Oxidation Under Reductive Conditions: From Benzylic Ethers to Acetals with Perfect Atom-Economy by Titanocene(III) Catalysis

Article abstract of DOI:10.1002/anie.202013561

Described here is a titanocene-catalyzed reaction for the synthesis of acetals and hemiaminals from benzylic ethers and benzylic amines, respectively, with pendant epoxides. The reaction proceeds by catalysis in single-electron steps. The oxidative addition comprises an epoxide opening. An H-atom transfer, to generate a benzylic radical, serves as a radical translocation step, and an organometallic oxygen rebound as a reductive elimination. The reaction mechanism was studied by high-level dispersion corrected hybrid functional DFT with implicit solvation. The low-energy conformational space was searched by the efficient CREST program. The stereoselectivity was deduced from the lowest lying benzylic radical structures and their conformations are controlled by hyperconjugative interactions and steric interactions between the titanocene catalyst and the aryl groups of the substrate. An interesting mechanistic aspect is that the oxidation of the benzylic center occurs under reducing conditions.

Full text of DOI:10.1002/anie.202013561

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Oxidizing under Reductive Conditions: From Benzylic Ethers to
Acetals with Perfect Atom-Economy by Titanocene(III)-Catalysis
Authors: Andreas Gansäuer, Stefan Grimme, Pierre Funk, Ruben
Richrath, and Fabian Bohle
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.202013561
Link to VoR: https://doi.org/10.1002/anie.202013561

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
Oxidizing under Reductive Conditions: From Benzylic Ethers to
Acetals with Perfect Atom-Economy by Titanocene(III)-Catalysis
Pierre Funk,[a] Ruben B. Richrath,[a] Fabian Bohle,[b] Stefan Grimme,[b] Andreas Gansäuer*[a]
[a]
P. Funk, Dr. R. B. Richrath, Prof. Dr. A. Gansäuer
Kekulé-Institut für Organische Chemie und Biochemie
Universität Bonn
Gerhard Domagk-Str. 1, D-53121 Bonn (Germany)
E-mail: andreas.gansaeuer@uni-bonn.de
F. Bohle, Prof. Dr. S. Grimme
[b]
Mulliken Center for Theoretical Chemistry
Institut für Physikalische und Theoretische Chemie
Universität Bonn
Beringstraße 4, D-53115 Bonn (Germany)
Supporting information for this article is given via a link at the end of the document.
Abstract: We describe a novel titanocene catalyzed method for the
synthesis of acetals and hemiaminals from benzylic ethers and
benzylic amines with pendant epoxides. The reaction proceeds by
catalysis in single electron steps. The oxidative addition is constituted
by an epoxide opening. A H-atom transfer to generate a benzylic
radical serves as radical translocation step, and an organometallic
oxygen rebound as reductive elimination. The reaction-mechanism
was studied by high level dispersion corrected hybrid functional DFT
with implicit solvation. The low-energy conformational space was
searched by the efficient CREST program. The stereoselectivity can
be deduced from the lowest lying benzylic radical structures. Their
conformations are controlled by hyperconjugative interactions and by
steric interactions between the titanocene-catalyst and the aryl groups
of the substrate. A particularly interesting mechanistical aspect is that
the oxidation of the benzylic center occurs under reducing conditions.
position that proceeds under reducing conditions. The oxidative
addition occurs by a titanocene(III) mediated epoxide opening[6]
through electron transfer to yield I. From I, radical translocation
occurs via the pivotal H-atom abstraction[7] to yield II. The driving
force for this step arises from the increased strength of the newly
formed C-H bond.[8] In II, the stage is set for the reductive
elimination of Cp2TiX that forms the desired product, acetal 2. In
this step that can be regarded as an organometallic oxygen
rebound, a second stereocenter is formed.[9] It remains to be seen
how selective the ring formation is.
Introduction
Catalysis in single electron steps[1] or metallo-radical catalysis[2] is
a concept that merges the concepts of radical chemistry with
those of transition metal catalysis. In this manner, the advantages
of both fields, such as the ease of radical generation, the high
functional group tolerance of radical reactions and the ability of a
metal complex to control the pertinent selectivities of
transformation can be combined to attractive novel catalytic
processes. An essential aspect for the success of catalysis in
single electron steps is the availability of oxidative additions and
reductions in single electron steps for radical and product
generation.[3] Both steps can be regarded as homolytic
substitution reactions[4] and require an efficient shuttling of the
catalyst between neighboring oxidation states.[5] During the
radical translocation, a new bond and radical is formed. This step
sets the scene for the ensuing reductive elimination.
Scheme 1: Atom-economic transformation of benzylic ethers into acetals by
titanocene(III) catalysis in single electron steps (SET).
Our results for finding suitable reaction conditions for the reaction
of 1a are summarized in Table 1. The active titanocene(III)
catalysts were prepared by in situ reduction of the titanocene(IV)
precatalysts with Zn. Without Zn, no reaction was observed. The
use of Mn is possible but less convenient since the reduction of
Ti(IV) requires more time.[6,10] In THF, Zn-(CH3C5H4)2TiCl2 and Zn-
Cp2TiCl2 resulted in the consumption of 1a. However, hardly any
2a (2% and 26%, respectively) was formed as judged by
1
quantitative H-NMR analysis of the reaction mixture. While Zn-
Cp2Ti(O2CCF3)2 gave an improved conversion to 2a, the isolated
[10]
yield was only 41%. Zn-Cp2Ti(O3SCH3)2
(O3SCH3= OMs)
provided the best catalyst and allowed the isolation of 2a in 67%
yield. Other solvents than THF lead to inferior results. The
superiority of Zn-Cp2Ti(O3SCH3)2 over Zn-(CH3C5H4)2TiCl2 and
Zn-Cp2TiCl2 suggests that the reductive elimination is the rate
determining step of the catalytic cycle because epoxide opening
is faster with more electron rich catalysts whereas reductive
eliminations are slowed down. In our case, the reductive
Results and Discussion
Here, we demonstrate how radical translocation via H-atom
abstraction from a benzylic C-H bond by an alkyl radical can be
included in a process proceeding via titanocene(III) catalysis in
single electron steps (Scheme 1). The overall reaction constitutes
an atom-economical oxidation at the benzylic
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
elimination constitutes a homolytic substitution at a Ti-O bond by
a benzylic radical. Due to the high stabilization of this radical, this
step seems particularly difficult.[4]
low lying vibrational frequencies and to reduce the effect of
numerical noise in the DFT. The vibrational frequencies are
scaled by 0.95 and the thermostatistical analysis is evaluated at
339.15 K. Free energies are calculated as the sum of electronic
energy, solvation- and thermostatistical terms. All DFT
calculations were performed with TURBOMOLE program
package 7.4.1.[20] The resolution of the identity approximation for
the Coulomb integrals is generally applied in combination with the
default auxiliary basis sets.[21] The numerical integration of the
exchange correlation contribution is performed with grids m4 for
PBEh-3c and grid m5 for PW6B95. Default convergence criteria
are applied for all single-point calculations (Econv 10-7 Eh). For
geometry optimizations Econv 2⸱10-5 Eh and Gconv
2⸱10-3 Eh/Bohr are used. The convergence criteria for the
transition-state optimizations were set to 5⸱10-6 Eh and 2⸱10-3
Eh/Bohr. The solvation contributions were calculated with the
conductor like screening model for real solvents (COSMO-RS)
using the BP_TZVP_C30_1601 parametrization and the
COSMOtherm2016 package.[22] The solvation contribution
incorporates the volume work RTln(Videal) (T = 339.15 K) for an
ideal gas at 1 bar to 1 mol L-1. Geometries and tabulated data are
provided in the SI.
Table 1. Screening of different catalysts and solvents to find the most suitable
reaction conditions for the conversion of 1a.
[Ti]
solvent
yield/ [%]
conversion[a]/ [%]
[b]
(CH3C5H4)2TiCl2
THF
THF
2[a]
26[a]
42
75
[b]
Cp2TiCl2
Cp2Ti(O2CCF3)2
Cp2Ti(O3SCH3)2
Cp2Ti(O3SCH3)2
Cp2Ti(O3SCH3)2
THF
41, 61[a]
67
91
THF
100
96
1,4-dioxane
acetone
38[a]
The free energy ΔG339.15 of the reaction 1a to 2a (-19.1 kcal mol-1
for cis-2a and -19.4 kcal mol-1 for trans-2a) is strongly negative.
This is a manifestation of the loss of the epoxide’s ring strain.[5]
The very similar values of ΔG339.15 for cis-2a and trans-2a also
show that the observed highly diastereoselective acetal formation
cannot be under thermodynamic control.
33[a]
41
[a] Yields and conversion are determined by quantitative 1H-NMR. [b] Requires
50 mol% Collidin*HCl to prevent a decomposition of the catalyst.
The other important question of the mechanism is the
understanding of the high diastereoselectivity of acetal formation
in the reductive elimination. To this end, we studied the preferred
conformation of the benzylic radical IIa and the structures and
energies of the transition-states leading to cis- and trans-2a.
We employed Cp2Ti(O3SCH3)2 that is a stronger oxidant than
Cp2TiCl2 due to the more electron withdrawing mesylate
ligand.[10d] In this manner, additional driving force for the reductive
elimination is provided. The results summarized in Table 1
therefore demonstrate that the tuning of the redox-properties of
the titanocenes is a crucial and powerful tool to adjust the
efficiency of the steps in catalytic cycle.[10] Remarkably, 2a is
formed as a single diastereoisomer (structure determined by 2D-
NMR spectroscopy, see Supporting Information).[11] To
understand this particular feature of the reaction and the
thermodynamics of the catalytic cycle (scheme 2), these were
studied computationally by applying the following protocol.
All initial geometries were prepared manually and meta-
dynamics-based conformer searches for all involved geometries
were performed using the CREST[12] code in combination with the
generic force field GFN-FF(GBSA[THF])[13] with implicit solvation
as implemented in the freely available xtb code[14d-f]. In case of
metal containing structures, a small harmonic bond lengths
constraint was applied during the minimum conformer search in
order to avoid dissociation. The conformer ensemble was
efficiently refined in multiple steps at DFT level
PBEh-3c/DCOSMO-RS(THF)[15] to obtain the lowest lying
conformer, facilitated by the ENSO code.[16] After identification of
the lowest lying conformer, each of the structures, was used to
calculate Gibbs free energies (G) to investigate the
thermodynamics of the catalytic cycle shown in Scheme 2. Herein,
PW6B95-D4/def2-QZVP[17] dispersion corrected, high-level
single-point energies and COSMO-RS[18] solvation contributions
were calculated on the PBEh-3c+DCOSMO-RS(THF) optimized
geometries. Thermostatistical contributions are obtained at the
PBEh-3c+COSMO(THF) level using the modified rigid rotor
harmonic oscillator (mRRHO) approach.[19] The mRRHO ansatz
is applied to reduce the error of the harmonic approximation for
Scheme 2: Atom-economic transformation of benzylic ethers into acetals by
titanocene(III) catalysis in single electron steps calculated at PW6B95-D4/def2-
QZVP + COSMO-RS(THF) // PBEh-3c/DCOSMO-RS(THF) level. Values in
parenthesis are calculated for the catalyst Cp2TiCl.
The oxidative addition of titanocene catalyst Cp2TiOMs to the
epoxide that leads to radical formation is the thermodynamically
critical step (ΔG339.15 = +13.9 kcal mol-1 for Cp2TiOMs and ΔG339.15
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
= +9.7 kcal mol-1 for Cp2TiCl). The more favorable reaction with
Cp2TiCl is in line with an earlier study on the thermodynamics of
epoxide opening.[1a] Experimental studies on the use of electron
deficient catalysts have demonstrated that they lead to
substantially lower rate constants (e.g. Cp2TiOMs: k293 = 0.013
M- 1 s-1) compared to Cp2TiCl (k293 = 1.1 M-1s-1).[10d] The HAT step
is strongly exergonic in line with the difference in stability of
tertiary and benzylic radicals. Here the conformational search for
the lowest lying conformer established that the HAT reaction from
Ia to IIa is accompanied by a conformational relaxation. This
structural relaxation results in an energy gain of 3.2 kcal mol-1
compared to a HAT where no conformational search has been
employed (see SI for further information). The identification of a
lower lying conformation affects all following reactions in the
catalytic cycle.
We found that IIa-cis (leading to cis-2a, Scheme 3) is the most
stable conformer of IIa. We attribute this stability to two
hyperconjugative interactions between the σ-orbitals of C-H
bonds and σ*–orbitals of C-O bonds (the C-O-Ti bond and the C-
O-C● bond). This results in the preferred gauche-orientation of
the two C-O bonds. Additionally, the ‘upper’ Cp-ligand and the –
OMs ligand of the catalyst forces the phenyl ring in a position
pointing away from ‘upper’ Cp-ligand and the H-substituent of the
benzylic radical towards the –OMs ligand. Natural bond orbitals
(see scheme 3) and second-order perturbative estimates of
donor-acceptor interactions in the NBO basis were calculated
using the PBEh-3c/CPCM(THF) density obtained with the ORCA
4.2.1 program package[24] and the NBO6 code.[23] The second
conformer IIa-trans (leading to trans-2a, Scheme 3) that we were
able to identify is 6.5 kcal mol-1 less favored than IIa-cis at
339.15 K. It arises from IIa-cis by a rotation around the O-C●
bond. The stabilization by the gauche-interactions in IIa-cis and
IIa-trans is essentially identical and, therefore, the difference in
stability is due to increased repulsive steric interactions between
the Ph-group and the ‘upper’ Cp-ligand and the –OMs-ligand. The
barrier for the rotation that transforms IIa-cis into IIa-trans is
estimated to be 7.1 kcal mol-1 (see SI).
From IIa-cis and IIa-trans we calculated the barriers for the
transformation to cis-2a (16.7 kcal mol-1) and trans-2a
(18.3 kcal mol-1), respectively.
Scheme 3: 3D-Structures of the radicals IIa-cis (a,b) and IIa-trans (e,f) and
their stabilization by gauche interactions, with side view. The natural bond
orbitals (NBO)[23] shown are obtained from a PBEh-3c+CPCM(THF) calculation.
In addition Newman projections (c,g) and wedge and dash structures (d,h) of
these radicals are given. They show the C-H bond orbital donating into the σ*-
orbital of the C-O bond, visualizing the gauche-effect. Only one of two effective
Scheme 4: Energy profile of the conversion of IIa-cis and IIa-trans to cis-2a
and trans-2a. All Gibbs free energies are shown in kcal mol-1. The transition-
state geometries were obtained with PBEh-3c/COSMO(THF). The catalytic
cycle has been investigated at PW6B95-D4/def2-QZVP + COSMO-RS(THF) //
PBEh-3c/ COSMO(THF) level of theory.
gauche interactions is shown. Isosurface value = 0.05 e-1/2 bohr-3/2
.
The initial reaction paths were calculated with GSM/GFN2-
xTB(GBSA[THF])[14] and subsequently refined at PBEh-
3c/COSMO(THF) level. The final transition-state is optimized and
verified for having only one imaginary mode. Free energies are
calculated using the multilevel approach discussed before
Finally, the reductive elimination that results in product formation
and catalyst regeneration is thermodynamically favorable for both
Cp2TiOMs (trans: ΔG339.15 = -18.1 kcal mol-1) and Cp2TiCl (trans:
ΔG339.15 = -14.8 kcal mol-1). The results suggest that increased
efficiency of Cp2TiOMs over Cp2TiCl in the reductive elimination
is the reason for the superiority of the former complex as catalyst.
(PW6B95-D4/def2-QZVP
+
COSMO-RS(THF) // PBEh-3c/
COSMO(THF)). The energy profile of the conversion of the
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
radicals IIa to products 2a is shown in Scheme 4. The selectivity
of acetal formation is mainly due to the difference in stability of
IIa-cis and IIa-trans. Both transition-states are ‘early’ with long
distances (TS2a-trans = 285 pm, TS2a-cis = 242 pm vs. 139 and
140 pm in trans-2a and cis-2a) of the forming C-O bonds and are,
therefore, structurally similar to IIa-cis and IIa-trans. Thus, the
conformations of radicals IIa-cis and IIa-trans and also those of
the transition-states TS2a-cis and TS2a-trans are controlled by
the same two factors. First, the dihedral angles in the O-CHiPr-
CH2-O chain are determined by hyperconjugative interactions
between the σ-orbitals of C-H bonds and σ*–orbitals of C-O bonds
(gauche-interactions). Second, the conformation at the benzylic
radical center is determined by the steric interactions of the Ph-
ring and the H-substituent with the ‘upper’ Cp-ligand and the OMs-
ligand.
occurring and provides further support for the proposed
mechanism (Scheme 2).
Scheme 5: Preparation of enantiomerically pure 2d from 1d.
The transition-state models for the formation of 2a provide an
explanation for the experimentally observed diastereoselectivity
with benzyl ethers. Our mechanistic analysis suggests that
replacing one –H atom at the benzylic center in 1 with bulkier alkyl
groups should have a profound impact on the reaction. This is
indeed the case as exemplified in the synthesis of ketals 4a – 4e
(Table 3). In all cases, diastereoselectivity was either non-existent
or low (50:50 – 72:38). We note that when applicable the major
diastereomer has a trans-orientation of the aryl and the iPr group.
For details of the structural assignment, see SI. While the
experiments were typically performed with 100 mg of the
substrate, the reaction of 4e was performed on a 1 g scale
indicating that a larger scale is not problematic.
This analysis suggests that acetals derived from benzylic ethers
(ArCH2O-groups) should be formed with high diastereoselectivity.
This is indeed the case as summarized in Table 2.
Diastereoselectivities of >99:<1 imply that the other isomer could
not be detected by 1H-NMR spectroscopy.
Table 2. Substrate scope of the 1,3-dioxolane synthesis by catalysis in single
electron steps.
Table 3. Substrate scope and diastereoselectivity of the 2,2-disubstituted 1,3-
dioxolane synthesis by catalysis in single electron steps.
[a] 19 mol% Cp2Ti(OMs)2 and 38 mol% Zn are used.
[a] = The experiment is performed on a 1 g scale.
The cis-2,4-disubstituted 1,3-dioxolanes 2a–2i were smoothly
obtained from the substituted benzyl ethers 1a–1i in reasonable
to high yield. M- and p-substituents are readily tolerated.
Diastereoselectivity is high (94:6 – 99:1) for 4-ipropyl substituted
dioxolanes. With a cyclohexyl substituent, the diastereoselectivity
is equally high. For the details of the determination of relative
configuration by NOESY-spectroscopy and the determination of
the diasteromeric ratios by 1H-NMR-spectroscopy see the SI.
The computed structures and energies of radicals IVc-cis and
IVc-trans highlight the reasons for the selectivity (Scheme 6). In
contrast to the situation for radical II (ΔG339.15 = 6.5 kcal mol-1), the
difference in the stability between the two conformers is low
(ΔG339.15 = -0.6 kcal mol-1) and IVc-trans is more stable than IVc-
cis. The small difference of -0.6 kcal mol-1 should therefore lead
to a slight excess of trans-4c compared to cis-4c, which indeed
is obtained for synthesis of ketals 4a-4e (Table 3).
Our method can also be employed to prepare cis-2,4-
disubstituted 1,3-dioxolanes in enantiomerically pure form
(Scheme 5). This experiment proves that no racemization is
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
that in agreement with the experiment the overall reaction should
proceed without problems.
Scheme 6: PBEh-3c/DCOSMO-RS(THF) structures of the radicals IVc-cis and
IVc-trans.
Finally, we investigated the synthesis of hemiaminals from
suitably substituted benzylamines (Table 4). In this manner, it can
be established if amines are tolerated under our conditions and
how this change in the substitution pattern affects selectivity.
Moreover, our method provides
a
convenient route to
hemiaminals and eventually carbonyl compounds in one step
from benzyl amines.
Table 4. Substrate scope and diastereoselectivity of the hemiaminal synthesis
Scheme 7: Atom-economic transformation of benzylamines into hemiaminals
by titanocene(III) catalysis in single electron steps. Calculated at PW6B95-
D4/def2-QZVP + COSMO-RS(THF) // PBEh-3c/DCOSMO-RS(THF) level of
theory.
by catalysis in single electron steps.
Radical translocation by H-atom abstraction is distinctly less
attractive (about 8 kcal mol-1) than in acetal formation. It seems
that this is not due to differences in the BDE at the benzylic C-H
bonds since the calculated BDE in benzylmethylether is only 0.2
– 0.9 kcal mol-1 lower than for benzyldimethylamine for 6 different
theoretical methods (see SI for details). Additionally, the
conformational investigation established that for the HAT from Va
to VIa only a small structural relaxation occurs, which further
explains the less attractive value for the HAT reaction compared
to the acetal formation (see SI). The reductive heterocycle
formation, is substantially more advantageous than for acetals.
Radical VIa-trans is slightly more stable than radical VIa-cis
(0.9 kcal mol-1) and thus a low diastereoselectivity is predicted in
agreement with the experiment. This is in line with our analysis of
ketal formation. Replacing oxygen with the bulkier alkylated
nitrogen results in similar steric interactions with the titanocene as
in radicals IV.
A practical difficulty associated with the hemiaminals obtained is
their lability towards chromatography on SiO2. The aldehyde
derived products 6a and 6b could be isolated in good yield. We
found that for the synthesis of the ketone derived hemiaminals 6c
– 6f dissolving the crude product in diethylether followed by a
filtration through a plug of cotton wool was sufficient to isolate the
desired products.
The thermodynamics of hemiaminal synthesis and of each step of
the catalytic cycle (Scheme 7) show that hemiaminal formation is
slightly less favorable than acetal formation. With the oxidative
addition being most likely rate-limiting, the calculations suggest
Scheme 8: Acetal and hemiaminal synthesis with aqueous acidic work up for
the synthesis of carbonyl compounds from benzylic ethers and amines.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
de Bruin, X. P. Zhang, J. Am. Chem. Soc. 2011, 133, 8518-8521; c) A.
Gansäuer, M. Klatte, G. M. Brändle, J. Friedrich, Angew. Chem. Int. Ed.
2012, 51, 8891-8894; Angew. Chem. 2012, 124, 9021-9024; d) D. S. G.
Henriques, K. Zimmer, S. Klare, A. Meyer, E. Rojo-Wiechel, M. Bauer, R.
Sure, S. Grimme, O. Schiemann, R. A. Flowers II, A. Gansäuer, Angew.
Chem. Int. Ed. 2016, 55, 7671-7675; Angew. Chem. 2016, 128, 7801-
7805; e) W. Hao, X. Wu, J. Z. Sun, J. C. Siu, S. N. MacMillan, S. Lin, J.
Am. Chem. Soc. 2017, 139, 12141-12144; f) W. Hao, J. H. Harenberg,
S. N. MacMillan, S. Lin, J. Am. Chem. Soc. 2018, 140, 3514-3517; g) H.-
M. Huang, J. J. W. McDouall, D. J. Procter, Nature Cat. 2019, 2, 211-
218; h) C. Yao, T. Dahmen, A. Gansäuer, J. Norton, Science 2019, 364,
764-767; i) M. Lankelma, A. M. Olivares, B. de Bruin, Chem. Eur. J. 2019,
25, 5658-5663; j) K. Lang, S. Torker, L. Wojtas, X. P. Zhang J. Am. Chem.
Soc. 2019, 141, 12388-12396; k) Y. Hu, K. Lang, C. Li, J. B. Gill, I. Kim,
H. Lu, K. B. Fields, M. K. Marshall, Q. Cheng, X. Cui, L. Wojtas, X. P.
Zhang, J. Am. Chem. Soc. 2019, 141, 18160-18169; l) M. Zhou, M.
Lankelma, J. I. van der Vlugt, B. de Bruin, Angew. Chem. Int. Ed. 2020,
59, 11073-11079; Angew. Chem. 2020, 132, 11166-11172.
An attractive feature of our method is that from the benzylamines
substrates, the corresponding ketones can be prepared when an
acidic aqueous work-up is carried with the crude product
(Scheme 8).
Conclusion
In summary, we have developed a novel titanocene catalyzed
method for the synthesis of acetals and hemiaminals from
benzylic ethers and benzylic amines with pending epoxides. The
reaction features catalysis in single electron steps, an epoxide
opening as oxidative addition, a H-atom transfer to generate a
benzylic radical as radical translocation step, and an
organometallic oxygen rebound as reductive elimination. The
thermochemistry
PW6B95-D4/def2-QZVP
of
the
+
reactions
COSMO-RS(THF) // PBEh-3c/
was
studied
at
[6]
a) W. A. Nugent, T. V. RajanBabu, J. Am. Chem. Soc. 1988, 110, 8561-
8562; b) K. Daasbjerg, H. Svith, S. Grimme, M. Gerenkamp, C. Mück-
Lichtenfeld, A. Gansäuer, A. Barchuk, F. Keller, Angew. Chem. Int. Ed.
2006, 45, 2041-2044; Angew. Chem. 2006, 118, 2095-2098; c) A.
Gansäuer, A. Barchuk, F. Keller, M. Schmitt, S. Grimme, M. Gerenkamp,
C. Mück-Lichtenfeld, K. Daasbjerg, H. Svith, J. Am. Chem. Soc. 2007,
129, 1359-1371.
DCOSMO-RS(THF) level of theory including full conformational
search for all stationary points. Theory and experiment are in
good qualitative agreement. The automatic exploration of
conformational space by the CREST program was essential for
identifying the lowest lying conformations and we recommend our
computational approach in general for such mechanistic studies.
The stereoselectivity can be deduced from the structure of the
lowest lying benzylic radicals. Their conformations are controlled
by hyperconjugative interactions and by steric interactions
between the catalyst and the aryl groups. A particularly interesting
aspect of our mechanism is that the oxidation of the benzylic
center occurs under reducing conditions. No external oxidizing
agent is required, making it a promising method for highly
selective oxidations in natural product and API synthesis.
Moreover, the corresponding carbonyl compounds can be directly
obtained by an acidic aqueous work-up of the reaction mixture.
[7]
[8]
For a recent review see: L. M. Stateman, K. M. Nakafuku, D. A. Nagib,
Synthesis 2018, 50, 1569-1586.
a) J. J. Brocks, H.-D. Beckhaus, A. L. J. Beckwith, C. Rüchardt, J. Org.
Chem. 1998, 63, 1935-1943; b) S. J. Blanksby, G. B. Ellison, Acc. Chem.
Res. 2003, 36, 255–263.
[9]
a) X. Huang, J. T. Groves, J. Biol. Inorg. Chem. 2017, 22, 185-207. (b) J.
P. T. Zaragoza, T. H. Yosca, M. A. Siegler, P. Moënne-Loccoz, M. T.
Green, D. P. Goldger, J. Am. Chem. Soc. 2017, 139, 13640-13643.
[10] a) A. Gansäuer, M. Behlendorf, D. von Laufenberg, A. Fleckhaus, C.
Kube, D. V. Sadasivam, R. A. Flowers II, Angew. Chem. Int. Ed. 2012,
51, 4739-4742; Angew. Chem. 2012, 124, 4819-4823; b) A. Gansäuer,
D. von Laufenberg, C. Kube, T. Dahmen, A. Michelmann, M. Behlendorf,
R. Sure, M. Seddiqzai, S. Grimme, D. V. Sadasivam, G. D. Fianu, R. A.
Flowers II, Chem. Eur.J. 2015, 21, 280-289; c) A. Gansäuer, S.
Hildebrandt, A. Michelmann, T. Dahmen, D. von Laufenberg, C. Kube, G.
D. Fianu, R. A. Flowers II Angew. Chem. Int. Ed. 2015, 54, 7003-7006;
Angew. Chem. 2015, 127, 7109-7112; d) R. B. Richrath, T. Olyschläger,
S. Hildebrandt, D. G. Enny, G. D. Fianu, R. A. Flowers II, A. Gansäuer,
Chem. Eur. J. 2018, 24, 6371-6379; e) F. Mühlhaus, H. Weißbarth, T.
Dahmen, G. Schnakenburg, A. Gansäuer, Angew. Chem. Int. Ed. 2019,
58, 14208-14212; Angew. Chem. 2019, 131, 14346-14350; f) Z. Zhang,
R. B. Richrath, A. Gansäuer, ACS Catal. 2019, 9, 3208-3212.g) H.
Weißbarth, F. Mühlhaus, A. Gansäuer, Synthesis 2020, 52, 2940-2947.
[11] a) We are not aware of any examples of our reaction or the postulated
mechanis in the literature. However, a stoichiometric reaction observed
in T. K. Chakraborty, A. K. Chattopadhyay, S. Ghosh, Tetrahedron Lett.
2007, 48, 1139-1142. may proceed by our mechanism. However, no
explanation for formation was givmen or proposed. We tank W. A Nugent
for bringing this reference to our attention. b) The formation of acetals
similar to the ones described here with acid catalysis under
thermodynamic control give essentially no diastereoselectivity: T. D. Inch,
N. Williams, J. Chem. Soc. (C) 1970, 263-269.
Acknowledgements
This work was supported by the DFG (Ga 619/12-2 to A. G.) and
in the framework of the “Gottfried Wilhelm Leibniz Prize” to S. G.
R. B. R. thanks the Konrad-Adenauer-Stiftung for a fellowship.
Keywords: catalysis • conformational search • mechanism •
radical • titanium
[1]
a) A. Gansäuer, A. Fleckhaus, M. Alejandre Lafont, A. Okkel, K. Kotsis,
A. Anoop, F. Neese, J. Am. Chem. Soc. 2009, 131, 16989-16999; b) A.
Gansäuer, S. Hildebrandt, E. Vogelsang, R. A. Flowers II, Dalton Trans.
2016, 45, 448-452.
[2]
[3]
a) J. I. van der Vlugt, Chem. Eur. J. 2019, 25, 2651-2662; b) H.-M. Huang,
M. H. Garduño-Castro, C. Morill, D. J. Procter, Chem. Soc. Rev. 2019,
48, 4626-4638.
[12] a) P. Pracht, F. Bohle, S. Grimme, Phys. Chem. Chem. Phys. 2020, 22,
7169-7192; b) S. Grimme J. Chem. Theory Coput. 2019, 155, 2847-2862.
[13] S. Spicher, S. Grimme, Angew. Chem. Int. Ed. 2020, 59, 15665-15673;
Angew. Chem. 2020, 132, 15795-15803.
a) S. Z. Zard, Radical Reactions in Organic Synthesis Oxford University
Press, Oxford, 2003; b) P. Renaud, M. P. Sibi, Radicals in Organic
Synthesis Wiley-VCH, Weinheim, 2001; c) D. P. Curran, N. A. Porter, B.
Giese, Stereochemistry of Radical Reactions Wiley-VCH, Weinheim,
1996.
[14] a) P. M. Zimmerman, J. Chem. Phys. 2013, 138, 184102; b) P. M.
Zimmerman, J. Chem. Theory Comput. 2013, 9, 3043-3050; c) S. Dohm,
M. Bursch, A. Hansen, S. Grimme J. Chem. Theory Comput. 2020, 16,
2002-2012; d) C. Bannwarth, S. Ehlert and S. Grimme., J. Chem. Theory
Comput. 2019, 15, 1652-1671; e) C. Bannwarth, E. Caldeweyher, S.
Ehlert, A. Hansen, P. Pracht, J. Seibert, S. Spicher, S. Grimme, WIREs
Comput Mol Sci. 2020, e01493; f) Semiempirical extended tight‐binding
[4]
[5]
a) A. Gansäuer, B. Rinker, M. Pierobon, S. Grimme, M. Gerenkamp, C.
Mück-Lichtenfeld, Angew. Chem. Int. Ed. 2003, 42, 3687-3690; Angew.
Chem. 2003, 115, 3815-3818; b) A. Gansäuer, B. Rinker, N. Ndene-
Schiffer, M. Pierobon, S. Grimme, M. Gerenkamp, C. Mück-Lichtenfeld,
Eur. J. Org. Chem. 2004, 2337-2351.
a) W. I. Dzik, X. Xu, X. P. Zhang, J. N. H Reek, B. de Bruin, J. Am. Chem.
Soc. 2010, 132, 10891-10902; b) H. Lu, W. I. Dzik, X. Xu, L. Wojtas, B.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
program package xtb, https://github.com/grimme-lab/xtb. Accessed
September 28, 2020.
[15] S. Grimme, J. G. Brandenburg, C. Bannwarth, A. Hansen, J. Chem. Phys.
2015, 143, 054107.
[16] S. Grimme, C. Bannwarth, S. Dohm, A. Hansen, J. Pisarek, P. Pracht, J.
Seibert, F. Neese, Angew. Chem. Int. Ed. 2017, 56, 14763-14769;
Angew. Chem. 2017, 129, 14958-4964.
[17] a) Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2005, 109, 5656-5667; b) E.
Caldeweyer, S. Ehlert, A. Hansen, H. Neugebauer, S. Spicher, C.
Bannwarth, S. Grimme, J. Chem. Phys. 2019, 150, 154122; c) F.
Weigend, R Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–3305.
[18] A. Klamt, J. Phys. Chem. 1995, 99, 2224-2235; b) A. Klamt, V. Jonas, T.
Bürger, J. C. Lohrenz, J. Phys. Chem. A 1998, 102, 5074-5085.
[19] S. Grimme, Chem. Eur. J. 2012, 18, 9955-9964; b) S. Spicher, S.
Grimme, J. Phys. Chem. Lett. 2020, 11, 6606–6611. (ASAP)
[20] TURBOMOLE V7.4.1 2019, a development of University of Karlsruhe
and Forschungszentrum Karlsruhe GmbH, 1989-2007, TURBOMOLE
GmbH, since 2007; available from http://www.turbomole.com.
[21] F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057-1065.
[22] F.Eckert, A. Klamt, AIChE J. 2002, 48, 369–385.
[23] NBO 6.0. E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter,
J. A. Bohmann, C. M. Morales, C. R. Landis, and F. Weinhold
(Theoretical Chemistry Institute, University of Wisconsin, Madison, WI,
2013); http://nbo6.chem.wisc.edu/.
[24] F. Neese, F. “The ORCA program system” Wiley Interdisciplinary
Reviews: Computational Molecular Science, 2012, Vol. 2, Issue 1, Pages
73–78.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202013561
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Insert graphic for Table of Contents here.
Oxidations with reducing agents? Fact and not fiction in a reaction featuring titanocene(III) catalysis, H-atom-transfer, and an
organometallic oxygen rebound. Acetals and hemiaminals can be readily prepared from benzylic ethers and amines under mild
conditions and in high yields. The structure of the radical intermediates was unraveled by the CREST program.
8
This article is protected by copyright. All rights reserved.