Modulating Electrostatic Interactions in Ion Pair Intermediates To Alter Site Selectivity in the C?O Deoxygenation of Sugars

Article abstract of DOI:10.1002/anie.202007415

Controlling which products one can access from the predefined biomass-derived sugars is challenging. Changing from CH₂Cl₂ to the greener alternative toluene alters which C?O bonds in a sugar are cleaved by the tris(pentafluorophenyl)borane/HSiR₃ catalyst system. This increases the diversity of high-value products that can be obtained through one-step, high-yielding, catalytic transformations of the mono-, di-, and oligosaccharides. Computational methods helped identify this non-intuitive outcome in low dielectric solvents to non-isotropic electrostatic enhancements in the key ion pair intermediates, which influence the reaction coordinate in the reactivity-/selectivity-determining step. Molecular-level models for these effects have far-reaching consequences in stereoselective ion pair catalysis.

Full text of DOI:10.1002/anie.202007415

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Modulating electrostatic interactions in ion pair intermediates to
alter site-selectivity in C–O deoxygenation of sugars
Authors: Jared Lowe, Bekah Bowers, Youngran Seo, and Michel R
Gagne
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202007415
Link to VoR: https://doi.org/10.1002/anie.202007415

10.1002/anie.202007415
Angewandte Chemie International Edition
COMMUNICATION
Modulating electrostatic interactions in ion pair intermediates to
alter site-selectivity in C–O deoxygenation of sugars
Jared M. Lowe, Bekah E. Bowers, Youngran Seo, and Michel R. Gagné*[a][b]
[a]
J. M. Lowe, B. E. Bowers, Dr. Y. Seo, Prof. Dr. M. R. Gagné
Department of Chemistry
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina, 27599, United States
E-mail: mgagne@unc.edu
Homepage: http://gagnegroup.web.unc.edu/
M.R.G. thanks the Department of Energy DE-FG02-05ER15630 for funding.
[b]
Supporting information for this article is given via a link at the end of the document.
Abstract: Controlling which products one can access from the pre-
defined biomass-derived sugars is challenging. Converting from
CH2Cl2 to a greener alternative, toluene, changes which C–O bonds
in a sugar are cleaved using the BCF/HSiR3 catalyst system. This
increases the diversity of high-value products that can be obtained
through 1-step, high-yielding, catalytic transformations on the base
mono-, di-, and oligo-saccharides. Computational methods helped
localize this non-intuitive outcome in low dielectric solvents to non-
isotropic electrostatic enhancements in the key ion-pair intermediates,
which influence the reaction coordinate in the reactivity/selectivity
determining step. Molecular-level models for these effects have
reaching consequences in stereoselective ion-pair catalysis.
e.g., Scheme 1a).[28,29] Low polarity solvents enhance the
electrostatic interaction between the cation and anion, and
where a tighter ion pair is beneficial to stereochemical
transfer, e.g., from chiral anion to cation, these solvents
enhance enantioselectivity.[30,31]
Tris(pentafluorophenyl)borane, BCF, has emerged as a
highly effective catalyst for the site-selective deoxygenation
of biomass-derived carbohydrates.[1–9] Choice of the sugar’s
protecting group and reductant (hydrosilane or
hydroborane) can tailor which C–O bonds are cleaved to
yield high-value chiral products.[7] Since the biorenewables
pool is fixed, new approaches to altering site-selectivity
diversifies the high-value compounds that can be accessed
from it.
Our synthetic and mechanistic experiments on complex
sugars are consistent with the Piers mechanism,[1,2,5] wherein
BCF activates the silane via an η1-adduct, and is in turn
heterolytically and reversibly cleaved by an oxygenated
nucleophile (the sugar) to yield a cationic silyloxonium that
is paired to the H–BCF– anion/reductant.[10–16] According to
Sakata and Fujimoto’s calculations,[17] these contact ion pairs
are dynamic and multiple orientations (local minima) can be
sampled prior to rate determining C–O reduction. The
rationale for site-selectivity in a sugar necessarily includes
which oxygen binds to the silylium along with the barriers of
hydride attack (within the ion pair) at each of these
positions. Since the key selectivity defining events occur at
the point of contact in a dynamic ion pair, they are under
Curtin-Hammett control. The added complication of
neighboring group participation multiplies the challenge of
predicting which site(s) will be reduced in polyols.[2,18–20]
Scheme 1. a) example of ion pairing in catalysis b) formation of an ion pair
involving 1-deoxyglucose.
As part of a broader effort to improve the greenness of site-
selective sugar deoxygenations,[6] we discovered that Et3Si-
1-deoxyglucose Et3Si-2 (generated in situ from Et3Si-MeO-
glucose)[5] provided an excellent yield of 1-deoxysorbitol 1
when the reaction was carried out in toluene (Scheme 1b).
This new product to BCF/HSiR3 catalysis was unexpected as
previous attempts to selectively reduce 1-deoxyglucose by
BCF/HSiR3 in CH2Cl2 failed due to the formation of an
unreactive (but spectroscopically characterizable) ion pair
[Si-1-deoxyglucose][H–BCF] (A, Scheme 1b) that
sequestered the catalytic quantities of silylium.[3,5] The
heightened reactivity of A in toluene was also apparent in
pentane and cyclohexane (Scheme 2). A mixture of 1 and 2
was observed for reactions in chloroform, in contrast to o-
dichlorobenzene and CH2Cl2, which do not proceed beyond
2. Heating the reaction in CH2Cl2 to 50˚C in a sealed NMR tube
The effect of ion pair structure on catalyst behavior has
been investigated in a number of scenarios, including single-
site Zeigler-Natta,[21–23] chiral ion-pair intermediates,[24–27]
and chiral ion-pair catalysts (organo and transition metal,
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202007415
Angewandte Chemie International Edition
COMMUNICATION
consumed Et3Si-2, but non-selective overreduction
dominated.
nicely converges both primary fragments to 6 in excellent
yields, and in the case of Entry 4, at a higher conversion. For
the oligo-saccharides in Entries 5-7, toluene more efficiently
provided 6 in each case. Overall, these reductions are
enhanced in toluene over CH2Cl2 and provide the means to
converge each sugar unit to a single product.
Since low dielectric solvents were expected to slow
catalysis by inhibiting the formation of ion pairs from neutral
precursors, these results prompted an in-depth investigation
of mechanism and synthetic utility. The result of these
studies not only enhances our understanding of site-
selective deoxygenation chemistry, but also ion-pair
catalysis more generally.
Table 1. Reduction of sugars in toluene and CH2Cl2.
Scheme 2. Solvent dependence in ring opening of Et3Si-MeO-glucose.
[a]Isolated yields. [b]Isolated as a mixture. [c]Ref. 5.
We first sought to determine how broadly this solvent
change effected selectivity in monosaccharides. Table 1
compares the reduction selectivity in CH2Cl2 and toluene for
Et3Si- (Entries 1-3) and Me3Si-protected (Entries 4-6)
hexoses and Et3Si-protected pentoses (Entries 7-8). On the
whole, toluene provides a different product than does CH2Cl2,
our originally reported conditions,[5] and Et3Si-protected
sugars (using Et3SiH) are less reactive than Me3Si-protected
sugars (using Me2EtSiH). For example, in CH2Cl2 Et3Si-1-
deoxyglucose (Entry 1) is unreactive while Me3Si-1-
deoxyglucose gives moderate conversions to 6 (Entry 4).
Although similar outcomes were apparent for CH2Cl2 and
toluene reductions of protected MeO-galactose, the other
hexoses had different selectivity profiles. Controllable
differences were achieved based on whether a Et3Si- or
Me3Si-protection scheme was employed in each solvent. In
this manner, up to 3 unique products could be obtained in
high yields and in one synthetic step from a single
biofeedstock through choice of solvent and silyl protecting
group. Notable cases include accessing 5,6-deoxytetraol 4[32]
(Entry 3) from Et3Si-MeO-α-D-mannose (rather than the
more precious 2-deoxyglucose),[5] which required a selective
2˚-reduction of Et3Si-5 (at C5). When Me3Si-protected
starting material is used, the tetraol is further reduced (at
C4) to 9 in good yields (toluene, Entry 6).
[a]Isolated yields. [b]Ref 5. [c]Isolated as (3:1) mixture with internal triol.
Our synthetic studies demonstrate the profound influence
on site-selectivity that a change from CH2Cl2 to toluene has
for BCF/silane catalysis. Molecular-level insights into the
role of solvent were obtained using DFT methods to model
the C1-reduction of Me3Si-1-deoxyglucose (see supporting
information for full computational details). In practice, this
required a full evaluation of the competing chair forms
In the pentose series, Et3Si-xylose provided 10 in both
solvents (Entry 7), while Et3Si-MeO-xylose (Entry 8) was
first demethoxylated in CH2Cl2 to produce the ring-opened
linear tetraol 11. This process likely proceeds via an
oxocarbenium intermediate (SN1). In toluene, however,
demethylation of the exocyclic -OMe occurs, which then ring
opened to 10. The in situ observation of methane leads to the
conclusion that toluene inhibits oxocarbenium formation
relative to CH2Cl2 and thereby shifts the selectivity to O–CH3
reduction of the exocyclic silyloxonium.
4
(Figure S1) as C1-reduction of the all-equatorial C1 form
(int1) was prohibitively high (+41.9 kcal/mol vs. starting
1
materials), while the all-axial C4 conformer int2 was more
reasonable (+19.4 kcal/mol, Figure 1). Computed barriers
for chair flipping Me3Si-1-deoxyglucose from 4C1 to 1C4 were
lower than either C–O bond reduction barrier (see
supporting information) and lower than that of similarly
calculated C–O bond reductions involving BCF/HSiR3.[33] The
data suggests that a chair flip to the 1C4 conformer is required
to minimize steric clashing between neighboring -OSi groups
We have reported that reduction of per-Si-disaccharides
with BCF and Me2EtSiH in CH2Cl2 provided a 1:1 ratio of the
recalcitrant 1-deoxyglucose and products derived from per-
Si-glucose (Table 2, Entries 1-3).[3] Switching to toluene
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202007415
Angewandte Chemie International Edition
COMMUNICATION
(Figure S2). An analysis of coupling constants (3JHH) showed
of the Me3Si-1-deoxyglucose to 6 in toluene and its sluggish
that Et3Si-1-deoxyglucose and A adopt the 4C1 and 1C4 conversion in CH2Cl2. Metrical parameters for int4 reveal
conformers, respectively (Figures S3 and S4), consistent that the distance in the nascent C1∙∙∙H bond shortens in
with computed conformer stabilities.
Table 2. Reduction of oligosaccharides in toluene and CH2Cl2.
progressively lower dielectric solvents (C1–O is largely
unchanged). This dielectric shortening of C1∙∙∙H appears to
advance the reaction coordinate towards a fully formed C1–
H in the product. Changes in the progression to products are
also apparent at the transition state, but in this case the
structure is slightly more product like in the high dielectric
cases. The sum changes in int 4 and TS1 reveal that net
movement of hydride in C–H bond formation is reduced in
lower dielectric solvents.
Figure 1. Free energy profiles for C1-reduction of Me3Si-1-deoxyglucose in the
1C4 conformer.
[a]Isolated yields. [b]Ref 3. [c]Reaction time 24 h.
Only two of the numerous possible ion pair local minima
are included in Figure 1. int3 is the first local minimum
formed upon silane heterolysis and int4 is that which
positions the anion B–H to reduce C1–O. The calculations of
Fujimoto, along with our own unpublished work, shows that
the barrier for interconverting contact ion pairs is low (<5
kcal/mol).[17] The barrier (TS1) for C1 reduction from int4 is
computed to be +19.4 kcal/mol using the CH2Cl2 dielectric
continuum. The structure of TS1 is shown in Figure 2 and
displays the expected alignment of the B–H nucleophile with
the C–OSi+ leaving group. Animation of the single imaginary
frequency shows it corresponds to C1–O cleavage.
Figure 2. Calculated structures of int4 and TS1.
These data demonstrate how low dielectric solvents create
electrostatic compressive forces that can shorten inter-ion
bond lengths in a contact ion pair. These strengthened
interactions have long been empirically used to enhance
stereochemical transfer in chiral ion pair catalysis. More
nuanced is the view that an ensemble of discrete but
dynamically interconverting ion-pair structures/transition
states control the overall selectivity under Curtin-Hammett
constraints. In the situation where charge neutralization
Repeating calculations of int4 and TS1 using dielectric
continuums corresponding to other solvents revealed
several fascinating observations related to ion pair structure
and energetics. As shown in Table 3, ΔG‡ increases with
increasing dielectric, consistent with the efficient conversion
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202007415
Angewandte Chemie International Edition
COMMUNICATION
occurs in
a selectivity determining step, we have
Co. KGaA, 2019, pp. 33–85.
demonstrated that such inter-ion forces might also help to
pick which structures dominate product output .
[16]
[17]
[18]
Y. Kim, S. Chang, Angew. Chemie - Int. Ed. 2016, 55, 218–222.
K. Sakata, H. Fujimoto, J. Org. Chem. 2013, 78, 12505–12512.
T. A. Bender, J. A. Dabrowski, H. Zhong, M. R. Gagné, Org. Lett.
2016, 18, 4120–4123.
Table 3. Computed reduction barriers for Me3Si-1-deoxyglucose.
[19]
[20]
[21]
T. A. Bender, J. A. Dabrowski, M. R. Gagné, ACS Catal. 2016, 6,
8399–8403.
int4
TS1
ΔG‡
(kcal/mol)
Solvent
ε
dC–O
dC–H
dC–O
dC–H
ΔdC–H
N. Drosos, B. Morandi, Angew. Chemie - Int. Ed. 2015, 54, 8814–
8818.
heptane
toluene
CH2Cl2
1.9
2.4
16.6
17.3
19.4
19.5
20.2
1.48 [a]
2.70
2.71
2.76
2.76
2.77
1.92
1.92
1.94
1.94
1.95
1.63
1.63
1.62
1.62
1.61
1.07
1.08
1.14
1.14
1.16
1.48
C. Zuccaccia, N. G. Stahl, A. Macchioni, M. C. Chen, J. A. Roberts,
T. J. Marks, J. Am. Chem. Soc. 2004, 126, 1448–1464.
L. Luo, T. J. Marks, Top. Catal. 1999, 7, 97–106.
E. Y. X. Chen, T. J. Marks, Chem. Rev. 2000, 100, 1391–1434.
K. Brak, E. N. Jacobsen, Angew. Chemie Int. Ed. 2013, 52, 534–
561.
8.9
1.48
o-C6H4Cl2
CH3CN
9.9
1.47
[22]
[23]
[24]
37.5
1.47
[a]Bond lengths in Å.
[25]
[26]
[27]
[28]
S. E. Reisman, A. G. Doyle, E. N. Jacobsen, J. Am. Chem. Soc.
2008, 130, 7198–7199.
Acknowledgements
D. D. Ford, D. Lehnherr, C. R. Kennedy, E. N. Jacobsen, ACS
Catal. 2016, 6, 4616–4620.
We thank UNC’s Department of Chemistry Mass Spectrometry
Core Laboratory, especially Dr. Brandie Ehrmann and Diane
Wallace, for their assistance with the mass spectrometry
analysis. All 1H and 13C{1H} NMR spectra (and data files)
reported herein are also available at our curated spectroscopic
S. M. Banik, A. Levina, A. M. Hyde, E. N. Jacobsen, Science (80-. ).
2017, 358, 761–764.
R. J. Phipps, G. L. Hamilton, F. D. Toste, Nat. Chem. 2012, 4, 603–
614.
database
(http://gagnegroup.web.unc.edu/sugars-
spectroscopy/sugars). This database currently contains high
resolution spectra and raw FID files for more than 75 biomass
derived, partially deoxygenated hexoses and pentoses.
[29]
[30]
M. Mahlau, B. List, Angew. Chemie - Int. Ed. 2013, 52, 518–533.
G. L. Hamilton, J. K. Eun, M. Mba, F. D. Toste, Science (80-. ).
2007, 317, 496–499.
[31]
[32]
C. Merten, C. H. Pollok, S. Liao, B. List, Angew. Chemie - Int. Ed.
2015, 54, 8841–8845.
Keywords: ion pairing • sugar • selective deoxygenation •
solvent • DFT
For convenience, we report deoxy carbon numberings that
correspond to which sites in the original cyclic sugars have become
deoxyegnated.
[1]
[2]
L. L. Adduci, M. P. McLaughlin, T. A. Bender, J. J. Becker, M. R.
Gagne, Angew. Chemie - Int. Ed. 2014, 53, 1646–1649.
L. L. Adduci, T. A. Bender, J. A. Dabrowski, M. R. Gagné, Nat.
Chem. 2015, 7, 576–581.
[33]
G. J. Cheng, N. Drosos, B. Morandi, W. Thiel, ACS Catal. 2018, 8,
1697–1702.
[3]
[4]
[5]
[6]
Y. Seo, M. R. Gagné, ACS Catal. 2018, 8, 81–85.
J. M. Lowe, Y. Seo, M. R. Gagné, ACS Catal. 2018, 8, 8192–8198.
Y. Seo, J. M. Lowe, M. R. Gagné, ACS Catal. 2019, 9, 6648–6652.
N. M. Hein, Y. Seo, S. J. Lee, M. R. Gagné, Green Chem. 2019, 21,
2662–2669.
[7]
T. A. Bender, J. A. Dabrowski, M. R. Gagné, Nat. Rev. Chem. 2018,
2, 35–46.
[8]
J. Zhang, S. Park, S. Chang, Angew. Chemie - Int. Ed. 2017, 56,
13757–13761.
[9]
I. Chatterjee, D. Porwal, M. Oestreich, Angew. Chemie - Int. Ed.
2017, 56, 3389–3391.
[10]
[11]
[12]
[13]
[14]
[15]
D. J. Parks, J. M. Blackwell, W. E. Piers, J. Org. Chem. 2000, 65,
3090–3098.
V. Gevorgyan, M. Rubin, S. Benson, J. X. Liu, Y. Yamamoto, J. Org.
Chem. 2000, 65, 6179–6186.
T. Robert, M. Oestreich, Angew. Chemie - Int. Ed. 2013, 52, 5216–
5218.
A. Y. Houghton, J. Hurmalainen, A. Mansikkamäki, W. E. Piers, H.
M. Tuononen, Nat. Chem. 2014, 6, 1–6.
M. Oestreich, J. Hermeke, J. Mohr, Chem. Soc. Rev. 2015, 44,
2202–2220.
D. Weber, M. R. Gagné, in Organosilicon Chem. Nov. Approaches
React. (Eds.: T. Hiyama, M. Oestreich), Wiley-VCH Verlag GmbH &
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202007415
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
Ion Pairs in Control. Certain ion pair
intermediates can be made more
reactive when electrostatic forces
between a reactive cation and anion
can couple to productive C–O bond
cleavage processes. Low dielectric
media can enhance such forces and
speed reactions.
1
This article is protected by copyright. All rights reserved.