
Albers, R. C. Rev. Mod. Phys. 2000, 72, 621.
(28) (a) MartinꢀDiaconescu, V.; Bellucci, M.; Musiani, F.; Ciurli,
S.; Maroney, M. J. J. Biol. Inorg. Chem. 2012, 17, 353. (b) Zambelꢀ
li, B.; Berardi, A.; MartinꢀDiaconescu, V.; Mazzei, L., Musiani, F.,
Maroney, M. J.; Ciurli, S. J. Biol. Inorg. Chem. 2014, 19, 319.
(29) Becke, A. D. Phys. Rev. A. Gen. Phys. 1988, 38, 3098.
(30) Neese, F. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2,
73.
(31) (a) Schäfer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
2571. (b) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005,
7, 3297.
(32) (a) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem.
Phys. 2010, 132, 154104. (b) Grimme, S.; Ehrlich, S.; Goerigk, L. J.
Comput. Chem. 2011, 32, 1456.
(33) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. J. Chem.
Phys. 2009, 356, 98.
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
10

Page 11 of 11
Journal of the American Chemical Society
1
2
3
TOC
4
5
R
2
= Ph
6
N
H NH
6
7
8
9
H
N
R2
N
Csp2-H
activation
N
X
X
Regioselective
C-C/C-N
coupling
R2
II
CoIII
Co
air
H
N
H
N H
+
-H
N
N
H
5
N
NH
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
R2
N
R
2
= 4-NO2Ph
Fullycharacterized
aryl-Co(III)
intermediates
ACS Paragon Plus Environment
11