Selective deoxygenation of gibberellic acid with fluoroarylborane catalysts

Article abstract of DOI:10.1016/j.tet.2019.130712

Reductive late-stage functionalization of gibberellic acid is reported using three fluoroarylborane Lewis acids; (B(C₆F₅)₃, B(3,5-C₆H₃(CF₃)₂), and B(2,4,6-C₆H₂F₃)₃) in combination with a tertiary silane and a borane (HBCat) reductant. In each case, C–O bond activation occurs, and different products are obtained depending on the reductant and catalyst employed.

Full text of DOI:10.1016/j.tet.2019.130712

Journal Pre-proof
Selective deoxygenation of gibberellic acid with fluoroarylborane catalysts
Youngran Seo, Anton Gudz, Jared M. Lowe, Michel R. Gagné
PII:
S0040-4020(19)31092-0
https://doi.org/10.1016/j.tet.2019.130712
TET 130712
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 26 August 2019
Revised Date: 14 October 2019
Accepted Date: 16 October 2019
Please cite this article as: Seo Y, Gudz A, Lowe JM, Gagné MR, Selective deoxygenation of gibberellic
acid with fluoroarylborane catalysts, Tetrahedron (2019), doi: https://doi.org/10.1016/j.tet.2019.130712.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Selective deoxygenation of gibberellic acid
with fluoroarylborane catalysts
Leave this area blank for abstract info.
Youngran Seo1, Anton Gudz1, Jared M. Lowe, and Michel R. Gagné*
Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina 27599, United States.

1
Tetrahedron
journal homepage: www.elsevier.com
Selective deoxygenation of gibberellic acid with fluoroarylborane catalysts
Youngran Seo1, Anton Gudz1, Jared M. Lowe, and Michel R. Gagné
Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina 27599, United States.
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Reductive late-stage functionalization of gibberellic acid is reported using three fluoroarylborane
Lewis acids; (B(C6F5)3, B(3,5-C6H3(CF3)2), and B(2,4,6-C6H2F3)3) in combination with a tertiary
silane and a borane (HBCat) reductant. In each case, C–O bond activation occurs, and different
products are obtained depending on the reductant and catalyst employed.
Available online
Keywords:
Late stage functionalization
Gibberellic acid
We are very pleased to dedicate this article to Prof. Steve Davies for his many contributions to
the Tetrahedron Journals and to the broader chemistry community.
Selective deoxygenation
Hydrosilylative deoxygenation
Fluoroarylborane
2009 Elsevier Ltd. All rights reserved
.
———
Corresponding author. e-mail: mgagne@unc.edu
1 These authors contributed equally to this work.

2
Tetrahedron
1. Introduction
Late stage functionalization (LSF) of natural products is a
useful medicinal chemistry tool, but is challenging to execute as
it requires a synthetic method able to chemo- and site-selectively
manipulate complex structures, which, by their nature, usually
contain a diverse array of off-target reactive groups.[1,2] Figure 1
demonstrates the LSF principle for the site-selective
deoxygenation of erythromycin A, enabled either by
selectively-defining O-thiocarbonylation or -phosphitylation with
peptide catalyst (Fig. 1(a)).[3,4] Metal catalysts for
a
a
deoxygenation are also feasible as recently demonstrated for the
ruthenium-catalyzed primary alcohol reduction of cholic alcohol
(Fig. 1(b)).[5] A similar preference for primary deoxygenation
(silane reductant) was demonstrated using the main group Lewis
acid B(C6F5)3 on a diosgenin derivative (Fig. 1(c)).[6] B(C6F5)3-
catalyzed hydrosilylative deoxygenations have additionally been
used to chemo-selectively reduce the hemiacetal of
dihydroartemisinin or to open the lactone ring of gibberellic acid
with an accompanying allylic transposition (Fig. 1(c)).[7]
Although the commercially available B(C6F5)3 has been the
workhorse catalyst, fluorinated arylboranes have also been
investigated for the reduction of diverse functional groups, with
one form typically exhibiting superior reactivity to the others.[8–
10] Our group recently discovered that tuning the position of
fluorination led to catalysts for site-selective deoxygenation that
yielded divergent products from a set of common cellulosic
starting materials.[11] Site-selective modifications in complex
Figure 1. Selective deoxygenation of natural products.
natural products such as natamycin have also been reported using
various fluoroaryl borane catalysts and silanes as reductants.[7]
The choice of reductant (e.g. borane versus silane) can also alter
which product is formed, as seen with carbohydrates under
B(C6F5)3-catalyzed conditions.[12] These results led us to ask
how modifications to the fluoroaryl borane catalyst and the
reductant influences the selectivity for the deoxygenation of a
multi-functional test molecule such as gibberellic acid.[13]
Trialkylsilanes (Me2EtSiH or Et3SiH) and catecholborane
(HBCat) were selected as the reductants while the following three
fluoroarylboranes were chosen based on their differing Lewis
acidity and steric profiles: B(C6F5)3, B((3,5-CF3)2C6H3)3 (BAr3,5-
CF3), and B(2,4,6-F3C6H2)3 (BAr2,4,6-F).
2. Results and Discussion
As previously computed by Heiden, the three fluoroaryl
borane Lewis acids in this study have hydride affinities
(kcal/mol) that decrease from B(C6F5)3 to BAr3,5-CF3 to BAr2,4,6-F
(Fig. 2(a)).[14] These hydricity values correspond to the negative
free energy of hydride formation from the Lewis acid and H–. For
the current study, we additionally computed electrostatic
potential surfaces to compare both the Lewis acid and conjugate
hydride forms; more negative regions of electron distribution are
visualized by red (Fig. 2(b)). Not surprisingly, the borohydrides
are overall more negative than the neutral boranes. Unexpected,
however, was the difference between BAr3,5-CF3 and BAr2,4,6-F
which have a similar hydride affinity. The computed surfaces
indicate that the boron center in BAr3,5-CF3 is the most electron
deficient of the three (most blue) while the hydride derived from
BAr2,4,6-F is the most negatively charged of all three (most red).
We conclude from these data that hydride affinity (a difference)
might not provide the full story of the reactivity of the Lewis
acids and their conjugate hydrides since it is possible that one
form might play a more important role in a given transformation.
Figure 2. Calculated hydride affinity (Heiden) and
electrostatic surfaces of the tested fluoroaryl boranes

3
In the study comparing the three catalysts for the site-selective
deoxygenation of cellulose-derived carbohydrates, the most
significant differences were noted between B(C6F5)3 and BAr3,5-
CF3.[11] For multiple carbohydrate starting materials the two
catalysts favored different products. Although no definitive
reasons were apparent, it was noted that the boron resting state
for B(C6F5)3-catalyzed reactions was B(C6F5)3−H– while in the
latter, it was the Lewis acid form. The higher nucleophilicity of
BAr3,5-CF3–H– presumably promoted its consumption. In this
observations agree with our previous DFT study on the
silyl/boryl oxonium ions that can be formed from 2-propanol.[12]
These calculations showed that a diboryl oxonium is significantly
higher in energy than a mixed silyl-boryl or disilyl oxonium ion.
In other words, boryl protected ethers are insufficiently Lewis
basic to support the formation of
a putative [diboryl
oxonium][(C6F5)3B−H–] Lewis pair. In this situation a Lewis acid
catalyzed allylic transposition becomes competitive over
heterolytic cleavage of HBCat. Thus, the formation of 1 through
C–O bond activation/cleavage predominates when relatively
more basic silyl ethers are present under B(C6F5)3-catalyzed
conditions, while non-reductive pathways dominate when the less
basic boryl ester intermediates are involved.
study BAr2,4,6-F was typically
B(C6F5)3.[15]
a less reactive version of
The previously reported B(C6F5)3-catalyzed hydrosilylative
deoxygenation of free gibberellic acid afforded the tetrasilyl-
protected diester Si-1 in 93% yield via
a cascade of
dehydrosilylation, and reductive olefin migration/ lactone
opening (Scheme 1(a)).[7] To enhance the starting material
solubility and to avoid vigorous hydrogen evolution, the allylic
alcohols of gibberellic acid were pre-silylated (Me2EtSi-, Et3Si-;
Si-Gibb), and tested under ambient reaction conditions with
excess Et3SiH. Unlike the results obtained with Gibb (Scheme
1(a)), Si-Gibb provided 1 (83%) along with the conjugated diene
2 (16%) (Scheme 1(b)).[16] Although it is unclear why pre-
silylated Si-Gibb promotes elimination to 2, all B(C6F5)3-
catalyzed deoxygenations favor product 1 as the major species.
Scheme 2. Deoxygenation with HBCat as reductant.
Next, the less Lewis acidic fluoroarylboranes, BAr2,4,6-F and
BAr3,5-CF3, were tested with Si-Gibb and silane reductants.
Although B(C6F5)3 generates a mixture of 1 and 2, each
compound can be prepared as the sole product by changing the
catalyst (Scheme 3(a)). With 10 mol% BAr2,4,6-F, Me2EtSi-
protected gibberellic acid was fully converted to 1 and isolated in
a good yield (82%). Even though the yield was a little lower than
that obtained with B(C6F5)3 and Gibb, 2 was not detectable by in
situ 13C{1H} NMR. In contrast, using 10 mol% of BAr3,5-CF3
resulted in allylic alcohol reduction coupled with lactone opening
and elimination. In situ 13C{1H} NMR monitoring indicated a
lack of detectable intermediates, with Si-Gibb being smoothly
converted to 2.[17]
We next sought to synthesize 2 from conjugated diene 4 via
Scheme 1. Hydrosilylative deoxygenation of Gibb and Si-
Gibb.
Since free gibberellic acid and silyl protected gibberellic acid
react slightly differently, both forms were tested with the
alternative reductant, catecholborane (HBCat). Interestingly,
different products, 1 or 3, were obtained from Si-Gibb and Gibb
after hydrolysis (Scheme 2). As evidenced from gas evolution on
mixing either Gibb or Si-Gibb with HBCat (no catalyst
necessary), the free alcohols and carboxylic acids were borylated,
making the resulting borylated compounds the actual starting
materials. Under these conditions isomerization to known 3
occurs in contrast to Si-Gibb, which converts to 1 (Scheme 2). In
situ monitoring of the reactions by 19F and 11B NMR
spectroscopy reveal that the fluoroarylborane catalyst rests as
(C6F5)3B−H– in the reaction using Si-Gibb whereas it rests as
various B(C6F5)3 Lewis adducts in reactions using Gibb. These

4
Tetrahedron
selective allylic alcohol reduction using a fluoroarylborane
SilaFlash P60 40-63 ꢀm (230-400 mesh). Thin layer
chromatography (TLC) was performed on SiliCycle Silica Gel 60
F254 plates and was visualized with ceric ammonium molybdate
(CAM) stain. All NMR spectra were recorded on a Bruker
Avance 600 MHz spectrometer at standard temperature and
pressure. All deuterated solvents were used as received from
Cambridge Isotope Laboratories, Inc. The residual solvent
protons (1H) or the solvent carbons (13C) were used as internal
standards. The following abbreviations are used in reporting
NMR data: s, singlet; d, doublet; dd, doublet of doublets; dt,
doublet of triplets; td, triplet of doublets; ddd, doublet of doublet
of doublets; and m, multiplet. Where necessary, 2D COSY, and
HSQC data were used for peak assignment. High Resolution
Mass spectra were obtained on Q ExactiveTM HF-X Hybrid
Quadrupole-OrbitrapTM Mass spectrometer.
catalyst and silane (Scheme 3(b)). Silyl protected diene 4 was
prepared via a previously reported procedure in 64% yield from
gibberellic acid.[13] Under B(C6F5)3-catalyzed hydrosilylative
deoxygenation conditions, 32% of reduced allylic alcohol 2 was
obtained after 24 h, with only trace amounts of 2 being formed
when BAr3,5-CF3 was the catalyst. Few examples of
hydrosilylative allylic alcohol reductions with fluoroarylborane
catalysts are known although several examples of C–O
reductions on cyclic ethers have been reported.[18,19]
Unlike B(C6F5)3 catalyzed reactions, in situ spectroscopic
studies of the BAr2,4,6-F and BAr3,5-CF3 reactions provided no clear
evidence for how the borane speciated. In an attempt to observe a
difference in the behavior of the catalysts an additional 20 mol%
of an external Lewis base (PPh3) was added, which can promote
the heterolysis of the silane into a borohydride/silylphosphonium
ion pair.[20] In the case of BAr2,4,6-F the rate of forming 1 and 3
slowed and resonances for H−BAr2,4,6-F– were observed in the 19F
NMR (-100.2 ppm for o-F, and -120.1 ppm for p-F) along with
All chemicals were used as received, or otherwise described
on how it was treated before use. Me2EtSiH and Et3SiH were
purchased from Gelest, and degassed via three freeze-pump-thaw
cycles and stored over molecular sieves in the glovebox.
Catecholborane was purchased from Sigma-Aldrich, distilled
prior to use, taken into a nitrogen filled glovebox, and stored at -
48 oC. B(C6F5)3 was purchased from Strem and used as received.
BAr2,4,6-F and BAr3,5-CF3 were synthesized via known
methods.[10,24]
its cation Me2EtSi-PPh3 (-3.29 ppm) by 31P NMR.[20,21] By
+
contrast, addition of 20 mol% PPh3 to the BAr3,5-CF3 catalyzed
reaction completely shut down its reduction to 2, and instead
provided only partial conversion of Si-Gibb to 3. As discussed in
Figure 1, despite having a similar net hydride affinity, both the
borohydrides and borane Lewis acids have different electrostatic
potential surfaces. In addition to these electrostatic differences,
the lack of ortho-F groups in BAr3,5-CF3 makes it more sterically
accessible to both the hydride source (Piers mechanism) and
other competing Lewis bases.[9,22–24] The balance of these
forces is especially complex in a multi-functional structure like
gibberellic acid and so the diverging reactivity patterns is
difficult to pin down to a specific feature of the catalysts.
4.2. B(C6F5)3 catalyzed reaction with Si-Gibb and silane
In a N2-filled glove box, B(C6F5)3 (4.9 mg, 0.010 mmol, 0.10
equiv) was placed in a 1 dram vial and dissolved in 0.2 mL of
CH2Cl2. To the catalyst solution was added Me2EtSiH (32 µL,
0.241 mmol, 2.50 equiv) and mixed. In a separate vial, Me2EtSi-
Gibb (50 mg, 0.096 mmol, 1.00 equiv) was diluted with 0.3 mL
of CH2Cl2. The catalyst and borane mixture was then added to the
substrate solution in one portion. The reaction mixture was
transferred to an NMR tube and sealed with a septum cap. After
24 h, the mixture was transferred to a vial and rinsed three times
with 0.5 mL of MeOH. After concentrating the resulting solution
in vacuo, the crude residue was purified by silica gel
chromatography (30:1 CH2Cl2/MeOH to 20:1 to 10:1 to 5:1) to
yield 1 (83%, 28 mg) and 2 (16%, 5.0 mg).
Scheme 3. Divergent deoxygenation with fluoroarylboranes.
3. Conclusion
Selective activation and cleavage of C–O bonds in gibberellic
acid has been achieved employing different fluoroarylborane
Lewis acid catalysts. Although the precise mechanistic reasons
for the diverging behavior could not be unambiguously
disentangled, a number of differences were noted, including
changes in the electrostatic surfaces of the borane Lewis acid and
their conjugate hydrides. Experimental studies noted that
B(C6F5)3, which has the highest hydride affinity, rests as the
borohydride. The partially fluorinated catalysts have lower
hydride affinities and neither build up significant quantities of
borohydride during catalysis. However, when PPh3 is added to a
reaction the BAr2,4,6-F catalyst converts to the borohydride in situ
while BAr3,5-CF3 does not. It is tempting to ascribe the difference
in reaction selectivities to the propensity of the catalysts to form
borohydride versus borane resting states, however, it is still too
early to tell if this is the ultimate source of the divergent
behavior.
1
Compound 1: H NMR (600 MHz, Acetone-d6) δ 5.19 (br s,
1H), 5.06 (br s, 1H), 4.90 (br s, 1H), 4.04 (dd, J = 4.3, 2.0 Hz,
1H), 3.17 (d, J = 5.5 Hz, 1H), 3.16 – 3.12 (m, 1H), 2.77 (dt, J =
18.2, 3.5 Hz, 1H), 2.68 (dt, J = 16.3, 3.0 Hz, 1H), 2.49 (br s, 1H),
2.28 – 2.23 (m, 1H), 2.13 (m, 1H), 1.96 – 1.88 (m, 1H), 1.74 (dd,
J = 11.0, 2.9 Hz, 1H), 1.69 – 1.58 (m, 2H), 1.49 – 1.39 (m, 2H),
1.32 (s, 3H); 13C NMR (151 MHz, Acetone-d6) δ 177.5, 176.2,
156.6, 142.1, 111.0, 105.7, 79.3, 70.5, 50.5, 50.0, 49.8, 49.7,
48.8, 47.2, 46.6, 40.1, 38.8, 33.4, 22.2, 19.4. HRMS (EI)
calculated for C19H24O6Na [M+Na]+:371.1465; found 371.1459.
Compound 2: 1H NMR (600 MHz, Methanol-d4) δ 5.85 (d, J =
9.8 Hz, 1H), 5.51 (d, J = 9.8 Hz, 1H), 5.10 (t, J = 2.5 Hz, 1H),
3.15 – 3.01 (m, 1H), 2.98 – 2.81 (m, 2H), 2.75 – 2.64 (m, 1H),
2.60 – 2.42 (m, 2H), 2.31 – 2.12 (m, 1H), 2.06 (dd, J = 10.1, 2.6
Hz, 1H), 2.03 – 1.89 (m, 1H), 1.73 (td, J = 11.9, 6.4 Hz, 1H),
1.64 (ddd, J = 10.7, 7.0, 2.6 Hz, 1H), 1.59 (dd, J = 10.1, 2.5 Hz,
1H), 1.21 (s, 3H); 13C NMR (151 MHz, Methanol-d4) δ 178.2,
177.8, 156.0, 136.5, 132.4, 128.9, 127.8, 105.8, 80.0, 57.1, 55.9,
53.9, 52.4, 40.6, 40.5, 26.1, 24.5, 21.6. HRMS (EI) calculated for
C19H22O5Na [M+Na]+: 353.1360; found: 353.1370.
4. Experimental section
4.1. General information
4.3. B(C6F5)3 catalyzed reaction with HBCat
All reactions were performed at ambient temperature (25 ºC,
RT) unless otherwise specified. All workup procedures were
performed under air with reagent grade reagents unless otherwise
specified. Column chromatography was performed using
In a N2-filled glove box, B(C6F5)3 (3.6 mg, 0.007 mmol, 0.10
equiv) was placed in a 1 dram vial and dissolved in 0.2 mL of
CH2Cl2. To the catalyst solution was added HBCat (37 µL, 0.350

5
[4]
S. Yoganathan, S.J. Miller, Structure diversification of
mmol, 5.00 equiv) and mixed. In a separate vial, Me2EtSi-Gibb
(36 mg, 0.070 mmol, 1.00 equiv) was diluted with 0.3 mL of
CH2Cl2. The catalyst and borane mixture was then added to the
substrate solution in one portion. The reaction mixture was
transferred to an NMR tube and sealed with a septum cap. After
24 h, the mixture was transferred to a vial and rinsed three times
with 0.5 mL of MeOH. After concentrating the resulting solution
in vacuo, the crude residue was purified by silica gel
chromatography (100% CH2Cl2 to 30:1 CH2Cl2/MeOH to 20:1 to
10:1 to 5:1) to yield 1 as a clear film in 90% yield (22.0 mg,
average over two runs).
vancomycin through peptide-catalyzed, site-selective lipidation: A
catalysis-based approach to combat glycopeptide-resistant
pathogens,
J.
Med.
Chem.
58
(2015)
2367–2377.
doi:10.1021/jm501872s.
[5]
[6]
X.-J. Dai, C.-J. Li, En Route to a Practical Primary Alcohol
Deoxygenation, J. Am. Chem. Soc. 138 (2016) 5433–5440.
doi:10.1021/jacs.6b02344.
M. Denancé, M. Guyot, M. Samadi, Short synthesis of 16β-
hydroxy-5α-cholestane-3,6-dione
oxysterol, Steroids, 71
10.1016/j.steroids.2006.03.002.
a
novel cytotoxic marine
(2006) 599–602.doi:
[7]
T.A. Bender, P.R. Payne, M.R. Gagné, Late-stage chemoselective
Functional-Group manipulation of bioactive natural products with
super-electrophilic silylium ions, Nat. Chem. 10 (2018) 85–90.
doi:10.1038/NCHEM.2863.
4.4. Selective deoxygenation with BAr2,4,6-F
In a N2-filled glove box, BAr2,4,6-F (3.2 mg, 0.008 mmol, 0.10
equiv) was placed in a 1 dram vial and dissolved in 0.2 mL of
CH2Cl2. To the catalyst solution was added Me2EtSiH (38 µL,
0.308 mmol, 4.00 equiv) and mixed. In a separate vial, Me2EtSi-
Gibb (40 mg, 0.077 mmol, 1.00 equiv) was diluted with 0.3 mL
of CH2Cl2. The catalyst and silane mixture was then added to the
substrate solution in one portion. The reaction mixture was
transferred to an NMR tube and sealed with a septum cap. After
24 h, the mixture was transferred to a vial and rinsed three times
with 0.5 mL of MeOH. After concentrating the resulting solution
in vacuo, the crude residue was purified by silica gel
chromatography (20:1 CH2Cl2/MeOH to 10:1 to 8:1 to 5:1) to
yield 1 as a clear film in 82% yield (22.0 mg).
[8]
G. Hamasaka, H. Tsuji, Y. Uozumi, Organoborane-Catalyzed
Hydrogenation of Unactivated Aldehydes with a Hantzsch Ester as
a Synthetic NAD(P)H Analogue, Synlett. 26 (2015) 2037–2041.
doi:10.1055/s-0034-1378846.
[9]
Q. Yin, S. Kemper, H.F.T. Klare, M. Oestreich, Boron Lewis Acid-
Catalyzed Hydroboration of Alkenes with Pinacolborane: BArF
3
Does What B(C6F5)3 Cannot Do!, Chem. - A Eur. J. 22 (2016)
13840–13844. doi:10.1002/chem.201603466.
J.R. Lawson, L.C. Wilkins, R.L. Melen, Tris(2,4,6-
trifluorophenyl)borane: An Efficient Hydroboration Catalyst,
[10]
Chem.
-
A
Eur.
J.
23
(2017)
10997–11000.
doi:10.1002/chem.201703109.
[11]
[12]
[13]
Y. Seo, J.M. Lowe, M.R. Gagné, Controlling Sugar Deoxygenation
Products from Biomass by Choice of Fluoroarylborane Catalyst,
ACS Catal. 9 (2019) 6648–6652. doi:10.1021/acscatal.9b01578.
J.M. Lowe, Y. Seo, M.R. Gagné, Boron-Catalyzed Site-Selective
Reduction of Carbohydrate Derivatives with Catecholborane, ACS
Catal. 8 (2018) 8192–8198. doi:10.1021/acscatal.8b02337.
R.W. Huigens III, K.C. Morrison, R.W. Hicklin, T.A. Jr Flood,
M.F. Richter, P.J. Hergenrother, A ring-distortion strategy to
construct stereochemically complex and structurally diverse
compounds from natural products, Nat. Chem. 5 (2013) 195–202.
doi:10.1038/nchem.1549.
4.5. Selective deoxygenation with BAr3,5-CF3
In a N2-filled glove box, BAr3,5-CF3 (4.5 mg, 0.007 mmol, 0.10
equiv) was placed in a 1 dram vial and dissolved in 0.2 mL of
CH2Cl2. To the catalyst solution, was added Et3SiH (28 µL, 0.174
mmol, 2.50 equiv) and mixed. In a separate vial, Et3Si-Gibb (40
mg, 0.070 mmol, 1.00 equiv) was diluted with 0.3 mL of CH2Cl2.
The catalyst and silane mixture was then added to the substrate
solution in one portion. The reaction mixture was transferred to
NMR tube and sealed with a septum cap. After 24 h, the mixture
was transferred to a vial and rinsed three times with 0.5 mL of
MeOH. After concentrating the resulting solution in vacuo, the
crude residue was purified by silica gel chromatography (20:1
CH2Cl2/MeOH to 10:1 to 8:1 to 5:1) to yield 2 as a clear film in
51% yield (15.7 mg, average over 3 runs).
[14]
[15]
[16]
[17]
[18]
Z.M. Heiden, A.P. Lathem, Establishing the hydride donor abilities
of main group hydrides, Organometallics. 34 (2015) 1818–1827.
doi:10.1021/om5011512.
I.B. Sivaev, V.I. Bregadze, Lewis acidity of boron compounds,
Coord.
Chem.
Rev.
270–271
(2014)
75–88.
doi:10.1016/j.ccr.2013.10.017.
Testing this reaction on the bench top under non-anhydrous
conditions gave similar results but required and extra 2 equiv. of
silane to “dry” the solvent.
Me2EtSi-Gibb with BAr3,5-CF3 and Me2EtSiH also gave the same
product 3 as the major compound but accompanied with a small
amount of 1 according to the in-situ 13C NMR analysis.
D.J. MacK, B. Guo, J.T. Njardarson, Synthesis of allylic and
homoallylic alcohols from unsaturated cyclic ethers using a mild
and selective C−O reduction approach, Chem. Commun. 48 (2012)
7844–7846. doi:10.1039/c2cc33551d.
T.A. Bender, J.A. Dabrowski, M.R. Gagné, Delineating The
Multiple Roles of B(C6F5)3 in the Chemoselective Deoxygenation
Acknowledgments
This work was exclusively supported by the National Institute
of General Medicine (5R01 GM130693). We thank UNC’s
Department of Chemistry Mass Spectrometry Core Laboratory,
especially Dr. Brandie Ehrmann for her assistance with the mass
spectrometry analysis.
[19]
[20]
[21]
of Unsaturated Polyols, ACS Catal.
doi:10.1021/acscatal.6b02551.
6 (2016) 8399–8403.
A. Gudz, P.R. Payne, M.R. Gagné, Phosphines as Silylium Ion
Carriers for Controlled C−O Deoxygenation: Catalyst Speciation
and Turnover Mechanisms, Organometallics. 36 (2017) 4047–4053.
doi:10.1021/acs.organomet.7b00689.
V. Morozova, P. Mayer, G. Berionni, Scope and Mechanisms of
Frustrated Lewis Pair Catalyzed Hydrogenation Reactions of
Electron-Deficient C=C Double Bonds, Angew. Chemie - Int. Ed.
54 (2015) 14508–14512. doi:10.1002/anie.201507298.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
http://doi.org/
References and notes
[22]
[23]
D.J. Parks, J.M. Blackwell, W.E. Piers, Studies on the mechanism
of B(C6F5)3-catalyzed hydrosilation of carbonyl functions, J. Org.
Chem. 65 (2000) 3090–3098. doi:10.1021/jo991828a.
[1]
[2]
C.R. Shugrue, S.J. Miller, Applications of Nonenzymatic Catalysts
to the Alteration of Natural Products, Chem. Rev. 117 (2017)
11894–11951. doi:10.1021/acs.chemrev.7b00022.
T. Cernak, K.D. Dykstra, S. Tyagarajan, P. Vachal, S.W. Krska,
The medicinal chemist’s toolbox for late stage functionalization of
drug-like molecules, Chem. Soc. Rev. 45 (2016) 546–576.
doi:10.1039/c5cs00628g.
P.A. Jordan, S.J. Miller, An approach to the site-selective
deoxygenation of hydroxy groups based on catalytic
phosphoramidite transfer, Angew. Chemie - Int. Ed. 51 (2012)
2907–2911. doi:10.1002/anie.201109033.
Q. Yin, Y. Soltani, R.L. Melen, M. Oestreich, BArF -Catalyzed
3
Imine Hydroboration with Pinacolborane Not Requiring the
Assistance of an Additional Lewis Base, Organometallics. 36
(2017) 2381–2384. doi:10.1021/acs.organomet.7b00381.
[24]
T.J. Herrington, A.J.W. Thom, A.J.P. White, A.E. Ashley, Novel
[3]
H2 activation by
a tris[3,5-bis(trifluoromethyl)phenyl] borane
frustrated Lewis pair, Dalt. Trans. 41 (2012) 9019–9022.
doi:10.1039/c2dt30384a.

6
Tetrahedron

• Gibberellic acid derivatives are synthesized with fluoroarylborane catalysts.
• An example of selective deoxygenation in a natural product
• Selective deoxygenation is possible with silane and borane reductants.

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: