Iridium-catalyzed C-H borylation of heterocycles using an overlooked 1,10-phenanthroline ligand: Reinventing the catalytic activity by understanding the solvent-assisted neutral to cationic switch

Article abstract of DOI:10.1021/om500420d

The preformed catalyst [Ir(Cl)(COD)(1,10-phenanthroline)] (2; COD = cyclooctadiene) was found to be highly effective in a model reaction for the borylation of N-Boc-indole at the 3-position with B₂pin₂ (pin = pinacolato) as the borylating agent to give consistently 99% yield with 0.5 mol % catalyst loading. The corresponding in situ formed catalyst from [Ir(Cl)(COD)]₂ and 1,10-phenanthroline provided very inconsistent results for the same reaction (0-94% conversion). We propose this to be due to the competing formation of a catalytically inactive cationic complex, [Ir(COD)(1,10-phenanthroline)]⁺Cl^- (1), in a noncoordinating solvent such as octane. Complexes 1 and 2 were characterized using solid-state NMR (¹³C and ³⁵Cl) in conjunction with XPS to be cationic and neutral, respectively. The X-ray crystal structure of a pentavalent neutral Ir complex, [Ir(Cl)(COD)(2,2′-bipyridine)] (3), was also obtained for comparison purposes. Using catalyst 2, the total synthesis of Meridianin G was accomplished in 87% overall isolated yield in a one-pot, three-step process.

Full text of DOI:10.1021/om500420d

Article
pubs.acs.org/Organometallics
Iridium-Catalyzed C−H Borylation of Heterocycles Using an
Overlooked 1,10-Phenanthroline Ligand: Reinventing the Catalytic
Activity by Understanding the Solvent-Assisted Neutral to Cationic
Switch
Carin C. C. Johansson Seechurn,† Vilvanathan Sivakumar,‡ Deepak Satoskar,‡ and Thomas J. Colacot*,§
†Johnson Matthey Catalysis & Chiral Technologies, Orchard Road, Royston SG8 5HE, U.K.
‡Johnson Matthey Catalysis & Chiral Technologies, Plot no. 6, MIDC Industrial Estate, Taloja, Maharashtra 410208, India
§Johnson Matthey Catalysis & Chiral Technologies, 2001 Nolte Drive, West Deptford, New Jersey 08066, United States
S
* Supporting Information
ABSTRACT: The preformed catalyst [Ir(Cl)(COD)(1,10-phenanthroline)] (2; COD = cyclooctadiene) was found to be highly
effective in a model reaction for the borylation of N-Boc-indole at the 3-position with B2pin2 (pin = pinacolato) as the borylating
agent to give consistently 99% yield with 0.5 mol % catalyst loading. The corresponding in situ formed catalyst from
[Ir(Cl)(COD)]2 and 1,10-phenanthroline provided very inconsistent results for the same reaction (0−94% conversion). We
propose this to be due to the competing formation of a catalytically inactive cationic complex, [Ir(COD)(1,10-
phenanthroline)]+Cl− (1), in a noncoordinating solvent such as octane. Complexes 1 and 2 were characterized using solid-
state NMR (13C and 35Cl) in conjunction with XPS to be cationic and neutral, respectively. The X-ray crystal structure of a
pentavalent neutral Ir complex, [Ir(Cl)(COD)(2,2′-bipyridine)] (3), was also obtained for comparison purposes. Using catalyst
2, the total synthesis of Meridianin G was accomplished in 87% overall isolated yield in a one-pot, three-step process.
it requires additional prefunctionalization steps, as the common
starting materials are typically bromo or iodo arenes.5
INTRODUCTION
■
In recent years, Suzuki−Miyaura (SM) coupling has become
the most prominent Pd-catalyzed cross-coupling reaction not
only in academia but also in the fine chemical, agrochemical,
electronics, and pharmaceutical industries, on the basis of a
recent literature search.1 This is mainly due to the fact that SM
coupling utilizes environmentally benign, air- and moisture-
stable organoboron reagents as coupling partners as well as the
fact that it has extremely high functional group tolerance.2
Therefore, the development of economically viable and
operationally simple processes for the synthesis of organoboron
reagents is highly desirable. The iridium-catalyzed direct C−H
borylation of arenes3 has emerged as an atom-economical green
chemical process, as it involves direct functionalization of C−H
bonds, as opposed to the classical methods which utilize either
Grignard- or organolithium-mediated processes.4 The Pd-
catalyzed Miyaura borylation is another catalytic process, but
The seminal reports of Ir-catalyzed processes were
independently disclosed by the research groups of Maleczka/
Smith III6 and Hartwig/Ishiyama.7 Maleczka/Smith III
employed phosphines as supporting ligands, while Hartwig/
Ishiyama used bipyridyl-type ligands. The regioselectivity of Ir-
catalyzed borylations is generally controlled by steric
interactions,8 thereby providing access to products which
would not have been possible through conventional aromatic
electrophilic substitution. The functional group tolerance
coupled with the complementary regioselectivity of Ir-catalyzed
C−H borylation has led to the development of one-pot
methods to convert these boronate esters to other useful
products such as boronic acids and trifluoroborates,9 1,10-
Received: April 23, 2014
Published: June 18, 2014
© 2014 American Chemical Society
3514
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
phenanthrolinols,10 aryl halides,11 aryl nitriles,12 and anilines.13
Ir-catalyzed C−H borylation and its subsequent functionaliza-
tion has been successfully applied to the total synthesis of
several complex natural products14 and organic materials.15
Even though the borylation methodologies reported so far
have found widespread use in the academic community, their
sensitivity to the order of addition of the reagents and the batch
to batch reproducibility dependence on the Ir precursors16
cause great concern when the process is scaled up, especially in
an industrial setup. During the past decade our group has been
successful in demonstrating the superiority of the preformed Pd
catalysts over in situ formed catalysts for various cross-coupling
reactions, where well-defined precatalysts have generated the
active catalytic species in a controlled manner.17 These readily
available preformed catalysts are now being used in multikilo-
gram quantities in fine chemical and pharmaceutical pro-
cesses.18 As a continuation of this theme, we were interested in
studying the effect of preformed vs in situ formed Ir catalysts
for the direct C−H borylation of arenes. We were particularly
interested in developing a robust and operationally simple
method for C−H borylation of nitrogen-containing hetero-
cycles, using relatively cheap ligands with low Ir loading. N-
Boc-indole was chosen as a model system for this study.
Scheme 1 provides the recently reported Ir-catalyzed
methods to make 3-borylated indoles. Hartwig and co-workers
reported the use of N-TIPS-indole and B2pin2 with in situ
formation of the catalyst from [Ir(Cl)(COD)]2 (COD =
cyclooctadiene) and dtbpy (4,4′-di-tert-butyl-2,2′-bipyridine) to
make the product in 83% yield (A).19 Maleczka, Smith, and co-
workers developed a process for the 3-borylation of N-Boc-
indole with HBpin using the [Ir(OMe)(COD)]2/dtbpy system
with 65% yield of the product (B).20 While they are significant
achievements, these methods employ relatively high loadings of
Ir catalyst and ligand, and the yields in some cases were
moderate. In addition, in the case of TIPS-indole, a 2-fold
excess of the heterocycle is required, which is not economical.
More recently, Krska, Maleczka, and Smith described a traceless
directing group approach for 3-borylation of indole (C).21
However, despite the lower loading of the Ir precursor, an
excess of relatively expensive Me4-1,10-phenanthroline ligand is
required. Depending on the desired subsequent reactions to be
carried out on the borylated product, the reaction in Scheme
1C may pose additional challenges due to the unprotected N−
H functionality.
In summary, currently the most effective Ir catalysts for C−H
borylation reactions are derived from readily available Ir
precursors such as [IrCl(COD)]2 and [Ir(OMe)(COD)]2 in
conjunction with electron-rich bipyridyl type ligands, in
particular dtbpy and Me4-1,10-phenanthroline.22 We were
intrigued, however, by one particular case reported by Hartwig,
where dtbpy and the simple 1,10-phenanthroline ligand
performed comparably in the borylation of benzylic C−H
bonds.23
A simple cost comparison may show the financial benefits of
using a catalyst formed from [Ir(Cl)(COD)]2 and 1,10-
phenanthroline. The dtbpy and the Me4-1,10-phenanthroline
ligands are respectively 5 and 13 times more expensive than
1,10-phenanthroline for research quantities (Table 1), while
[Ir(OMe)(COD)]2 is 1.5 times more expensive than the
[Ir(Cl)(COD)]2 precursor.
a
Scheme 1. Reported 3-Borylations of Indole19−21
Table 1. Cost Comparisons of Reagents
reagent
cost for 5 g (£)
rel cost
1,10-phenanthroline
9.80
50.00
1
5
b
dtbpy
Me4-1,10-phenanthroline
[Ir(Cl)(COD)]2
123.00
471.00
13
1
[Ir(OMe)(COD)]2
676.00
1.5
a
b
Alfa Aesar UK prices as of November 2013. Sigma-Aldrich UK price
as of November 2013.
With the aim of studying the effect of preformed vs in situ
formed catalysts in Ir-catalyzed C−H borylation reactions, we
initially focused our attention on preparing preformed catalysts
containing the cheaper 1,10-phenanthroline ligand for the
direct borylation of N-Boc-indole as a model system. We
developed a robust, operationally simple process with relatively
low Ir loading and later extended the study to a few other
nitrogen-containing heterocycles.
RESULTS AND DISCUSSION
■
Catalyst Synthesis and Characterization. The known
cationic complex [Ir(COD)(phen)]+Cl− (1; phen = 1,10-
phenanthroline; Scheme 2) was synthesized according to
literature procedures (Scheme 2).24 This complex has
previously only been investigated for oxidative addition
reactions with a number of reagents: e.g., MeCN and
MeOH.24 During the characterization of 1, it was observed
that this green solid gives a purple solution on dissolution in
CDCl3.
As was observed in the literature procedure,24 during the
preparation of this complex in CH2Cl2, initially a purple
3515
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
Scheme 2. Preparation of Ir Complex 1
Table 2. Solvent-Dependent Preparation of 1 and 2
entry
solvent
complex
appearance
purple solution
time before color change (min)
structure
1
2
3
4
5
6
THF
2
1
1
1
1
1
N/A
10
0
neutral
cationic
cationic
cationic
cationic
cationic
CH2Cl2
hexane
toluene
NMP
purple solution to green solid
green solid
green solid
0
purple solution to green solid
purple solution to green solid
75
10
CH3CN
reaction mixture, mostly in the solution phase, was obtained,
from which a green solid precipitated after 10 min of stirring.
The purple solution was proposed24 to be the neutral
pentavalent complex [Ir(Cl)(COD)(phen)] (2). However, to
the best of our knowledge this neutral Ir complex has not
previously been isolated or characterized. With the aim of
isolating and characterizing this proposed species, a solvent
screen was carried out (Table 2). It was revealed that, by
reacting [Ir(Cl)(COD)]2 with 1,10-phenanthroline in THF,
purple complex 2 was isolated as a stable product (entry 1), in
contrast to the reported reaction in CH2Cl2, which produced a
green product 1 (entry 2). Importantly, in hexane, the typical
solvent of choice for the C−H activated borylation reactions,
the product formed was the green cationic complex 1 (entry 3).
A similar observation (instantaneous green product formation)
was observed when the reaction was carried out in toluene
(entry 4). Although both NMP (N-methylpyrrolidone) and
CH3CN produced the green cationic product at different
intervals of time (Table 2, entries 5 and 6) the reaction
mixtures were initially purple, suggesting the conversion of the
species from neutral to cationic. We have not carried out any
detailed concentration studies to understand the neutral to
cationic switch. This study suggests that polar solvents,
preferably those with coordinating ability, tend to favor
formation of the neutral complex, which in some cases
transforms into the thermodynamically stable cationic complex.
The exact mechanism of this solvent-assisted switch is yet to be
determined.
THF), making solution-state characterization extremely
challenging.
The solid-state 13C NMR spectrum, on the other hand,
showed a clear structural difference between green complex 1
and purple complex 2 (Figure 1).
Solid-state 35Cl NMR experiments supported the ionic
nature of complex 1 and the covalent structure of complex 2
(Figure 2). In the 35Cl spectra, a signal is clearly observed in the
case of cationic [Ir(COD)(phen)]+Cl− (1), whereas no signals
can be seen in the case of covalent [Ir(Cl)(COD)(phen)]
As mentioned previously, interestingly the green cationic
complex 1, on dissolution in CDCl3, gave a purple solution.
1
Therefore, complexes 1 and 2 displayed identical H NMR
spectra when they were recorded in this solvent, consistent with
the proposed structure 2 (Table 2). However, with time,
precipitation of a green solid occurs from the purple solution.
In addition, complex 1 was found to be insoluble, or
decomposed, in most other solvents (H2O, alcohols, DMSO,
Figure 1. Solid-state 13C NMR spectra of 1 (cationic) and 2 (neutral),
showing the differences. Spinning side bands are marked with asterisks.
Peaks corresponding to solvent molecules (THF) are marked with the
letter S.
3516
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
borylating agent to produce the expected product 3-Bpin-N-
Boc-indole. The results are summarized in Table 3.
Table 3. Evaluation of Preformed Catalysts 1 and 2 in
Borylation of N-Boc-indole
[Ir]
B
solvent
(0.17 M)
NMR conversn
a
entry catalyst
structure
source
(%)
1
2
3
4
5
6
7
8
1
2
1
2
1
2
2
2
cationic
neutral
cationic
neutral
cationic
neutral
neutral
neutral
B2pin2
B2pin2
B2pin2
B2pin2
B2pin2
B2pin2
HBpin
HBpin
octane
0
octane
100 (99)
Figure 2. Normalized 35Cl solid-state static NMR spectra of 1 and 2,
2-MeTHF
2-MeTHF
C2H4Cl2
C2H4Cl2
octane
0
alongside a simulated 35Cl line shape.
94
b
0
(2).25 This behavior is consistent with what is typically
observed for ionic and covalent chlorine sites, respectively.26
XPS studies further supported 1 being of ionic nature and 2
containing a covalent Ir−Cl bond (see the Supporting
Information).
0
89
c
octane
b
93
a
Isolated yield given in parentheses. Reaction mixture instanta-
neously turned purple (indicating formation of neutral Ir species);
however, no reaction took place. 0.33 M.
c
Therefore, it appears that THF is unique for isolation of the
neutral species, while other nonpolar solvents (hexane, toluene)
prefer the formation of the cationic complex. Polar solvents
such as CH2Cl2, NMP, and CH3CN favor initially the
formation of the neutral species, which eventually changes to
the cationic species with time. This switching of neutral to
cationic in certain solvents was observed for the first time in
this study, which helped us in understanding the poor activity
of the 1,10-phenanthroline ligand. This phenomenon, resulting
in a significant destructive effect on the borylation reaction
when the catalyst is formed in situ, will be discussed later.
Due to the instability of 2 in solution over prolonged periods
of time and the complete insolubility of 1, it was not possible to
grow crystals suitable for single-crystal X-ray diffraction.
Therefore, an analogous bpy complex, [Ir(Cl)(COD)(bpy)]
(3; bpy = 2,2′-bipyridine), was prepared in THF for a single-
crystal X-ray determination. The crystal structure of the purple
complex 3 revealed a pentavalent neutral Ir complex, further
supporting the existence of such a species (Figure 3).
The cationic complex 1 resulted in no product formation at
all, independent of the solvent used for the reaction (octane, 2-
MeTHF, or C2H4Cl2; Table 3, entries 1, 3 and 5). The reason
for the catalytic inactivity of the ionic complex is not clear at the
moment and is currently being investigated. Notably, 0.5 mol %
of neutral [Ir(Cl)(COD)(1,10-phenanthroline)] (2) provided
the borylated indole in 100% conversion by 1H NMR with 99%
isolated yield, from the reaction carried out in octane (entry 2).
To the best of our knowledge, this is the best conversion with
isolated yield reported for the borylation of the model
substrate, with the lowest Ir loading. A simple filtration of the
reaction mixture on silica gel followed by removal of volatiles
under reduced pressure provided pure 3-Bpin-N-Boc-indole
(4a). No borylation of octane was observed.28
The reaction using 2 in 2-MeTHF provided the product in
94% conversion (Table 3, entry 4), while no product formation
was observed when the reaction was carried out in C2H4Cl2
(entry 6).
HBpin has been described as superior to B2pin2 as a
borylating reagent in many previous cases. It is proposed to
result in faster generation of the catalytically active species,
thereby providing enhanced results for the overall borylation
reaction.29 In this case, however, no beneficial effects were
observed when changing B2pin2 for 2 equiv of HBpin (Table
3). Under standard reaction conditions, using 0.5 mol % of 2,
only 89% conversion was observed (entry 7). A very slight
improvement was achieved by increasing the reaction
concentration to 0.33 M (93%, entry 8). We have not made
any effort to further optimize the reaction.
Figure 3. X-ray crystal structure of neutral [Ir(Cl)(COD)(bpy)]
complex, 3.
Interestingly, the crystallized structure of 3 is that of a slightly
distorted square pyramidal Ir complex. The Ir−Cl bond is
slightly longer (2.611 Å) in comparison to previously reported
covalent Ir−Cl bonds (2.36−2.38 Å).27
Catalytic Activity Studies. With preformed catalysts 1 and
2 in hand, we evaluated their use in the direct C−H borylation
of N-Boc-indole. Either B2pin2 or HBpin could be used as
Subsequently, the performance of preformed [Ir(Cl)(COD)-
(phen)] (2) was compared to that of the current state of the art
in situ formed catalyst systems (Table 4). Both Me4-1,10-
phenanthroline and dtbpy have previously been demonstrated
to outperform the unsubstituted simple 1,10-phenanthroline
and bpy when they are employed in conjunction with
[Ir(Cl)(COD)]2 or [Ir(OMe)(COD)]2.30
3517
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
Both 5-chloro- and 5-Bpin-N-Boc-indole resulted in the
formation of the 3-borylated products 4b,c in excellent yields
(96 and 93%; Table 5, entries 2 and 3), providing useful
functionalities for further regioselective transformations.
Increased Ir loading had to be used in order to form 2-
borylated pyrrole 4d (Table 5, entry 4) in moderate yield. With
a slight excess of B2pin2 with respect to pyrrole, 2,5-bis-
borylated pyrrole 4e (entry 5) was formed in excellent yield
(96%), suggesting that the presence of free N−H functionalities
does not influence the catalysis. Non-nitrogen-containing
heterocycles, such as benzofuran and benzothiophene, did
not undergo any borylation reaction using 2 under standard
reaction conditions. The reaction mixture turned green
instantaneously in both cases, suggesting that even some
substrates are capable of turning covalent 2 into cationic 1 prior
to the desired borylation reaction.
Table 4. In Situ Formed Ir Catalysts
NMR conversn
(%)
entry
[Ir]
ligand
1
2
3
4
[Ir(Cl)(COD)]2
dtbpy
87
94
[Ir(OMe)(COD)]2 dtbpy
[Ir(Cl)(COD)]2 Me4-1,10-phenanthroline
[Ir(OMe)(COD)]2 Me4-1,10-phenanthroline
a
>90
a
>90
a
Side products observed upon 1H NMR analysis of the crude reaction
mixture, proposed to be the result of multiple borylations.
As mentioned previously, when aliphatic solvents are used for
the borylation reaction, in the presence of the [Ir(Cl)-
(COD)]2/1,10-phenanthroline system (in situ) catalytically
inactive cationic species 1 (see Table 2) may form. Our effort
to optimize the reaction conditions for the 3-borylation of N-
Boc-indole under in situ conditions was not successful, as it is
difficult to reliably control the formation of the desired
catalytically active covalent species (Scheme 3). This might
be the reason the cheaper 1,10-phenanthroline ligand was
previously reported to underperform, in comparison to the
more costly Me4-1,10-phenanthroline or dtbpy ligand, by earlier
researchers during the ligand screens and hence was overlooked
in subsequent studies. This hypothesis has been tested in the
current study, where inconsistent results have been obtained
while repeating the in situ catalysis experiment several times,
using [Ir(Cl)(COD)]2/1,10-phenanthroline for the 3-boryla-
tion of N-Boc-indole. Even when the reactions are carried out
under identical conditions (addition of the solvent to a flask
containing [Ir(Cl)(COD)]2, 1,10-phenanthroline, B2pin2, and
N-Boc-indole), conversions vary from 0 to 94%, suggesting the
sensitivity of the in situ conditions. However, no such
inconsistencies were observed when the preformed neutral
complex was used as a catalyst.
In the current study, the in situ formed catalysts were less
active than the preformed catalyst 2 (see Table 3, entry 2).
Both [Ir(Cl)(COD)]2 and [Ir(OMe)(COD)]2 in conjunction
with dtbpy provided the 3-borylated product in conversions of
87 and 94%, respectively (Table 4, entries 1 and 2). Using Me4-
1,10-phenanthroline as ligand gave conversions greater than
90%; however, multiple side products were observed in the H
NMR spectra (entries 3 and 4), suggesting the formation of a
less selective catalyst.
Having identified [Ir(Cl)(COD)(phen)] (2) as the best-
performing precatalyst, we continued to evaluate its use in the
direct borylation of a few other nitrogen-containing hetero-
cycles to establish the generality (Table 5).
1
Table 5. Borylation of Nitrogen-Containing Heterocycles
a
using Neutral [Ir(Cl)(COD)(phen)] (2)
Since there have been some reports on the importance of the
order of addition of reagents in room-temperature boryla-
tions,22 we also investigated the effect in our 80 °C reaction.
The results are summarized in Table 6. Conditions A involved
adding the ligand to a suspension of the Ir precursor, B2pin2,
and N-Boc-indole in octane. Conditions B involved adding the
N-Boc-indole to a suspension of the Ir dimer precursor, ligand,
and B2pin2 in octane.
None of the desired product was formed in the [Ir(Cl)-
(COD)]2/1,10-phenanthroline case, when the ligand was
added last (Table 6, entry 1). However, using [Ir(OMe)-
(COD)]2/1,10-phenanthroline under conditions A did provide
a catalytically active species, and 97% conversion was observed
1
by H NMR (entry 2). Notably, a number of side products
were observed in the 1H NMR spectrum after filtration on silica
gel.25
When the order of addition was reversed, and the aryl
reagent was added last, the in situ [Ir(Cl)(COD)]2/1,10-
phenanthroline system did result in good catalytic activity: 96%
conversion to the product (Table 6, entry 3, reaction carried
out in duplicate). Notably, this is slightly less active than the
preformed 2, which consistently provided 100% conversion to
the product. Using conditions B, the in situ catalyst formed
from [Ir(OMe)(COD)]2 and 1,10-phenanthroline provided
a
Reaction conditions: 1 mol % of [Ir(Cl)(COD)(phen)] (2),
heteroarene (1.0 mmol), B2pin2 (1 equiv), octane (6 mL), 80 °C,
16 h. 0.5 mol % of 2. 3 mol % of 2, 10 mmol of pyrrole, 1 mmol of
B2pin2. 3 mol % of 2, 1.5 equiv of B2pin2.
b
c
d
3518
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
Scheme 3. Reaction between Ir Dimer Precursor and N,N Ligands in Octane
Table 6. Investigation of In Situ Conditions at 0.5 mol % Ir
a b
,
Loading
order of
addition
NMR conversn
(%)
entry
catalyst
[Ir(Cl)
ligand
1
2
3
4
1,10-
phenanthroline
A
A
B
B
0
(COD)]2
c
[Ir(OMe)
(COD)]2
1,10-
phenanthroline
97
Figure 4. Catalytic intermediate proposed by Hartwig.29
[Ir(Cl)
(COD)]2
1,10-
phenanthroline
96
c
[Ir(OMe)
(COD)]2
1,10-
phenanthroline
97
In this study, however, prestirring 2 with B2pin2 for 10 min at
80 °C before adding N-Boc-indole gave only 56% conversion to
the product (Scheme 4).
a
Reaction conditions: N-Boc-indole (1 mmol), B2pin2 (1 mmol), [Ir]
catalyst (0.5 mol %), ligand (0.5 mol %), octane (0.17 M), 80 °C, 16 h.
b
Order of addition: (conditions A) (i) Ir dimer, B2pin2, (ii) N-Boc-
Scheme 4. Stepwise Addition of Reagents using Precatalyst 2
indole, solvent, (iii) ligand; (conditions B) (i) Ir dimer, B2pin2, ligand,
c
(ii) solvent, (iii) N-Boc-indole. Side products observed upon 1H
NMR analysis of the crude reaction mixture.
results comparable to those for the reaction under conditions A
(entry 4).
On the basis of preliminary results, the reason for the order
of addition of reagents and catalysts not affecting the borylation
reaction when the [Ir(OMe)(COD)]2 precursor is used can be
tentatively explained by the fact that the [Ir(OMe)(COD)]2/
1,10-phenanthroline combination does not form a cationic
complex but rather a neutral tetravalent Ir complex where the
ligand is monodentate (see the Supporting Information). The
same observation can also be used to explain why the order of
addition for catalyst systems formed in situ from [IrCl(COD)]2
and dtbpy or Me4-1,10-phenanthroline ligand does not affect
the outcome of the borylation reaction. When catalysts are
prepared from this precursor and dtbpy or Me4-1,10-
phenanthroline, the precatalysts formed in both THF and
CH2Cl2 are identical on the basis of solid-state 13C NMR
studies and are therefore proposed only to exist as neutral
complexes (see the Supporting Information). These results
indicate that the [Ir(Cl)(COD)]2/1,10-phenanthroline combi-
nation is unique in terms of its dependence on the order of
addition. When the ligand is added last, formation of the
desired catalytically active species is inhibited. When the aryl
reagent is added last, the catalytically active species is formed in
order for the borylation reaction to take place. This suggests
that the aryl reagent in some way may hinder the formation of
the catalytically active neutral species. This can be avoided by
using the precatalyst [Ir(Cl)(COD)(phen)] (2), where the
ligand is already bound to Ir.
This suggests that the reaction between 2 and B2pin2 leads to
a detrimental side reaction, producing a boron species that is
unreactive in the subsequent borylation of N-Boc-indole.31 It is
therefore not clear whether or not the preformed complex
[Ir(Cl)(COD)(phen)] (2) follows the same initial reaction
pathway as in the case of the in situ system. Further studies are
currently being undertaken to gain more insight into under-
standing the reaction pathway.
On the basis of these results, the protocol for using
preformed 2 has to involve a one-time addition of Ir catalyst
2, B2pin2, and aryl reagent followed by solvent.
In addition to the above advantages, preformed Ir catalyst 2
is easy to handle and stable to storage for a longer period of
time under inert conditions. Shelf life studies indicate that after
storage under air for 60 days, a slight drop in conversion in a
test reaction from 100 to 95% has been observed, with 0.5 mol
% Ir loading.
Total Synthesis of Meridianin G. A one-pot C−H
borylation followed by Suzuki−Miyaura coupling and a
deprotection methodology was demonstrated using the same
catalytic system to synthesize the natural product meridianin G
in excellent overall yield (Scheme 5).32,33 The three-step
reaction sequence uses only commercially available starting
materials. The 3-borylated indole 4a was generated using the
standard Ir-catalyzed procedure. The volatiles were subse-
quently removed, and the resulting crude residue was used
directly in the subsequent Suzuki coupling reaction, catalyzed
by the preformed Pd complex dtbpfPdCl2 (dtbpf = di-tert-
butylphosphinoferrocene) with trade name Pd-118.34 Under
In previous mechanistic work, it has been suggested that the
Ir dimer reacts with the boron reagent prior to reacting with the
ligand.29 Complex 5, formed from this reaction, is proposed to
be an important intermediate in the catalytic cycle, generating
the catalytically active species by dissociation of the COE ligand
(Figure 4).
3519
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
Scheme 5. One-Pot Total Synthesis of Meridianin G
temperature for 3 h. Subsequently, the solvent was removed in vacuo
to provide the desired product.
the chosen reaction conditions, with Cs2CO3 as a base in
aqueous isopropyl alcohol solvent, the Boc protecting group
was also removed and the desired product meridianin G was
obtained in 87% overall yield, over three reaction steps.
Further application studies using preformed catalyst 2 and
related examples are in progress.
[Ir(COD)(phen)]Cl (1):24 [IrCl(COD)]2 (0.250 g, 0.372 mmol);
1,10-phenanthroline (0.135 g, 0.748 mmol); CH2Cl2 (10 mL). The
product was obtained as a green solid (98% yield). 13C SS-NMR (100
MHz): δ 150.3, 144.7, 140.4, 129.2, 128.6, 124.0, 73.0, 67.1, 35.7, 31.6.
Anal. Calcd for C20H20ClIrN2: C, 46.55; H, 3.91; N, 5.43. Found: C,
46.12; H, 3.85; N, 5.27.
CONCLUSIONS
[Ir(Cl)(COD)(phen)] (2): [IrCl(COD)]2 (0.250 g, 0.372 mmol);
1,10-phenanthroline (0.135 g, 0.748 mmol); THF (10 mL). The
■
[Ir(Cl)(COD)(phen)] (2) was identified as an efficient
precatalyst for the direct 3-borylation of N-Boc-indole and
other nitrogen-containing heterocycles. This system provides a
cost-effective and highly active alternative to the well-
established in situ [Ir(OMe)(COD)]2/dtbpy system. The in
situ protocol using [Ir(Cl)(COD)]2 and 1,10-phenanthroline
was found to give irreproducible results due to the competing
formation of catalytically inactive cationic Ir species [Ir(COD)-
(phen)]+Cl− (1) in aliphatic solvents. The neutral to cationic
switch was established for the first time to understand the
structure−activity relationship of the catalyst and the reaction
conditions.
1
product was obtained as a dark purple solid (99% yield). H NMR
(CDCl3, 400 MHz): δ 8.69 (dd, J 5.6, 1.2, 2H), 8.54 (dd, J 8.0, 1.2,
2H), 7.97 (s, 2H), 7.84 (dd, J 8.0, 5.2, 2H), 3.95 (m, 4H), 2.49−2.47
(m, 4H), 1.80−1.76 (m, 4H). 13C NMR (CDCl3, 100 MHz): δ 147.9,
146.5, 135.0, 131.2, 127.4, 125.3, 59.8, 31.9. 13C SS-NMR (100 MHz):
δ 148.5, 146.4, 138.1, 136.2, 133.0, 131.6, 128.9, 127.9, 125.0, 33.5,
30.3. Anal. Calcd for C20H20ClIrN2: C, 46.55; H, 3.91; N, 5.43. Found:
C, 46.49; H, 3.89; N, 5.39.
[Ir(Cl)(COD)(bpy)] (3): [IrCl(COD)]2 (0.250 g, 0.372 mmol); 2,2′-
bipyridyl (0.117 g, 0.748 mmol); THF (10 mL). The product was
obtained as a purple solid (90% yield). 1H NMR (CDCl3, 400 MHz):
δ 8.32 (4H, m), 8.08 (2H, m), 7.53 (2H, m), 3.80 (4H, m), 2.46 (4H,
m), 1.79 (4H, m). Anal. Calcd for C18H20ClIrN2: C, 43.94; H, 4.10; N,
5.69. Found: C, 43.43; H, 4.21; N, 5.21; CCDC 985451.
EXPERIMENTAL SECTION
■
General Method for Ir-Catalyzed Borylation Reactions. The
Ir catalyst, ligand (if applicable), B2pin2, and heteroaryl species (if a
solid) were weighed and transferred to a Schlenk flask. The flask was
evacuated and refilled with nitrogen five times. Then the heteroaryl
species (if a liquid) was added, immediately followed by the solvent.
The reaction mixture was stirred at 80 °C for 16 h, cooled to room
temperature, and subsequently passed through silica gel to remove
iridium catalyst residue. The silica gel was washed with 2 × 2 mL of
dichloromethane. The combined fractions were evaporated to dryness
using a rotary evaporator to provide the desired borylated compound.
If required, the product was further purified by column chromatog-
raphy.
General Considerations. Unless otherwise stated, all reactions
were carried out under a dry nitrogen atmosphere using standard
Schlenk techniques. Iridium catalyst preparation reactions were carried
out in 50 mL Schlenk tubes with magnetic stirring bars. Catalyst
screening reactions were carried out in individual Schlenk tubes with
magnetic stirring. Anhydrous tetrahydrofuran, dichloromethane, n-
hexane, octane, toluene, and NMP were purchased from commercial
sources in Sure/Seal bottles and used as received. N-Boc-indole and
4,4′-di-tert-butylbipyridine were purchased from Sigma-Aldrich and
used as received. Bis(pinacolato)diboron, 2,2′-bipyridine, 1,10-
phenanthroline, and 3,4,7,8-tetramethyl-1,10-phenanthroline were
purchased from Alfa Aesar and used as received. [Ir(Cl)(COD)]2
and [Ir(OMe)(COD)]2 are available from Johnson Matthey Catalysis
& Chiral Technologies, as are precatalysts 1−3.
3-Bpin-N-Boc-indole (4a):20 N-Boc-indole (202 μL, 1.0 mmol);
B2pin2 (254 mg, 1.0 mmol); 2 (2.6 mg, 0.005 mmol); octane (6 mL).
The product was obtained as a white solid (342 mg, 99%). No further
The identity of known isolated products was confirmed by
1
purification required. H NMR (CDCl3, 400 MHz): δ 8.16 (m, 1H),
1
comparison with reported literature spectroscopic data. Both H and
8.0 (s, 1H), 7.99 (dd, J 6.8, 0.8, 1H), 7.32−7.24 (m, 2H), 1.66 (s, 9H),
1.37 (s, 12H).
13C{1H} NMR spectra were recorded on a 400 MHz spectrometer
1
referenced to the residual H and 13C{1H} signals of the solvents.
3,5-di-Bpin-N-Boc-indole (4b): 5-Bpin-N-Boc-indole (343 mg, 1.0
mmol); B2pin2 (254 mg, 1.0 mmol); 2 (5.2 mg, 0.01 mmol); octane (6
mL). The product was obtained as a white solid (438 mg, 93%). No
further purification required. 1H NMR (CDCl3, 400 MHz): δ 8 40 (s,
1H), 8.17 (d, J 8.4, 1H), 7.99 (s, 1H), 7.76 (dd, J 8.0, 0.8, 1H), 1.65 (s,
9H), 1.39 (s, 12H), 1.38 (s, 12H). 13C NMR (CDCl3, 100 MHz): δ
151.3, 140.3, 137.4, 134.8, 132.7, 131.4, 116.2, 85.9, 85.5, 85.4, 30.1,
26.9, 26.8. mp 204−209 °C. Anal. Calcd for C25H37B2NO6: C, 64.00;
H, 7.95; N, 2.99. Found: C, 63.81; H, 7.97; N, 2.96.
3-Bpin-5-Cl-N-Boc-indole (4c): 5-Cl-N-Boc-indole (251 mg, 1.0
mmol); B2pin2 (254 mg, 1.0 mmol); 2 (5.2 mg, 0.01 mmol); octane (6
mL). The product was obtained as a white solid (361 mg, 96%). No
further purification required. 1H NMR (CDCl3, 400 MHz): δ 8.08 (d,
J 8.8, 1H), 8.01 (s, 1H), 7.95 (d, J 2.0, 1H), 7.26 (dd, J 8.8, 2.0, 1H),
1.65 (s, 9H), 1.37 (s, 12H). 13C NMR (CDCl3, 100 MHz): δ 149.1,
136.1, 134.7, 128.6, 124.4, 122.2, 115.9, 84.3, 28.2, 24.9. mp 179−183
NMR multiplicities are abbreviated as follows: s, singlet; d, doublet; t,
triplet; q, quartet; m, multiplet; br, broad signal. Coupling constants J
are given in Hz.
Solid-state NMR spectra were recorded at a static magnetic field
strength of 9.4 T (υ0(1H) = 400.16 MHz). For 13C, the probe was
tuned to 100.63 MHz and the spectra were referenced to the alanine
CH3 signal at 20.5 ppm. For 35Cl, the probe was tuned to 39.21 MHz
and referenced to KCl.
The radiation for the XPS studies was monochromatized aluminum
Kα radiation with a 650 μm spot size. Charge compensation was
provided by the in-lens electron flood gun at a 2 eV setting and the
“401” unit for “zero energy” argon ions.
General Method for Preparation of Ir Complexes. [IrCl-
(COD)]2 (0.372 mmol) and the bidentate N−N ligand (0.748 mmol)
were weighed and transferred to a 50 mL Schlenk tube. Then 10 mL of
solvent was added and the reaction mixture was stirred at room
3520
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
°C. Anal. Calcd for: C19H25BClNO4: C, 60.42; H, 6.67; N, 3.71.
Found: C, 60.15; H, 6.72; N, 3.69.
(2) (a) Halford, B. Chem. Eng. News 2012, 90, 40. (b) Roughley, S.
D.; Jordan, A. M. J. Med. Chem. 2011, 54, 3451. (c) Colacot, T. J.
Platinum Met. Rev. 2011, 55, 84.
(3) (a) Hartwig, J. F. Acc. Chem. Res. 2012, 45, 864. (b) Hartwig, J. F.
Chem. Soc. Rev. 2011, 40, 1992. (c) Mkhalid, I. A. I.; Barnard, J. H.;
Marder, T. B.; Murphy, J. M.; Hartwig, J. F. Chem. Rev. 2010, 110, 890.
(d) Ishiyama, T.; Miyaura, N. Pure Appl. Chem. 2006, 78, 1369.
(4) Leermann, T.; Leroux, F. R.; Colobert, F. Org. Lett. 2011, 13,
4479.
(5) Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem. 1995, 60,
7508.
(6) Cho, J.-Y.; Tse, M. K.; Holmes, D.; Maleczka, R. E., Jr.; Smith, M.
R., III. Science 2002, 295, 305.
(7) Ishiyama, T.; Takagi, J.; Ishida, K.; Miyaura, N.; Anastasi, N. R.;
Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 390.
(8) Green, A. G.; Liu, P.; Merlic, C. A.; Houk, K. N. J. Am. Chem. Soc.
2014, 136, 4575.
(9) Murphy, J. M.; Tzschucke, C. C.; Hartwig, J. F. Org. Lett. 2007, 9,
757.
(10) Maleczka, R. E., Jr.; Shi, F.; Holmes, D.; Smith, M. R., III J. Am.
Chem. Soc. 2003, 125, 7792.
(11) (a) Fier, P. S.; Luo, J.; Hartwig, J. F. J. Am. Chem. Soc. 2013, 135,
2552. (b) Partridge, B. M.; Hartwig, J. F. Org. Lett. 2013, 15, 140.
(c) Murphy, J. M.; Liao, X.; Hartwig, J. F. J. Am. Chem. Soc. 2007, 129,
15434.
2-Bpin-pyrrole (4d):35 Pyrrole (693 μL, 10 mmol); B2pin2 (254 mg,
1.0 mmol); 2 (15.5 mg, 0.03 mmol); octane (6 mL). Crude product
was purified by flash column chromatography (0−20% MTBE/PE) to
1
give the product as a colorless oil (91 mg, 47%). H NMR (CDCl3,
400 MHz): δ 8.82 (s, 1H), 7.01−6.99 (m, 1H), 6.86−6.85 (m, 1H),
6.31−6.29 (m, 1H), 1.32 (s, 12H).
2,5-di-Bpin-pyrrole (4e):36 Pyrrole (69 μL, 1.0 mmol); B2pin2 (381
mg, 1.5 mmol); 2 (15.5 mg, 0.03 mmol); octane (6 mL). The product
was obtained as a white solid (306 mg, 96%). No further purification
required. 1H NMR (CDCl3, 400 MHz): δ 9.28 (s, 1H), 6.83 (d, J 2.4,
2H), 1.31 (s, 24H).
Meridianin G:33 4a (1.0 mmol) was prepared according to the
general method for Ir-catalyzed C−H borylation reactions described
above. After completion of the reaction by GC monitoring, octane was
removed under reduced pressure. To the resulting residue was added
4-chloropyrimidin-2-amine (130 mg, 1.0 mmol), Cs2CO3 (0.815 g, 2.5
mmol), and dtbpfPdCl2 (6.4 mg, 0.01 mmol), followed by the addition
of a 1:1 mixture of IPA and water (6 mL). This mixture was heated to
80 °C with stirring for 16 h under an N2 atmosphere. Then the
reaction mixture was extracted with water (6 mL) and brine (6 mL).
The organic phase was dried over sodium sulfate, filtered, and
concentrated under reduced pressure. The crude product was purified
by column chromatography on silica gel with a solvent mixture of 5%
MeOH and 95% hexane. The desired product was obtained as a pale
1
(12) Liskey, C. W.; Liao, X.; Hartwig, J. F. J. Am. Chem. Soc. 2010,
132, 11389.
(13) Tzschucke, C. C.; Murphy, J. M.; Hartwig, J. F. Org. Lett. 2007,
9, 761.
yellow powder (0.183 g, 87% overall yield). H NMR (CDCl3, 400
MHz): δ 11.66 (s, 1H), 8.58 (d, J 7.8, 1H), 8.19 (d, J 2.9, 1H), 8.10 (d,
J 5.3, 1H), 7.44 (d, J 7.8, 1H), 7.14 (m, 2H), 7.01 (d, J 5.3, 1H), 6.41
(s, 2H).
(14) (a) Beck, E. M.; Hatley, R.; Gaunt, M. J. Angew. Chem., Int. Ed.
2008, 47, 3004. (b) Fischer, D. F.; Sarpong, R. J. Am. Chem. Soc. 2010,
132, 5926. (c) Liao, X.; Stanley, L. M.; Hartwig, J. F. J. Am. Chem. Soc.
2011, 133, 2088. (d) Leal, R. A.; Beaudry, D. R.; Alzghari, S. K.;
Sarpong, R. Org. Lett. 2012, 14, 5350. (e) Han, S.; Morrison, K. C.;
Hergenrother, P. J.; Movassaghi, M. J. Org. Chem. 2014, 79, 473.
(15) (a) Keller, S. N.; Veltri, N. L.; Sutherland, T. C. Org. Lett. 2013,
15, 4798. (b) Nakano, M.; Shinamura, S.; Sugimoto, R.; Osaka, I.;
Miyazaki, E.; Takimiya, K. Org. Lett. 2012, 14, 5448. (c) Shinamura, S.;
Sugimoto, R.; Yanai, N.; Takemura, N.; Kashiki, T.; Osaka, I.;
Miyazaki, E.; Takimiya, K. Org. Lett. 2012, 14, 4718. (d) Liu, Z.; Wang,
Y.; Chen, Y.; Liu, J.; Fang, Q.; Kleeberg, C.; Marder, T. B. J. Org.
Chem. 2012, 77, 7124. (e) Eliseeva, M. N.; Scott, L. T. J. Am. Chem.
Soc. 2012, 134, 15169. (f) Yamaguchi, R.; Hiroto, S.; Shinokubo, H.
Org. Lett. 2012, 14, 2472.
ASSOCIATED CONTENT
* Supporting Information
■
S
A table regarding lower catalyst loading, text giving additional
details on SS NMR 35Cl and XPS spectra of 1 and 2, figures
giving 1H and 13C NMR spectra for compounds 2, 3, 4a−e, and
meridianin G, DSC traces of 2 and 3, comparison of solution-
1
state H NMR spectra of 2 and [Ir(OMe)(COD)(phen)],
solution and SS NMR spectra of Ir(Cl)(COD)(Me4phen) and
Ir(Cl)(COD)(dtbpy), and the 11B NMR spectrum of the
reaction of [Ir(Cl)(COD)(dtbpy)] with B2pin2, and a CIF file
giving X-ray crystallographic data for 3. This material is
available free of charge via the Internet at http://pubs.acs.org.
(16) This has in particular been observed when using [Ir(OMe)
(COD)]2 as a precursor: see ref 22b.
(17) Li, H.; Seechurn, C. C. C. J.; Colacot, T. J. ACS Catal. 2012, 2,
1147.
AUTHOR INFORMATION
Corresponding Author
*E-mail for T.J.C.: Thomas.Colacot@jmusa.com.
■
(18) www.jmcct.com.
(19) Takagi, J.; Sato, K.; Hartwig, J. F.; Ishiyama, T.; Miyaura, N.
Tetrahedron Lett. 2002, 43, 5649.
Notes
The authors declare the following competing financial interests:
The catalysts used in this study are commercially available from
Johnson Matthey Catalysis and Chiral Technologies.
(20) Kallepalli, V. A.; Shi, F.; Paul, S.; Onyeozili, E. N.; Maleczka, R.
E., Jr.; Smith, M. R., III J. Org. Chem. 2009, 74, 9199.
(21) Preshlock, S. M.; Plattner, D. L.; Maligres, P. E.; Krska, S. W.;
Maleczka, R. E., Jr.; Smith, M. R., III Angew. Chem., Int. Ed. 2013, 52,
12915.
(22) (a) Larsen, M. A.; Hartwig, J. F. J. Am. Chem. Soc. 2014, 136,
4287. (b) Preshlock, S. M.; Ghaffari, B.; Maligres, P. E.; Krska, S. W.;
Maleczka, R. E., Jr.; Smith, M. R., III J. Am. Chem. Soc. 2013, 135,
7572.
(23) Cho, S. H.; Hartwig, J. F. J. Am. Chem. Soc. 2013, 135, 8157.
(24) Mestroni, G.; Camus, A.; Zassinovich, G. J. Organomet. Chem.
1974, 73, 119.
(25) See the Supporting Information for more details.
(26) Perras, F. A.; Bryce, D. L. Angew. Chem., Int. Ed. 2012, 51, 4227.
(27) (a) Lokare, K. S.; Nielsen, R. J.; Yousufuddin, M.; Goddard, W.
A., III; Periana, R. A. Dalton Trans. 2011, 40, 9094. (b) Thai, T.-T.;
ACKNOWLEDGMENTS
■
We thank Dr. M. Netaji at IISc Bangalore for collecting X-ray
crystal structure data. In addition, we thank Dr. Jonathan
Bradley (JM Technology Centre U.K.) for carrying out SS
NMR studies and Dr. Richard Smith (JM Technology Centre
U.K.) for doing the XPS analysis, both providing valuable
discussions. We also thank Fred Hancock, Technical Director
(JMCCT), and Gerard Compagnoni, General Manager
(JMCCT), for supporting this work.
REFERENCES
■
(1) Johansson Seechurn, C. C. C.; Kitching, M. O.; Colacot, T. J.;
Snieckus, V. Angew. Chem., Int. Ed. 2012, 51, 5062.
Therrien, B.; Suss-Fink, G. Inorg. Chem. Commun. 2009, 12, 806.
̈
3521
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522

Organometallics
Article
(28) Hartwig has noted borylation of octane using a catalyst formed
in situ from [Ir(OMe)(COD)]2 and Me4-1,10-phenanthroline: Liskey,
C. W.; Hartwig, J. F. J. Am. Chem. Soc. 2012, 134, 12422.
(29) Boller, T. M.; Murphy, J. M.; Hapke, M.; Ishiyama, T.; Miyaura,
N.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127, 14263.
(30) (a) Ishiyama, T.; Takagi, J.; Hartwig, J. F.; Miyaura, N. Angew.
Chem., Int. Ed. 2002, 41, 3056. (b) ref (22b).
(31) See the Supporting Information for 11B NMR studies.
Speculatively, the unreactive boron species is hydrolysis product
pinBOH. For spectral data of this compound, see: Bettinger, H. F.;
Filthaus, M.; Bornemann, H.; Oppel, I. M. Angew. Chem., Int. Ed. 2008,
47, 4744.
(32) (a) Lebar, M. D.; Baker, B. J. Aust. J. Chem. 2010, 63, 862.
(b) Seldes, A. M.; Brasco, M. F. R.; Franco, L. H.; Palermo, J. A. Nat.
́
Prod. Res. 2007, 21, 555. (c) Franco, L. H.; Bal de Kier Joffe, E.;
Puricelli, L.; Tatian, M.; Seldes, A. M.; Palermo, J. A. J. Nat. Prod.
1998, 61, 1130.
(33) Merkul, E.; Schafer, E.; Muller, T. J. J. Org. Biomol. Chem. 2011,
9, 3139.
(34) Colacot, T. J. e-EROS Encyclopedia of Reagents for Organic
Synthesis 2009, DOI: 10.1002/047084289X.rn01062.
(35) Mertins, K.; Zapf, A.; Beller, M. J. Mol. Catal. A 2004, 207, 21.
(36) Miyaura, N.; Ishiyama, T. EP 1481978 A1, 2004.
3522
dx.doi.org/10.1021/om500420d | Organometallics 2014, 33, 3514−3522