High-throughput optimization of Ir-catalyzed C-H borylation: A tutorial for practical applications

Article abstract of DOI:10.1021/ja400295v

With the aid of high-throughput screening, the efficiency of Ir-catalyzed C-H borylations has been assessed as functions of precatalyst, boron reagent, ligand, order of addition, temperature, solvent, and substrate. This study not only validated some acce

Full text of DOI:10.1021/ja400295v

Article
pubs.acs.org/JACS
High-Throughput Optimization of Ir-Catalyzed C−H Borylation: A
Tutorial for Practical Applications
Sean M. Preshlock,† Behnaz Ghaffari,† Peter E. Maligres,‡ Shane W. Krska,*,‡ Robert E. Maleczka, Jr.,*,†
and Milton R. Smith, III*,†
†Department of Chemistry, Michigan State University, East Lansing, Michigan 48824-1322, United States
‡Department of Process Chemistry, Merck Research Laboratories, Rahway, New Jersey 07065, United States
S
* Supporting Information
ABSTRACT: With the aid of high-throughput screening, the efficiency
of Ir-catalyzed C−H borylations has been assessed as functions of
precatalyst, boron reagent, ligand, order of addition, temperature, solvent,
and substrate. This study not only validated some accepted practices but
also uncovered unconventional conditions that were key to substrate
performance. We anticipate that insights drawn from these findings will
be used to design reaction conditions for substrates whose borylations are
difficult to impossible using standard catalytic conditions.
utilizing [Ir(cod)(μ2-OMe)]2 (2, cod = 1,5-cyclooctadiene) as
the precatalyst, dtbpy as the ligand, B2pin2 as the borylating
agent, and hexane as the solvent gave excellent results for a
panel of electron-deficient substrates. In this report, it was
noted that reaction conversions were solvent dependent,
following the order hexanes > DME > DMF. The ability to
carry out the reaction with an air-stable boron reagent (B2Pin2)
using the precatalyst [Ir(cod)(μ2-OMe)]2, which has been
described as being air stable, has made the ITHM protocol the
preferred method for carrying out Ir-catalyzed C−H
borylation.7
INTRODUCTION
■
The development of iridium catalyzed C−H borylation over
the past decade has provided a simple, atom economical route
to arylboronate esters.1 Studies of precatalysts such as (η6-
2
mesitylene)Ir(Bpin)3 and (dtbpy)Ir(Bpin)3(coe)3 (1, dtbpy
=4,4′-di-tert-butyl-2,2′-bipyridine, coe = cyclooctene) support a
mechanism where species of the formula L2IrIIIBpin3 are the
intermediates that mediate C−H cleavage (Scheme 1). This
Scheme 1. Mechanism for Ir-Catalyzed C−H Borylation
To date, many unique combinations of precatalyst, boron
reagent, ligand, and solvent have been described in the
literature.8 Not surprisingly, many of these are variants of the
ITHM protocol, where modifications have been made based on
mechanistic work and/or empirical findings. Since the initial
reports, the substrate scope for C−H borylation has expanded
considerably. Yet, there has been no comprehensive study of
the synergistic roles that substrate, solvent, precatalyst,
borylating agent, temperature, order of addition, etc. might
have on borylations. Such a study is important because it is
likely that conditions that are optimal for one substrate class
(e.g., electron-deficient arenes) might have limited success, or
fail entirely, for another substrate (e.g., an azaindole).
Comprehensive studies of this type are challenging for a
number of reasons. First, exploration of n variables creates an n-
dimensional reaction space to be surveyed, making most
undertakings labor intensive. Second, collection and processing
of experimental data must be reliable and fast to avoid an
analysis bottleneck. Third, the analysis and interpretation of n-
dimensional data are nontrivial, and it is critical that they are
presented such that underlying themes are clear so scientists
picture has been corroborated by computational studies4 and
more recently by the isolation of 16-electron complexes
(dippe)IrBpin3 and (dcpe)IrBpin3 that react directly with C−
H bonds at room temperature.5
In an early communication by Ishiyama, Takagi, Hartwig, and
Miyaura (hereafter ITHM), a set of precatalysts and ligands
were screened for borylations.6 From this screen, conditions for
Received: January 21, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
can use the information to solve specific problems. Lastly,
variability in any experimental endeavor is unavoidable. This
means that multiple runs must be carried out for conclusions
from the data to be made with confidence.
dispensed as stock solutions in the designated solvent. After the
catalyst components were combined, aliquots from a stock
solution of the substrate and internal standard (typically
dodecahydrotriphenylene) were added to the well plates. The
reaction plate was then sealed and stirred at the desired
temperature for an allotted amount of time, at which point the
reaction was cooled to room temperature and quenched by
exposure to atmospheric O2.
The advent of high-throughput reaction screening techniques
and modern analytical methods has provided a means of
overcoming the obstacles listed above, enabling a comprehen-
sive study to be undertaken with a reasonable investment of
time and resources.9 Given the potential value of Ir-catalyzed
C−H borylation, a comprehensive study would provide a
roadmap for practitioners of this method. For these reasons, we
have undertaken a systematic study of multiple reaction
variables affecting Ir catalyzed C−H borylation. While this
study validates some accepted practices, there are some
surprising findings and several cases where unconventional
conditions are the key to substrate performance.
The four precatalysts used in the screens were the
commercially available compounds 2, 3, Ir(acac)(cod) (4),
and (Ind)Ir(cod) (5) (Chart 1). The ligands chosen were the
Chart 1
RESULTS
■
Methods. To expedite the process of running the number
of reactions required in this study, reactions were conducted in
a glovebox using microscale 96-well plate reactors. To ensure
that results from this study mirrored typical laboratory
applications, commercially supplied anhydrous solvents and
reagents were used as received, except for THF and 2-methyl-
tetrahydorfuran, which were distilled from Na/benzophenone
to remove inhibitors. Reagents were dispensed to each well
from stock solutions to accurately control stoichiometries, and
reactions were run in duplicate on each plate to test for
consistency. Furthermore, entire reaction screens were periodi-
cally repeated to further certify reproducibility. Temperatures
and time points were chosen so that the fastest reactions were
stopped before completion. Reaction products were verified by
commonly employed dtbpy and the electron-rich ligand 3,4,7,8-
tetramethyl-1,10-phenanthroline (tmphen).10 The reaction in
the initial screen was the borylation of 3-bromotoluene carried
out in THF at room temperature. The results are shown in
Figure 1.
1
comparing H NMR spectra of crude mixtures to data of
authentic compounds. Product yields were determined from
HPLC data calculated from peak areas, relative to an internal
standard.
Order of Addition Effects. Compound 1 is most likely the
major catalyst resting state in borylations utilizing 2 as the Ir
precatalyst. In the synthesis of 1 from 2 it was noted that the
order in which HBpin and dtbpy were added greatly affected
isolated product yields.3 Consequently, it might be expected
that orders of addition might impact borylation efficiencies for
in situ generated catalysts. The nature of the borane reagent has
also been shown to influence relative rates of borylation6a
somewhat surprising finding given that kinetic studies show
catalytic reactions with B2pin2 are zero order in [B2pin2].3
Variations have also been observed depending on the nature of
the precatalyst. For example, borylations with B2pin2 using the
precatalyst [Ir(μ2-Cl)(cod)]2 (3) exhibit induction periods that
can be eliminated when catalytic amounts of HBpin are added
to the reaction mixture.3
The present study uses catalysts generated in situ because
this is the typical practice reported for C−H borylations. Given
this and the influences of reagents and catalysts noted above,
examination of the effects of the order of addition of
precatalyst, ligand, and boron reagent on the outcome of C−
H borylation was deemed the most logical starting point. Two
conditions for in situ catalyst formation were tested and found
to significantly impact borylation efficiency. These are
designated as conditions A and B in the text. For condition
A, the order of addition is precatalyst, boron reagent, and
ligand. The order for condition B is precatalyst, ligand, and
boron reagent. The precatalyst, boron reagent, and ligand were
Figure 1. Order of addition effects for room temperature C−H
borylations. Data plotted are product yields determined by HPLC.
B
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Some general trends emerge. First, the relative ordering of
precatalyst activity is 2 > 4 > 3 > 5. This is consistent with the
literature for 2, 3, and 5, with 5 being ineffective at room
temperature.2 While precatalyst 4 has not been used in any
borylations to date, it outperforms more commonly employed
3 in most cases and is more effective than Ir(OAc)(cod).6
The order of addition had little effect on the performance of
the precatalyst/ligand combination 2 or 4/dtbpy, which is
somewhat unexpected given the influence of addition order in
synthesis of 1 from 2.3 For 2/dtbpy borylation conversions with
B2pin2 were ∼50% greater than those with HBpin, while HBpin
was ∼50% more effective than B2pin2 for precatalyst 4/dtbpy.
Surprisingly, the order of addition greatly affected the
performance of precatalyst 2/HBpin when the more electron-
rich ligand tmphen was used. For example, the performance of
the combination 2/tmphen/HBpin rivals that of 2/dtbpy/
B2pin2 when the borane is added before the ligand. This
reactivity is attenuated 2.5-fold when tmphen is added before
HBpin. In contrast, borylations with the combination 2/
tmphen/B2pin2 were insensitive to the order of addition and
had conversions that were intermediate to those for HBpin. It is
particularly noteworthy that the highest borylation conversions
for 2/tmphen are achieved with HBpin, even though the
thermodynamic driving force is less than the analogous reaction
with B2pin2.
As expected, conversions at 80 °C were much greater than
those at 25 °C. In contrast to the observations made with room
temperature borylation of 3-bromotoluene, conversions in the
present case were only slightly influenced by order of addition
with condition B (precatalyst, ligand, then boron reagent)
giving marginally higher conversions. At 80 °C the pairing of
ligand and boron reagent had pronounced effects. For the
typically employed dtbpy ligand, conversions with HBpin were
considerably lower than those for B2pin2. Borylation con-
versions with tmphen-ligated catalysts were superior to those
for dtbpy, and the choice of boron reagent was less important
with HBpin and B2pin2 performing similarly.
To more broadly assess the effects of elevated temperatures
on C−H borylation, the reaction of 3-methyl-N,N-dimethylani-
line was screened at 80 °C. The combination of precatalysts,
boron reagent, and orders of addition was identical to that in
Figure 1. Because this substrate is particularly unreactive, the
data in Figure 3 were recorded after 8 h reaction time.
Of the precatalysts that operated at room temperature, 3 was
most sensitive to addition order and borane reagent.
Specifically, the only combination that gave conversion at
room temperature was condition A with HBpin. The
effectiveness of HBpin over B2pin2 is consistent with the
previously noted induction period for borylations employing
the latter with precatalyst 3;3 however, the influence of addition
order with HBpin was unexpected.
Temperature Effects. Substrate electronic effects influence
Ir-catalyzed C−H borylation rates with electron-deficient
substrates being the most reactive.4a Consequently, room
temperature borylations of electron-rich substrates are less
practical. This is illustrated by reactivity of 1,2,3-trimethox-
ybenzene. In Figure 2, borylation conversions are given at 25
and 80 °C. In all cases the reaction duration was 2.5 h, and
precatalyst 2 was used. In addition to temperature, the ligand,
boron reagent, and order of addition were varied.
Figure 3. Effects of precatalyst, boron reagent, and order of addition,
on C−H borylation of 3-methyl-N,N-dimethylaniline at 80 °C.
The effects of the ligand were dramatic with tmphen
outperforming dtbpy with every precatalyst except for 3. For all
other catalysts conversions with tmphen were 1.5−4.5 times
greater than those with dtbpy. This is consistent with previous
observations that more electron-rich catalysts are more
reactive.6
The data in Figure 3 differ significantly from those in Figure
1 in several respects. For example, the order of addition did not
affect conversions at elevated temperature, consistent with the
observations on the borylation of 1,2,3-trimethoxybenzene
(vide supra). Also the choice of precatalyst is less important
when dtbpy is used. Specifically, conversions with precatalysts
2−5 with dtbpy/HBpin gave uniformly low conversions. For
the combination dtbpy/B2Pin2, precatalysts 2−4 perform
Figure 2. Temperature and ligand effects on borylation of 1,2,3-
trimethoxybenzene with precatalyst 2. Data plotted are product yields
determined by HPLC.
C
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
equally well and superior to reactions with 5. As was the case in
Figure 1, the combination dtbpy/B2Pin2 outperforms dtbpy/
HBpin, but the increase (3-fold on average) was more
pronounced at elevated temperature for precatalysts 2−4.
In contrast, conversions with tmphen are sensitive to the
choice of precatalyst. The performance of precatalyst 3 is
particularly poor, mirroring the situation for borylation at room
temperature. For the combination tmphen/HBpin the
precatalyst activity is 2 > 4 > 5 ≫ 3 when the order of
addition is condition A. For condition B, the activity of 2
decreases and 3 increases such that 2 ∼ 4 > 5 > 3. When B2Pin2
is used with tmphen, the order of addition does not matter.
Interestingly, the activities of 3 and 5 increase relative to 4 such
that the overall ordering is 2 > 4 ∼ 5 > 3 with the performance
of 4 and 5 being ∼85% of that for 2.
candidates for C−H borylation.6 Since the order of addition
effects can be attributed to efficiency of in situ catalyst assembly
(vide infra), we wondered whether the poor performance in
polar solvents might be due to inefficient catalyst assembly
rather than performance of the catalyst itself. To address this
question we synthesized precatalyst 1 according the literature
procedure.3 The tmphen analog (6) was prepared similarly.
Compound 1 is one of the most efficient borylation precatalysts
reported to date,7 generating the active catalytic intermediate 7
by dissociation of cyclooctene. Intermediate 8 would be
generated in analogous fashion from tmphen complex 6
(Scheme 2).
Scheme 2. Isolable Precatalysts and Equilibria Generating
Active Intermediates
Ligand Effects. Phosphine ligands have been used in
catalytic borylations.2 In practice, the catalyst ensemble that
results is less active than the in situ generated nitrogen chelate
ligands. Since quantitative comparisons have not been made,
the borylation activity of 3-methyl-N,N-dimethylaniline was
screened using the bidentate phosphine ligands in Chart 2.
Chart 2
Figure 5 compares borylation conversions for 3-bromoto-
luene in 11 different solvents with a wide range of polarities.11
The performance of catalysts generated in situ from precatalyst
Borylations with dmpe, dppe, and dmpbz in THF at 80 °C
did not give product for any combination of ligand, precatalyst
and boron reagent. In contrast, borylation with HBpin and
dppbz gave appreciable yields of product. The activity of dppe,
dtbpy, dppbz, and tmphen are compared in Figure 4. The best
Figure 4. Comparison between dipyridyl and diphosphine ligated
catalysts. Data plotted are product yields determined by HPLC.
Results for dtbpy and tmphen are the same as shown in Figure 3.
precatalyst for dppbz is compound 5. In fact, for the borylation
of this electron-rich substrate with HBpin, when used with 5
dppbz outperforms all combinations of precatalyst and dtbpy, is
competitive with tmphen/3 or 5 but is less reactive than the
pairing of tmphen with precatalysts 2 or 4.
Solvent Effects. From the early report that solvent effects
for borylation follow the order hexanes > DME > DMF it
generally has been accepted that polar solvents are poor
Figure 5. Solvent effects on activity in situ generated catalysts and
isolated precatalysts 1 and 6. Solvents are ordered by increasing
dielectric constant.
D
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
2 using HBpin and ligand (Condition A) were compared to
borylations catalyzed by preassembled (ligand)Ir(coe)(Bpin)3
precatalysts 1 and 6.
ligand to metal ratio on the borylation of 3-methyl-N,N-
dimethylaniline. The results are shown in Figure 7.
The data in Figure 5 convincingly show that the in situ
catalyst generation can be problematic since conversions for in
situ generated catalysts are uniformly lower than those for 1
and 6. Moreover, the efficiency of in situ catalyst generation
varies considerably with the nature of the solvent. The best
illustration of this is the comparison between the results
obtained with in situ catalyst generation from precatalyst 2 and
dtbpy as ligand versus those obtained with complex 1, both in
hexanes and dioxane as solvent. In line with the original
literature report, when in situ catalyst generation is employed,
hexanes is almost twice as effective as dioxane as solvent;
however, borylation conversions in dioxane are higher in
dioxane than hexane when preformed 1 is the precatalyst.
While we have confirmed that borylations in DMF were poor
for both in situ and preformed catalysts, N-methylpyrolidone
(NMP) is an excellent solvent for borylations with preformed
1. In fact, seven solvents, including THF, NMP, and isopropyl
acetate (Ipac), are superior to hexane for borylations with
preformed 1. Thus, the data in Figure 5 dispel the notion that
polar solvents are generally unsuitable for borylation.
Figure 7. Effects of ligand to precatalyst ratio on performance of in
situ generate C−H borylation catalysts. The highest yield for each
respective catalyst loading is highlighted in boldface. Data plotted are
product yields determined by HPLC.
Since it is desirable to carry out reactions with more electron-
rich substrates at elevated temperatures, we examined
borylations of dimethylresorcinol at 80 °C for a subset of
solvents. These results are shown in Figure 6. For dtbpy, the
For both dtbpy and tmphen, conversions were dependent on
the ligand to metal ratio as a function of the catalyst loading.
Specifically, increasing the ligand to metal ratio gives optimum
yields as the catalyst loading is lowered. The effects for tmphen
were more dramatic than for dtbpy. For example, identical
conversions can be obtained with a 10-fold decrease in catalyst
loading by simply doubling the ligand to metal ratio. It is
noteworthy that tmphen conversions diminish when the ligand
to metal ratio exceeds the optimal value. In contrast, exceeding
the optimal value had less effect on conversions for dtbpy.
Precatalyst Stability. While precatalyst 2 is described as
being air stable, we have observed that its activity diminishes
when it is left on the benchtop. This prompted us to examine
the effects of air exposure on the activity of precatalysts in
Chart 1. This was accomplished by storing the catalyst for 100
days at ambient temperature at different relative humidity
levels. The precatalysts were then assayed for activity by
carrying out room temperature borylations of 3-bromotoluene
using dtbpy as the ligand and comparing their performance that
of control samples that were stored in a nitrogen filled drybox.
As shown in Figure 8, catalytic activity of precatalyst 2
diminishes upon exposure to air. The fact that performance
deteriorates as humidity increases indicates that moisture
sensitivity of 2 is the primary issue. In contrast, the
performance of precatalyst 3 was not compromised under the
same conditions. Lastly, we note that the performance of
precatalyst 2 was most batch dependent. This observed
decrease in reactivity for samples of 2 upon aging in air was
accompanied by the development of a green-yellow color of the
solid, in contrast to the lemon-yellow color of fresh samples of
2. These partially degraded samples of 2 were judged to be
Figure 6. Solvent effects on activity at 80 °C of in situ generated
catalysts and isolated precatalysts 1 and 6. Data plotted are product
yields determined by HPLC.
best solvents are THF and NMP, whereas for tmphen, THF,
hexanes, and Hunig’s base had superior performance. For both
̈
ligands, isopropyl acetate had the lowest conversions. Most
importantly, the data clearly show that at elevated temperatures
the performance of the in situ generated catalysts improved to
the point that it rivaled that of the preformed complexes. This
is consistent with previous observations for borylations carried
out in neat, excess benzene-d6.3
1
∼95% pure by H NMR.
Open vs Closed Systems. In this study the reactions were
carried out in closed systems. Borylations with B2pin2 generate
HBpin as shown in eq 1. The HBpin can effect a second
borylation that generates H2 (eq 2). The H2 generated will
likely react with Ir−boryl complexes to form hydrides according
to eq 3. Since effects from H2 would be more pronounced in
Ligand to Precatalyst Ratio Effects. For precatalysts like
those depicted in Scheme 1, stoichiometry requires 1 equiv of
bidentate ligand per Ir. It is not unusual to find that an excess of
ligand gives optimum performance in reactions mediated by in
situ generated catalysts. Thus, we examined the effect of the
E
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 8. Aging effects on catalyst performance. Samples were stored
at ambient temperature for 100 days. The data reported at 0% relative
humidity refers to precatalyst samples stored under dry nitrogen.
Other samples were stored under air at the reported relative
humidities.
closed systems, we compared reactivities between open and
closed systems for HBpin and B2pin2, and the results are shown
in Figure 9. For HBpin, borylations in open systems were
slightly more efficient than closed ones, while performance of
open and closed systems was essentially identical for B2pin2.
pinB−Bpin + Ar−H → Ar−Bpin + pinB−H
(1)
Figure 9. Comparisons between borylations with HBpin and B2pin2 in
open and closed systems. Red and blue symbols represent tmphen and
dtbpy, respectively; + = open system, o = closed system.
pinB−H + Ar−H → Ar−Bpin + H−H
(2)
[Ir]−Bpin + H−H ⇌ H−Bpin + [Ir]−H
(3)
The kinetic behavior of the reactions with B2pin2 is
interesting. In the case of dtbpy, the product initially forms
more rapidly than when tmphen is the borylating ligand.
Moreover, the B2pin2 reaction plateaus once the B2pin2 has
been consumed. In contrast, the tmphen conversion is initially
slower, which may indicate an induction period for the
formation of active catalyst, but ultimately more product is
formed because the tmphen system is more efficient for
borylation with HBpin. While the detailed picture may be more
complex and beyond the scope of this work, the improved
conversion for HBpin has practical consequences for the boron
atom economy of the reaction (vide infra).
metal boryl complexes. These putative complexes could then
react with chelating ligands to generate reactive boryl
intermediates of the formula (ligand)Ir(H)x(Bpin)3−x, where
x = 0−2. Even though oxidative addition of B2pin2 to 2 and 3
should be thermodynamically favorable, no reaction was
observed at room temperature. Interestingly, complex 3 was
ineffective for room temperature borylations with B2pin2.
When dtbpy and tmphen are added to precatalyst 3 an
1
immediate reaction occurs. H NMR experiments (THF-d8)
show that a new complex forms when dtbpy is added to 3. A
single high-field resonance for the olefinic cod protons and
chemically equivalent dtbpy t-Bu groups are consistent with
generation of the 18-electron complex IrCl(κ2-dtbpy)(κ2-cod)
(9).12 Oxidative addition to this coordinatively saturated
complex will be slow, which accounts for the lack of reactivity.
A similar complexation is seen in the reaction between tmphen
and 3.13 The reactions of dtbpy and tmphen with precatalyst 3
are detrimental to catalyst performance at room temperature.
Under these conditions it appears that generation of boryl
intermediates prior to addition of the chelating ligand is
essential to catalytic activity. If so, the success of HBpin over
B2pin2 in the room temperature borylations with 3 can be
attributed to the reactivity of the former over the latter.
DISCUSSION
■
Order of Addition Effects. The most pronounced effects
were for borylations carried out at room temperature, where
the choice of precatalyst, ligand, and boron reagent, combined
with order of addition, can be critical. From Figure 1 it is clear
that the order of addition has no effect on borylations carried
out using B2pin2. In contrast, borylations with HBpin are
sensitive to the order of addition, being more effective when the
order is precatalyst, borane, followed by ligand (condition A).
The behavior can be attributed to precatalyst reactivity with
HBpin and ligand. Precatalysts 2 and 3 react rapidly with
HBpin. While the structures of the resulting iridium complexes
are not known, oxidative additions of boranes typically generate
For the combination of precatalyst 2 and dtbpy, the order of
1
addition had no effect on reactivity. H NMR experiments
(THF-d8) showed no evidence for complexation. Thus, dtbpy
F
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
does not interfere with generation of active catalyst. In contrast,
tmphen reacts rapidly with 2 to generate a green precipitate
from THF, which explains the diminished reactivity when the
order of addition is 2, tmphen, followed by HBpin.
Scheme 3. Precatalyst/Ligand Complexes Relevant to Order
of Addition Effects
With respect to precatalyst, compound 2 had superior
performance, regardless of ligand, boron reagent, or order of
addition. This is consistent with the original report for the
ITHM system. For the combination of dtbpy/2, order of
addition had little effect on conversion (Figure 1). This is
somewhat surprising given the report that yields in the
synthesis of compound 1, presumably an active species in the
in situ generated reaction, were sensitive to the order of
addition.
For the combination tmphen/2, Figure 1 shows that the
order of addition is important when HBpin is the boron
reagent. Specifically, conversion diminishes when the addition
order is precatalyst, ligand, and boron reagent. When tmphen is
added to solutions of 2, a dark-green solid precipitates. While
insolubility has hampered attempts to characterize the species
that form, the low conversion under these conditions likely
arises from inefficient generation of the homogeneous catalyst
from the heterogeneous mixture.
Room temperature borylations with precatalyst 3 were most
sensitive to order of addition, as no conversion was observed
for the order: precatalyst, ligand, and boron reagent.
Conversely, oxidative addition of HBpin to 16-electron
complex 3 is more viable, and the boryl and hydride ligands
will labilize ligands trans to them. This will facilitate
coordination of dtbpy or tmphen ligands to generate active
catalysts. Induction effects have been observed for borylations
with precatalyst when the borane reagent is B2pin2. Presumably
oxidative addition of B2pin2 to 3 is thermodynamically viable
but kinetically slow.
The performance of precatalyst 4, the second most reactive
at room temperature, was not affected by order of addition.
Compound 4 does not react with dtbpy, but it forms an
insoluble precipitate with tmphen, which accounts for the low
conversions at room temperature.
HBpin at 80 °C compared to the 1.4-fold improvement for
B2pin2 over HBpin at room temperature in Figure 2.
The difference between the dtbpy and tmphen data is
striking. In all cases, tmphen outperforms dtbpy, which is
consistent with tmphen being the more electron-rich ligand.
The surprising feature of the tmphen reactivity is that
borylation conversions for HBpin rival those for B2pin2. This
difference is more pronounced for the borylations of N,N-
dimethyl-m-toluidine carried out with precatalyst 2 in Figure 3.
For more electron-rich substrates the tmphen/2 combination
improves the atom economy for B2pin2 by utilizing both boron
equivalents for borylation.
The order of addition effects at room temperature correlates
with interactions between the precatalysts and the nitrogen
chelate ligands. When complexation between the Ir precatalyst
and ligand occurs, catalysis is inhibited if the boron reagent is
added after the chelating ligand. This is the case when tmphen
is paired with precatalyst 2, 3, or 4 and for dtbpy/3 as
summarized in Scheme 3. For room temperature borylations
with these precatalyst ligand pairings, the best conversions are
obtained when the order of addition is precatalyst, boron
reagent, then ligand. Conversely, borylation conversions for
combinations of precatalyst 2−5 with dtbpy and tmphen
carried out at elevated temperature (Figures 2 and 3) are not
affected by order of addition.
Precatalysts that performed poorly at ambient temperature
were more viable at 80 °C. In particular, 3 performed just as
well as 2 in borylations with dtbpy; however, complex 3
performed poorly with tmphen. The improvement for 3/dtbpy
can be rationalized as follows (Scheme 4). At elevated
temperature pre-equilibrium dissociation of one dipyridyl arm
in complex could generate 16-electron intermediate IrCl(κ1-
dtbpy)(κ2-cod) (10) which can undergo oxidative addition with
HBpin or B2pin2. In addition to being a stronger donor,
Scheme 4. Accessibility of κ1-Intermediates for Dipyridyl
and Phenanthroline Ligands
Temperature Effects. There are many cases where it is
desirable to carry out borylations at elevated temperatures to
minimize reaction times, particularly in the case of electron-rich
substrates. As most of the C−H borylations with the 2/dtbpy
precatalyst/ligand combination have been carried out at room
temperature, it is important to assess catalytic viability at
elevated temperatures. This is demonstrated for the borylations
of 1,2,3-trimethoxybenzene in Figure 2, where conversions
clearly improve with increased temperature.
As the room temperature data in Figure 2 show, conversions
were more sensitive to the choice of boron reagent. Specifically,
conversions with B2pin2 were 2.0 times higher than those for
G
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
tmphen has a constrained geometry that will disfavor
preequilibrium dissociation to IrCl(κ1-tmphen)(κ2-cod).
Catalysis with Constrained Geometry Ligands. Despite
the fact that five-coordinate complexes with the formulas (κ2-
R2PCH2CH2PR2)Ir(Bpin)3 (R = i-Pr (11) and Cy) borylate
aromatic and heteroaromatic substrates at room temperature,
dppe was ineffective for the borylation of N,N-dimethyl-m-
toluidine at 80 °C. Attempts to prepare dppe analogs of 11
from dppe and (MesH)Ir(Bpin)3 (12) yield the six-coordinate
complex (κ2-dppe) (κ1-dppe)Ir(Bpin)3 (13).5,14 Complex 13 is
much less reactive than five-coordinate complexes 11 because a
phosphine ligand must dissociate to generate 16-electron
intermediate (κ2-dppe) Ir(Bpin)3 that effects borylation. In
contrast, analogs of dppbz react with (MesH)Ir(Bpin)3 to
afford five-coordinate structures (κ2-dppbz)Ir(Bpin)3 (14) that
are reactive. Improved access to five-coordinate structures for
dppbz relative to dppe is consistent with the enhanced
reactivity of dppbz (Scheme 5).
conformational rigidity of tmphen that enforces κ2 coordination
to the Ir center.
Solvent Effects. The notion that nonpolar solvents are
privileged for Ir-catalyzed C−H borylation arose because a
limited sample space was examined in early studies. While
relative conversions for in situ generated catalyst for 2/dtbpy
follow the order hexanes > DME > DMF, the data in Figure 5
show no clear correlation between solvent polarity and
conversions. In fact, the best solvent for room temperature
borylations with in situ generated 2/dtbpy is THF, which is one
of the most polar of those surveyed. In contrast, the catalyst
generated in situ from 2/tmphen performs best in nonpolar
solvents for room temperature borylations.
It has been shown in other catalytic systems that precatalysts
where coligands are bound to the metal can be more effective
than cases where ligand complexation is performed in situ.17
Likewise, the data for preformed catalysts 1 and 6 clearly show
that solvent effects for the in situ generated catalysts stem from
catalyst generation as opposed to the borylation reaction itself.
In fact, borylations with dtbpy ligated catalyst 1 become more
efficient as solvent polarity increases. This trend is not followed
for tmphen analog 6, where hexanes is the best solvent at room
temperature. Nevertheless, the borylation efficiency in polar
solvents like NMP is acceptable for synthetic applications.
At elevated temperatures (Figure 6), precatalyst 1 still
outperforms the in situ generated catalyst from 2/dtbpy, and
the more polar solvents, THF and NMP, are the best.
Conversely, there is no particular advantage to using precatalyst
6 as conversions for in situ generated catalysts from 2/tmphen
were virtually identical. In addition, the performance of the
tmphen ligated catalysts in NMP decreases at elevated
temperatures.
Scheme 5. Effects of Ligand Geometries on Catalysis
In the second column in Figure 5, triethylamine and
methylene chloride are the only outliers to the correlation
between conversion and solvent polarity. Low conversions from
triethylamine may be due solvent coordination to reactive
intermediates 7 and 8. Two observations support this notion.
First, borylation activity improves for sterically hindered
Hunig’s base. Second the relative order THF > NEt
≫
̈
3
tetrahydrothiophene (not shown in Figure 5) is consistent with
inhibition increasing with the softness of the donor atom (O <
N < S).
It is instructive to comment on solvents that are poor
candidates for borylations. For example, solvents with acidic
hydrogens like alcohols (ROH) are ineffective because they
react with the Ir−Bpin bonds to form ROBpin and Ir hydrides.
Thus, C−H acidity of acetonitrile makes it a poor solvent for
C−H borylation. Similarly, reactivity of the acetyl methyl
protons may account for the relatively poor performance of
isopropyl acetate in Figures 5 and 6. The poor conversions in
methylene chloride result from reactions between the active
catalysts and the solvent that generate catalytically inactive
species. Other side reactions can interfere with borylation for
solvents like DMSO where 11B NMR spectra indicate formation
of borates, presumably products of O-atom transfer from the
solvent.
Despite these limitations, this study shows that a number of
solvents, which practitioners of borylations might have not
considered, are viable, and in some cases preferred options. The
data in Figure 5 also show that poor performance for certain
solvents for in situ generated catalysts is the result of inefficient
catalyst generation. Given the number of steps involved in
It is interesting that the ligands where the dihedral angles in
the chelate backbone are constrained, tmphen and dppbz, are
more active than their unconstrained counterparts, dtbpy and
dppe. The difference between tmphen and dtbpy could be due
to the fact that tmphen is a stronger donor than dtbpy. If this
were the main contributing factor, then 4,4′-bis-
(dimethylamino)-2,2′-bipyridine (dmabpy) should be even
more reactive because it is more electron rich than dtbpy or
tmphen.15,16 A comparison of relative conversions for
borylation of 2,6-dimethylanisole with HBpin follows the
order dtbpy < dmabpy < tmphen, indicating that tmphen
outperforms a ligand that is more electron rich. Thus, the
superior performance of tmphen may not solely be due to
electronic effects. Perhaps, one contributing factor is the
H
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
generating active catalyst from 2 + ligand, solvent effects are not
surprising.
Scheme 6. Borylation of a Polar Heterocycle in
Dichloromethane
Ligand to Precatalyst Ratio Effects. For in situ generated
catalysts, employing a 1:1 ratio of ligand to Ir frequently results
in lower activity compared to preassembled catalysts. The data
in Figure 7 clearly show that conversions at lower loading can
be improved by increasing the equivalents of ligand per Ir
precatalyst. Remarkably, comparable conversions can be
obtained with a 10-fold reduction in catalyst loading by
doubling the ligand to compound 2 ratio. Excess dtbpy does
not inhibit borylation significantly, while for tmphen there is
clearly an optimum ratio for a particular loading after which
addition of more tmphen reduces conversion.18
When borylations were carried out using trisboryl catalysts 1
and 6, conversions were not affected significantly by addition of
dtbpy or tmphen. This suggests that there is a delicate balance
between the in situ formation of active catalytic structures, such
as 1 and 6, and deleterious side reactions of dtbpy, and tmphen
in particular, with Ir intermediates formed en route to 1 and 6.
Precatalyst Stability. Despite reports to the contrary, the
data in Figure 8 clearly show that 2 is not air stable. As relative
humidities increase, the performance of precatalyst 2 markedly
decreases. It is noteworthy that the precatalysts aged in dry air
were as active as samples stored in a glovebox, showing that
catalyst decomposition is accelerated by water. As the
precatalyst 2 decomposes in the solid state, the color changes
from lemon yellow to green and the solubility in CDCl3
decreases. The latter factor may explain why samples with
relatively clean NMR spectra can have marginal activity.
In contrast, precatalyst 3 was stable to air under identical
conditions. Because 3 does not require special precautions and
is prepared directly from IrCl3·(H2O)x, there are situations
(e.g., the borylations with dtbpy in Figure 3) in which it may be
the preferred precatalyst.
Open vs Closed Systems. The data in Figure 9 show that
borylation reactivity in closed systems is diminished when
HBpin is the borylating reagent. The higher H2 concentrations
in a closed system make the generation of hydride
intermediates (eq 2) more likely. While it is not obvious a
priori that these intermediates would be less reactive in
borylation reactions, the propensity for Ir hydrides to bridge is
well-known, and the resulting oligomers are typically less
reactive than monomeric species.19 The data in Figure 9 also
highlight the fact that both HBpin and B2pin2 react similarly
when tmphen is the coligand.
Borylation Case Studies. Solvent Effects. Relatively few
borylations have been described for heteroatom rich substrates.
This is unfortunate because many of these compounds are
important pharmacophores. For example, tetrazolo[1,5-a]-
pyridine scaffolds exhibit interesting antibacterial and anti-
inflammatory properties.20 Consequently, borylations of
tetrazolo[1,5-a]pyridines could yield desirable building blocks
for medicinal chemistry. However, borylation of tetrazolo[1,5-
a]pyridine 15, under conditions where boron reagents were
limiting, gave only traces of products after 24 h when the
reaction was carried out in hexane (Scheme 6). In contrast, the
analogous borylation in CH2Cl2 gave complete conversion of
boron reagents with formation of mono (16) and diborylated
(17) products after 4.5 h (see SI for detailed analysis of
products). Thus, despite the fact that methylene chloride gave
the lowest conversions in Figure 5, it is an excellent solvent for
borylation of 15. Under tetrazole limiting conditions, two
diborylated compounds form (17a,b Scheme 6) with complete
conversion of tetrazole within 1 h in CH2Cl2. At higher B2pin2
concentrations, reaction rates in hexane, MTBE, and THF
improved slightly; however, the reaction times required were
still 10 times longer than those for reactions in CH2Cl2.
Based on electronic effects, the 5-position of 15 is expected
to be the preferred site for the first borylation.4a Subsequent
borylation at the 7- or 8-position would generate the 17a and
17b, respectively. The combination of nonselective borylation
and modern separation methods is a potentially attractive
approach for generating diverse compound libraries. In the case
of 15, the key to efficient functionalization is using a solvent
that would normally be eschewed for the borylation reaction.
B2pin2 Atom Economy. Borylations of electron-rich
structures are the most difficult, and it is desirable to carry
them out as efficiently as possible. For the ITHM protocol, 1.0
equiv of B2pin2 per arene and higher catalyst loadings are
typically used to achieve high conversion. This is because the
HBpin that is generated is much less reactive when dtbpy is the
ligand used for catalysis. This is evident from the data in Figure
9 where borylation of 1,3-diisoproylbenzene with 0.75 equiv
B2pin2 plateaus at 75% conversion. For tmphen ligated catalysts
the situation is different, as Figures 2 and 3 show that the rates
of reaction for HBpin and B2pin2 are similar. Thus, using
tmphen should make it possible to reduce the amount of B2pin2
to 0.5 equiv of per arene. Furthermore, the data in Figure 7
show that the Ir loading can be reduced if the ligand to Ir ratio
is increased.
Scheme 7 shows results for borylating an electron-rich arene
with 0.5 equiv of B2pin2. In this case, complete conversion to
product requires efficient borylation by both B2pin2 and the
HBpin that is produced when B2pin2 is consumed. The results
clearly show that borylation with 2/tmphen greatly improves
the boron atom economy of the reaction when compared to the
more widely used 2/dtbpy combination.21 It is noteworthy that
the Ir catalyst loading with tmphen can be reduced 6-fold when
I
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 7. Low Loading Borylation of an Electron-Rich
Substrate
ACKNOWLEDGMENTS
■
We thank the NSF (GOALI-1012883), Merck Research
Laboratories New Technologies Review & Licensing Commit-
tee (NT-RLC), and the ACS Green Chemistry Institute
Pharmaceutical Roundtable for generous financial support.
REFERENCES
■
(1) Mkhalid, I. A. I.; Barnard, J. H.; Marder, T. B.; Murphy, J. M.;
Hartwig, J. F. Chem. Rev. 2010, 110, 890.
(2) Cho, J. Y.; Tse, M. K.; Holmes, D.; Maleczka, R. E., Jr.; Smith, M.
R., III Science 2002, 295, 305.
(3) Boller, T. M.; Murphy, J. M.; Hapke, M.; Ishiyama, T.; Miyaura,
N.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127, 14263.
(4) (a) Vanchura, B. A., II; Preshlock, S. M.; Roosen, P. C.; Kallepalli,
V. A.; Staples, R. J.; Maleczka, R. E., Jr.; Singleton, D. A.; Smith, M. R.,
III Chem. Commun. 2010, 46, 7724. (b) Tamura, H.; Yamazaki, H.;
Sato, H.; Sakaki, S. J. Am. Chem. Soc. 2003, 125, 16114.
(5) Chotana, G. A.; Vanchura, B. A., II; Tse, M. K.; Staples, R. J.;
Maleczka, R. E., Jr.; Smith, M. R., III Chem. Commun. 2009, 5731.
(6) Ishiyama, T.; Takagi, J.; Hartwig, J. F.; Miyaura, N. Angew. Chem.,
Int. Ed. 2002, 41, 3056.
(7) Hartwig and co-workers have shown that the combination of
precatalyst 3 and dtbpy outperforms complex 1 as well as in situ
generated catalyst from complex 2 and dtbpy in the borylation of
benzene-d6 with B2pin2 at 100 °C (ref 3). In contrast to the current
study, the reactions were carried out with scrupulously dried reagents
and solvents.
setting the ligand to Ir ratio to 2:1 according to the data in
Figure 7.
SUMMARY
■
The following conclusions can be drawn from this study:
(1) Efficiencies of room temperature borylations are sensitive
to order of addition as a function of precatalyst and
boron reagent. Specifically, borylations with [IrCl(cod)]2
are effective only with HBpin, which must be added to
the precatalyst prior to addition of dipyridyl coligands.
(2) At elevated temperatures, order of addition had minimal
influence on borylation efficiency.
(3) The most commonly used dipyridyl ligand, dtbpy, is
outperformed by 3,4,7,8-tetramethyl-1,10-phenanthroline
and in one case by a 1,2-diphosphino benzene,
particularly for borylation of electron-rich substrates at
elevated temperatures.
(8) (a) The following articles on C−H borylation have been
published since the review article cited in ref 1: Kallepalli, V. A.;
Sanchez, L.; Li, H.; Gesmundo, N. J.; Turton, C. L.; Maleczka, R. E.,
Jr.; Smith, M. R., III Heterocycles 2010, 80, 1429. (b) Liskey, C. W.;
Liao, X. B.; Hartwig, J. F. J. Am. Chem. Soc. 2010, 132, 11389.
(c) Robbins, D. W.; Boebel, T. A.; Hartwig, J. F. J. Am. Chem. Soc.
2010, 132, 4068. (d) Yamazaki, K.; Kawamorita, S.; Ohmiya, H.;
Sawamura, M. Org. Lett. 2010, 12, 3978. (e) Itoh, H.; Kikuchi, T.;
Ishiyama, T.; Miyaura, N. Chem. Lett. 2011, 40, 1007. (f) Li, G. Q.;
Kiyomura, S.; Yamamoto, Y.; Miyaura, N. Chem. Lett. 2011, 40, 702.
(g) Liao, X. B.; Stanley, L. M.; Hartwig, J. F. J. Am. Chem. Soc. 2011,
133, 2088. (h) Nguyen, D. H.; Perez-Torrente, J. J.; Lomba, L.;
Jimenez, M. V.; Lahoz, F. J.; Oro, L. A. Dalton Trans. 2011, 40, 8429.
(i) Norberg, A. M.; Smith, M. R., III; Maleczka, R. E., Jr. Synthesis
2011, 857. (j) Ozawa, R.; Yoza, K.; Kobayashi, K. Chem. Lett. 2011, 40,
941. (k) Ros, A.; Estepa, B.; Lopez-Rodriguez, R.; Alvarez, E.;
Fernandez, R.; Lassaletta, J. M. Angew. Chem., Int. Ed. 2011, 50, 11724.
(l) Teraoka, T.; Hiroto, S.; Shinokubo, H. Org. Lett. 2011, 13, 2532.
(m) Cheng, J. H.; Yi, C. L.; Liu, T. J.; Lee, C. F. Chem. Commun. 2012,
48, 8440. (n) Crawford, A. G.; Liu, Z. Q.; Mkhalid, I. A. I.; Thibault,
M. H.; Schwarz, N.; Alcaraz, G.; Steffen, A.; Collings, J. C.; Batsanov,
A. S.; Howard, J. A. K.; Marder, T. B. Chem.Eur. J. 2012, 18, 5022.
(o) Eliseeva, M. N.; Scott, L. T. J. Am. Chem. Soc. 2012, 134, 15169.
(p) Hartwig, J. F. Acc. Chem. Res. 2012, 45, 864. (q) Hitosugi, S.;
Nakamura, Y.; Matsuno, T.; Nakanishi, W.; Isobe, H. Tetrahedron Lett.
2012, 53, 1180. (r) Hume, P.; Furkert, D. P.; Brimble, M. A.
Tetrahedron Lett. 2012, 53, 3771. (s) Liskey, C. W.; Hartwig, J. F. J.
Am. Chem. Soc. 2012, 134, 12422. (t) Liu, T. F.; Shao, X. X.; Wu, Y.
M.; Shen, Q. L. Angew. Chem., Int. Ed. 2012, 51, 540. (u) Ohmura, T.;
Torigoe, T.; Suginome, M. J. Am. Chem. Soc. 2012, 134, 17416.
(v) Robbins, D. W.; Hartwig, J. F. Org. Lett. 2012, 14, 4266.
(w) Roering, A. J.; Hale, L. V. A.; Squier, P. A.; Ringgold, M. A.;
Wiederspan, E. R.; Clark, T. B. Org. Lett. 2012, 14, 3558. (x) Roosen,
P. C.; Kallepalli, V. A.; Chattopadhyay, B.; Singleton, D. A.; Maleczka,
R. E., Jr.; Smith, M. R., III J. Am. Chem. Soc. 2012, 134, 11350. (y) Ros,
A.; Lopez-Rodriguez, R.; Estepa, B.; Alvarez, E.; Fernandez, R.;
Lassaletta, J. M. J. Am. Chem. Soc. 2012, 134, 4573. (z) Wang, C.;
Sperry, J. J. Org. Chem. 2012, 77, 2584. (aa) Bruck, A.; Gallego, D.;
Wang, W. Y.; Irran, E.; Driess, M.; Hartwig, J. F. Angew. Chem., Int. Ed.
2012, 51, 11478. (ab) Partridge, B. M.; Hartwig, J. F. Org. Lett. 2012,
(4) Borylations with tmphen are highly efficient with B2pin2
or HBpin. Thus, reactions with B2pin2 are more atom
economical with tmphen because both boron equivalents
can be transferred.
(5) Polar solvents can be excellent candidates for C−H
borylation.
(6) Ir loadings for in situ generated catalysts can be lowered
significantly if the number of ligand equivalents per Ir is
increased.
(7) Ligands with constrained geometries exhibit superior
performance.
(8) By using appropriate precatalysts, ligands, boron
reagents, solvents, and conditions, substrates that
performed poorly under standard practices could be
borylated efficiently.
ASSOCIATED CONTENT
* Supporting Information
Full characterization, copies of all spectral data, and
experimental procedures. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
shane_krska@merck.com; maleczka@chemistry.msu.edu;
smithmil@msu.edu
■
Notes
The authors declare no competing financial interest.
́ ́
15, 140. (ac) Lopez-Rodríguez, R.; Ros, A.; Fernandez, R.; Lassaletta, J.
J
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
M. J. Org. Chem. 2012, 77, 9915. (ad) Robbins, D. W.; Hartwig, J. F.
Angew. Chem., Int. Ed. 2013, 933. (ae) Liskey, C. W.; Hartwig, J. F. J.
Am. Chem. Soc. 2013, 135, 3375. (af) Lee, C.-I.; Zhou, J.; Ozerov, O.
V. J. Am. Chem. Soc. 2013, 135, 3560. (ag) Kawamorita, S.; Murakami,
R.; Iwai, T.; Sawamura, M. J. Am. Chem. Soc. 2013, 135, 2947.
(ah) Chang, Y.; Lee, H. H.; Kim, S. H.; Jo, T. S.; Bae, C.
Macromolecules 2013, 46, 1754. (ai) Tajuddin, H.; Harrisson, P.;
Bitterlich, B.; Collings, J. C.; Sim, N.; Batsanov, A. S.; Cheung, M. S.;
Kawamorita, S.; Maxwell, A. C.; Shukla, L.; Morris, J.; Lin, Z.; Marder,
T. B.; Steel, P. G. Chem. Sci. 2012, 3, 3505.
(9) (a) Dreher, S. D.; Dormer, P. G.; Sandrock, D. L.; Molander, G.
A. J. Am. Chem. Soc. 2008, 130, 9257. (b) Shultz, C. S.; Krska, S. W.
Acc. Chem. Res. 2007, 40, 1320.
(10) Recently, tmphen has been employed in borylations of sp3 C−H
bonds (refs 8s, u, and ae).
(11) Abboud, J. L. M.; Notario, R. Pure Appl. Chem. 1999, 71, 645.
(12) Conductivity data suggest that the Cl− ligand in (phen)
IrCl(cod) is labile. In CH2Cl2 the Cl− appears to be coordinated, while
in H2O it dissociates to generate (phen)Ir(cod)+: Mestroni, G.;
Camus, A.; Zassinovich, G. J. Organomet. Chem. 1974, 73, 119.
(13) Filipuzzi, S.; Farnetti, E. J. Mol. Catal. A: Chem. 2005, 238, 111.
(14) Vanchura, B. A., II, Ph.D. Thesis, Michigan State University:
East Lansing, MI, 2010.
(15) The electron-donating ability of dipyridyl (dpy) ligands to
transition metals can be estimated from relative νCO frequencies in dpy
metal carbonyl complexes or by evaluating their effects on redox
potentials in dpy complexes. From the average νCO frequencies in
(dpy)Cr(CO)4 complexes(refs 16a and d) and E° values for Ru2+/3+ in
2+
Ru(dpy)3 compounds (refs 16b and c), the electron-donor ability
increases in the order dtbpy < tmp ≪ dmabpy.
(16) (a) Connor, J. A.; Overton, C. J. Organomet. Chem. 1983, 249,
165. (b) Leidner, C. R.; Murray, R. W. J. Am. Chem. Soc. 1984, 106,
1606. (c) Nazeeruddin, M. K.; Zakeeruddin, S. M.; Kalyanasundaram,
K. J. Phys. Chem. 1993, 97, 9607. (d) Johnson, R.; Madhani, H.;
Bullock, J. P. Inorg. Chim. Acta 2007, 360, 3414.
(17) Biscoe, M. R.; Fors, B. P.; Buchwald, S. L. J. Am. Chem. Soc.
2008, 130, 6686.
(18) (a) The origins of tmphen inhibition are not clear; however, bis
and tris(dipyridyl) complexes of IrIII have been structurally
characterized. The failure to observe inhibition with dtbpy would be
consistent with it being a weaker donor than tmphen; (b) Hazell, A.
C.; Hazell, R. G. Acta Crystallogr., Sect.C: Cryst. Struct. Commun. 1984,
40, 806. (c) Andersen, P.; Josephsen, J. Acta Chem. Scand. 1971, 25,
3255. (d) Yoshikawa, N.; Sakamoto, J.; Kanehisa, N.; Kai, Y.;
Matsumura-Inoue, T. Acta Crystallogr., Sect. E: Struct. Rep. Online 2003,
59, m155.
(19) Dervisi, A.; Carcedo, C.; Ooi, L. Adv. Synth. Catal. 2006, 348,
175.
(20) (a) Upadhayaya, R. S.; Shinde, P. D.; Sayyed, A. Y.; Kadam, S.
A.; Bawane, A. N.; Poddar, A.; Plashkevych, O.; Foeldesi, A.;
Chattopadhyaya, J. Org. Biomol. Chem. 2010, 8, 5661. (b) Bekhit, A.
A.; El-Sayed, O. A.; Aboulmagd, E.; Park, J. Y. Eur. J. Med. Chem. 2004,
39, 249.
(21) The efficiency of HBpin in borylations with hemilabile
pyridylhydrazones can be enhanced by increasing the electron-donor
ability of the pyridyl moeity (ref 8ac).
K
dx.doi.org/10.1021/ja400295v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX