Regioselective, borinic acid-catalyzed monoacylation, sulfonylation and alkylation of diols and carbohydrates: Expansion of substrate scope and mechanistic studies

Article abstract of DOI:10.1021/ja302549c

Synthetic and mechanistic aspects of the diarylborinic acid-catalyzed regioselective monofunctionalization of 1,2- and 1,3-diols are presented. Diarylborinic acid catalysis is shown to be an efficient and general method for monotosylation of pyranoside derivatives bearing three secondary hydroxyl groups (7 examples, 88% average yield). In addition, the scope of the selective acylation, sulfonylation, and alkylation is extended to 1,2- and 1,3-diols not derived from carbohydrates (28 examples); the efficiency, generality, and operational simplicity of this method are competitive with those of state-of-the-art protocols including the broadly applied organotin-catalyzed or -mediated reactions. Mechanistic details of the organoboron-catalyzed processes are explored using competition experiments, kinetics, and catalyst structure-activity relationships. These experiments are consistent with a mechanism in which a tetracoordinate borinate complex reacts with the electrophilic species in the turnover-limiting step of the catalytic cycle.

Full text of DOI:10.1021/ja302549c

Article
pubs.acs.org/JACS
Regioselective, Borinic Acid-Catalyzed Monoacylation, Sulfonylation
and Alkylation of Diols and Carbohydrates: Expansion of Substrate
Scope and Mechanistic Studies
Doris Lee, Caitlin L. Williamson, Lina Chan, and Mark S. Taylor*
Department of Chemistry, University of Toronto, 80 St. George Street, Toronto ON M5S 3H6, Canada
S
* Supporting Information
ABSTRACT: Synthetic and mechanistic aspects of the
diarylborinic acid-catalyzed regioselective monofunctionaliza-
tion of 1,2- and 1,3-diols are presented. Diarylborinic acid
catalysis is shown to be an efficient and general method for
monotosylation of pyranoside derivatives bearing three
secondary hydroxyl groups (7 examples, 88% average yield).
In addition, the scope of the selective acylation, sulfonylation,
and alkylation is extended to 1,2- and 1,3-diols not derived
from carbohydrates (28 examples); the efficiency, generality,
and operational simplicity of this method are competitive with
those of state-of-the-art protocols including the broadly applied organotin-catalyzed or -mediated reactions. Mechanistic details of
the organoboron-catalyzed processes are explored using competition experiments, kinetics, and catalyst structure−activity
relationships. These experiments are consistent with a mechanism in which a tetracoordinate borinate complex reacts with the
electrophilic species in the turnover-limiting step of the catalytic cycle.
draws on previous work by Aoyama and co-workers in which
stoichiometric quantities of organoboron reagent were used to
activate carbohydrate derivatives toward regioselective alkyla-
tion and glycosylation through the formation of putative
tetracoordinate boronate complexes.15 The broad scope of
these processes, the low toxicity and ease of handling of the
borinic acid-derived precatalyst, and the mild reaction
conditions employed are features that compare favorably with
those of existing methods for regioselective carbohydrate
protection. The regioselective glycosyl acceptor activation
represents the first example of such a process using a synthetic
(nonenzyme) catalyst and may offer new prospects for the
efficient preparation of oligosaccharide derivatives.
Herein, we report expansions of the scope of the borinic
acid-catalyzed diol activation developed in our laboratory
including applications of this method to noncarbohydrate-
derived 1,2- and 1,3-diols, and the development of a
regioselective sulfonylation variant that is applicable to both
carbohydrate derivatives and simpler substrates. We also
describe competition experiments, kinetics studies, and catalyst
structure−activity relationships that shed light on the
mechanism of these transformations.
INTRODUCTION
■
The selective manipulation of hydroxyl (OH) groups in di- and
polyols is a frequently encountered problem in organic
synthesis, in contexts ranging from reactions of simple
hydroxylated feedstocks to the preparation of complex, highly
functionalized natural products. This problem is of central
importance in carbohydrate chemistry, where regioselectivity is
the major obstacle to the preparation of novel medicinal agents
or probes of biological function from readily available
monosaccharide derivatives. Many protocols for selective
protection of hydroxyl groups have been developed,1 but
procedures that display high levels of substrate generality
(especially in the context of more complex substrates such as
carbohydrates), as well as step and atom economy, remain an
unmet need. Catalysis offers considerable promise in this
regard, and several examples of catalyst-controlled regio- or
stereoselective transformations of polyols, including carbohy-
drates, have been disclosed.2−7
We recently described a new mode of organoboron catalysis
in which a diarylborinic acid derivative mediates the acylation,8
alkylation,9 and glycosylation10 of a wide range of carbohydrate
derivatives, with high levels of regioselectivity for the equatorial
OH group of cis-1,2-diol pairs.11 The regiochemical outcome
was interpreted in terms of a mechanism involving selective
complexation of the cis-vicinal diol pairs generating tetracoor-
dinate borinate complexes that display enhanced nucleophilicity
at oxygen.12 This proposed catalyst−substrate interaction is
related to the reversible, covalent bonding between boronic
acids and diols that has been studied extensively in molecular
recognition and chemical sensing of carbohydrates.13,14 It also
RESULTS AND DISCUSSION
Regioselective Sulfonylation of Carbohydrate Deriv-
atives. Sulfonylation of carbohydrates is of interest in light of
■
Received: March 15, 2012
Published: April 25, 2012
© 2012 American Chemical Society
8260
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
the utility of arenesulfonate esters as protective groups,1b as
well as the potential for carrying out nucleophilic displacement
reactions of such compounds.16 Regioselective sulfonylation of
sugar derivatives can be accomplished by two-step procedures
involving stoichiometric generation of stannylene acetals,
followed by electrophilic attack by a sulfonylating agent,
typically p-toluenesulfonyl chloride (TsCl).17 The groups of
Martinelli,18 Burke,19 and Onomura4b developed regioselective
methods for sulfonylation of di- and polyols using catalytic
Bu2SnO or Me2SnCl2, thus reducing the amount of toxic,
lipophilic Sn(IV) waste generated. The fluorous dialkyltin
catalyst of Curran and co-workers represents another important
development from the latter perspective.20
Given the utility of borinic ester 1a as a precatalyst for
regioselective reactions of carbohydrate derivatives with acyl,
alkyl, and glycosyl halides, and, considering the ease of handling
and favorable safety profile of this organoboron derivative, we
have studied its application in regioselective sulfonylation
reactions. The results (Table 1) indicate that 1a promotes
selective monotosylation of cis-vicinal diol moieties, with the
equatorial O3 group undergoing selective tosylation in
pyranoside derivatives of galactose (2a, 2b, 2c), mannose
(2d), fucose (2e), and rhamnose (2f). A thioglycoside
functional group is tolerated by this catalyst system (2a), as
is variation of the stereochemistry of the anomeric substituent
(2b vs 2c). Using 1,6-anhydrogalactopyranose (2g) resulted in
sulfonylation at O4 providing a switch in selectivity from the
unbridged galactopyranoside. These regiochemical outcomes
are consistent with our previous results using catalyst 1a;
computational studies suggest that electronic effects contribute
to differences in nucleophilicity between the two boron-bound
oxygen atoms in a tetracoordinate borinate complex.8,9
However, given that the electrophile is delivered to the more
accessible equatorial position, the possibility of steric effects on
the regioselectivity of these transformations cannot be
discounted.21 Extensions of this method to diols not derived
from carbohydrates reveal several instances in which
regioselectivity appears to be under steric control (below). In
comparison to previously reported organotin-catalyzed mono-
tosylations of carbohydrate derivatives, the borinic acid-
catalyzed protocol displays a broad substrate scope and is
notable for its high selectivity for functionalization of cis-diol
motifs.
Monofunctionalization of Non-Carbohydrate-Derived
Diols. Whereas regioselective reactions of carbohydrate
derivatives are of particular interest (given both the
fundamental challenges involved and the utility of the products
in oligosaccharide synthesis), the monoprotection of simpler
diol substrates is also an important problem. We have
investigated the extension of the borinic acid-catalyzed
protocols to a range of acyclic and cyclic 1,2- and 1,3-diols,
and have found that they are useful for monoprotection of such
substrates. The results are described in detail below.
Table 1. Borinic Acid-Catalyzed Monotosylation of
Carbohydrate Derivatives
a
Acylation and Sulfonylation. Diverse methods for mono-
acylation of 1,2-diols have been developed: systems that deliver
high selectivity for mono- over bis-functionalized products
(especially when a sterically unhindered acylating agent is
employed), or that enable useful levels of regiocontrol for
substrates lacking a strong steric bias for one of the two OH
groups, are of particular interest. Successful approaches include
the use of solid-supported substrates or reagents,22 ring-
opening of cyclic acetals and related species,23 and enzymatic,24
nucleophilic, or Lewis acid catalysis4a,25,26 including enantiose-
lective variants.27 In contrast, fewer methods for monosulfony-
lation of 1,2-diols have been developed, despite the utility of
this transformation for the preparation of enantioenriched
epoxides. The organotin-mediated and -catalyzed protocols
described above represent the state-of-the-art methods; Ag2O-
promoted28 and montmorillonite-catalyzed29 monosulfonyla-
tions, as well as enantioselective variants with Lewis acid30 and
Lewis base catalysts,31 have also been developed.
Using diarylborinic acid derivative 1a, various 1,2-diols were
monoacylated and -tosylated efficiently (Table 2). The
selectivity for mono- over bis-functionalization was high: for
example, the reaction of ethylene glycol with 1.5 equiv of
benzoyl chloride generated 2-hydroxyethyl benzoate in 84%
yield (entry 1). For diols containing both primary and
secondary hydroxyl groups, the primary OH was selectively
functionalized over the secondary position (4b, 4c, 4k).
Cycloalkanediols of various ring sizes were selectively
monofunctionalized with high yields (4d, 4e, 4f).8 Sterically
hindered (−)-pinanediol was monotosylated at the secondary
over the tertiary OH group (4i). In addition to vicinal diols,
1,3-diols 4j and 4k were also activated by precatalyst 1a.
a
Reaction conditions: substrate (1.0 mmol), catalyst 1a (10 mol %),
TsCl (1.5 mmol), i-Pr2NEt (1.5 mmol), CH3CN (5 mL), 23 °C. See
the Supporting Information. Isolated yield.
b
8261
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
catalysts for monoalkylation of (R,R)-hydrobenzoin and 1,2-
cyclohexanediol (both the cis and trans diastereomers).37
With the aim of developing a protocol that would avoid the
use of Ag(I)-based activators (such as Ag2O, which was
employed in the borinic acid-catalyzed alkylation of carbohy-
drate derivatives9), we investigated the use of halide additives. A
screen of iodide salts revealed that one equivalent of KI, in the
presence of potassium carbonate as base, enabled efficient
monoalkylation of 1,2-diols with BnBr in acetonitrile at 60 °C
(Table 2). Under these conditions, 4-fluorophenylboronic acid
(the optimal organoboron catalyst from Onomura’s study) was
less efficient than diphenylborinic acid derivative 1a, yielding
products 7c and 7k in 56% and 11% yields, respectively. The
optimized conditions were tolerant of a wide range of
substrates, including 1,3-diols, and were suitable for installation
of substituted benzyl ethers such as the 4-methoxybenzyl
(PMB) and 2-naphthylmethyl (Nap) protective groups. In
certain cases, the monobenzylation may be conducted at low
catalyst loadings: Table 3 indicates that as little as 0.5 mol % of
1a may be employed for efficient monoalkylation of cis-
cyclohexanediol.
Table 2. Borinic Acid-Catalyzed Monofunctionalization of
1,2- and 1,3-Diols
a
Table 3. Effects of Catalyst Loading on the Monobenzylation
of 4d
a
entry
mol % la
yield (%)
1
2
3
4
5
0
0.5
2
15
90
>99
95
5
10
95
a
Yield determined by 1H NMR with mesitylene as a quantitative
internal standard.
a
Reaction conditions. For acylation and sulfonylation: substrate (1.0
mmol), catalyst 1a (1−10 mol %), electrophile (1.1−1.5 mmol), i-
Pr2NEt (1.1−1.5 mmol), CH3CN (5 mL), 23 °C. For alkylation:
substrate (0.2 mmol), catalyst 1a (10 mol %), BnBr (0.3 mmol), KI
(0.2 mmol), K2CO3 (0.22 mmol), CH3CN (1 mL), 60 °C. See the
Competition Experiments. Competition experiments
were undertaken to evaluate the ability of catalyst 1a to
distinguish between structurally similar alcohol substrates. In
each case, the electrophile was employed as the limiting reagent
so that the product ratios would reflect relative rates of
activation by the organoboron catalyst. In our previous studies
of acylations catalyzed by 1a, similar competition experiments
revealed a high level of selectivity for cis- versus trans-1,2-
cycloalkanediols, an observation that formed the basis for
applications in carbohydrate activation.8 The competition
experiments conducted in the present study indicate that
diols are activated in preference to simple alcohols (Scheme 1,
eqs 1, 2), with 1,2-diols undergoing sulfonylation at
considerably higher rates than 1,3-diols (eq 3). In a competition
experiment between (R,R)- and meso-hydrobenzoin, a 6:1 ratio
favoring benzylation of the (R,R)-diastereomer was observed
(eq 4). Selective alkylation of syn- over anti-1,3-diols was
promoted by 1a (eqs 5, 6); in the last of these two examples,
the less sterically hindered OH group of a nonsymmetrical syn-
1,3-diol was selectively benzylated.
b
Supporting Information. Isolated yield. Entries 9, 14, and 17 were
c
previously reported in ref 8. Product isolated along with 19% of the
d
regioisomeric product. Product isolated along with 6% of the bis-
e
tosylated product. Product isolated along with 11% of the
regioisomeric product.
Alkylation. Selective installation of the benzyl ether
protective group and its substituted congeners is a useful
transformation: benzyl ethers are robust, do not readily migrate,
and are easily removed under mild conditions. Several methods
have been developed for the selective installation of benzyl and
related ether groups. One of the earlier reports of selective
monoalkylation of diols involves the use of polymer supports.32
Reductive ring-opening of cyclic acetals33 and alkylation of
stannylene acetals34 are other established and broadly applied
methods. In addition to methods that employ stoichiometric
quantities of metal additives (such as Ag(I) salts),35 several
transition metal-36 and Lewis acid-catalyzed processes have
been developed.37 Other catalytic strategies include the use of
crown ethers38 and phase transfer agents.39 Of particular
relevance to this work is a recent report by the group of
Onomura in which arylboronic acids were employed as
The selectivities described above can be rationalized
according to known trends in the thermodynamics of binding
of diols to organoboron compounds.13 Boronic acids show a
strong preference for binding of 1,2-diols to generate 1,2-
8262
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
Scheme 1. Competition Experiments
Scheme 2. Proposed Catalytic Cycle for Sulfonylation
Promoted by 1a
spectroscopy with mesitylene as a quantitative internal
standard. In cases where excess TsCl and i-Pr2NEt were
employed, the product of bis-sulfonylation could not be
1
detected by H NMR.
iii). Entry of 1a and Ph2BOH into the Catalytic Cycle, and
Determination of the Ph2BOH−4d Association Constant.
The ability of substrate 4d to displace the ethanolamine ligand
from precatalyst 1a was probed by NMR experiments and rate
studies. Experiments in which 0.5−20 equiv of diol 4d were
added to a solution of 1a in CD3CN (either in the presence or
absence of i-Pr2NEt) provided no evidence for diplacement of
ethanolamine. Thus, whereas ethanolamine interacts weakly
with Ph2BOH in pH 7.0 phosphate buffer (Ka < 5 M−1),40 its
displacement by substrate under the conditions of catalysis
appears to be difficult. In contrast, similar experiments probing
the interaction of Ph2BOH and 4d (CD3CN solvent, in the
presence of 5 equiv of i-Pr2NEt relative to 4d) resulted in the
dioxaborolanes, over binding of 1,3-diols to form 1,3-
dioxaborinanes. Selective complexation of (R,R)-hydrobenzoin
over its meso diastereomer results from destabilization of the
latter complex by eclipsing of the phenyl groups. Similarly,
selective complexation of syn- over anti-1,3-diols is expected
because both substituents can occupy equatorial positions in
the resulting dioxaborinane. The observed relationship between
the thermodynamics of diol binding and rates of activation is
consistent with our studies of reaction rates, which indicate
saturation kinetics in the diol substrate (below). Given the
extensive literature describing the thermodynamics of organo-
boron−diol interactions, relationships of this type may be
useful in planning applications of this method to more complex
polyol substrates. In particular, the ability to activate syn-1,3-
diol units preferentially may be of utility in selective
manipulation of building blocks for polyketide synthesis.
Kinetics of Borinic Acid-Catalyzed Sulfonylation.
i). Proposed Catalytic Cycle. Kinetics experiments are
consistent with the proposed catalytic cycle for monosulfony-
lation of cis-1,2-cyclohexanediol (4d) depicted in Scheme 2.
Entry of aminoethyl diphenylborinate 1a into the catalytic cycle
is triggered by irreversible bis-sulfonylation of the ethanolamine
ligand and binding of the diol to form an activated ‘ate’
complex. The reaction of this tetracoordinate adduct with TsCl
is the turnover-limiting step. Rapid, thermodynamically
favorable displacement of the product by diol returns the
catalyst to its resting state. Details of the experimental
observations that support this mechanistic proposal are
described below.
1
appearance of new signals in the H NMR spectrum within a
few minutes of mixing the two components, consistent with
slow binding on the NMR time scale (part a of Figure 1).
Analysis by the method of continuous variation (Job plot, part
b of Figure 1) was consistent with a 1:1 stoichiometry of
binding. From these data, an association constant of 70 20
M−1 was estimated for the Ph2BOH−4d interaction under
these conditions. In the absence of i-Pr2NEt, the spectral
changes corresponding to complexation of 4d by Ph2BOH were
not observed, consistent with previous studies of the pH-
dependence of organoboron−diol interactions.41
The experiments described in the preceding paragraph
indicate that kinetic and/or thermodynamic barriers prevent
the rapid entry of 1a into the catalytic cycle through
displacement of ethanolamine by 4d. However, sulfonylation
reactions catalyzed by 1a displayed no detectable induction
period, and reaction rates using 1a and Ph2BOH as
sulfonylation catalysts were essentially identical (Figure 2).42
Moreover, inhibition by added ethanolamine was not observed.
These results indicate that both 1a and Ph2BOH enter rapidly
into the catalytic cycle. The contrast between the NMR
experiments and rate studies can be understood according to a
mechanism in which the ethanolamine ligand reacts with TsCl
prior to displacement by 4d. Indeed, subjecting 1a to TsCl and
i-Pr2NEt in acetonitrile resulted in the formation of the
ii). Experimental Details. Rates of the borinic acid-catalyzed
sulfonylation of cis-cyclohexanediol (4d) were determined by
1H NMR. Aliquots from reaction mixtures carried out in
acetonitrile at 23 °C were quenched by addition of excess
methanol and the product concentration was assessed by NMR
8263
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
sulfonamido tosylate derived from N,O-bis-tosylation of
1
ethanolamine, which could be detected by H NMR. Similarly,
N,O-bis-benzoylation of precatalyst was observed previously in
acylation reactions catalyzed by 1a.8
iv). Kinetic Orders in Diol 4d, Ph2BOH, TsCl, and i-Pr2NEt.
As is evident in Figure 2, the rates of reactions carried out with
5 equiv of base and TsCl, using 1 mol % catalyst (either
Ph2BOH or 1a), were found to be effectively invariant until
roughly 60% of the substrate had been consumed (i.e., the
reactions displayed apparent zero-order kinetics in substrate).
Monitoring reactions carried out with TsCl as the limiting
reagent and Ph2BOH as catalyst, using varying (excess)
quantities of 4d, provided additional support for this
contention (Figure 3). These observations are consistent with
Figure 3. Product concentration versus time for Ph2BOH-catalyzed
reactions of TsCl (1 equiv) with varying concentrations of diol 4d: 1
■
◆
▲
equiv 4d ( , blue), 2 equiv 4d ( , red), and 5 equiv 4d ( , green).
Reaction conditions: TsCl (0.5 mmol), 4d (1−5 equiv), 1b (1 mol %),
i-Pr2NEt (5 equiv), CH3CN (10 mL), 23 °C.
Figure 1. (a) Changes in the 1H NMR spectrum (400 MHz, CD3CN,
295 K) of cis-1,2-cyclohexanediol (4d) upon addition of Ph2BOH
(1b), in the presence of i-Pr2NEt (5 equiv). From bottom to top, the
spectra of 4d in the presence of 0 equiv, 1 equiv, and 3 equiv of
Ph2BOH are depicted. The signals shown correspond to the methine
groups of free and bound 4d. (b) Job plot for the 4d−1b interaction:
concentration of the 4d−1b complex, as determined by integration of
saturation kinetics, a scenario that is encountered quite
frequently in catalysis.43 The Ph2BOH−4d association constant
determined by H NMR (Ka = 70 M−1, above) is compatible
1
1
the H NMR spectrum, is plotted against the mole fraction χ of 4d.
with the observation of saturation kinetics: based on this value
of Ka, catalyst−substrate complex would be expected to account
for roughly 80% of the total Ph2BOH concentration at the
outset of a sulfonylation of 4d under the standard conditions.
Kinetic orders in catalyst, TsCl, and base were determined by
measurements of initial rates of reactions carried out using one
equivalent of substrate and excess base/TsCl (2−13 equiv).
Given the lack of an observable induction period, and the fact
that the reaction rates were essentially constant until relatively
high conversions due to zero-order kinetics in 4d (above), the
use of initial rates appears to be justified. Plots of initial rates
versus concentration revealed a first-order kinetic dependence
on Ph2BOH concentration (Figure 4), first-order kinetics in
TsCl (Figure 5), and zero-order kinetics in i-Pr2NEt (Figure 6).
The rate law implied by these kinetic orders is consistent with a
mechanism involving turnover-limiting sulfonylation of the
borinic acid−diol adduct.
The total concentration [4d] + [1b] was held constant at 16.7 mM.
Figure 2. Effect of catalyst identity on the rate of sulfonylation of 4d:
v). Comparison with Kinetics of Selective Glycosyl
Acceptor Activation. We previously carried out kinetics
◆
diphenylborinic acid 1b ( , green), 2-aminoethyl diphenylborinate 1a
■
▲
(
, red), and [1a + ethanolamine] ( , blue). Reaction conditions: 4d
1
experiments, using initial rates determined by H NMR, for
(0.5 mmol), catalyst (1 mol %), TsCl (5 equiv), i-Pr2NEt (5 equiv),
CH3CN (10 mL), 23 °C.
the regioselective glycosylation of a mannopyranoside substrate
catalyzed by borinic ester 1a. The rate law for sulfonylation
described above differs from that determined previously for
glycosylation: for the latter process, first-order kinetics in
8264
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
halide-abstracting agent and Brønsted base.44 The saturation
kinetics in substrate observed for the sulfonylation reaction as
opposed to first-order kinetics for the glycosylation reaction is
the distinguishing factor.
The dependence of the Ph2BOH−diol association constant
on base concentration, along with the distinct natures of the
Brønsted bases employed for sulfonylation and glycosylation (i-
Pr2NEt and Ag2O, respectively), hints at an explanation for the
observed difference in kinetic behavior. In the absence of a
soluble Brønsted base, catalyst−substrate binding is relatively
weak under the conditions of glycosylation leading to first-
order kinetics in substrate. The presence of i-Pr2NEt in the
sulfonylation reaction enhances the catalyst−substrate associ-
ation constant to the point that saturation kinetics in substrate
are observed.
vi). Electronic Effects on Catalyst Activity. The rates of
sulfonylation reactions catalyzed by substituted arylborinic acids
1c and 1d were investigated. Both the 4-methoxy- and 3,5-
bis(trifluoromethyl)-substituted derivatives provided qualita-
tively similar rate profiles to that observed using diphenylbor-
inic acid showing no evident induction period and apparent
zero-order kinetics in substrate; however, 1c and 1d provided
lower rate constants than did the unsubstituted catalyst 1b
(Figure 7). The fact that both electron-donating and electron-
withdrawing substituents on the diarylborinic acid resulted in
lower sulfonylation activties was somewhat surprising.
Figure 4. Dependence of the initial rate of sulfonylation of 4d (kobs
)
on the concentration of Ph2BOH. Reaction conditions: 4d (0.2
mmol), 1b (0.1−1.5 mol %), TsCl (1.1 equiv), i-Pr2NEt (1.1 equiv),
CH3CN (2 mL), 23 °C.
Figure 5. Dependence of the initial rate of sulfonylation of 4d (kobs
)
on the concentration of TsCl. Reaction conditions: 4d (0.2 mmol), 1b
(1 mol %), TsCl (1−5 equiv), i-Pr2NEt (5 equiv), CH3CN (2 mL), 23
°C.
Figure 7. Electronic effects on the activity of diarylborinic acid
◆
catalysts 1b−1d for sulfonylation of diol 4d: catalyst 1b ( , green), 1c
■
▲
(
, red), and 1d ( , blue). Reaction conditions: 4d (0.5 mmol),
catalyst (1 mol %), TsCl (5 equiv), i-Pr2NEt (5 equiv), CH3CN (10
mL), 23 °C.
Given the rate law obtained for diol sulfonylation catalyzed
by 1b, we have attempted to interpret this effect in terms of the
influence of substituents on the reactivity of the tetracoordinate
‘ate’ complexes toward sulfonylation. To this end, gas-phase
DFT calculations (B3LYP/6-31+G(d,p)) of the anionic
adducts of ethylene glycol with 1b−1d were undertaken. We
have previously found that B3LYP/6-31+G(d,p) gas-phase
calculations of Fukui indices for monosaccharide-derived
borinates are consistent with the observed regiochemical
Figure 6. Dependence of the initial rate of sulfonylation of 4d (kobs
)
on the concentration of i-Pr2NEt. Reaction conditions: 4d (0.2 mmol),
1b (1 mol %), TsCl (1 equiv), i-Pr2NEt (1−13 equiv), CH3CN (2
mL), 23 °C.
glycosyl acceptor, glycosyl donor, and catalyst were observed,
along with zero-order kinetics in Ag2O, which plays the role of
8265
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
outcomes of organoboron-catalyzed transformations suggesting
that this level of theory is appropriate for modeling electronic
effects in such systems.8 Here, the calculated Mulliken charges
of the ethylene glycol-derived oxygen atoms were found to
decrease in absolute value in the order 1b > 1c > 1d, mirroring
the trend in catalytic activity. Calculated absolute values of the
molecular electrostatic potential at oxygen followed the same
order. Thus, calculations appear to indicate that both the 3,5-
bis(trifluoromethyl) and 4-methoxy groups exert inductive
electron-withdrawing effects in the context of borinate esters,
despite the negative value of σ− for the latter (σ− for 4-OCH3 is
−0.2).45 It should be noted that we cannot discount the
possibility that the methoxy group impairs substrate binding
rather than attenuating the nucleophilicity of the catalyst−
substrate complex.
diphenylborinic acid. NMR experiments have enabled the
determination of a diphenylborinic acid−diol association
constant under the conditions of catalysis. The rate law (first-
order kinetics in catalyst and electrophile, zero-order kinetics in
Brønsted base, saturation kinetics in diol substrate) is
consistent with the reaction of such an adduct in the
turnover-limiting step of the catalytic cycle. Computational
modeling of the tetracoordinate borinates provides data that are
consistent with experimentally observed regiochemical out-
comes and catalyst substituent effects. These results may
provide a useful starting point for efforts toward identifying
novel diorganoboron catalysts that exhibit increased activity or
altered regioselectivity for reactions of this type, and for
extending these methods to more complex polyol substrates.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data for all new
compounds, details of kinetics experiments, binding studies and
calculations. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: mtaylor@chem.utoronto.ca.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was funded by NSERC (Discovery Grants Program,
Graduate Scholarships to D. L. and L. C.), the Canadian
Foundation for Innovation, the Province of Ontario (Early
Researcher Award to M.S.T.), and Boehringer Ingelheim
Canada Ltd. Peter Dornan (Department of Chemistry,
University of Toronto) is acknowledged for advice regarding
the kinetics experiments.
Figure 8. Calculated geometries of the anionic adducts of ethylene
glycol with catalysts 1b−1d, and average Mulliken charges of ethylene
glycol-derived oxygen atoms.
CONCLUSIONS
■
REFERENCES
■
The results described herein emphasize that diphenylborinic
acid catalysis is a useful method for selective monofunction-
alization of diols. The selective activation of equatorial OH
groups of cis-1,2-diol moieties has been successfully applied to
sulfonylation, enabling the high-yielding synthesis of mono-
tosylates derived from diverse pyranoside substrates. This mode
of catalysis has been extended to the selective acylation,
sulfonylation, and alkylation of simpler 1,2- and 1,3-diol
substrates, and provides significant advantages in terms of
efficiency and operational simplicity for this important class of
transformations. In particular, the high-yielding, diphenylbor-
inic acid-catalyzed monosulfonylation of 1,2-diols may be of
significant interest as an alternative to well-established
protocols requiring the use of diorganotin catalysts. Similarly,
the borinic acid-catalyzed monobenzylation of 1,2-diols
compares favorably to existing methods and employs relatively
benign and inexpensive reagents. These results should
encourage applications of this method outside the area of
carbohydrate chemistry.
́
(1) (a) Gοmez, A. M. In Glycoscience; Fraser-Reid, B., Tatsuka, K.,
Thiem, J., Eds.; Springer-Verlag: Berlin, 2008; pp 103−177. (b) Wuts,
P. G. M.; Greene, T. W. Greene’s Protective Groups in Organic Synthesis,
4th ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2007.
(2) (a) Therisod, M.; Klibanov, A. M. J. Am. Chem. Soc. 1987, 109,
3977−3981. (b) Wang, Y.-F.; Lalonde, J. J.; Momongan, M.;
Bergbreiter, D. E.; Wong, C.-H. J. Am. Chem. Soc. 1988, 110, 7200−
7205. (c) Lewis, J. C.; Bastian, S.; Bennett, C. S.; Fu, Y.; Mitsuda, Y.;
Chen, M. M.; Greenberg, W. A.; Wang, C.-H.; Arnold, F. H. Proc. Natl.
Acad. Sci. U.S.A. 2009, 106, 16550−16555.
(3) (a) Griswold, K. S.; Miller, S. J. Tetrahedron 2003, 59, 8869−
8875. (b) Kawabata, T.; Muramatsu, W.; Nishio, T.; Shibata, T.;
Schedel, H. J. Am. Chem. Soc. 2007, 129, 12890−12895. (c) Sculim-
brene, B. R.; Miller, S. J. J. Am. Chem. Soc. 2001, 123, 10125−10126.
(d) Zhao, Y.; Rodrigo, J.; Hoveyda, A. H.; Snapper, M. L. Nature 2006,
443, 67−70.
(4) (a) Iwasaki, F.; Maki, T.; Onomura, O.; Nakashima, W.;
Matsumura, Y. J. Org. Chem. 2000, 65, 996−1002. (b) Demizu, Y.;
Kubo, Y.; Miyoshi, H.; Maki, T.; Matsumura, Y.; Moriyama, N.;
Onomura, O. Org. Lett. 2008, 10, 5075−5077.
(5) Dhiman, R. S.; Kluger, R. Org. Biomol. Chem. 2010, 8, 2006−
2008.
Mechanistic experiments have provided strong support for
the proposal that anionic, tetracoordinate borinate esters are
key intermediates in this class of processes. Competition
experiments establish a link between the thermodynamics of
two-point covalent binding of diols to organoboron compounds
and the relative rates of diol activation in the presence of
(6) (a) Evtushenko, E. V. Synth. Commun. 2006, 36, 1593−1599.
(b) Evtushenko, E. V. J. Carbohydr. Chem. 2010, 29, 369−378.
(7) Sun, X.; Worthy, A. D.; Tan, K. L. Angew. Chem., Int. Ed. 2011,
50, 8167−8171.
(8) Lee, D.; Taylor, M. S. J. Am. Chem. Soc. 2011, 133, 3724−3727.
8266
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267

Journal of the American Chemical Society
Article
(9) Chan, L.; Taylor, M. S. Org. Lett. 2011, 13, 3090−3093.
(10) Gouliaras, C.; Lee, D.; Chan, L.; Taylor, M. S. J. Am. Chem. Soc.
2011, 133, 13926−13929.
(36) Sirkecioglu, O.; Karliga, B.; Talinli, N. Tetrahedron Lett. 2003,
44, 8483−8485.
(37) Maki, T.; Ushijima, N.; Matsumura, Y.; Onomura, O.
Tetrahedron Lett. 2009, 50, 1466−1468.
(11) Lee, D.; Newman, S. G.; Taylor, M. S. Org. Lett. 2009, 11,
5486−5489.
(38) Bessodes, M.; Boukarim, C. Synlett 1996, 11, 1119−1120.
(39) de la Zerda, J.; Barak, G.; Sasson, Y. Tetrahedron 1989, 45,
1533−1536.
(40) Chudzinski, M. G.; Chi, Y.; Taylor, M. S. Aust. J. Chem. 2011,
64, 1466−1469.
(41) Springsteen, G.; Wang, B. Tetrahedron 2002, 58, 5291−5300.
(42) In principle, Ph2BOH is itself a precatalyst, from which water
must be displaced for substrate binding. The NMR titrations indicate
that this displacement is relatively rapid. In addition, no increase in
initial rate was observed when Ph2BOH was subjected to TsCl/i-
Pr2NEt for 1 h before adding 4d.
(43) (a) Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109,
1746−1754. (b) Vachal, P.; Jacobsen, E. N. J. Am. Chem. Soc. 2002,
124, 10012−10014. (c) Pluth, M. D.; Bergman, R. G.; Raymond, K. N.
J. Am. Chem. Soc. 2008, 130, 11423−11429.
(12) Boronic acids have been employed as catalysts for dehydrative
acylation of amines and alcohols with carboxylic acids. The proposed
mechanism involves the formation of acyloxyboranes (i.e., electrophile
activation, rather than nucleophile activation). See: (a) Al-Zoubi, R.
M.; Marion, O.; Hall, D. G. Angew. Chem., Int. Ed. 2008, 47, 2876−
2879. (b) Maki, T.; Ishihara, K.; Yamamoto, H. Tetrahedron 2007, 63,
8645−8657.
(13) (a) James, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. Angew.
Chem., Int. Ed. 1996, 35, 1910−1922. (b) Davis, A. P.; James, T. D. In
Functional Synthetic Receptors; Schrader, T., Hamilton, A. D., Eds.;
Wiley: Weinheim, 2005; pp 45−109.
(14) For a review of catalytic reactions based on reversible, covalent
interactions: Tan, K. L. ACS Catal. 2011, 1, 877−886.
(15) (a) Oshima, K.; Kitazono, E.-i.; Aoyama, Y. Tetrahedron Lett.
1997, 38, 5001−5004. (b) Oshima, K.; Aoyama, Y. J. Am. Chem. Soc.
1999, 121, 2315−2316.
(16) Barton, D. H. R.; Ferreira, J. A.; Jazsberenyi, J. C. In Preparative
Carbohydrate Chemistry; Hanessian, S., Ed.; Marcel Dekker: New York,
1997; pp 85−101.
(44) Our ability to carry out kinetic studies of the glycosylation
reaction has been somewhat limited due to oxidation of the
diphenylborinic acid catalyst by Ag2O under the reaction conditions.
This catalyst decomposition pathway poses a challenge for monitoring
reactions to completion and for assessing the relative reactivities of
different organoboron catalysts.
(17) (a) Munavu, R. M.; Szmant, H. H. J. Org. Chem. 1976, 41,
1832−1836. (b) Shanzer, A. Tetrahedron Lett. 1980, 21, 221−222.
(c) Thiem, J.; Wessel, H.-P. Tetrahedron Lett. 1980, 21, 3571−3574.
(d) Tsuda, Y.; Nishimura, M.; Kobayashi, T.; Sato, Y.; Kanemitsu, K.
Chem. Pharm. Bull. 1991, 39, 2883−2887. (e) Boons, G.-J.; Castle, G.
H.; Chase, J. A.; Grice, R.; Ley, S. V.; Pinel, C. Synlett 1993, 12, 913−
914. (f) David, S.; Hanessian, S. Tetrahedron 1985, 41, 643−663.
(g) Grindley, T. B. Adv. Carbohydr. Chem. Biochem. 1998, 53, 17−142.
(18) Martinelli, M. J.; Vaidyanathan, R.; Pawlak, J. M.; Nayyar, N. K.;
Dhokte, U. P.; Doecke, C. W.; Zollars, L. M. H.; Moher, E. D.; Van
Khau, V.; Kosmrjl, B. J. Am. Chem. Soc. 2002, 124, 3578−3585.
(19) Voight, E. A.; Rein, C.; Burke, S. D. J. Org. Chem. 2002, 67,
8489−8499.
(45) Ritchie, C. D.; Sager, W. F. Prog. Phys. Org. Chem. 1964, 2, 323−
400.
(20) Bucher, B.; Curran, D. P. Tetrahedron Lett. 2000, 41, 9617−
9621.
(21) For discussions of the origins of regioselectivity in reactions of
stannylene acetals, see ref 17g and: (a) Grindley, T. B.; Thangarasa, R.
Can. J. Chem. 1990, 68, 1007−1019. (b) Kong, X.; Grindley, T. B. Can.
J. Chem. 1994, 72, 2405−2415. (c) Dong, H.; Zhou, Y.; Pan, X.; Cui,
F.; Liu, W.; Liu, J.; Ramstrom, O. J. Org. Chem. 2012, 77, 1457−1467.
̈
(22) Nishiguchi, T.; Taya, H. J. Am. Chem. Soc. 1989, 111, 9102−
9103.
(23) Zhu, P. C.; Lin, J.; Pittman, C. U., Jr. J. Org. Chem. 1995, 60,
5729−5731.
(24) Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in Organic
Synthesis, Wiley-VCH: Weinheim, 2006.
(25) (a) Clarke, P. A. Tetrahedron Lett. 2002, 43, 4761−4763.
(b) Clarke, P. A.; Kayaleh, N. E.; Smith, M. A.; Baker, J. R.; Bird, S. J.;
Chan, C. J. Org. Chem. 2002, 67, 5226−5231.
(26) Wurz, R. P. Chem. Rev. 2007, 107, 5570−5595.
(27) Matsumura, Y.; Maki, T.; Murakami, S.; Onomura, O. J. Am.
Chem. Soc. 2003, 125, 2052−2053.
́
(28) Bouzide, A.; Sauve, G. Org. Lett. 2002, 4, 2329−2332.
(29) Choudary, B. M.; Chowdari, N. S.; Kantam, M. L. Tetrahedron
2000, 56, 7291−7298.
(30) Demizu, Y.; Matsumoto, K.; Onomura, O.; Matsumura, Y.
Tetrahedron Lett. 2007, 48, 7605−7609.
(31) Fiori, K. W.; Puchlopek, A. L. A.; Miller, S. J. Nature Chem.
2009, 1, 630−634.
(32) Wong, J. Y.; Leznoff, C. C. Can. J. Chem. 1973, 51, 2452−2456.
(33) (a) Garegg, P. J. Pure Appl. Chem. 1984, 56, 845−858.
(b) Barton, D. H. R.; Zhu, J. Tetrahedron 1992, 48, 8337−8346.
(34) Nagashima, N.; Ohno, M. Chem. Pharm. Bull. 1991, 39, 1972−
1982.
́
(35) Bouzide, A.; Sauve, G. Tetrahedron Lett. 1997, 38, 5945−5948.
8267
dx.doi.org/10.1021/ja302549c | J. Am. Chem. Soc. 2012, 134, 8260−8267