Deprotection of pinacolyl boronate esters via hydrolysis of intermediate potassium trifluoroborates

Article abstract of DOI:10.1016/j.tetlet.2005.09.101

An efficient two-step procedure for the deprotection of pinacolyl organoboronate esters is described. Reaction with excess potassium hydrogen fluoride produces the corresponding stable, crystalline potassium organotrifluoroborate salts. Treatment of the trifluoroborates with either inorganic base or trimethylsilyl chloride and water affords the corresponding organoboronic acid in high yield.

Full text of DOI:10.1016/j.tetlet.2005.09.101

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 7899–7903
Deprotection of pinacolyl boronate esters via hydrolysis
of intermediate potassium trifluoroborates
Alexander K. L. Yuena and Craig A. Huttona,b,c,
*
aSchool of Chemistry, The University of Sydney, NSW 2006, Australia
bSchool of Chemistry, The University of Melbourne VIC 3010, Australia
cBio21 Molecular Science and Biotechnology Institute, The University of Melbourne VIC 3010, Australia
Received 28 July 2005; revised 6 September 2005; accepted 16 September 2005
Available online 3 October 2005
Abstract—An efficient two-step procedure for the deprotection of pinacolyl organoboronate esters is described. Reaction with excess
potassium hydrogen fluoride produces the corresponding stable, crystalline potassium organotrifluoroborate salts. Treatment of the
trifluoroborates with either inorganic base or trimethylsilyl chloride and water affords the corresponding organoboronic acid in high
yield.
ꢀ 2005 Elsevier Ltd. All rights reserved.
The utility of organoboronic acids in organic synthesis
has flourished in recent years, particularly through
developments in the Miyaura–Suzuki coupling,1 allyl-
boration,2,3 copper-catalysed arylboronic acid–hetero-
atom coupling4 and the Petasis reaction.5,6 Miyauraꢀs
protocol7 for the palladium-catalysed boronation of aryl
and vinyl halides with bis(pinacolato)diboron has
become one of the most popular methods for incorpo-
rating boron into functionalised substrates under mild
conditions. The resultant pinacolyl boronate esters have
the advantage of being stable compounds, which gener-
ally can be purified by chromatography.
higher yielding and more general with boronic acid sub-
strates.10 Hence, the conversion of pinacolyl organoboro-
nate esters to the corresponding organoboronic acids is
of much interest.
Methods currently available for the deprotection of
pinacolyl esters include destructive procedures such as
the use of periodate to cleave the protecting diol oxida-
tively,11 or hydrolytic protocols, for example, displace-
ment of the diol with diethanolamine to form an
intermediate ꢁate complexꢀ, which is then hydrolysed
under acidic or basic conditions.12 Transesterification
of pinacolyl esters to yield the free boronic acids has
been achieved previously using mild conditions under
phase-transfer13 or homogeneous conditions,9 but such
methods normally suffer from incomplete reaction or
problems in separating the desired boronic acid from
a large excess of the transesterification partner. We
recently developed an improved transesterification
procedure, which employs excess polymer-supported
boronic acid to deprotect a variety of arylboronic pinac-
olyl esters in good to excellent yields. Ease of purifica-
tion and the ability to regenerate the solid-supported
reagent were also advantages of this approach.14 The
only current limitation of this procedure, however,
remains the inconsistent commercial availability of the
polymer-supported boronic acid.
Although pinacolyl boronate esters are useful substrates
for cross-coupling chemistry in their own right, many
procedures are more efficient with or have an absolute
requirement for free boronic acids in order to proceed.
The arylation of phenols and a variety of nitrogen nucleo-
philes in the presence of arylboronic acids and Cu(I) or
Cu(II) catalysts has been fine-tuned over recent years to
represent one of the best methods for heteroatom aryl-
ation.4 However, these reactions proceed in very low
yields with pinacolyl boronate esters.8,9 Similarly, the
three-component coupling of an aldehyde, amine and
organoboron species (Petasis reaction) can proceed with
a pinacolyl boronate ester in some circumstances, but is
*
We report herein an alternative method for pinacolyl
Corresponding author. Tel.: +61 3 8344 2393; fax: +61 3 9347
8124; e-mail: chutton@unimelb.edu.au
ester deprotection using a two-step process that
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.09.101

7900
A. K. L. Yuen, C. A. Hutton / Tetrahedron Letters 46 (2005) 7899–7903
proceeds via a readily isolable potassium trifluorobo-
rate. Aryl trifluoroborate salts are readily prepared from
the corresponding boronate esters and are crystalline,
air and moisture stable, and can be stored indef-
initely without oligomerisation.15 Liberation of the
desired free boronic acid can be achieved effectively
using either basic hydrolysis with a variety of alkali
metal salts or using trimethylsilyl chloride as an effective
fluorophile.
potassium or sodium hydroxide in converting the tri-
fluoroborate to the boronic acid. The efficacy of lithium
hydroxide at hydrolysing the trifluoroborate may be in
part due to the large lattice enthalpy of the lithium fluo-
ride formed (1047 kJ molÀ1
)
and its low solubility in
18
water (ca. 1.3 mg/mL at 25 ꢁC),19 compared with either
sodium or potassium fluoride. Vedejs et al.17 noted in
their work concerning aryltrifluoroborates that the
counter-ion present is largely responsible for the stabil-
ity of the complex. While potassium, and to a lesser
extent sodium, led to the most stable trifluoroborates,
lithium and magnesium counterions caused trifluoro-
borate decomposition in solution due to their fluorophilic
natures.
We initially investigated the conversion of phenyl pinac-
olylboronate 1 to the corresponding boronic acid 3 via
potassium phenyltrifluoroborate 2. The pinacolyl boro-
nate was converted to the trifluoroborate by treatment
with aqueous potassium hydrogen fluoride according
to Geneˆtꢀs16 modification of Vedejsꢀ17 procedure
(Scheme 1). The resultant crude potassium trifluorobo-
rate was recrystallised from acetone/diethyl ether to
afford the pure product, free of pinacol.
Despite the difference in reactivities between lithium,
sodium and potassium hydroxides, the corresponding
carbonates were all effective bases in the hydrolysis of
the phenyl trifluoroborate (entries 4–6). Additionally,
sodium bicarbonate was found to be a reasonably effec-
tive base, with potassium bicarbonate being less effective
(entries 7–9). Thus, while generally basic conditions lead
to hydrolysis, it is important that the cation be chosen to
ensure completion, with lithium being slightly more
effective than sodium, and potassium being the least
effective. Acetonitrile was found to be a better co-solvent
than acetone, despite the increased solubility of the sub-
strates in the latter (entries 4 and 6).
Hydrolysis of phenyl trifluoroborate was then investi-
gated with a range of aqueous alkali metal hydroxides
and carbonates in mixtures of aqueous acetone or aceto-
nitrile (Table 1). Solvent choice was limited by the solu-
bility of the trifluoroborate, with DMSO being avoided
for ease of work-up and alcohols being incompatible
due to the subsequent formation of boronate esters.
Comparison of entries 1–3 in Table 1 clearly indicates
that lithium hydroxide is much more effective than either
The hydrolysis of phenyltrifluoroborate 2 presumably
proceeds via the hydroxydifluoro 4 and dihydroxyfluoro
5 borate species, as outlined in Scheme 2. Previous work
by the Molander20,21 and Batey22,23 groups sought to
identify the nature of hydrolytic intermediates of aryl-
trifluoroborates, as part of research into their utility in
palladium-catalysed cross-coupling reactions. Although
unable to reveal the identities of the hydrolytic interme-
diates unequivocally, NMR studies20–22 and model cou-
plings of trifluoroborates, with and without added
base,23 led both groups to conclude that exchange of
fluoride for hydroxide ligands occurred and was neces-
sary for successful cross-coupling. We conducted 19F
NMR studies of the hydrolysis of potassium phenyltri-
fluoroborate, which supports these suggestions (Figs. 1
O
base
KHF2
BF3K
B(OH)2
B
O
solvent/H2O
1
2
3
Scheme 1.
Table 1. Hydrolysis of potassium phenyltrifluoroborate 2 to phen-
ylboronic acid 3
Entry
Base
Equiv
% Conversion
CH3CN/H2Oa Acetone/H2Ob
1
2
3
4
5
6
7
8
9
LiOH
NaOH
KOH
Li2CO3
Na2CO3
K2CO3
NaHCO3
KHCO3
KHCO3
3.0
3.0
3.0
1.5
1.5
1.5
3.0
3.0
5.0
100
84
74
100
100
100
—
—
—
—
80
—
84
84
74
74
—
—
Figure 1. 19F NMR spectra of PhBF3K 2 in D2O (a), and after
addition of 1.0 (b) and 2.0 (c) equivalents of LiOH.
a PhBF3K (1 mmol) in acetonitrile (10 mL) and water (9 mL), 4–9 h.
b PhBF3K (1 mmol) in acetone (5 mL) and water (5 mL), 7–24 h.
3
Ph B(OH)2
+
KOH
LiOH
LiOH
LiOH
Ph BF3 K
Ph B(OH)F2 K
Ph B(OH)2F K
Ph B(OH)3 K
(– LiF)
(– LiF)
(– LiF)
2
4
5
6
Scheme 2. Presumed potassium phenyltrifluoroborate hydrolysis pathway.

A. K. L. Yuen, C. A. Hutton / Tetrahedron Letters 46 (2005) 7899–7903
7901
and 2). The first NMR experiment was conducted by the
sequential addition of aqueous lithium hydroxide to a
solution of potassium phenyltrifluoroborate in D2O
(Fig. 1). The phenyltrifluoroborate anion gives rise to
a
multiplet (arising from B–F coupling) at
d
À135 ppm. After addition of 1 equiv of lithium hydrox-
ide two new broad peaks were observed at d À119 and
À132 ppm. After the addition of 2 equiv of LiOH, only
fluoride was observed as a sharp singlet at d À119 ppm,
1
and phenylboronic acid was observed by H and 11B
NMR.
Figure 3. ESI MS of potassium phenyltrifluoroborate treated with
1.0 equiv of LiOH.
In contrast to the above experiment, addition of 3 equiv
of KHCO3 over a 10 h time-period failed to effect com-
plete hydrolysis of potassium phenyltrifluoroborate, but
rather, an equilibrium was reached in which all three
species are present, with the formation of the boronic
acid favoured (Fig. 2).
Having established that lithium bases were well suited to
the task of trifluoroborate hydrolysis, various aryl pin-
acolyl boronate substrates were first converted to the
corresponding trifluoroborates and then subjected to
basic hydrolysis with lithium hydroxide in acetonitrile/
water (Table 2).
Under both sets of conditions, a broad peak at d
À132 ppm was observed in the 19F NMR spectra, pre-
sumably attributable to intermediate 4 or 5. Neither
19F nor 11B NMR were able to identify the nature of
the intermediate, but ESI mass spectrometry was able
to identify the presence of difluorohydroxy species 4
under similar conditions: a solution of potassium
phenyltrifluoroborate in methanol was treated with
1 equiv of lithium hydroxide, and the solution analysed
by negative ion electrospray ionisation (ESI) mass spec-
trometry (Fig. 3). The three main species present were
identified as the trifluoroborate 2, which gave rise to a
peak at m/z 145, the difluorohydroxy species 4, which
gave rise to a peak at m/z 143, and the difluoromethoxy
species (m/z 157), which arises from exchange of –OH
for –OMe. The peaks at m/z 325–353 correspond to
potassium-linked homo- and hetero-dimers of these
anions. These results indicate that the difluorohydroxy
species 4 is stable enough to be observed by mass
spectrometry and is presumably the species that gives
rise to the signal at d À132 ppm in the 19F spectra in
Figures 1 and 2. It cannot be ruled out however that
the broadness of these signals is due to ligand exchange
and that both 4 and 5 exist in equilibrium in solution,
with 4 being the only species stable enough to withstand
ESI conditions.
Conversion of the pinacolyl boronates to the trifluoro-
borates proceeded in good to excellent yields under the
same conditions described previously.24 Hydrolysis of
the trifluoroborates to the boronic acids proceeded well
in most cases; however, mixed results were obtained for
2-substituted aryltrifluoroborates. While both the 2-tos-
ylate and 2-formyl derivatives (entries 7 and 8) were
hydrolysed completely, hydrolysis of the 2-hydroxy
derivative resulted in decomposition of the starting
material to a dark polymeric mixture (entry 4) and in
the case of the 2-methoxy derivative only a low yield
of the boronic acid was obtained (entry 5).
An alternative approach to the conversion of the triflu-
oroborates to the corresponding boronic acids was then
investigated, based on the work of Vedejs et al.17 con-
cerning the conversion of trifluoroborates to reactive
difluoroborane species. Vedejs had shown that treat-
ment of an aryl trifluoroborate with 1 equiv of trimeth-
ylsilyl chloride, under anhydrous conditions, led to the
rapid formation of a difluoroborane, with concomitant
formation of potassium chloride and trimethylsilyl fluo-
ride. Matteson and Kim27 were able to use a modified
version of this approach with silicon tetrachloride to
prepare cyclic, secondary amines from alkyltrifluorobo-
rates via intramolecular cyclisation of the intermediate
difluoroborane onto a pendant azido group. It was rea-
soned that treatment of aryltrifluoroborates with excess
trimethylsilyl chloride in the presence of water would
lead to initial formation of the difluoroborane 7, which
would react with water to give the hydroxydifluoroboro-
nate species 4. Sequential removal of the remaining
fluorides by TMS-Cl and in situ trapping with water
would lead to the boronic acid 3 (Scheme 3).
To our delight, treatment of the aryltrifluoroborates
with 3 equiv of trimethylsilyl chloride and 3 equiv of
water in acetonitrile resulted in rapid conversion to the
boronic acid in all cases (Table 2, method B). Yields
were good to excellent and the work-up procedure was
simpler than for the corresponding reactions with
Figure 2. 19F NMR of PhBF3K 2 + 3 equiv of KHCO3; (a) t = 10 min;
(b) t = 2 h; (c) t = 6 h; (d) t = 10 h.

7902
A. K. L. Yuen, C. A. Hutton / Tetrahedron Letters 46 (2005) 7899–7903
Table 2. Conversion of pinacolyl esters to boronic acids via the corresponding trifluoroborates
method A;
LiOHaq, CH3CN
O
B
B(OH)2
BF3K
KHF2 (4.5 M)
MeOH, r.t.
O
R
R
R
OR method B;
3
1
2
TMS-Cl, H2O
Entry
R
H
2-Me
4-Me
2-OH
2-OMe
4-OMe
2-OTs
2-CHO
2-Cl
Yield of 2 (%)24
Yield of 3 (%) (% 2 remaining in brackets)
Method A25
Method B26
1
2
3
4
5
6
7
8
73
97
77
87
97
95
63
95
95
80
67
94
Quant.
71 (18)
85 (10)
0 (decomp.)
49 (12)
95 (2)
—
Quant.
—
90
70
—
Quant.
Quant.a
21a(75)
Quant.
—
Quant.
Quant.
Quant.
—
9
10
11
12
4-F
4-Br, 3-NHCbz
4-Br, 3-CO2Me
93
—
Quant.
a 3.5 equiv of LiOH used.
H2O
5. Petasis, N. A.; Zavialov, I. A. J. Am. Chem. Soc. 1997,
119, 445–446.
6. Petasis, N. A.; Yudin, A. K.; Zavialov, I. A.; Prakash, G.
K. S.; Olah, G. A. Synlett 1997, 606–608.
7. Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem.
1995, 60, 7508–7510.
TMS-Cl
Ar BF3 K
(– TMS-F)
2
Ar BF2
7
+
KCl
Ar B(OH)F2 K
(– HCl)
4
TMS-Cl,
H2O
3
Ar B(OH)2
8. Chan, D. M. T.; Monaco, K. L.; Li, R.; Bonne, D.; Clark,
C. G.; Lam, P. Y. S. Tetrahedron Lett. 2003, 44, 3863–
3865.
Scheme 3.
9. Decicco, C. P.; Song, Y.; Evans, D. A. Org. Lett. 2001, 3,
1029–1032.
10. Koolmeister, T.; Sodergren, M.; Scobie, M. Tetrahedron
Lett. 2002, 43, 5965–5968.
11. Nakamura, H.; Fujiwara, M.; Yamamoto, Y. J. J. Org.
Chem. 1998, 63, 7529–7530.
12. Song, Y.-L.; Morin, C. Synlett 2001, 266–268.
13. Coutts, S. J.; Adams, J.; Krolikowski, D.; Snow, R.
Tetrahedron Lett. 1994, 35, 5109–5112.
lithium hydroxide. The only by-products were inorganic
salts and the volatile trimethylsilyl fluoride. Noteworthy
is the fact that these conditions are compatible with base
sensitive substrates, such as the free phenol (compare
methods A and B, entry 4) and that the procedure led
to an improved isolated yield of difficult substrates such
as those containing ortho electron-donating substituents
(compare methods A and B, entry 5).
14. Pennington, T. E.; Hardiman, C.; Hutton, C. A. Tetra-
hedron Lett. 2004, 45, 6657–6660.
In conclusion, implementation of a practical two-step
protocol for the mild deprotection of pinacolyl boro-
nate esters has been demonstrated. Following conver-
sion of the pinacolyl boronate to the corresponding
trifluoroborate with potassium hydrogen fluoride, sub-
sequent fluoride removal/hydroxylation was achieved
with alkali metal bases or more effectively using tri-
methylsilyl chloride and water. These mild conditions
tolerated a range of functional groups and were rela-
tively unaffected by the steric or electronic properties
of the aromatic ring.
15. Molander, G. A.; Rivero, M. R. Org. Lett. 2002, 4, 107–
109.
16. Darses, S.; Geneˆt, J.-P. Eur. J. Org. Chem. 2003, 4313–
4327.
17. Vedejs, E.; Chapman, R. W.; Fields, S. C.; Lin, S.;
Schrimpf, M. R. J. Org. Chem. 1995, 60, 3020–3027.
18. Aylward, G.; Findlay, T. SI Chemical Data, 3rd ed.;
Jacaranda Wiley Ltd: Berlin, 1994; p 125.
19. The Merck Index of Chemicals and Drugs; Stecher, P. G.,
Ed.; Merck & Co: Rahway, NJ, USA, 1960; p 613.
20. Molander, G. A.; Ito, T. Org. Lett. 2001, 3, 393–396.
21. Molander, G. A.; Biolatto, B. Org. Lett. 2002, 4, 1867–
1870.
22. Thadani, A. N.; Batey, R. A. Org. Lett. 2002, 2, 3827–
3830.
References and notes
23. Batey, R. A.; Quach, T. D. Tetrahedron Lett. 2001, 42,
9099–9103.
1. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457–2483.
2. Roush, W. R.; Adam, M. A.; Walts, A. E.; Harris, D. J. J.
Am. Chem. Soc. 1986, 108, 3422–3434.
3. Flamme, E. M.; Roush, W. R. J. Am. Chem. Soc. 2002,
124, 13644–13645.
24. Representative procedure for preparation of potassium
aryltrifluoroborates 2 (Table 2, entry 12): to a solution of
4-bromo-3-methoxycarbonylphenyl pinacolylboronate (814
mg, 2.39 mmol) in methanol (7 mL) was added aqueous
potassium hydrogen fluoride (3.0 mL, 4.5 M, 13.5 mmol).
The resulting white slurry was stirred at room temperature
4. Ley, S. V.; Thomas, A. W. Angew. Chem., Int. Ed. 2003,
42, 5400–5449.

A. K. L. Yuen, C. A. Hutton / Tetrahedron Letters 46 (2005) 7899–7903
7903
for 15 min, concentrated in vacuo and dissolved in hot
acetone. The mixture was filtered, the filtrate was concen-
trated in vacuo and the residue recrystallised from a
minimal amount of hot acetone and ether (20 mL), to
afford the corresponding potassium trifluoroborate as a
crystalline solid (721 mg, 94%); mp 226–228 ꢁC; Found: C,
28.78, H, 2.36; C8H6BF3BrO2KÆH2O requires C, 28.35, H,
2.38; 1H NMR (300 MHz, acetone-d6) d 7.87 (1H, d,
J = 1.1 Hz), 7.49 (1H, dd, J = 1.1, 7.8 Hz), 7.42 (1H, d,
J = 7.8 Hz), 3.84 (3H, s); 13C NMR (75 MHz, acetone-d6)
d 168.4, 137.1, 135.2, 132.6, 131.6, 118.1, 52.1 (C ipso to B
not observed); 19F NMR (282 MHz, acetone-d6) d À142.9
(br m); MS (ESI, negative ion); found m/z 280.9619,
C8H611B79BrF3O2 requires m/z 280.9611.
with ethyl acetate (3 · 10 mL). The combined organic
extracts were dried over sodium sulfate, filtered and
concentrated in vacuo to afford 4-fluorophenylboronic
acid as a white solid (140 mg, 100%).
26. Representative procedure for hydrolysis of aryltrifluoro-
borates 2 with TMS-Cl/H2O (Table 2 entry 12): to a
solution of potassium 4-bromo-3-methoxycarbonyl-phen-
yltrifluoroborate (107 mg, 0.33 mmol) in acetonitrile
(3 mL) and water (18 lL, 1.00 mmol), was added trimeth-
ylsilyl chloride (127 lL, 1.00 mmol). The resulting suspen-
sion was stirred for 1 h, quenched with saturated aqueous
sodium hydrogencarbonate solution (0.5 mL) and dried
over sodium sulfate. The mixture was filtered and the
filtrate concentrated in vacuo to afford the corresponding
boronic acid as a white solid (86 mg, 100%); 1H NMR
(400 MHz, acetone-d6) d 8.24 (1H, d, J = 1.7 Hz), 7.87
(1H, dd, J = 1.7, 8.0 Hz), 7.71 (1H, d, J = 8.0 Hz), 3.90
(3H, s); 13C NMR (75 MHz, acetone-d6) d 167.3, 139.0,
137.4, 134.2, 132.9, 123.9, 52.6 (C ipso to B not observed);
11B NMR (128 MHz, acetone-d6) d 26.2 (br s).
25. Representative procedure for hydrolysis of aryltrifluoro-
borates 2 with LiOH (Table 2, entry 10): to a mixture of
potassium 4-fluorophenyl trifluoroborate (202 mg,
1.00 mmol) and lithium hydroxide (84 mg, 3.50 mmol)
was added water (5 mL) and acetonitrile (10 mL). The
resulting solution was stirred at room temperature for
20 h, acidified with saturated aqueous ammonium chloride
(8 mL) and 1 M hydrochloric acid (2 mL), then extracted
27. Matteson, D. S.; Kim, G. Y. Org. Lett. 2002, 4, 2153–
2155.