Br?nsted Acid and H-Bond Activation in Boronic Acid Catalysis

Article abstract of DOI:10.1002/chem.202001902

Boronic acid catalysis has emerged as a mild method for promoting a wide variety of reactions. It has been proposed that the mode of catalysis involves Lewis acid or covalent activation of hydroxyl groups by boron, but limited mechanistic evidence exists. In this work, representative boronic acid catalyzed reactions of alcohols and oximes have been reinvestigated. A series of control experiments with boronic and Br?nsted acids were interpreted along with correlations between their reactivity and their acidity measured by the Gutmann–Beckett method. Overall, it was concluded that the major modes of catalysis involve either dual H-bond catalysis or Br?nsted acid catalysis. Strong Br?nsted acids were shown to be generated in situ from covalent assembly of the boronic acids with hexafluoroisopropanol, explaining why the solvent had such a major impact on the reactivity. This new insight should guide the future development of boronic acid catalysis, where the diverse and solvent-specific nature of catalytic modes has been overlooked.

Full text of DOI:10.1002/chem.202001902

Accepted Article
Accepted Article
Title: Boronic Acids as Lewis Acid Catalysts: Myth or Reality?
Authors: Shaofei Zhang, David Lebœuf, and Joseph Moran
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202001902
Link to VoR: https://doi.org/10.1002/chem.202001902
01/2020

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
Brønsted Acid and H-bond Activation in Boronic Acid Catalysis
Shaofei Zhang, David Lebœuf and Joseph Moran*
S. Zhang, Dr. D. Lebœuf and Prof. Dr. J. Moran
University of Strasbourg, CNRS, ISIS UMR 7006
67000 Strasbourg, France
E-mail: moran@unistra.fr
Supporting information for this article is given via a link at the end of the document.
Abstract: Boronic acid catalysis has emerged as a mild method for
promoting a wide variety of reactions. It has been proposed that the
mode of catalysis involves Lewis acid or covalent activation of
hydroxyl groups by boron, but limited mechanistic evidence exists.
Here, we reinvestigate representative boronic acid-catalyzed
reactions of alcohols and oximes. We interpret a series of control
experiments with boronic and Brønsted acids along with correlations
between their reactivity and their acidity measured by the Gutmann-
Beckett method. Overall, we conclude that the major modes of
catalysis involve either dual H-bond catalysis or Brønsted acid
catalysis. We show that strong Brønsted acids are generated in situ
from
covalent
assembly
of
the
boronic
acids
with
hexafluoroisopropanol, explaining why the solvent had such a major
impact on the reactivity. This new insight should guide the future
development of boronic acid catalysis, where the diverse and solvent-
specific nature of catalytic modes has been overlooked.
Boronic acids have emerged as a promising class of catalysts[1]
that
enable
dehydrative
nucleophilic
substitution
or
rearrangements of alcohols,[2] Beckmann rearrangement of
oximes,[3] and various reactions involving either carboxylic acids[4]
or epoxides[5] under mild conditions. In depth studies into the
catalytic mechanism have been performed in the case of
carboxylic acids,[4c,4h] whereas only preliminary mechanistic
evidence exists for the reactions of alcohols and oximes.[2d,2j,3b]
The arylboronic acid catalyst systems required for reactions
involving alcohols and oximes (B1-B3, Scheme 1) are
substantially more electrophilic than those used for the activation
of carboxylic acids, the former requiring either multiple electron-
withdrawing groups, cationic boronic acids or complexation with
highly electronically deactivated diols. Another critical parameter
in these reactions is the solvent. Our group as well as many others
have pointed out the enabling effect of solvents, such as
hexafluoroisopropanol (HFIP)[6-7] and nitromethane (MeNO2),[8] on
Brønsted and Lewis acid-catalyzed reactions through the
formation of an H-bond network. In the case of HFIP, we
emphasized that the role of the catalysts was to significantly
increase the acidity of an H-bond cluster of HFIP, which was the
true catalytically active species.[7i,7m] Our reflections on the
reactivity of arylboronic acids started during our investigations on
the TfOH-catalyzed arylative ring-opening of unactivated
cyclopropanes in HFIP (Table 1, entry 1.1).[7m] We were puzzled
by the reactivity of two catalyst systems typically used for
activation of alcohols and oximes (B1 and B3), as they were able
to trigger the ring-opening of phenylcyclopropane to generate
product 1 (entries 1.2 and 1.5). Given the absence of an OH
functional group in the substrate and the lack of Frustrated Lewis
Scheme 1. Representative boronic acid catalyzed transformations of alcohols
and oximes.
Pairs,[9] we surmised that it would be unlikely that this reaction
features covalent or Lewis acid catalysis. One plausible
mechanism would entail catalysis by a Brønsted acid, which
would be generated through the coordination of the boronic acid
(in the case of B1) or the boronate ester (in the case of B3) with
HFIP.[10] Indeed, the presence of 2,6-di-tert-butylpyridine, a bulky
Brønsted base commonly used to distinguish between Lewis and
Brønsted acid catalysis,[11] completely shut down the reactivity in
both cases, consistent with Brønsted acid catalysis (entries 1.3
and 1.6). Additionally, Brønsted acids weaker than TfOH were not
effective catalysts (entries 1.10-1.13), leading us to suspect that
very strong Brønsted acids might have been produced from the
boronic acids under the reaction conditions. These observations
led us to wonder whether certain previously reported boronic acid
catalyzed reactions might also be the result of Brønsted
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
Table 1. Comparison between boronic and Brønsted acids for the catalytic ring-
12.1 ppm further downfield than the signal generated due to the
same experiment carried out with B(C6F5)3•H2O (78.2 ppm). The
influence of B3 in HFIP/MeNO2 (4:1) thus appears to be stronger
than B(C6F5)3•H2O, and comparable to that of HCl (90.7 ppm) or
CSA (91.3 ppm), in close agreement with the ability of B3 to
promote the opening of unactivated cyclopropanes (see Table 1).
This is very different from the situation in toluene, where CSA shifts
the signal of TEPO nearly 20 ppm further downfield than B3. At
least one species produced from the components of B3,
presumably a highly electrophilic hexafluoropinacol boronate ester,
can serve to generate a strong Brønsted acid in HFIP. It should be
highlighted that the diol component of B3 is essential here, as no
shift was observed with the boronic acid alone (B3’). In line with
these suggestions, it was established that a strongly Brønsted
acidic species is formed from the covalent assembly of
pentafluorophenylboronic acid and oxalic acid, another electron-
poor bidentate species.[14] Likewise, the shift produced by the
addition of B1 (84.5 ppm) is significantly higher than that produced
by B(C6F5)3•H2O, congruent with its demonstrated reactivity (see
Table 1). Although this experiment does not distinguish whether
Brønsted or Lewis acids are causing the observed shifts per se,
strong boron Lewis acids such as B(C6F5)3 are well known to
rapidly react with adventitious water to form hydrates that are
strong Brønsted acids.[15] In a similar way, the large magnitude of
the observed shift in the 31P NMR means that strong Brønsted
acids are almost certainly produced by B1 and B3 in HFIP. For
these reasons, the mild pKa values established for these boronic
acids in DMSO or water cannot be transposed to reactions carried
out in HFIP and HFIP/MeNO2 in order to predict reactivity. Indeed,
none of the shifts corresponding to the boronic acids in the
absence of HFIP exceeded 61.5 ppm (see the Supporting
Information for details). Lastly, we found that boronic acid B2 (66.8
ppm in HFIP/MeNO2 4:1) induces a shift in the 31P NMR
comparable to oxalic acid (69.5 ppm), in agreement with the lack
of reactivity observed in the ring-opening transformation. Of note,
in the absence of the diol component of B3, the shift is similar to
that of B2 (66.9 ppm).
opening hydroarylation of phenylcyclopropane.
Entry
Catalyst
Yield [%][a]
1.1
TfOH
B1
97
1.2
68
1.3
B1
<1[b]
<1
96
1.4
B2
1.5
B3
1.6
B3
<1[b]
1.7
HCl
94
1.8
H2SO4
CSA
92
1.9
95
1.10
1.11
1.12
1.13
CF3CO2H
(COOH)2
CH3CO2H
36
12
<5
<5
B(OH)3
[a] Yields were determined by 1H NMR using hexamethyldisiloxane as an
external standard. [b] In the presence of 15 mol% 2,6-di-tert-butylpyridine. CSA
= camphorsulfonic acid.
acid catalysis.
Herein, we address the above issues by comparing boronic and
Brønsted acid catalysts with nine formerly disclosed boronic acid
catalyzed reactions spanning the four representative studies
summarized in Scheme 1. We will provide a correlation between
the reactivity profiles generated by the various boronic and
Brønsted acids and their influence on triethylphosphine oxide
(Gutmann-Beckett method).[12] We found that in eight of the nine
reactions that were probed, Brønsted acid or H-bond catalysis,
rather than a Lewis acid or covalent activation mechanism, is
likely the main source of reactivity.
At the outset, to correlate the catalytic effects observed for boronic
and Brønsted acids with their physicochemical properties, we
elected to compare their interaction with triethylphosphine oxide
(TEPO), a representative weak Lewis base, in various solvents
(Figure 1). The strength of the interaction between the additive and
TEPO can be inferred from the change in chemical shift in the
corresponding 31P NMR spectrum, compared to that obtained from
a control experiment performed in the absence of additive and
HFIP (i.e., 46.1 ppm in toluene-d8). The control experiments
confirmed our hypothesis regarding the pivotal role of the solvent.
Indeed, in the presence of MeNO2 or HFIP/MeNO2 (4:1), we
observed substantial shifts in the 31P NMR signal (53.0 and 67.1
ppm, respectively), indicating that the solvents are non-innocent in
the activation of alcohols, even in the absence of Lewis or
Brønsted acids. This does not come as a surprise since we and
others have noticed similar reactivity trends in the past for HFIP
and MeNO2. The former solvent forms aggregates that are
excellent H-bond donors,[7,8] while the latter templates the
formation of similar aggregates through interactions with
molecules such as water.[8a,13] In the case of the HFIP/MeNO2
mixture, adding 2,6-di-tert-butylpyridine did not affect the 31P NMR
shift, confirming that, without catalyst, no Brønsted acid is
generated. Mixing catalyst system B3 with TEPO gave rise to a
few new resonances, the highest frequency of which (90.3 ppm) is
Figure 1. Gutmann-Beckett plot showing the influence of an additive (3 equiv)
on TEPO (1 equiv) in HFIP/MeNO2 (4:1) as expressed by the variations in
chemical shift of the highest frequency signal observed in the 31P NMR spectrum
when compared to the reference TEPO in toluene-d8. B3’ = B3 in the absence
of diol.
Based on these results, the first set of transformations examined
was the Friedel-Crafts reaction of primary benzylic alcohols 2a-2c
catalyzed by hexafluoroantimonate salt B1 in HFIP/MeNO2
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
(4:1).[2g] The three substrates reacted as previously described
(Table 2, entry 2.1) but not in the presence of 2,6-di-tert-
butylpyridine (entry 2.2).[16] Replacing the cationic boronic acid by
Brønsted acids spanning a wide pKa range revealed that, in all
cases, the boronic acid may be substituted with a Brønsted acid
with similar results, although the strength of the required acid
varies with the nature of the electronic substituents on the
benzylic alcohol. The conjugate acid of the boronic acid catalyst,
HSbF6, catalyzed the reaction of all three substrates (entry 2.3),
albeit in lower yields. In the case of benzylic alcohols 2a and 2b,
the weak Brønsted acid oxalic acid was surprisingly able to
promote the reaction under otherwise identical conditions (entries
2.9-2.11). Electronically deactivated alcohol 2c needed a stronger
Brønsted acid but its transformation could still be effectively
catalyzed by H2SO4 (entry 2.6). Previously, the possibility of
Brønsted acid catalysis was ruled out on the basis of: 1) the lack
of reactivity of 2b with CF3CO2H and 2) the fact that a different
boronic acid (B2), which has a comparable pKa in H2O and DMSO
to the catalyst used (B1),[2h,17] does not facilitate the reaction in
HFIP/MeNO2. However, as the data provided in Figure 1
emphasized, B2 and CF3CO2H do not produce a suitably strong
Brønsted acid in HFIP/MeNO2 and thus are not suitable points of
reference. The reduced reactivity of B3 compared to B1, might be
explained by the sequestration of the diol component of B3 by H-
bonding with an excess of benzylic alcohol substrate,[18] leaving
behind boronic acid B3’ as a much weaker catalyst as expected
from Figure 1.
transformations catalyzed by B3 were efficient in each instance
(Table 3, entry 3.3) but not in the presence of 2,6-di-tert-
butylpyridine (entry 3.4). In the case of 4a, none of the Brønsted
acids tested promoted the rearrangement. In the catalytic
experiments with B3, the reaction kinetics for 4a did display an
induction period consistent with a slow catalyst formation and the
previously proposed covalent mechanism. However, since all
prior optimization and mechanistic studies were conducted on 4a,
this might have led the authors to conclusions about the
mechanism which do not hold for other substrates. For substrates
4b-4d, a screen of Brønsted acids revealed several to be capable
of promoting the reaction with either a similar efficiency or even
more effectively than the B3 catalyst system (entries 3.5-3.9).
CSA proved to be particularly effective in the case of substrates
4b and 4c, while CF3CO2H was able to promote the reaction with
4d (entries 3.8-3.9). Indeed, the kinetic profile of the reaction of
oxime 4b catalyzed by B3 proved to be nearly identical to that
catalyzed by CSA, with no observation of an induction period
expected for a mechanism involving slow formation of a
catalytically active acyl oxime species (see Supporting
Information for details). These experiments thus support Brønsted
acid catalysis, rather than covalent catalysis, being the dominant
mechanism for substrates 4b-4d. Covalent activation and
Brønsted acid catalysis therefore appear to be competitive
catalytic mechanisms in the Beckmann rearrangement, with three
of the four representative substrates dominated by Brønsted acid
catalysis.
Table 2. Comparison between boronic and Brønsted acids for the catalytic
Table 3. Comparison between boronic acids and Brønsted acids for the catalytic
dehydrative Friedel-Crafts reactions of benzylic alcohols.[a]
Beckmann rearrangement.[a]
Yield 3a
[%][b]
Yield 3b
[%][b]
Yield 3c
[%][c]
Entry
Catalyst
2.1
B1
95
<1[d]
<1
62
79
98
90
95
95
92
92
85
15
94
<1[d]
<1
19
40
98
92
94
96
<1
47
<1
<1
40
<1[e]
<1
5
2.2
B1
Yield 5a
[%][b]
Yield 5b
[%][c]
Yield 5c
[%][c]
Yield 5d
[%][b]
2.3
B2
Entry
Catalyst
2.4
B3
3.1
B1
<1
<1
98
<1[d]
<1
<1
<1
<1
<1
<1
<1
<1
20
<1
94
<1[e]
40
16
<1
90
36
<1
<1
<1
<10
<1
85
<1
<1
45
<1[d]
<1
<1
<1
20
78
<1
<1
<1
2.5
HSbF6∙6H2O
TfOH
45
95
<1
90
<1
<1
<1
<1
<1
3.2
B2
2.6
3.3
B3
2.7
HCl
3.4
B3
<1[e]
2.8
H2SO4
CSA
3.5
TfOH
HCl
96
2.9
3.6
26
2.10
2.11
2.12
2.13
CF3CO2H
(COOH)2
CH3CO2H
B(OH)3
3.7
H2SO4
CSA
<1
86
3.8
3.9
CF3CO2H
(COOH)2
CH3CO2H
B(OH)3
21
3.10
3.11
3.12
<1
<1
<1
[a] Yields were determined by 1H NMR using hexamethyldisiloxane as an
external standard.. [b] 10 mol% catalyst, 50 ºC, 24 h. [c] 20 mol% catalyst, 80
ºC, 48 h. [d]With 15 mol% 2,6-di-tert-butylpyridine. [e]With 30 mol% 2,6-di-tert-
butylpyridine.
[a] Yields were determined by 1H NMR using hexamethyldisiloxane as an
external standard. [b] 30 mol% catalyst and perfluoropinacol, 80 ºC, 24 h. [c] 5
mol% catalyst and perfluoropinacol, 25 ºC, 24 h. [b] 30 mol% catalyst and
perfluoropinacol, 80 ºC, 24 h. [d] With 45 mol% 2,6-di-tert-butylpyridine. [e] With
7.5 mol% 2,6-di-tert-butylpyridine.
The second transformation that we examined was the Beckmann
rearrangement of oximes to amides catalyzed by B3. Previous
studies on acetophenone oxime (4a) supported a mechanism
involving slow formation of a catalytically competent O-boronyl
oxime ester.[3b] We tested three representative aryl-alkyl (4b),
aryl-aryl (4c) and alkyl-alkyl oximes (4d) from the original
publication, as well as 4a, all under the reported conditions. The
The third reaction that we analyzed was the carbocyclization of
allylic alcohols, reported to be catalyzed by boronic acid B2 in
MeNO2.[2f] The transformation occurred as described (Table 4,
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
entry 4.2), but was inefficient in the presence of 2,6-di-tert-
butylpyridine (entry 4.3). Likewise, B1 and B3 have a relatively
similar efficacy (entries 4.1 and 4.4). In contrast, a Brønsted acid
as weak as oxalic acid enabled the reaction under otherwise
identical conditions (entries 4.11). Both oxalic acid and boronic
acids are known to act as dual H-bond donors,[19] and likely act as
H-bond catalysts here. The possibility of H-bond activation may
have been previously overlooked, since in the original disclosure,
control experiments designed to compare B2 to Brønsted acids
were performed with p-TsOH only (entry 4.8).
5.4
B3
<1
6
5.5
TfOH
5.6
HCl
16
44
85
18
11
<1
<1
5.7
H2SO4
CSA
5.8
5.9
CF3CO2H
(COOH)2
CH3CO2H
B(OH)3
5.10
5.11
5.12
[a] Yields were determined by 1H NMR using hexamethyldisiloxane as an
external standard. [b] Yield reported in reference 2d. [c] In the presence of 15
mol% 2,6-di-tert-butylpyridine.
Table 4. Comparison between boronic acids and Brønsted acids for the catalytic
carbocyclization of allylic alcohols.
To summarize, this study sheds light on the activation mode of
boronic acid catalysis of alcohols and oximes, showing that
Brønsted acid and H-bond catalysis, rather than Lewis acid or
covalent activation, are likely responsible for the observed
reactivity in nearly all the representative examples studied.
Specifically, catalysts B1 and B3 produce strong Brønsted acids
in the presence of HFIP, and catalyst B2 likely acts as a H-bond
catalyst in MeNO2. We have concluded the above on the basis of
the following key findings: 1) Boronic acids B1 and B3 are able to
rupture unactivated cyclopropanes in HFIP. 2) The hindered
Brønsted base 2,6-di-tert-butylpyridine entirely inhibits reactivity
of the boronic acids in the reactions investigated. 3) Boronic acid
catalyzed reactions of alcohols and oximes can be promoted by
Brønsted acids or H-bond catalysts. 4) Brønsted acids that
facilitate those transformations exert a deshielding influence on
TEPO comparable to the boronic acids that promote the same
transformations. 5) For three of the four representative oximes
studied, the data collected are inconsistent with a covalent
mechanism. Moving forward, these insights should be useful for
the rational design of second-generation catalysts for dehydrative
nucleophilic substitution or for oxime rearrangements, whether or
not they are based on boron. Finally, this work cautions that a
wide range of control experiments are necessary to rule out a
catalytic role for Brønsted acid and H-bond donors, taking into
consideration the important numerous roles played by the solvent.
Entry
Catalyst
Yield [%]
4.1
B1
72
52
<1[a]
67
81
<1
36
24
<1
12
50
12
<1
4.2
B2
4.3
B2
4.4
B3
4.5
TfOH
HCl
4.6
4.7
H2SO4
p-TsOH
CSA
4.8
4.9
4.10
4.11
4.12
4.13
CF3CO2H
(COOH)2
CH3CO2H
B(OH)3
[a] Yields were determined by 1H NMR using hexamethyldisiloxane as an
external standard. [b] In the presence of 15 mol% 2,6-di-tert-butylpyridine.
The last process explored was the 1,3-allylic transposition of 1,1-
diphenyl allyl alcohol, also reported to be catalyzed by B2 (Table
5).[2d] This reaction proved to be more challenging to rationalize
as we faced a major hurdle to reproduce the published results.
Using either commercially available or freshly prepared and
recrystallized catalyst B2, yields never exceeded 20%. We
suspect that the reported success of this transformation might
result from the presence of an impurity in the way that B2 was
synthesized (a borinic acid or traces of another acid, for instance).
Even a trace impurity might not be negligible given the catalyst
loading of 20 mol%. In turn, stronger Brønsted acids enabled the
reaction to a limited extent (entries 5.5-5.7), but CSA was highly
effective, delivering the product in 85% yield (entry 5.8). This
result is similar to those reported in the literature, which again
strongly suggest Brønsted acid catalysis.
Acknowledgements
J.M. thanks the European Research Council (ERC) (grant
agreement n° 639170), ANR LabEx “Chemistry of Complex
Systems” (ANR-10-LABX-0026 CSC) and CNRS. S. Z. thanks the
China Scholarship Council for a fellowship. We thank Prof. D. G.
Hall for discussion and comments.
Notes
The authors declare no competing financial interest.
Acknowledgements
Table 5. Comparison between boronic acids and Brønsted acids for the catalytic
1,3-allylic transposition of allylic alcohols.[a]
J.M. thanks the ANR LabEx “Chemistry of Complex Systems”
(ANR-10-LABX-0026 CSC) and CNRS. S. Z. thanks the China
Scholarship Council for a fellowship. We thank Prof. D. G. Hall for
discussion and comments.
Entry
Catalyst
Yield [%]
5.1
5.2
5.3
B1
B2
B2
46
20 (80)[b]
<1[c]
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
Gassman, D. A. Singleton, J. Am. Chem. Soc. 1984, 106, 7993-7994; c)
C. Bergquist, B. M. Bridgewater, C. J. Harlan, J. R. Norton, R. A. Friesner,
G. Parkin, J. Am. Chem. Soc. 2000, 122, 10581-10590; d) M. Yasuda, T.
Somyo, A. Baba, Angew. Chem. Int. Ed. 2006, 45, 793-796; e) M. Vayer,
R. Guillot, C. Bour, V. Gandon, Chem Eur. J. 2017, 23, 13901-13905.
[12] a) V. Gutmann, Coordination Chemistry Reviews 1976, 18, 225-255; b)
H. Großekappenberg, M. Reißmann, M. Schmidtmann, T. Mꢁller,
Organometallics 2015, 34, 4952-4958; c) M. A. Beckett, D. S.
Brassington, S. J. Coles, M. B. Hursthouse, Inorg. Chem. Commun. 2000,
3, 530-533; d) G. C. Welch, L. Cabrera, P. A. Chase, E. Hollink, J. D.
Masuda, P. R. Wei, D. W. Stephan, Dalton Trans. 2007, 3407-3414; e)
J. Alarcón-Espósito, R. Contreras, R. A. Tapia, P. R. Campodónico,
Chem. Eur. J. 2016, 22, 13347-13351.
[13] J. J. Montalvo-Acosta, M. Dryzhakov, E. Richmond, M. Cecchini, J.
Moran, under revisions.
[14] a) S. Estopiña-Durán, L. J. Donnelly, E. B. Lclean, B. M. Hockin, A. M. Z.
Slawin, J. E. Taylor, Chem. Eur. J. 2019, 25, 3950-3956; b) E. Wolf, E.
Richmond, J. Moran, Chem. Sci. 2015, 6, 2501-2505.
[15] T. Beringhelli, D. Maggioni, G. D’Alfonso, Organometallics 2001, 20,
4927-4938.
[16] The possibility that the inhibiting effect of DTBP might be due to
poisoning of the boronic acid by alkoxide generated from deprotonation
of HFIP seems unlikely due to the following experiment: we found the
boronic acid catalyzed reaction also occurred in the absence of DTBP in
the aprotic solvent DCE under slightly higher temperatures. It is also
inhibited by DTBP under these conditions, where no alkoxides are
present (see the Supporting Information for more information).
[17] A. N. J. Moore, D. D. M. Wayner, Can. J. Chem. 1999, 77, 681-686.
[18] A. Shokri, X.-B. Wang, S. R. Kass, J. Am. Chem. Soc. 2013, 135, 9525-
9530.
Keywords: boronic acid • hexafluoroisopropanol • Bønsted acid
• Lewis acid • dual H-bond
[1]
[2]
For reviews on boronic acid catalysis, see: a) E. Dimitrijeviꢀ, M. S. Taylor,
ACS Catal. 2013, 3, 945-962; b) H. C. Zhang, D. G. Hall, Aldrichim. Acta.
2014, 47, 41-51; c) D. G. Hall, Chem. Soc. Rev. 2019, 48, 3475-3496.
For catalytic activation of alcohols by boronic acids, see: a) J. A.
McCubbin, O. V. Krokhin, Tetrahedron Lett. 2010, 51, 2447-2449; b) J.
A. McCubbin, H. Hosseini, O. V. Krokhin, J. Org. Chem. 2010, 75, 959-
962; c) J. A. McCubbin, C. Nassar, O. V. Krokhin, Synthesis 2011, 19,
3152-3160; d) H. Zheng, M. Lejkowski, D. G. Hall, Chem. Sci. 2011, 2,
1305-1310; e) J. A. McCubbin, C. Nassar, O. V. Krokhin, Synthesis 2011,
3152-3160; f) H. C. Zheng, S. Ghanbari, S. Nakamura, D. G. Hall, Angew.
Chem. Int. Ed. 2012, 51, 6187-6190; g) X. B. Mo, J. Yakiwchuk, J.
Dansereau, J. A. McCubbin, D. G. Hall, J. Am. Chem. Soc. 2015, 137,
9694-9703; h) C. L. Ricardo, X. B. Mo, J. A. McCubbin, D. G. Hall, Chem.
Eur. J. 2015, 21, 4218-4223; i) W.-B. Tang, K.-S. Cao, S.-S. Meng, W.-
H. Zheng, Synthesis 2017, 49, 3670-3675; j) H. T. Ang, J. P. G. Rygus,
D. G. Hall, Org. Biomol. Chem. 2019, 17, 6007-6014.
[3]
[4]
For catalytic activation of oximes by boronic acids, see: a) S.
Chandrasekhar, K. Gopalaiah, Tetrahedron Lett. 2002, 43, 2455-2457;
b) X. B. Mo, T. D. R. Morgan, H. T. Ang, D. G. Hall, J. Am. Chem. Soc.
2018, 140, 5264-5271
For catalytic activation of carboxylic acids by boronic acids, see: a) R. M.
Al-Zoubi, O. Marion, D. G. Hall, Angew. Chem. Int. Ed. 2008, 47, 2876-
2879; b) A. Sakakura, T. Ohkubo, R. Yamashita, M. Akakura, K. Ishihara,
Org. Lett. 2011, 13, 892-895; c) N. Gernigon, R. M. Al-Zoubi, D. G. Hall,
J. Org. Chem. 2012, 77, 8386-8400; d) C. Wang, H. Z. Yu, Y. Fu, Q. X.
Guo, Org. Biomol. Chem. 2013, 11, 2140-2146; e) N. Gernigon, H. C.
Zheng, D. G. Hall, Tetrahedron Lett. 2013, 54, 4475-4478; f) K. Ishihara,
Y. H. Lu, Chem. Sci. 2016, 7, 1276-1280; g) K. Wang, Y. H. Lu, K.
Ishihara, Chem. Commun. 2018, 54, 5410-5413; h) S. Arkhipenko, M. T.
Sabatini, A. S. Batsanov, V. Karaluka, T. D. Sheppard, H. S. Rzepa, A.
Whiting, Chem. Sci. 2018, 9, 1058-1072.
[18] a) K. M. Diemoz, A. K Franz, J. Org. Chem. 2019, 84, 1126-1138; b) G.
Lv, X. Sun, C. Zhang, M. Lei, Atmos. Chem. Phys. 2019, 19, 2833-2844.
[5]
[6]
a) J. Liu, H. Yao, C. Wang, ACS Catal. 2018, 8, 9376-9381; b) H. Yao, J.
Liu, C. Wang, Org. Biomol. Chem. 2019, 17, 1901-1905.
For recent reviews on HFIP, see: a) T. Sugiishi, M. Matsugi, H.
Hamamoto, H. Amii, RSC Adv. 2015, 5, 17269-17282; b) J. Wencel-
Delord, F. Colobert, Org. Chem. Front. 2016, 3, 394-400; c) I. Colomer,
A. E. R. Chamberlain, M. B. Haughey, T. J. Donohoe, Nat. Rev. Chem.
2017, 1, 0088; d) S. K. Sinha, T. Bhattacharya, D. Maiti, React. Chem.
Engl. 2019, 4, 244-253.
[7]
For recent examples of enhancement of reactivity via the use of HFIP,
see: a) A. Berkessel, J. A. Adrio, D. Hüttenhain, J. M. Neudörfl, J. Am.
Chem. Soc. 2006, 128, 8421-8426; b) A. Berkessel, J. A. Adrio, J. Am.
Chem. Soc. 2006, 128, 13412-13420; c) H. F. Motiwala, C. Fehl, S.-W.
Li, E. Hirt, P. Porubsky, J. Aubé, J. Am. Chem. Soc. 2013, 135, 9000-
9009; d) R. H. Vekariya, J. Aubé, Org. Lett. 2016, 18, 3534-3537; e) X.
Zeng, S. Liu, B. Xu, Org. Lett. 2016, 18, 4770-4773; f) I. Colomer, R. C.
Barcelos, K. E. Christensen, T. J. Donohoe, Org. Lett. 2016, 18, 5880-
5883; g) T. Kamitanaka, K. Morimoto, K. Tsuboshima, D. Koseki, H.
Takamuro, T. Dohi, Y. Kita, Angew. Chem. Int. Ed. 2016, 55, 15535-
15538; h) I. Colomer, C. Batchelor-McAuley, B. Odell, T. J. Donohoe, R.
G. Compton, J. Am. Chem. Soc. 2016, 138, 8855-8861; i) D. Lebœuf, L.
Marin, B. Michelet, A. Perez-Luna, R. Guillot, E. Schulz, V. Gandon,
Chem. Eur. J. 2016, 22, 16165-16171; j) V. D. Vukoviꢀ, E. Richmond, E.
Wolf, J. Moran, Angew. Chem. Int. Ed. 2017, 56, 3085-3089; k) C. Qi, F.
Hasenmaile, V. Gandon, D. Lebœuf, ACS Catal. 2018, 8, 1734-1739; l)
Z. Tao, K. A. Robb, K. Zhao, S. E. Denmark, J. Am. Chem. Soc. 2018,
140, 3569-3573; m) E. Richmond, J. Yi, V. D. Vukoviꢀ, F. Sajadi, C. N.
Rowley, J. Moran, Chem. Sci. 2018, 9, 6411; n) C. Qi, V. Gandon, D.
Lebœuf, Angew. Chem. Int. Ed. 2018, 57, 14245-14249; o) Y. Zhu, I.
Colomer, A. L. Thompson, T. J. Donohoe, J. Am. Chem. Soc. 2019, 141,
6489-6493.
[8]
[9]
a) M. Dryzhakov, M. Hellal, E. Wolf, F. C. Falk, J. Moran, J. Am. Chem.
Soc. 2015, 137, 9555-9558; b) M. Dryzhakov, J. Moran, ACS Catal. 2016,
6, 3670-3673.
J. G. M. Morton, M. A. Dureen, D. W. Stephan, Chem. Commun. 2010,
46, 8947-8949.
[10] For selected papers where transition metal-promoted reactions were
later found to be Brønsted acid catalyzed processes, see: a) B.
Schlummer, J. F. Hartwig, Org. Lett. 2002, 4, 1471-1474; b) T. C.
Wabnitz, J.-Q. Yu, J. B. Spencer, Chem. Eur. J. 2004, 10, 484-493; c) D.
C. Rosenfeld, S. Shekhar, A. Takemiya, M. Utsunomiya, J. F. Hartwig,
Org. Lett. 2006, 8, 4179-4182; d) P. N. Liu, Z. Y. Zhou, C. P. Lau, Chem.
Eur. J. 2007, 13, 8610-8619; e) T. Jin, F. Yang, C. Liu, Y. Yamamoto,
Chem. Commun. 2009, 3533-3535; f) T. T. Dang, F. Boeck, L.
Hintermann, J. Org. Chem. 2011, 76, 9353-9361; g) H. J. Jin, J. H. Kim,
E. J. Kang, Synthesis 2017, 49, 3137-3144; h) E. T. Sletten, Y.-J. Tu, H.
B. Schlegel, H. M. Nguyen, ACS Catal. 2019, 9, 2110-2123.
[11] The bulky pyridine 2,6-di-tert-butylpyridine (DTBP) is commonly used to
distinguish between boron or other Lewis acids and Brønsted acids
because it does not complex most Lewis acids, including boron: a) H. C.
Brown, B. Kanner, J. Am. Chem. Soc. 1966, 88, 986-992; b) P. G.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202001902
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
To B or not to B: We shed new light on the mechanism of boronic acid catalysis for reactions of alcohols or oximes, which were
thought to involve the boron atom acting as a Lewis acid or to feature transient covalent bonds to boron. The present study suggests
that, in most cases, H-bond catalysis or Brønsted acid catalysis are the major activation modes.
Institute and/or researcher Twitter usernames: @djpleboeuf and @MoranLabUdS
6
This article is protected by copyright. All rights reserved.