Kinetic Study of Carbophilic Lewis Acid Catalyzed Oxyboration and the Noninnocent Role of Sodium Chloride

Article abstract of DOI:10.1021/acs.organomet.5b00939

In the present study, the oxyboration reaction catalyzed by IPrAuTFA in the presence and absence of NaTFA has been examined with kinetic studies, mass spectrometry, and ¹H NMR and ¹¹B NMR spectroscopy. Data from monitoring the reacti

Full text of DOI:10.1021/acs.organomet.5b00939

Article
pubs.acs.org/Organometallics
Kinetic Study of Carbophilic Lewis Acid Catalyzed Oxyboration and
the Noninnocent Role of Sodium Chloride
Joel S. Johnson, Eugene Chong, Kim N. Tu, and Suzanne A. Blum*
Department of Chemistry, University of California, Irvine, California 92697-2025, United States
*
S Supporting Information
ABSTRACT: In the present study, the oxyboration reaction
catalyzed by IPrAuTFA in the presence and absence of NaTFA
has been examined with kinetic studies, mass spectrometry, and
1
11
H NMR and B NMR spectroscopy. Data from monitoring the
reactions over the temperature range from 30 to 70 °C, the
catalyst range from 1.3 to 7.5 mol %, and the NaTFA additive
range from 2.5 to 30 mol % suggests a mechanism that involves rate-determining catalyst generation. Data from additive studies
that replaced NaTFA with NaBARF (BARF = tetrakis[3,5-bis(trifluoromethyl)phenyl]borate) or Bu NTFA as an alternative
4
additive suggest that catalyst quenching from residual NaCl remaining from a one-pot substrate synthesis/reaction method is the
cause of this effect, despite the low solubility of this NaCl byproduct in toluene. Material produced through an alternative,
sodium chloride-free substrate synthesis exhibited faster reaction rates, consistent with a change in rate-determining step that
depended on the substrate synthesis route.
INTRODUCTION
■
Despite major progress in developing B−X element addition
1
−3
reactions to carbon−carbon π systems,
the corresponding
B−O additionoxyborationremained unknown until it
4
yielded to gold catalysis in 2014. The role of the gold catalyst
was proposed to be activation of the π system, a strategy
orthogonal to other metal-catalyzed B−X σ-bond addition
reactions which typically proceed through oxidative addition or
σ-bond metathesis of the B−X bond (Scheme 1). This gold-
functionalized heterocycles. A practical feature of the originally
reported method was the ability to generate the requisite B−O
bond of the starting material from phenols and B-
chlorocatecholborane (ClBcat) in one pot, without the need
to isolate this moisture-sensitive intermediate. The kinetic data
show consistency with the proposed carbophilic activation
pathway but identify an unanticipated kinetic dependence on
the reagents used and route of synthesis in the prior one-pot
substrate-construction step. More specifically, the presence of
NaCl generated during the one-pot substrate synthesis can be
Scheme 1. Traditional Role of Metal Catalyst To Activate
B−X Bonds vs New Role of Metal Catalyst To Activate C−C
π Bonds, in B−X Addition Reactions to C−C π Bonds
13
noninnocent to the reaction,1 analogous to the reported silver
4−16
salt effects on gold catalysis.
Similar to the case for silver
halides, sodium chloride has a low solubility in organic solvents,
yet we show that sufficient quantities remain in solution to
influence reactivity. A possible explanation for the role of
NaTFA additive as a chloride scavenger and transmetalation
partner with gold chloride is thus provided herein. It is
envisioned that these studies will assist in the development of
methods for addition reactions of these previously recalcitrant
B−X bonds to C−C π systems and thus enable rapid access to
heterocyclic borylated building blocks for drug discovery and
materials synthesis. Further, these studies may inform the
catalysis working mechanistic strategy has since been
5
4,6
successfully applied to B−N and B−O
bond addition
reactions to make organoboron heterocyclic building blocks;
yet the rationale for the proposed pathway has been largely
limited to analogy to existing carbophilic Lewis acid
7
−12
catalysis
rather than experimental observations (Scheme
). Additionally, the working mechanistic postulate for the
4
1
oxyboration of alkynes included possible participation of
trifluoroacetate (TFA) additive.
We herein investigate the kinetics of the gold-catalyzed
oxyboration reaction in the synthesis of borylated benzofurans
Received: November 10, 2015
(eq 1) that serve as building blocks for the synthesis of
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
development of dual-metal/metalloid catalytic chemistry for
17
gold that currently requires halide-free conditions.
RESULTS AND DISCUSSION
■
Carbophilic Lewis Acid Activation Mechanistic Pro-
posal with Bifunctional Activation. The proposed mech-
anism for the oxyboration reaction in the synthesis of borylated
benzofurans is shown in Scheme 2. The original mechanistic
catechol borenium adduct 7, which serves as a precursor to
the borylated benzofuran product 2. This pathway would be
increasingly plausible with a less coordinating counterion on
a
Scheme 2. Proposed Mechanism for Oxyboration
25−27
gold.
The counterion TFA is intermediate in coordinating
ability, raising both possibilities.
Initial analysis of the reaction mixture of the oxyboration
reaction in eq 1 prior to reaction completion by ESI-HRMS
revealed the presence of molecular ion peaks consistent with
−
the TFA-coordinated tetraborate component ([M] , m/z
4
25.0820) of intermediate 5, or possibly in the cyclized
+
intermediate 11, and organogold intermediate 8 ([M + H] , m/
z 779.3294). These initial findings suggested the viability of
gold and trifluoroacetate intermediates in the oxyboration
reaction, as proposed in Scheme 2; the simple detection of
compounds, however, does not imply kinetic relevance. To
probe the role of gold and the counterion, we therefore next
determined the reaction order with respect to substrate 1, gold
catalyst, and trifluoroacetate, via kinetic studies of the
transformation of 1 to 2 as described in detail in the following
sections.
Kinetic Studies under Standard Conditions (NaCl
Present). For the first set of kinetic experiments, we used
4
the standard synthesis of substrate 1 (eq 3) from the
a
Originally proposed structures in the catalytic cycle from ref 4 shown
in black; additional data/possible intermediates from this work shown
in purple (Y = BF , SbF , TFA).
4
6
deprotonation of phenol 13 with NaH, followed by the
reaction of the resulting sodium phenolate with ClBcat to
eliminate NaCl byproduct and obtain 1 (>95%). A series of
hypothesis for oxyboration featured IPrAuTFA 4 as a
bifunctional Lewis acidic/basic catalyst, in which the IPrAu+
moiety participates in carbophilic Lewis acid activation of the
alkyne in 1 and the TFA counterion coordinates to boron, to
assemble intermediate 5. Nucleophilic attack on the Au-
activated C−C π bond with the TFA-activated B−O bond in 5,
through transition state 6, generates organogold intermediate 8
1
initial-rate kinetic experiments were conducted using H NMR
spectroscopy to monitor the formation of 2 with respect to
28
time at 50 °C using the ERETIC method. Error is reported as
the standard deviation between triplicate runs.
For these kinetics experiments, one change was made in
comparison to the previously reported synthetic procedure (of
eq 1): the presumed active catalyst, IPrAuTFA 4, was prepared
ex situ via silver salt metathesis between IPrAuCl and AgTFA
and then added to the reaction mixture, rather than producing
it in situ via salt metathesis between IPrAuCl and NaTFA. This
change was made to avoid convolution of the kinetic data via a
concurrent catalyst generation step and to allow for the
manipulation of known amounts of the anticipated active
catalyst 4. We discovered later, however, that speciation of gold
was depended on substrate synthesis (vide infra) and that,
therefore, the route of synthesis of each reagent would thus
deserve additional scrutiny. In the absence of catalyst, no
product formation occurred under otherwise identical con-
ditions. This lack of conversion established that boron alone is
insufficiently electrophilic to activate the alkyne toward
nucleophilic attack in this system, perhaps in part because it
4
17,18
and catB(TFA) 9. Subsequent Au-to-B transmetalation
between 8 and 9 affords the borylated benzofuran 2 and
regenerates the IPrAuTFA catalyst. To our knowledge, our
initial report of oxyboration was the first example of organogold
4
−6,17−19
to boron transmetalation;
been reported.
the reverse reaction had
2
0−24
Given that the forward and reverse
transmetalation reactions have both been reported, it seems
plausible that small differences in the structure of the reagents
affect which side of the transmetalation reaction is thermody-
namically favored. Prior studies in our laboratory identified that
some transmetalation systems with gold and boron are in
1
8
equilibrium.
Carbophilic Lewis Acid Activation Mechanistic Pro-
posal without Bifunctional Activation. Alternatively,
substrate 1 could be undergoing a carbophilic Lewis acid
gold-mediated cyclization, wherein trifluoroacetate serves as an
outer-sphere ion (eq 2). A nucleophilic attack of the oxygen
lone pair to the IPrAu-activated C−C π bond could generate
the key alternative intermediate organogold benzofuran/
6
,29−31
contains three oxygen ligands.
Consistent with the proposed mechanism (Scheme 2), the
reaction exhibited a first-order kinetic dependence on the
B
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
concentration of added IPrAuTFA over the range of 1.3−7.5
mol % (Table 1, entries 1−4; Figure 1a).
°C with a fixed 7.5 mol % of IPrAuTFA and 30 mol % of
⧧
NaTFA gave the activation parameters ΔH = 13 ± 2 kcal
−
1
⧧
−1 −1
⧧
mol , ΔS = −40 ± 5 cal mol K , and ΔG = 26 ± 2 kcal
−1
⧧
mol . The large negative value of ΔS indicates a bimolecular
rate-determining step. To maintain consistency with the
observed zero-order substrate kinetics, this bimolecular step
cannot involve substrate−catalyst binding. In the considered
mechanisms, this data would be consistent with a rate-
determining generation of the active catalyst from a
Table 1. Oxyboration of 1 at Varied Concentrations of
Catalyst and NaTFA: Investigation of Kinetic Order
3
3,34
precatalyst
which would suggest that the active catalyst
kobs (μM s−1
)
was not present from t = 0 after allor with bimolecular
entry
IPrAuTFA (mol %)
NaTFA (mol %)
recombination of 8 and 9 (Scheme 2).
1
2
3
4
5
6
7
1.3
2.5
5.0
7.5
5.0
5.0
5.0
30
30
30
30
2.5
5.0
15
2.1 ± 0.3
5.5 ± 0.5
9.2 ± 1.1
11 ± 1.1
1.4 ± 0.7
4.5 ± 0.6
9.9 ± 0.9
Kinetic Dependence on NaTFA. In order to differentiate
between these hypotheses, the kinetic dependence on the
NaTFA additive was examined next. The reaction exhibited
first-order kinetics with respect to NaTFA over the range of 0−
a
1
5 mol % (Figure 1b). However, additional NaTFA after this
point had no detectable effect on the rate. This absence of a
consistent kinetic effect is seen through comparison of entries 3
a
The reaction is sufficiently slow that integration of small quantities of
product produces a large error.
and 7 in Table 1, showing similar k values between 15 mol %
obs
−1
of NaTFA (kobs = 9.9 ± 0.9 μM s ) and 30 mol % of NaTFA
−1
(
kobs = 9.2 ± 1.1 μM s ). The observed first-order dependence
on NaTFA from 0 to 15 mol % and non-first-order dependence
at 30 mol % are possibly due to the low solubility of NaTFA in
toluene at higher concentrations, which could produce similar
effective concentrations of NaTFA under these two conditions.
The data support the role of sodium, trifluoroacetate, or both in
the rate-determining step.
To investigate the solubility and speciation of NaTFA in
−
1
toluene under the reaction conditions, saturated solutions of
Figure 1. (a) Plot of zero-order kobs (μM s ) vs [IPrAuTFA] showing
1
9
−
1
NaTFA at 50 °C in toluene were examined by F NMR
spectroscopy. First, a saturated sample of only NaTFA in
toluene was examined. This sample produced no observable
signal in the 19F NMR spectrum, indicating low solubility of
this salt in the absence of other reaction components. Second, a
sample was prepared identically, but with a stoichiometric
quantity of starting material 1 present. In this case, sufficient
TFA was solubilized for 19F NMR detection, as observed by a
small signal centered at δ −75 ppm that is within the range of
first-order dependence. (b) Plot of zero-order kobs (μM s ) vs
NaTFA] showing first-order dependence up to 15 mol %.
[
Monitoring of the chemical reaction kinetics over the course
of the full reaction (Figure 2) indicated zero-order substrate
35
literature values of compounds containing TFA. Presumably
the coordination of TFA to the empty p orbital on boron assists
in its solubility in toluene, which could resemble the proposed
double-activation intermediate 5. The viability of coordination-
assisted solubility of NaTFA was further supported in a simpler
model system, wherein a solution of ClBcat and NaTFA in
toluene at 50 °C produced sufficient quantities of
Figure 2. Plot of formation of 2 versus time using 7.5 mol % of
IPrAuTFA and 30 mol % of NaTFA at 70 °C, showing a zero-order
rate dependence on substrate concentration over the course of
approximately 2 half-lives.
−
−
[
(TFA) Bcat] 10 and [(TFA) B] to be detected by MS
2
4
−
(
10, m/z 345; [(TFA) B] , m/z 463) with an isotopic pattern
4
consistent with boron.
Thus, these results establish that trifluoroacetate can
coordinate to boron as shown in intermediates 5 and 9−11;
however, the catalytic relevance of such species is not
necessarily implied. In order to probe the catalytic relevance
of such TFA-coordinated boron species, the kinetics of the
reaction in the absence of added sodium, TFA, or both was
next examined.
Kinetic Studies under Modified Reagent Conditions.
In attempt to examine the accelerating effect of added NaTFA,
a systematic variation of the additive was carried out (Table 2).
Changing the cation of the trifluoroacetate additive from
sodium (Table 2, entry 1) to tetrabutylammonium (Table 2,
entry 2) resulted in no formation of 2. Further examination of
additives by substituting the TFA with the noncoordinating
kinetics over the majority of the reaction (approximately 2 half-
lives) and showed that initial-rate kinetic studies would provide
information similar to that for the full course of the experiment.
For this reason, initial-rate kinetics were then followed for the
majority of the measurements.
Over this IPrAuTFA catalyst loading range (1.3−7.5 mol %),
the reaction displayed these zero-order substrate kinetics in the
32
presence of NaCl byproduct from substrate synthesis. These
saturation kinetics pinpointed a rate-determining step that did
not involve catalyst−substrate binding.
Eyring Analysis. To gain additional insight into the rate-
determining step, we examined the energetic parameters of the
reaction. Rates measured over the temperature range of 30−70
C
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Variation of Additives and Its Effect on
Oxyboration
With substrate 1 in hand using eq 5, the kinetics of NaCl-free
oxyboration reaction was examined (Table 3). With 5 mol % of
Table 3. Oxyboration of NaCl-Free Generated Substrate 1
and the Effect of NaTFA Additive on This Reaction
entry
additive
NaTFA
kobs (μM s−1)
1
2
3
4
9.9 ± 0.9
Bu NTFA
no formation of 2
no formation of 2
no formation of 2
4
a
NaBARF
entry
IPrAuTFA (mol %)
NaTFA (mol %)
k
obs (μM s−1)
no additive
a
BARF = tetrakis[3,5-bis(trifluoromethyl)phenyl]borate.
1
2
3
0
0
no formation of 2
54 ± 6
19 ± 1
5.0
5.0
0
anion BARF (Table 2, entry 3), resulted in no formation of
product. Importantly, no reactivity was observed with the
presumed catalyst IPrAuTFA alone in the absence of additive
5.0
IPrAuTFA catalyst alone in the absence of any additive, a
significantly faster reaction rate for the formation of 2 was
observed with kobs = 54 ± 6 μM s− (Table 3, entry 2), in
comparison to the rate observed with the previously reported
NaCl-present conditions using 5 mol % of IPrAuTFA and the
largest amount of additive, 30 mol % of NaTFA (k = 9.2 ±
.1 μM s ; Table 1, entry 3). These drastically different k
values observed between NaCl-free and NaCl-present con-
ditions support the hypothesis that NaCl is indeed inhibiting
gold catalysis, presumably due to the chloride quenching of the
(
Table 2, entry 4), suggesting that catalyst inhibition was
1
occurring from the beginning of the reaction, presumably by
+
quenching the catalytically active IPrAu species. It was evident
that the NaTFA additive could play a critical role in
overcoming such catalyst inhibition through scavenging residual
chloride and/or serving as a transmetalation partner to
regenerate the active catalyst, thus requiring further evaluation.
We considered the possibility that the presence of NaCl
byproduct, generated during the preparation of 1 (eq 3), might
have a detrimental effect on the reaction. As the IPrAuCl
complex was previously shown to be an incompetent catalyst
obs
−1
1
obs
+
catalytically active IPrAu species into inactive IPrAuCl.
Under the conditions of Table 3, the reaction no longer
displayed zero-order substrate kinetics. This result was
consistent with a change in rate-determining step such that
the rate was no longer limited by catalyst generation.
4
for this reaction, we hypothesized that NaCl, although only
limitedly soluble in toluene, might not be benign. It might
cause catalyst inhibition from counterion exchange with the
IPrAuTFA catalyst into the inactive IPrAuCl complex (eq 4),
such that the rate-determining step under the reaction
In order to examine the kinetic dependence on substrate and
catalyst concentration more clearly under NaCl-free conditions,
the catalyst loading was varied from 0 to 1.25, 2.5, 5, and 7.5
mol % and the fit of the reaction kinetics was examined. Zero-,
first-, and second-order kinetics did not fit the full reaction
course (example, 2.5 mol % of catalyst, average of triplicate
runs, Figure 3a). Specifically, the first-order assessment plot
displayed a slight data curvature that deviated from the linear fit
for both 2.5 and 5 mol % of catalyst Figure 3b,d). The second-
order assessment plot deviated slightly at early reaction times
with 2.5 mol % of catalyst but fit well for the full measured
reaction course for 5.0 mol % (Figure 3c,e). Because the rate of
product formation was slower than that required for a good fit
at early times, these fits may imply a catalyst induction period.
In the case of 7.5 mol %, wherein the reaction was sufficiently
fast that initial rates were not measured, the kinetics best fit first
order (see the Supporting Information for data and fits from
33,34
conditions was regeneration of the active catalyst.
We
4
,5
further hypothesized that the additional NaTFA might be
preventing this catalyst quenching by chloride through the
common ion effect to push the equilibrium of eq 4 to the left to
increase the concentration of the active IPrAuTFA catalyst.
Hence, a chloride-free synthetic route to starting material 1 was
examined next.
IPrAuTFA +
NaCl
⇌ IPrAuCl + NaTFA
active
byproduct from 1 inactive
(4)
Change in Route of Synthesis of Substrate 1: Sodium
Chloride-Free Generation. The treatment of 13 with 1 equiv of
ClBcat (eq 5, top arrow), followed by removal of the HCl
1
.25 and 7.5 mol % catalyst). This deviation at early
conversions precluded using initial rates to determine the
order in the gold catalyst. In the absence of catalyst no product
formation occurred (Table 3, entry 1).
Interestingly, the addition of 5 mol % of NaTFA additive
under NaCl-free reaction conditions lead to a decreased
reaction rate (kobs = 19 ± 1 μM s ; Table 3, entry 3); under
these conditions, the previously accelerating additive had a
decelerating effect. The increased availability of TFA in solution
−1
byproduct under vacuum, afforded 1 in quantitative yield via a
route suitable for small-scale mechanistic experiments; This
approach is less practical from a synthetic standpoint than the
previous substrate synthesis due to the formation of corrosive
HCl gas and the requirement to trap this gas under vacuum.
Unfortunately, the exploration of an alternative synthesis of 1
from the reaction of 13 and HBcat (eq 5, bottom arrow) that
could generate less corrosive hydrogen gas as the byproduct
was unsuccessful.
may shift the equilibrium between the proposed species
−
(TFA)Bcat 9 and [(TFA) Bcat] 10 toward 10 (Scheme 2).
2
The plausibility of such an equilibrium and proposal was
supported by the detection of 10 by MS in a related model
reaction between commercially available ClBcat, as a surrogate
boron electrophile to (TFA)Bcat, and NaTFA. Therefore,
D
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 4. Examination of Gold Catalysts with Varying
Counterions
a
entry
catalyst (5 mol %)
conditions
conversion (%)
1
2
3
4
IPrAuTFA
50 °C, 18 h
50 °C, 18 h
50 °C, 18 h
80 °C, 19 h
23 °C, 44 h
>95
>95
<5
IPrAu(NCMe)BF4
IPrAu(NCMe)SbF6
IPrAu(NCMe)SbF6
77
b
5
IPrAu(NCMe)SbF6
66
a
1
Determined by H NMR spectroscopy using 1,3,5-triisopropylben-
b
zene as an internal standard. 20 mol % of NaTFA added.
catalyzed reaction (Table 4, entry 4). The differing reactivities
observed between these two catalysts can be attributed to the
lesser solubility of IPrAu(NCMe)SbF6 in comparison to
IPrAu(NCMe)BF , on the basis of qualitative assessment of
4
the weaker intensity of IPr ligand signals in d -toluene for the
8
1
former than for the latter complex in H NMR spectra of these
37
reaction mixtures. Presumably, heating the reaction mixture
to 80 °C allows for increased solubility of IPrAu(NCMe)SbF6
in toluene for subsequent participation in the oxyboration
reaction.
Furthermore, the catalytic activity of IPrAu(NCMe)SbF was
6
Figure 3. (a) Plot of formation of 2 versus time using 2.5 mol % of
IPrAuTFA at 50 °C, showing a non-zero-order rate dependence on
substrate concentration. (b) First-order fit plot with 2.5 mol % of
catalyst. (c) Second-order fit plot with 2.5 mol % of catalyst. (d) First-
order fit plot with 5.0 mol % of catalyst. (e) Second-order fit plot with
also revived, even at room temperature (Table 4, entry 5), upon
1
the addition of 20 mol % of NaTFA (66% H NMR
spectroscopic yield of 2 in 44 h). This observation can be
attributed to either NaTFA participating in the proposed
bifunctional activation of the substrate 1, as observed by the
appearance of 11B NMR spectroscopy signal at δ ∼1 ppm in
this sample that may correspond to tetraborate species, or
simply acting as an agent that helps solubilize the catalytically
5
.0 mol % of catalyst.
NaTFA additive is necessary to overcome catalyst inhibition by
NaCl, but an additional 1 equiv of NaTFA respective to gold
can impede catalysis.
Now that the role of NaTFA was established, we next probed
whether the oxyboration reaction can also be catalyzed by
carbophilic Lewis acid catalysis in the absence of assistance by
TFA. For this study, substrate 1 was synthesized under NaCl-
free conditions. Two other (NHC)gold(I) complexes with
weakly coordinating counterions, IPrAu(NCMe)SbF6 and
+
active IPrAu species for carbophilic Lewis catalysis.
Experimental results up to this point suggested that
oxyboration of 1 could plausibly be catalyzed by bifunctional
Lewis acid/base catalysis using IPrAuTFA and/or simply by
+
carbophilic Lewis acid catalysis with the IPrAu species alone.
In order to compare these two proposed pathways, the reaction
of 1 with 5 mol % of IPrAuTFA or IPrAu(NCMe)BF was
4
1
monitored side by side over time by H NMR spectroscopy
IPrAu(NCMe)BF , were prepared using literature proce-
(Figure 4). In the first 380 min of the reaction at ambient
temperature, the IPrAuTFA-catalyzed reaction (pale blue line)
resulted in slightly faster formation of 2 in comparison to the
4
3
6
dures. These counterions would be less likely to coordinate
to boron. These experiments were performed to assess the
accessibility of a mechanism proceeding through intermediate
IPrAu(NCMe)BF -catalyzed reaction (pale red line). Com-
4
7
, which has no inner-sphere assistance in the B−O bond
cleavage step from TFA.
Under NaCl-free conditions, the reaction mixture of 1 with
these gold catalysts (5 mol %) were heated to 50 °C for 18 h
(
Table 4). The catalytic use of IPrAu(NCMe)BF gave full
4
conversion to product 2 under these conditions (Table 4, entry
). These results indicated that the TFA counterion is not
2
required for oxyboration, which suggests that carbophilic Lewis
+
acid catalysis with IPrAu species can mediate this reaction
independent of TFA (Scheme 2 via intermediate 7). In
contrast, the reaction with IPrAu(NCMe)SbF6 gave trace
quantities of 2 (<5%) under the same conditions (Table 4,
entry 3). Heating this reaction mixture at the higher
temperature of 80 °C for 19 h resulted in the formation of 2
Figure 4. Monitoring the formation of 2 over time under NaCl-free
conditions using IPrAuTFA (in blue) and IPrAu(NCMe)BF (in red).
4
Conditions: 23 °C from 0 to 380 min (in paler shade); 50 °C from
380 to 700 min (in brighter shade).
(77%), which is also consistent with a carbophilic Lewis acid
E
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
parable reactivity is observed between these two complexes
when the temperature is raised to 50 °C for an additional 320
min of reaction time (blue and red lines between 380 and 700
min).
Given these results with chloride-free conditions, the
catalysts with and without the possibility of Lewis base
assistance display reactivity sufficiently similar for the
conclusion that Lewis base assistance is not a requirement for
oxyboration; nevertheless, the operation of such an assisted
mechanism with the TFA counterion when it is present cannot
be ruled out.
alternative silver salts have already been established to be
noninnocent. An understanding of the reaction mechanism for
Lewis acid catalyzed oxyboration is similarly anticipated to
assist in the development of new B−X addition reactions to
alkynes and alkenes.
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were conducted
under a nitrogen atmosphere using a glovebox or standard Schlenk
techniques unless stated otherwise. All chemicals were used as received
from commercial sources unless otherwise noted. Sodium trifluor-
oacetate (NaTFA) was dried at 130 °C under vacuum (10 mTorr) for
24 h before use. Toluene, CH Cl , and Et O were purified by passage
In the scenario that the oxyboration mechanism is indeed
operated independently by carbophilic Lewis acid catalysis by
2
2
2
+
IPrAu species without the inner-sphere assistance of the
through an alumina column under argon pressure on a push-still
−
solvent system. d -Toluene was dried over CaH , degassed using three
counterion (Y ), we propose that the reaction follows the
8
2
1
freeze−pump−thaw cycles, and vacuum-transferred prior to use. H,
B, and F NMR spectra were recorded on a Bruker AVANCE-600
revised pathway shown in Scheme 3. One possible trans-
11
19
or Bruker DRX-400 spectrometer. All chemical shifts (δ) are reported
Scheme 3. Revised Proposed Mechanism for Oxyboration
in parts per million (ppm) and referenced to the residual proton
1
solvent peak (δ 7.26 ppm for CDCl , δ 2.08 ppm for d -toluene in H
3
8
NMR spectra). 11B and F NMR spectroscopy experiments are
referenced to the absolute frequency of 0 ppm in the H dimension
19
1
according to the Xi scale. Low- and high-resolution mass spectrometry
data were obtained at the University of California, Irvine.
Synthesis of 2-(Phenylethynyl)phenol (13). A flame-dried 1 L
round-bottom flask equipped with a stir bar was charged with 2-
iodophenol (14.0 g, 63.6 mmol), CuI (1.09 g, 5.71 mmol), and
Pd(PPh ) Cl (1.33 g, 1.89 mmol). In this flask, toluene (320 mL),
3
2
2
diisopropylamine (8.90 mL, 63.6 mmol), and phenylacetylene (10.5
mL, 95.4 mmol) were added sequentially. The reaction mixture was
stirred for 8.5 h at 25 °C. Ethyl acetate (100 mL) was added, and the
organic layer was washed with saturated ammonium chloride solution
(50 mL) and then brine (50 mL) and dried over sodium sulfate. The
organic layer was filtered and concentrated, and the resulting dark
brown oil was purified by flash chromatography (12% EtOAc in
hexanes). The resultant oil was further recrystallized from hexanes to
1
afford a brown crystalline solid (6.82 g, 55%). H NMR (600 MHz,
metalation mechanism is that, after intermediate 12 undergoes
IPrAu-catalyzed cyclization, the resulting two cyclized inter-
mediates 7 could undergo intermolecular transmetalation to
afford product 2 and regenerate the catalyst. While the
concentration of two activated complexes might be low,
analogous gold-activated cyclization intermediates have been
observed by NMR to build up in other systems. An
alternative consistent mechanism that avoids the need to
bring two activated molecules together involves intermediate 7
reacting with starting material 1 in a metal/metalloid transfer
step. This alternative mechanism is also plausible.
CDCl ): δ 7.58−7.56 (m, 2H), 7.45 (dd, J = 7.6, 1.6 Hz, 1H), 7.40−
3
7
.39 (m, 3H), 7.31−7.28 (m, 1H), 7.02 (dd, J = 8.3, 0.8 Hz, 1H), 6.94
(
td, J = 7.5, 1.0 Hz, 1H), 5.89 (s, 1H). This spectrum is in agreement
with previously reported spectral data.
4
Synthesis of IPrAuTFA (4). IPrAuCl (124 mg, 0.200 mmol) and
AgTFA (44.2 mg, 0.200 mmol) were weighed in separate vials.
19
IPrAuCl was dissolved in CH
Cl (1 mL), and this solution was
2 2
transferred to the vial containing AgTFA. Additional CH Cl (1 mL)
2
2
was used to rinse the vial during the transfer. The vial was capped with
a Teflon-coated cap and covered in aluminum foil, and the reaction
mixture was stirred for 30 h in the glovebox. The resulting reaction
mixture was then filtered through glass fiber filter paper. The filtrate
was concentrated in vacuo to afford a colorless solid (110 mg, 79%).
CONCLUSIONS
1
■
This complex was used without further purification, and its5
H
H
1
In conclusion, these results show that counterions capable of
coordinating and activating boron in the carbophilic Lewis acid
catalyzed oxyboration reaction are not required for oxyboration
reactivity. A revised mechanism was proposed that maintains
the original Au-catalyzed alkyne activation but does not require
binding by the counterion. Kinetic measurements identified a
dependence on the route of substrate synthesis. When residual
sodium chloride was present, the catalyst was quenched, despite
the low solubility of this ionic compound in toluene. However,
chloride-induced catalyst inhibition of cationic gold can be
circumvented by using the NaTFA additive, and its
substoichiometric addition was beneficial to reviving the
catalytic activity of gold. Determination of the noninnocent
role of sodium chloride is expected to facilitate catalytic
reaction development because sodium salts are increasingly
attractive for late-metal catalyst generation, given that the
spectrum is in agreement with previously reported spectral data.
NMR (400 MHz, d -toluene): δ 7.13 (t, J = 7.8 Hz, 2H), 6.97 (d, J =
8
7
6
.8 Hz, 4H), 6.30 (s, 2H), 2.47 (septet, J = 6.8 Hz, 4H), 1.36 (d, J =
.8 Hz, 12H), 1.03 (d, J = 6.8 Hz, 12H).
General Procedure for the Preparation of Kinetic Experi-
ment Samples. Example Given for Standard Synthesis of 1 (Eq 3)
and Subsequent Oxyboration Reaction using 7.5 mol % of
IPrAuTFA and 30 mol % of NaTFA. 13 (34.0 mg, 0.175 mmol),
89% NaH (4.7 mg, 0.18 mmol), B-chlorocatecholborane (27.0 mg,
0
.175 mmol), and NaTFA (7.2 mg, 0.053 mmol) were weighed
individually in separate dram vials. Substrate 13 was dissolved in d8-
toluene (250 μL) and transferred to the vial containing NaH. This
mixture was stirred by shaking the vial by hand, allowed to sit for 15
min, and then transferred to the vial containing B-chlorocatecholbor-
ane. Additional d -toluene (2 × 250 μL) was used to rinse the first two
8
vials during the transfer. The reaction mixture was allowed to sit for 30
min with occasional stirring by shaking the vial by hand. To this
mixture was added NaTFA as a suspension in d -toluene (650 μL),
8
F
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
which gave the total volume as 1400 μL. From this boric ester stock
solution, 400 μL was transferred to a J. Young NMR tube and capped
with a rubber septum. This was repeated a total of three times to
create three identical samples. To prepare the 7.5 mol % IPrAuTFA
stock solution, the IPrAuTFA catalyst (9.2 mg, 0.013 mmol) was
(152 mg, 0.43 mmol). In this vial was placed MeCN (2 mL), and the
reaction mixture was stirred for 5 min. Workup identical with that for
the synthesis of IPrAu(NCMe)BF was used to obtain the product as a
4
1
white solid (344 mg, 99%). H NMR (600 MHz, CDCl ): δ 7.59 (t, J
3
= 7.9 Hz, 2H), 7.36 (d, J = 7.8 Hz, 4H), 7.33 (s, 2H), 2.45 (septet, J =
6.9 Hz), 2.33 (s, 3H), 1.32 (d, J = 6.9 Hz, 12H), 1.25 (d, J = 6.8 Hz,
12H). This spectrum is in agreement with previously reported spectral
dissolved in d -toluene (350 μL).
8
Example Given for NaCl-Free Synthesis of 1 (eq 5, Top Arrow)
and Subsequent Oxyboration Reaction using 5 mol % of IPrAuTFA.
3
6
data.
1
3 (34.0 mg, 0.175 mmol) was dissolved in d -toluene (250 μL) and
Examination of Gold Catalysts with Varying Counterions
Using NaCl-Free Synthesis of 1. Preparation of 1. To a cooled
8
placed in a vial containing B-chlorocatecholborane (27.0 mg, 0.175
mmol). Additional d -toluene (250 μL) was used to rinse and transfer.
The solution was allowed to sit for 20 min, and the HCl byproduct
and solvent were removed under vacuum for 40 min. The resulting
solution of 13 (0.265 g, 1.36 mmol) in Et
mL Schlenk tube was slowly added a solution of B-chlorocatecholbor-
ane (0.210 g, 1.36 mmol) in Et O (5 mL). The reaction mixture was
2
O (5 mL) at −78 °C in a 25
8
2
compound 2 was dissolved in d -toluene (1400 μL). From this boric
stirred while it was warmed to room temperature for 1 h. The HCl
byproduct and the solvent were removed in vacuo to quantitatively
8
ester stock solution, 400 μL was transferred to a J. Young NMR tube
and the tube was capped with a rubber septum. This was repeated a
total of three times to create three identical samples. To prepare the 5
mol % IPrAuTFA stock solution, the IPrAuTFA catalyst (6.1 mg,
yield 1 as a yellow oil, which was directly used to prepare a 0.272 M
1
stock solution of 1 in d
NMR (600 MHz, d
-toluene using a 5 mL volumetric flask. H
8
-toluene): δ 7.38 (dd, J = 7.7, 1.7 Hz, 1H), 7.24−
8
0
.0088 mmol) was dissolved in d -toluene (350 μL).
General Procedure for Kinetic Experiments. A J. Young NMR
7.20 (m, 2H), 6.99−6.97 (m, 1 H), 6.92 (td, J = 7.6, 1.7 Hz, 1H),
8
6.86−6.84 (m, 3H), 6.83−6.80 (m, 2H), 6.75 (td, J = 7.6, 1.2 Hz, 1H),
11
tube containing 2 (from the above procedure) and one gastight
syringe with 100 μL of the IPrAuTFA solution, capped with a rubber
stopper, were removed from the glovebox. The rubber septum on the
J. Young NMR tube was wrapped in Parafilm. The sample in the J.
Young NMR tube was immediately transported to the NMR
spectrometer, which underwent temperature calibration before sample
injection. The IPrAuTFA solution was injected using the gastight
syringe into the J. Young NMR tube. This tube was shaken and
inserted into the NMR spectrometer. Acquisition of spectra
immediately followed. The acquisition time was set to 6 s. A 90°
pulse was used. The line broadening was set to 1 Hz. There was one
scan taken per experiment, and there were no dummy scans . A total of
6.69−6.65 (m, 2H). B NMR (193 MHz, d -toluene): δ 23.1. These
8
4
spectra are in agreement with previously reported spectral data.
General Procedure for Catalyst Screen. A dram vial was charged
with the gold complex (0.0068 mmol, 5 mol %). In this vial were
placed 1 (500 μL, 0.136 mmol, 0.272 M) and 1,3,5-triisopropylben-
zene (11.0 μL, 0.0453 mmol) as an internal standard using a gastight
syringe. The suspension was mixed and transferred to a J. Young NMR
tube using a Pasteur pipet. The reaction was either kept at ambient
temperature or heated at a specified temperature. The reaction was
1
monitored by H NMR spectroscopy over time by observing growing
signals of the product: doublet centered at δ 8.35 ppm and multiplet
centered at δ 8.14 ppm.
5
0 experiments were conducted sequentially with 34 s delays between
experiments.
Synthesis of Bu NTFA. The synthetic procedure was adapted
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.5b00939.
■
4
S
*
38
from the literature. Trifluoroacetic acid (0.04 mL, 0.5 mmol) was
added dropwise into a 10 mL round-bottom flask containing trimethyl
phosphate (0.7 mL, 6.0 mmol). To this stirred solution was added
Bu NBr (161 mg, 0.500 mmol). The reaction mixture was stirred at 60
4
°
C overnight. Excess trimethyl phosphate was removed via vacuum
Additional kinetic plots and NMR spectra (PDF)
distillation at 85 °C. The resultant residue was dissolved in CH Cl (2
mL), followed by addition of 25 drops of a 10% (w/v) NH OH
aqueous solution. Once the pH of the solution was neutralized (pH 7),
water (2 mL) was added. The organic layer was separated, and the
aqueous layer was extracted with CH Cl (2 × 2 mL). The combined
2
2
4
AUTHOR INFORMATION
Corresponding Author
*E-mail for S.A.B.: blums@uci.edu.
■
2
2
organic layers were concentrated and dried under high vacuum at 100
C overnight. The product was isolated as a beige solid (133 mg, 75%)
Notes
°
1
The authors declare the following competing financial
interest(s): US Pat. No. 9,238,661.
and used without further purification. H NMR (600 MHz, CDCl ): δ
3
3
.32−3.27 (m, 8H), 1.68−1.61 (m, 8H), 1.44 (sextet, J = 7.4 Hz, 8H),
1
.01 (t, J = 7.4 Hz, 12H). 19F NMR (376 MHz, CDCl ): δ −75.1.
3
ACKNOWLEDGMENTS
This work was supported by a grant from the NIH
1R01GM098512-01) and by the University of California,
These spectra are in agreement with previously reported spectral
■
(
38
data.
Synthesis of IPrAu(NCMe)BF . The synthetic procedure was
adapted from the literature. A dram vial equipped with a magnetic
stir bar was charged with IPrAuCl (300 mg, 0.480 mmol) and AgBF4
4
36
Irvine.
(
94 mg, 0.48 mmol). In this vial was placed MeCN (2 mL), and the
REFERENCES
reaction mixture was stirred for 5 min. For workup, the reaction
solvent was separated from the AgCl byproduct and concentrated in
vacuo. The resulting crude residue was diluted in CH Cl (2 mL) and
filtered through a short plug of silica gel to remove residual silver salts.
Additional CH Cl (2 × 2 mL) was used to ensure complete elution of
the product from the silica gel. After concentration of the filtrate in
■
(1) Miyaura, N. Hydroboration, Diboration, Silylboration, and
Stannylboration. In Catalytic Heterofunctionalization; Togni, A.,
Grutzmacher, H., Eds.; Wiley-VCH: Weinheim, Germany, 2001; pp
2
2
̈
1−45.
2
2
(2) Suginome, M. Chem. Rec. 2010, 10, 348−358.
(3) Barbeyron, R.; Benedetti, E.; Cossy, J.; Vasseur, J.-J.; Arseniyadis,
S.; Smietana, M. Tetrahedron 2014, 70, 8431−8452.
(4) Hirner, J. J.; Faizi, D. J.; Blum, S. A. J. Am. Chem. Soc. 2014, 136,
4740−4745.
1
vacuo, the product was obtained as a white solid (340 mg, 99%). H
NMR (600 MHz, CDCl ): δ 7.58 (t, J = 7.9 Hz, 2H), 7.37 (s, 2H),
3
7
.35 (d, J = 7.8 Hz, 4H), 2.44 (septet, J = 6.9 Hz), 2.41 (s, 3H), 1.30
(
d, J = 6.9 Hz, 12H), 1.25 (d, J = 6.8 Hz, 12H). This spectrum is in
36
agreement with previously reported spectral data.
(5) Chong, E.; Blum, S. A. J. Am. Chem. Soc. 2015, 137, 10144−
10147.
(6) Tu, K. N.; Hirner, J. J.; Blum, S. S. Org. Lett. 2016, 18, 480.
Synthesis of IPrAu(NCMe)SbF . The synthetic procedure was
6
36
adapted from the literature. A dram vial equipped with a magnetic
stir bar was charged with IPrAuCl (250 mg, 0.40 mmol) and AgSbF6
(7) Dorel, R.; Echavarren, A. M. Chem. Rev. 2015, 115, 9028−9072.
G
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
8) Hashmi, A. S. K.; Frost, T. M.; Bats, J. W. Org. Lett. 2001, 3,
769−3771.
9) Hashmi, A. S. K.; Enns, E.; Frost, T. M.; Schaf
Rominger, F. Synthesis 2008, 2008, 2707−2718.
10) Zhang, Y.; Xin, Z.-J.; Xue, J.-J.; Li, Y. Chin. J. Chem. 2008, 26,
461−1464.
11) Hashmi, A. S. K.; Ramamurthi, T. D.; Rominger, F. Adv. Synth.
Catal. 2010, 352, 971−975.
12) Abbiati, G.; Marinelli, F.; Rossi, E.; Arcadi, A. Isr. J. Chem. 2013,
3, 856−868.
13) Hirner, J. J.; Roth, K. E.; Shi, Y.; Blum, S. A. Organometallics
012, 31, 6843−6850.
14) Shi, Y.; Peterson, S. M.; Haberaecker, W. W.; Blum, S. A. J. Am.
Chem. Soc. 2008, 130, 2168−2169.
15) Lu, Z.; Han, J.; Hammond, G. B.; Xu, B. Org. Lett. 2015, 17,
534−4537.
16) Wang, D.; Cai, R.; Sharma, S.; Jirak, J.; Thummanapelli, S. K.;
3
(
̈
er, S.; Frey, W.;
(
1
(
(
5
(
2
(
(
4
(
Akhmedov, N. G.; Zhang, H.; Liu, X.; Petersen, J. L.; Shi, X. J. Am.
Chem. Soc. 2012, 134, 9012−9019.
(
17) Hirner, J. J.; Shi, Y.; Blum, S. A. Acc. Chem. Res. 2011, 44, 603−
6
13.
(
(
18) Hirner, J. J.; Blum, S. A. Tetrahedron 2015, 71, 4445−4449.
19) For transmetalation from gold to other metals, see ref 16 and:
Shi, Y.; Roth, K. E.; Ramgren, S. D.; Blum, S. A. J. Am. Chem. Soc.
009, 131, 18022−18023.
20) Forward, J. M.; Fackler, J. P., Jr.; Staples, R. J. Organometallics
995, 14, 4194−4198.
21) Partyka, D. V.; Zeller, M.; Hunter, A. D.; Gray, T. G. Angew.
Chem., Int. Ed. 2006, 45, 8188−8191.
22) Partyka, D. V.; Zeller, M.; Hunter, A. D.; Gray, T. G. Inorg.
Chem. 2012, 51, 8394−8401.
23) Hofer, M.; Gomez-Bengoa, E.; Nevado, C. Organometallics
014, 33, 1328−1332.
24) Smith, D. A.; Ros ç a, D.-A.; Bochmann, M. Organometallics 2012,
2
(
1
(
(
(
2
(
3
(
(
(
1, 5998−5600.
25) Strauss, S. H. Chem. Rev. 1993, 93, 927−942.
26) Beck, W.; Suenkel, K. Chem. Rev. 1988, 88, 1405−1421.
27) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066−
090.
2
(
28) Akoka, S.; Barantin, L.; Trierweiler, M. Anal. Chem. 1999, 71,
554−2557.
29) Hansmann, M. M.; Melen, R. L.; Rudolph, M.; Rominger, F.;
Wadepohl, H.; Stephan, D. W.; Hashmi, A. S. K. J. Am. Chem. Soc.
015, 137, 15469−15477.
30) Warner, A. J.; Lawson, J. R.; Fasano, V.; Ingleson, M. J. Angew.
Chem., Int. Ed. 2015, 54, 11245−11249.
31) Chen, C.; Kehr, G.; Frohlich, R.; Erker, G. J. Am. Chem. Soc.
010, 132, 13594−13595.
32) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic
Chemistry; University Science Books: Sausalito, CA, USA, 2006; p 396.
33) Sanford, M. S.; Ulman, M.; Grubbs, R. H. J. Am. Chem. Soc.
001, 123, 749−750.
34) Rosner, T.; Pfaltz, A.; Blackmond, D. G. J. Am. Chem. Soc. 2001,
23, 4621−4622.
35) Wang, F.-P.; Chen, D.-L.; Deng, H.-Y.; Chen, Q.-H.; Liu, X.-Y.;
Jian, X.-X. Tetrahedron 2014, 70, 2582−2590.
36) de Fremont, P.; Marion, N.; Nolan, S. P. J. Organomet. Chem.
2
(
2
(
(
̈
2
(
(
2
(
1
(
(
́
2
009, 694, 551−560.
(
37) See the Supporting Information.
(
38) Jeon, J. Y.; Varghese, J. K.; Park, J. H.; Lee, S.-H.; Lee, B. Y. Eur.
J. Org. Chem. 2012, 2012, 3566−3569.
H
DOI: 10.1021/acs.organomet.5b00939
Organometallics XXXX, XXX, XXX−XXX