N. Soga, T. Yoshiki, A. Sato et al.
Tetrahedron Letters 69 (2021) 152977
as BH3-THF [9] or BH3-SMe2 [10]. These complexes are stable and
easy to use; however, they are not inexpensive and need to be
manipulated under inert atmosphere. Once the bottles storing
these complexes are opened, they begin to decompose because of
the exposure to air and moisture. As a result, the purity of the
reagents decreases. Another method to employ borane is generat-
ing it in situ from borohydride salts with oxidants such as sulfuric
acid, iodide, or alkyl iodides [11]. Borohydrides should be reacted
gradually and carefully with oxidants under inert atmosphere,
which is a tedious and exhausting procedure. We herein propose
the in situ generation of borane using the PV method with a test
tube (Fig. 1).
yields of the desired alcohols (2j, 2k) decreased (Table 1, Entries
10 and 11, Condition A). A possible reason could be that some of
the acid reacted with the substrates in the organic phase, thereby
decreasing the amount of borane evolved. Therefore, we attempted
to generate borane from NaBH4 and I2 (Condition B, optimisation of
this condition is summarised in Table S2 in the Supplementary
Material). Diffusion of I2 alone into the fluorous phase was too slow
to generate borane; therefore, I2 was dissolved in 1,2-dibro-
moethane (
q
= 2.17 g cmÀ3) [4a] and FC-72 was employed as the
fluorous phase because of its low viscosity. The PV systems
employing I2 at the initial and final stages of hydroboration are
shown in Fig. S2a and S2b in the Supplementary Material, respec-
tively. These systems afforded the desired alcohols in better yields
with many substrates compared to that obtained using H2SO4. As
shown in Table 1, I2 can be used as the oxidant in the PV method
for borane evolution with a wide range of substrates. However,
their isolated yields sometimes significantly decrease compared
to the NMR yields. In addition, products obtained using the PV sys-
tem with I2 are always contaminated with 1,2-dibromoethane.
Therefore, it is necessary to separate the compound during the
workup, which often further lowered the isolated yields of the
desired products (Table 1, Entries 1, 10–13, Condition B). Hence,
we explored alternative oxidants. As alkyl iodides are known to
react with the borohydride and form borane, we examined methyl
Results and discussion
Hydroboration of alkenes using borane-evolution PV method
Hydroboration of 4-methylstyrene was carried out using the PV
method for borane gas evolution (1a). The optimisation of reaction
conditions has been provided in detail in the Supplementary Mate-
rial. 1,2-Dimethoxyethane (DME;
as the organic phase (top layer), while FC-72 (
den HT-135 (
= 1.72 g cmÀ3), or HT-200 (
q
= 0.868 g cmÀ3) was employed
q
= 1.68 g cmÀ3), Gal-
q
= 1.79 g cmÀ3) was
q
used as the fluorous phase (middle layer) [12]. NaBH4 was used
iodide (q q
= 2.28 g cmÀ3) or ethyl iodide ( = 1.94 g cmÀ3) as oxi-
as the borane source, because it is stable, easy to handle, and inex-
dants because these iodides are denser than fluorous solvents.
Methyl iodide diffused quickly into the fluorous phase (even into
the more viscose Galden HT-200), evolving a large amount of bor-
ane in a short time. However, it was not suitable for the PV hydrob-
oration reaction because a substantial amount of borane escaped
from the organic layer without being used in the reaction. Ethyl
iodide diffuses into the fluorous phase more slowly than methyl
iodide, and hence, it is preferred for application in the PV method.
Indeed, when a mixture of Galden HT-135 and HT-200 (2:1 v/v)
was employed as the fluorous phase (Condition C, optimisation
of this condition is summarised in Table S3 in the Supplementary
Material), ethyl iodide worked very well, affording the desired
hydroboration products in good yields. The initial and final stages
of PV hydroboration using ethyl iodide are shown in Fig. 2a and b,
respectively. The organic by-product of the PV reaction using ethyl
iodide was ethane. The workup procedure was simpler and easier
than that involving iodine. In addition, nBu4NBH4 was employed
instead of NaBH4 as the borane source (Condition D). THF was
employed as the organic phase in this case because the product
yields in DME were slightly lower than those in THF. It should be
noted that nBu4NBH4 dissolved in the solvent. Hence, the PV
method with this reagent has three phases at the initial stage of
hydroboration (Fig. S3a in the Supplementary Material), and the
bottom layer vanishes in 2 h (Fig. S3b in the Supplementary Mate-
rial). After oxidation with H2O2 under basic conditions, the desired
alcohols were obtained in good yields as listed in Table 1. PV
hydroboration with nBu4NBH4 afforded the alcohols in similar
yields as that obtained using ethyl iodide.
pensive. Fortunately, the density of the reagent is 1.074 g cmÀ3
,
and hence, it can float at the interface of the fluorous and organic
phases. Oxidants should sink to the bottom layer, as they are den-
ser than the fluorous phase. Under the standard conditions (Condi-
tion A), we used H2SO4 (q
= 1.83 g cmÀ3) as the oxidant and Galden
HT-135 as the fluorous phase. However, we slightly modified the
conditions from A to E; the results are summarised in Table 1.
When Condition A was applied, the reaction afforded a quad-
riphasic system comprising H2SO4, Galden HT-135, NaBH4, and
DME solution of 1a, as shown in Fig. S1a in the Supplementary
Material (detailed optimisation of this condition is summarised
in Table S1 in the Supplementary Material). After degassing the
test tube with a syringe, the bottom layer was gently stirred at
25 °C, taking care not to mix the layers. Continuous borane gas evo-
lution was observed, and the bottom layer disappeared after 20 h
(Fig. S1b in the Supplementary Material). At a glance, H2SO4
worked well to evolve borane, and the desired product, 2-(p-
tolyl)-ethan-1-ol (2a), was obtained in an acceptable yield after
the oxidation of the reaction mixture with H2O2 under basic condi-
tions (Table 1, Entry 1, Condition A), compared with the yield of the
product obtained using BH3-THF as the borane source (Table 1,
Entry 1, Condition E). However, when alkenes bearing a hydroxy
or an ester group (1j, 1k) were employed as the substrate, the
b
b
a
a
l
l
l
o
l
o
o
o
n
n
As summarised in Table 1, we investigated the substrate scope
for the hydroboration–oxidation reaction of alkenes using the PV
method. Not unexpectedly, anti-Markovnikov products 2 were
obtained as the major products, while the Markovnikov products
3 were sometimes observed in trace amounts except for some
styrenes described below. Styrene derivatives bearing an elec-
tron-donating group afforded the corresponding alcohols in good
yields (Entries 1, 2), while non-substituted styrene 1c provided
the desired alcohols in similar yields (Entry 3). Styrenes with a
halogen or an electron-withdrawing group afforded the corre-
sponding products in moderate-to-good yields (Entries 4–6).
Because hydroboration is an electrophilic reaction, styrene deriva-
tives with an electron-withdrawing group provided slightly lower
Substrate
Product
gentle stirring
BH 3
Organic solvent
NaBH4
Fluorous solvent
Oxidant
Stirrer bar
Fig. 1. Concept of borane evolution PV method.
2