Dearomative Photocatalytic Construction of Bridged 1,3-Diazepanes

Article abstract of DOI:10.1002/anie.201914390

The construction of diverse sp³-rich skeletal ring systems is of importance to drug discovery programmes and natural product synthesis. Herein, we report the photocatalytic construction of 2,7-diazabicyclo[3.2.1]octanes (bridged 1,3-diazepanes) via a reductive diversion of the Minisci reaction. The fused tricyclic product is proposed to form via radical addition to the C4 position of 4-substituted quinoline substrates, with subsequent Hantzsch ester-promoted reduction to a dihydropyridine intermediate which undergoes in situ two-electron ring closure to form the bridged diazepane architecture. A wide scope of N-arylimine and quinoline derivatives was demonstrated and good efficiency was observed in the construction of sterically congested all-carbon quaternary centers. Computational and experimental mechanistic studies provided insights into the reaction mechanism and observed regioselectivity/diastereoselectivity.

Full text of DOI:10.1002/anie.201914390

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Dearomative Photocatalytic Construction of Bridged 1,3-
Diazepanes
Authors: Jamie Alexander Leitch, Tatiana Rogova, Fernanda Duarte,
and Darren Dixon
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.201914390
Angew. Chem. 10.1002/ange.201914390
Link to VoR: http://dx.doi.org/10.1002/anie.201914390
http://dx.doi.org/10.1002/ange.201914390

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Dearomative Photocatalytic Construction of Bridged 1,3-
Diazepanes
Jamie A. Leitch,[a] Tatiana Rogova, [a] Fernanda Duarte,[a]* and Darren J. Dixon[a]*
Abstract: The construction of diverse sp3-rich skeletal ring systems
is of importance to drug discovery programmes and natural product
synthesis. Herein, we report the photocatalytic construction of 2,7-
diazabicyclo[3.2.1]octanes (bridged 1,3-diazepanes) via a reductive
diversion of the Minisci reaction. The fused tricyclic product is
proposed to form via radical addition to the C4 position of 4-
substituted quinoline substrates, with subsequent Hantzsch ester-
radical functionalization at the C4 position (without blocking the
C2 position) still remains a challenge.[9] We reasoned that
judicious choice of reaction conditions could permit an α-amino
radical – created from PCET of an imine derivative enabled by a
Hantzsch ester reductant and photocatalyst[10] – to add to a
suitable Minisci acceptor such as lepidine (4-methylquinoline,
Scheme 1B). Due to the reducing nature of the reaction medium
required to form the α-amino radical (e. g. stoichiometric
quantities of reductant), a question of regioselectivity arose with
two possible products potentially accessible under the conditions.
Firstly a classical redox neutral Minisci functionalization at C2 or
secondly – in the case of C4 addition where rearomatization is not
possible – the net reductive dearomatized dihydropyridine.[11]
Either way, unlocking new reactivity with abundant amine,
aldehyde, and quinoline reagents to generate any C–C linked
structures is of general synthetic interest, and herein we wish to
report our findings.
promoted reduction to
a dihydropyridine intermediate which
undergoes in situ two-electron ring closure to form the bridged
diazepane architecture. A wide scope of N-arylimine and quinoline
derivatives was demonstrated and good efficiency was observed in
the construction of sterically congested all-carbon quaternary centers.
Computational and experimental mechanistic studies provided
insights
into
the
reaction
mechanism
and
observed
regioselectivity/diastereoselectivity.
Introduction
Photocatalytic one-electron activation of organic substrates has
granted access to complementary and/or unprecedented
reactivity to established two-electron chemistry.[1] In this context,
the generation and manipulation of -amino radicals has become
a prominent research focus in recent years.[2] Furthermore, the
photocatalytic single electron reduction – often proton coupled
electron transfer (PCET) reduction – of classically electrophilic
imine derivatives,[3] has emerged as a new pathway to create
such key α-amino radicals. These nucleophilic intermediates have
been shown to then engage in a variety of transformations
including radical-radical coupling[4] and reaction with electrophilic
species.[5] The latter of these methods represents challenging
umpolung cross-electrophile coupling,[6] nevertheless recent
research has established this technique as a valuable avenue
towards decorated α-functionalized amines (Scheme 1A).
Notwithstanding these advances, current methods largely rely on
the use of classical electrophilic Michael acceptors such as
acrylates, and (hetero)styrene derivatives, and accordingly,
adapting this chemistry towards new classes of electrophiles such
as heteroaromatic ring systems could bring new synthetic
opportunities.
The functionalization of quinoline motifs has become a
commonplace target in Minisci-type chemistry,[7] and reactivity at
both C2 and C4 positions has been established, although often
with selectivity issues arising in unbiased systems.[7d],[8] Whilst
recent photocatalytic approaches have enabled selective C2
functionalization through hydrogen bonding networks,[8a] selective
Scheme 1. Photoredox catalysis in reverse polarity synthesis, and proposed
concept for quinoline functionalization
[a]
Dr J. A. Leitch, T. Rogova, Prof. Dr. F. Duarte, Prof. Dr. D. J. Dixon
Department of Chemistry, Chemical Research Laboratory, University of
Oxford, 12 Mansfield Road, Oxford (UK)
E-mail: fernanda.duartegonzalez@chem.ox.ac.uk,
darren.dixon@chem.ox.ac.uk
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Results and Discussion
Although bridged 1,3-diazepane structures have been shown to
possess in vivo anti-cancer activity;[17] and are present in natural
products and corresponding analogues,[18] the heterocycle
remains synthetically challenging to access. Accordingly, we were
eager to further develop this quinoline dearomatization system as
a novel access point to this underexplored architecture.[19]
Previously, our group has reported that the use of bespoke
Hantzsch ester reductants can modulate reactivity and in turn
increase product yield and / or improve diastereoselectivity in
reductive functionalization of imines.[5d],[5f],[20] A survey of Hantzsch
ester reductants (Scheme 2B, entries 2-4) demonstrated that
methyl carboxyphenyl derivative (HE4) led to increased
conversion to product (90%) and to a higher diastereomeric ratio
(1.9:1, for further optimization details, see supporting information).
From a variety of additives, we were pleased to observe that
certain Lewis acids increased the diastereoselectivity of this
transformation, with zinc triflimide performing most effectively
(entries 5-7). Sterically demanding Lewis acids – such as
methylaluminium bis(2,6-di-tert-butyl-4-methylphenoxide) (MAD)
– unfortunately led to a complex mixture of products.
We began our investigation using fluorine tagged imine (1a), 4-
methylquinoline (lepidine, 2a), [Ir(dFCF3(ppy))2(dtbbpy)]PF6 (1
mol%) as photocatalyst, the commercial Hantzsch ester (HE1) as
stoichiometric reductant, in DMSO, under blue light irradiation
(Scheme 2A). Excitingly, good reactivity was observed from
preliminary experiments. Interestingly however, neither C2-
Minisci product nor C4-dihydropyridine were formed; the major
product
isolated
was
a
fused
tricyclic
2,7-
diazabicyclo[3.2.1]octane (or bridged 1,3-diazepane), the
cycloisomerized adduct of the anticipated C4-dihydropyridine.
This transformation constitutes a formal diversion from classical
Minisci chemistry towards the dearomatization of the quinoline
heterocycle, forming an unusual fused tricyclic framework
possessing four new sp3 carbon centres. It also exemplifies the
synthetic utility that dearomative photochemistry can provide as a
tool in upgrading abundant two-dimensional feedstock chemicals
into
frameworks,[12] a concept thoroughly explored in seminal work by
Sarlah[13] others.[14]
and Furthermore, photocatalytic
structurally
complex
sp3-rich
three-dimensional
dearomatization methods have potential benefits over other
complementary approaches using the exhaustive hydrogenation
of (hetero)arenes, notably in the formation of bridged
heterocycles.[15],[16]
Scheme 2. Preliminary studies and subsequent optimization of photocatalytic dearomative construction of bridged 1,3-diazepanes.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 3. Substrate scope for the dearomative photocatalytic construction of bridged 1,3-diazepanes
Importantly, control experiments elucidated that the iridium
photocatalyst, the Hantzsch ester, and blue light irradiation
were all essential for reactivity (>95% imine 1a remained
unreacted, entries 8-10). This methodology was also shown to
be compatible with in situ formation of the imine (entry 11).[21]
Furthermore, isopropyl (3ae) and cyclohexyl (3af) derivatives
were well-tolerated, with product structures obtained in high
yields and excellent diastereoselectivity. When exploring
substitution on the aryl moiety of the quinoline, this chemistry
was shown to be amenable to functionalization at the 7-
position (3ag-3aj), with starting materials readily prepared
from commercially available 4,7-dichloroquinoline. Notably,
the pinacol boronate derivative (3ai) was tolerated.
Gratifyingly, C6-substitution was also amenable to substitution
with 6-bromolepidine performing well, further incorporating
functional handles into the product (3ak).
With optimal conditions established, we explored how
substitution patterns across the quinoline moiety affected
product formation (Scheme 3A).[22][23] Pleasingly with
increased steric demand at the 4-position, ethyl (3ab), n-butyl
(3ac), and i-butylquinoline (3ad) substrates performed with
greater diastereoselectivity and reaction efficiency was
maintained (for unsuccessful/low yielding substrates, please
see supporting information)
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Substituted arenes (3o-3p) also performed efficiently as did a
thiophene derivative (3q), albeit in reduced yield.
As previous reports on quinoline dearomatization have
required pre-activation of the quinoline substrate as N-alkyl
quinolinium salt to permit reaction efficiency,[17b],[19c] we
reasoned that this salt could also engage in this chemistry.
Pleasingly when N-benzyl lepidinium bromide (2l) was
exposed to the dearomatization conditions, the N-benzyl
diazepane structure was also observed, although any
diastereoselectivity witnessed previously was suppressed (3al,
Scheme 4A). The addition of the Lewis acid remained
beneficial to this transformation, suggesting that zinc
coordination to the nitrogen atom on the -amino radical rather
than the quinoline moiety could be relevant.
Recent research in the area has demonstrated that
aldehydes react similarly in reductive photocatalytic conditions
to imine substrates.[4],[5] Accordingly, we investigated whether
an aldehyde substrate was also amenable to this dearomative
chemistry to give the corresponding hemiaminal adduct
(Scheme
4B).
Interestingly,
however,
when
4-
fluorobenzaldehyde (4a) was submitted to the reaction
conditions a fused tetracyclic structure 5a was observed as the
major product amongst a mixture of other compounds. This
complex molecular architecture is derived from the formal
addition of two lepidine molecules to an α-oxoradical. We
propose that this takes place initially via formation of the
hemiaminal product (analogous to that observed in the case
of the imine). Subsequent direct or indirect net loss of a
hydrogen atom provides the open shell α-amino radical which
can then participate in C–C bond formation at the C4 position
of a secondary lepidine molecule. Subsequent reduction
followed by in situ cyclization affords the structurally complex
sp3-rich fused hexacyclic 5a. Single crystal X-ray diffraction
analysis of this structure confirmed the bridgehead
connectivity, with notably two C–C bond formations at C4
taking place, along with the construction of two connected
quaternary centres.
Until this point in our studies using 4-alkylquinoline
substrates, we observed a diastereomeric preference for the
endo isomer. Interestingly, however, when we employed 4-
phenylquinoline (2m) as coupling partner, we observed a
notable shift in selectivity and the exo diastereomer now
predominated in a 4.5:1 dr (3am). To further confirm this
switch in selectivity, we conducted the reaction with two further
4-arylquinolines. Pleasingly, the exo-selectivity was
maintained in these experiments (3an-3ao), and
computational
insights
on
the
origins
of
this
diastereoselectivity switch are provided below.
Scheme 4. (A) Use of the quinolinium salt. (B) Use of α-oxoradical in the
Mechanistic Studies
methodology. (C) Use of 4-arylquinolines as coupling partners.
Determination of active radical species. Whilst the single
electron reduction of imine derivatives has become well
established,[4],[5] a few recent reports[9b],[24] have demonstrated
that quinoline / pyridinium molecules can be activated in redox
pathways to create heteroarene-centred radicals. For these
reasons, an alternative mechanism could be plausible, in
which proton coupled electron transfer of a lepidine molecule
would result in a C4-centred radical on the lepidine and which
could viably add to an imine molecule (Scheme 5, left).
The imine component was then studied, initially with variation
to the aniline fragment (Scheme 3B). A variety of functionality
was well-tolerated in this methodology including
trifluoromethoxyarene (3e) and iodoarenes (3f-3g).
Furthermore the aldehyde fragment of the imine was varied,
and a wide electronic profile tolerance varying from electron-
releasing methoxy (3h) to electron withdrawing trifluoromethyl
(3m) and ester derivatives (3n), was observed. meta-
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 5. Study into the nature of the radical species operating in the mechanism.
To study this, first we conducted a radical clock experiment
using 4-cyclopropyllepidine (2ap) where the presence of a
stabilized C4-radical would lead to favourable fragmentation of
this cyclopropyl unit to the primary radical. Interestingly, the
diazepane product was still formed in good yield and dr (3ap)
with the cyclopropyl unit remaining intact and no fragmentation
adducts were observed.[25]
the partially oxidized Hantzsch ester radical cation (pKa = 7.2)
or the zinc additive are (Lewis) acidic enough to facilitate the
electron transfer event to deliver the key nucleophilic α-amino
radical (A1, ω = 0.98,[28] Scheme 6A, left).
C4/C2 selectivity. Having identified the active radical species
we then investigated the regioselectivity of C–C bond
formation via computational methods. Although partial
protonation may be relevant for the transformation; for
simplicity, our initial model considered radical addition to a
neutral lepidine molecule.[29]
Furthermore we examined the redox potentials of 2a and 1a
relative to the catalyst both computationally and
experimentally (Scheme 5, right, see supporting information
for further details, electrochemical potentials are given vs.
saturated calomel electrode [SCE]). Values of –2.22 V
(computed)/–2.14 V (measured), and –1.97 V (computed)/–
1.95 V (measured) were obtained for 2a and 1a respectively.
Initially the addition of the α-amino radical to either the C2 or
C4 position of the lepidine coupling partner was explored.
Density Functional Theory (DFT) calculations at the
SMD(DMSO)-ωB97X-D/6-311++G(d,p)//SMD(DMSO)-
ωB97X-D/6-31G(d) level of theory show that the C2 position is
These
are
outside
the
reduction
capacity
of
slightly preferred, albeit within computational error (ΔG‡
=
(C2)
[Ir(dF(CF3)ppy)2(dtbbpy)]PF6 (calculated as Ered [IrIII/IrII] = –
1/2
21.7 vs ΔG‡
= 22.4 kcal mol-1).[30] The resulting radical
1.51 V in DMSO (vs. –1.37 V in MeCN).[1a],[26] This suggests
that H-bond interactions and/or a certain degree of protonation
(or Lewis acid assistance) is essential for raising the reduction
potential of these molecules into an accessible range.[10]
Whilst predicting the extent of imine protonation at the event
of electron transfer is challenging,[27] calculation of the
reduction potentials for both the imine (1a) and lepidine
species (2a), in both their neutral and protonated forms,
provides an estimate of the operating window. As expected,
protonation raises the reduction potentials (easier to reduce)
for both 1a and 2a species into the operating range of the
photocatalyst [Ir]. The fact that the reduction potential of
iminium 1a[H+] is 0.30 V higher than the protonated 2a[H+]
counterpart strongly suggests that the active radical species is
formed from the imine starting material 1a rather than the
lepidine coupling partner 2a.
(C4)
aromatic intermediates A2-C4 & A2-C2 are high in energy (cf.
A1 and 2a), with the C4 intermediate slightly more stable (16.5
vs 15.0 kcal mol-1 for C2 and C4 respectively, Scheme 6A).
These results suggest a reversible radical addition, where both
pathways could be populated under thermodynamic control.
Subsequently, the radical aromatic intermediates A2-C4& A2-
C2 undergo formal addition of a hydrogen atom to afford the
dihydropyridine. The redox potentials associated with this
process showed that A2-C4 is more readily reduced than A2-
C2 (Ered = –0.85 V vs –0.95 V for A2-C4 and A2-C2
1/2
respectively). A similar trend is observed for the protonated A2
intermediate (A2[H+]) (Ered = –0.16 V vs -0.31 V for A2-
1/2
C4[H+] and A2-C4[H+] respectively). It is noteworthy that all
four are sufficiently oxidizing to take an electron from the
Hantzsch ester intermediate HEH• (Eox = –1.15 V). This
1/2
transformation likely takes place as a concerted PCET, with
the true reduction potentials for A2-C2 and A2-C4 lying
between the two calculated values.
This result along with experimental studies allows us to rule
out the presence of a lepidine based radical. On studying the
pKBH+ value of the substrate (1a, pKBH+ = 7.4) we suggest that
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 6. Mechanistic studies into C4 vs. C2 regioselectivity with in silico insights: (A) Radical addition. (B) Fates of both C2 and C4 dihydropyridine adducts.
Furthermore, we investigated the plausible fates of
dihydropyridines A3-C4 and A3-C2 (Scheme 6B). We observed
that the tautomerization and subsequent ring closure of A3-C4 to
form 3aa-C4 is energetically facile (ΔG° = –11.2 kcalmol-1),
indicating why the dihydropyridine structure is not detected.
Control experiments deduced that both diastereomers of this
framework are stable to further re-subjection to the reaction
conditions with or without Hantzsch ester. In contrast,
tautomerization from A3-C2 is substantially less favourable (ΔG°
= +21.8 kcalmol-1), suggesting this structure has no plausible
downhill two-electron pathway. Despite this, from our previous
calculations we were aware that this substrate can undergo single
reasons, we hypothesize that A3-C2 will competitively quench the
photoexcited iridium(III) species (Eox1/2 = +1.21 V) in favour of the
Hantzsch ester. As the reaction conditions are net reducing we
expect no further oxidation to the Minisci product. Following the
single electron oxidation, there is a low energetic barrier to
fragmentation of this radical cation intermediate to the
corresponding α-amino radical and lepidine (ΔG‡ = +5.2 kcalmol-1,
ΔG° = –16.5 kcalmol-1). Aligned to previous reports detailing the
reversibility in Minisci-type radical additions, we conclude that this
fragmentation pathway is fully plausible.[7d],[8a]
As the reducing iridium(II) species formed above can then
reduce a further molecule of 1a through a PCET mechanism (with
the proton lost in fragmentation), this feedback loop reforms an
equivalent of the α-amino radical with no net exhaustion of
reaction components.
electron oxidation to the corresponding radical cation (Eox
=
1/2
+0.31 V). In fact this substrate is more readily oxidized than the
Hantzsch ester reductant (HE4) (Eox = +0.61 V). For these
1/2
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 7. Computed origins of diastereoselectivity
Whilst fragmentation is also feasible for A3-C4 (Eox1/2 = +0.12 V),
as no viable downstream path for A3-C2 exists, we consider this
mechanism responsible for the amplification of regioselectivity,
with yields up to 95% of the experimentally observed C4 product.
The degree of π-orbital overlap that could be achieved in the
transition states of the 3am exo and endo structures possibly has
a
greater effect on the energy difference than for the
diastereomers of 3aa. This change in atomic arrangement of
radical addition accounts for the switch in
diastereoselectivity.[31],[32]
Diastereoselectivity. Despite endo selectivity predominating
throughout the scope with 4-alkylquinolines, as mentioned above,
when 4-phenylquinoline (1am) was introduced as the coupling Conclusion
partner a switch in diastereoselectivity was observed (Scheme 7),
with the exo isomer favoured in a 4.5:1 dr (3am). The
diastereomeric preference of the reaction is set by the initial
addition of the α-amino radical A1 into the respective quinolines
2a/2m (TS1). DFT calculations demonstrate that for 3aa (derived
from lepidine), the transition state energy difference was
negligible between the endo and exo diastereomers. Conversely,
In conclusion, a mild and practical method for the photocatalytic
coupling of simple N-arylimines and quinolines into bridged 1,3-
diazepane frameworks has been developed. This was achieved
via diverting the classical Minisci reaction using net reducing
conditions to permit dearomatization and cyclization events. The
optimized method was shown to be tolerant of a range of
functional groups with the formation of 31 examples of the fused
tricyclic structure with yields up to 95% and diastereoselectivity
up to 8.0:1 dr. Excellent regioselectivity for C–C bond formation
at C4 was observed throughout the study. These results were
for substrate 3an (derived from 4-phenylquinoline), TS1-
is
exo
lower in energy by 1.9 kcal mol-1 with respect to TS1-endo. A visual
comparison (Scheme 7, top right) of the transition structures
reveals that the phenyl substitution on quinoline 2m, results in a
change of orientation of the free α-amino radical species. This
alternative conformation maximises π-π stacking interactions
between the phenyl substituent on 2m and the N-aryl group on
the α-amino radical.
rationalised
via
computational
studies
whereby
a
fragmentation/recycling method for unobserved C2 functionalized
products into the C4 diazepane was postulated. This method
offers a valuable step in the rapid construction of complex sp3-rich
heterocycles with demonstrable efficiency and regioselectivity.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
54, 8828−8832; e) E. Fava, A. Millet, M. Nakajima, S. Loescher, M.
Rueping, Angew. Chem., Int. Ed. 2016, 55, 6776−6779; f) M. Chen, X.
Zhao, C. Yang, W. Xia, Org. Lett. 2017, 19, 3807−3810; g) H.-H. Zhang,
S. Yu, J. Org. Chem. 2017, 82, 9995-10006; h) J. Rong, P. H. Seeberger,
K. Gilmore, Org. Lett. 2018, 20, 4081−4085; i) N. Zhou, X.-A. Yuan, Y.
Zhao, J. Xie, C. Zhu, Angew. Chem., Int. Ed. 2018, 57, 3990−3994.
a) L. Qi, Y. Chen, Angew. Chem., Int. Ed. 2016, 55, 13312−13315; b) A.
L. Fuentes de Arriba, F. Urbitsch, D. J. Dixon, Chem. Commun. 2016, 52,
14434−14437; c) K. N. Lee, Z. Lei, M.-Y. Ngai, J. Am. Chem. Soc. 2017,
139, 5003−5006; d) J. A. Leitch, A. L. Fuentes de Arriba, J. Tan, O. Hoff,
C. M. Martinez, D. J. Dixon, Chem. Sci. 2018, 9, 6653-6658; e) A.
Trowbridge, D. Reich, M. J. Gaunt, Nature 2018, 561, 522-527; f) T.
Rossolini, J. A. Leitch, R. Grainger, D. J. Dixon, Org. Lett. 2018, 20,
6794-6798; g) N. J. Floden, A. Trowbridge, D. Willcox, S. M. Walton, Y.
Kim, M. J. Gaunt, J. Am. Chem. Soc. 2019, 141, 8426-8430; h) D. Vasu,
A. L. Fuentes de Arriba, J. A. Leitch, A. de Gombert, D. J. Dixon, Chem
Sci 2019, 10, 3401-3407; i) K. Cao, S. M. Tan, R. Lee, S. Yang, H. Jia,
X. Zhao, B. Qiao, Z. Jiang, J. Am. Chem. Soc. 2019, 141, 5437-5443; j)
R. Wang, M. Ma, X. Gong, X. Fan, P. J. Walsh, Org. Lett. 2019, 21, 27-
31.
Further studies into new reactivity of related radical precursors are
currently underway, and the results will be disclosed in due course.
Acknowledgements
J.A.L. thanks the Leverhulme Trust (RPG-2017-069) for a
research fellowship. T.R. is supported by the EPSRC Centre for
Doctoral Training in Synthesis for Biology and Medicine
(EP/L015838/1), generously supported by AstraZeneca, Diamond
Light Source, Defence Science and Technology Laboratory,
Evotec, GlaxoSmithKline, Janssen, Novartis, Pfizer, Syngenta,
Takeda, UCB, and Vertex. T. R. also thanks the Royal
Commission of 1851 Industrial Fellowship and NSERC PGS-D.
We thank Heyao Shi for X-ray structure determination and Dr
Amber L. Thompson and Dr Kirsten E. Christensen (Oxford
Chemical Crystallography Service) for X-ray mentoring and help.
We also thank Thomas Rossolini (University of Oxford) for select
relevant experiments, Prof. John Brown (University of Oxford) for
helpful discussions, and Dr. Andrew Frawley, Dr. Virginia Wycisk,
and Kathryn Leslie University of Oxford) for assistance in
photophysical and electrochemical measurements.
[5]
[6]
[7]
D. A. Everson, D. J. Weix, J. Org. Chem. 2014, 79, 4793-4798.
a) F. Minisci, F. Fontana, E. Vismara, J. Heterocycl. Chem. 1990, 27, 79-
96; b) R. S. J. Procter, R. J. Phipps, Angew. Chem. Int. Ed. 2019, 58,
13666-13699; c) A. C. Sun, R. C. McAtee, E. J. McClain, C. R. J.
Stephenson, Synthesis, 2019, 51, 1063-1072; d) F. Minisci, E. Vismara,
F. Fontana, G. Morini, M. Serravelle, J. Org. Chem. 1987, 52, 736-740.
a) R. S. J. Proctor, H. J. Davis, R. J. Phipps, Science 2018, 360, 419-
422; b) J. Jin, D. W. C. MacMillan, Nature 2015, 525, 87-90; c) J. Jin, D.
W. C. MacMillan, Angew. Chem. Int. Ed. 2015, 54, 1565-1569; d) D. A.
DiRocco, K. Dykstra, S. Krska, P. Vachal, D. V. Conway, M. Tudge,
Angew. Chem. Int. Ed. 2014, 53, 4802-4806; e) G.-X. Li, C. A. Morales-
Rivera, Y. Wang, F. Gao, G. He, P. Liu, G. Chen, Chem. Sci. 2016, 7,
6407-6412; f) Y. Zhang, K. B. Teuscher, H. Ji, Chem. Sci. 2016, 7, 2111-
2118; g) S. Devari, B. A. Shah, Chem. Commun. 2016, 52, 1490-1493;
h) A. Arora, J. D. Weaver, Org. Lett. 2016, 18, 3996-3999; i) J. K. Matsui,
D. N. Primer, G. A. Molander, Chem. Sci. 2017, 8, 3512-3522; j) R. A.
Garza-Sanchez, A. Tlahuext-Aca, G. Tavakoli, F. Glorius, ACS Catal.
2017, 7, 4057-4061; k) P. Nuhant, M. S. Oderinde, J. Genovino, A.
Juneau, Y. Gagné, C. Allais, G. M. Chinigo, C. Cjoi, N. W. Sach, L.
Bernier, Y. M. Fobian, M. W. Bundesmann, B. Khunte, M. Frenette, O. O.
Fadeyi, Angew. Chem. Int. Ed. 2017, 56, 15309-15313; l) P. Liu, W. Liu,
C.-J. Li, J. Am. Chem. Soc. 2017, 139, 14315-14321.
Keywords: Dearomatization, photocatalysis, quinoline, sp3
[8]
framework, diazepane
[1]
For reviews see: a) C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem.
Rev. 2013, 113, 5322-5363; b) K. L. Skubi, T. R. Blum, T. P. Yoon, Chem.
Rev. 2016, 116, 10035-10074; c) J. M. Narayanam, C. R. J. Stephenson,
Chem. Soc. Rev. 2011, 40, 102-113; d) A. Studer, D. P. Curran, Angew.
Chem. Int. Ed. 2016, 55, 58-102. e) M. Silvi, P. Melchiorre, Nature 2018,
554, 41-49. In building block chemistry see: f) J. Wu, A. Noble, V. K.
Aggarwal, J. Am. Chem. Soc. 2018, 140, 10700-10704; g) A. Fawcett, J.
Pradeilles, Y. Wang, T. Mutsuga, E. L. Myers, V. K. Aggarwal, Science
2017, 357, 283-286; h) J. L. Shwarz, F. Shäfers, A. Tlahuext-Aca, L.
Lückemeier, F. Glorius, J. Am. Chem. Soc. 2018, 140, 12705-12709; i)
F. J. R. Klauck, M. J. James, F. Glorius, Angew. Chem. Int. Ed. 2017, 56,
12336-12339; j) L. Candish, M. Teders, F. Glorius, J. Am. Chem. Soc.
2017, 139, 7440-7443. In natural product chemistry see: k) S. P. Pitre, N.
A. Weires, L. E. Overman, J. Am. Chem. Soc. 2019, 141, 2800-2813; l)
M. D. Kärkäs, J. A. Porco, C. R. J. Stephenson, Chem. Rev. 2016, 116,
9683-9747. In derivatization of biologically relevant compounds see: m)
J. J. Douglas, M. J. Sevrin, C. R. J. Stephenson, Org. Process Res. Dev.
2016, 20, 1134-1147; n) D. A. Nagib, D. W. C. MacMillan, Nature 2011,
480, 224-228; o) N. A. Romero, K. A. Margrey, N. E. Tay, D. A. Nicewicz,
Science 2015, 349, 1326-1330.
[9]
For selected examples see: a) X. Ma, H. Dang, J. A. Rose, P. Rablen; S.
B. Herzon, J. Am. Chem. Soc. 2017, 139, 5998-6007; b) I. Kim, G. Kang,
K. Lee, B. Park, D. Kang, H. Jung, Y.-T. He, M.-H. Baik, S. Hong, J. Am.
Chem. Soc. 2019, 141, 9239-9248; c) S. Jung, H. Lee, Y. Moon, H.-Y.
Jung, S. Hong, ACS Catal. 2019, 9, 9891-9896. For non-radical based
approaches see: d) M. Nagase, Y. Kuninobu, M. Kanai, J. Am. Chem.
Soc. 2016, 138, 6103-6106; e) M. C. Hilton, R. D. Dolewski, A. McNally,
J. Am. Chem. Soc. 2016, 138, 13806-13809; f) T. Iwai, M. Sawamura,
ACS Catal. 2015, 5, 5031-5040.
[2]
For reviews see: a) J. W. Beatty, C. R. J. Stephenson, Acc. Chem. Res.
2015, 48, 1474-1484; b) K. Nakajima, Y. Miyake, Y. Nishibayashi, Acc.
Chem. Res. 2016, 49, 1946-1956; c) K. N. Lee, M.-Y. Ngai, Chem.
Commun. 2017, 53, 13093−13112. For select papers see: d) Z. Zuo, D.
T. Ahneman, L. Chu, J. A. Terrett, A. G. Doyle, D. W. C. MacMillan,
Science, 2014, 345, 437-440; e) A. Noble, D. W. C. MacMillan, J. Am.
Chem. Soc. 2014, 136, 11602-11605; f) L. R. Espelt, I. S. McPherson, E.
M. Wiensch, T. P. Yoon, J. Am. Chem. Soc. 2015, 137, 2452-2455.
a) M. Ischay, T. P. Yoon, Eur. J. Org. Chem. 2012, 2012, 3359-3372; b)
T. Morack, C. Mück‐Lichtenfeld, R. Gilmour, Angew. Chem. Int. Ed. 2019,
58, 1208-1212; c) W. Xu, J. Ma, X.-A. Yuan, J. Dai, J. Xie, C. Zhu, Angew.
Chem. Int. Ed. 2018, 57, 10357-10361.
[10] E. C. Gentry, R. R. Knowles, Acc. Chem. Res. 2016, 49, 1546-1556.
[11] For two interesting recent reports on reductive Minisci reactions which
afford quinoline products through spin-centre shifts see: a) B. Bieszczad,
L. A. Perego, P. Melchiorre, Angew. Chem. Int. Ed. 2019, 58, 16878-
16883; b) Z. Wang, Q. Liu, X. Ji, G.-J. Deng, H. Huang, ACS Catal. 2020,
10, 154-159.
[12] For benefits on sp3-rich frameworks see: a) H. Lachance, S. Wetzel, K.
Kumar, H. Waldmann, J. Med. Chem. 2012, 55, 5989-6001; b) F.
Lovering, J. Bikker, C. Humbelt, J. Med. Chem. 2009, 52, 6752-6756; c)
F. Lovering, Med. Chem. Commun. 2013, 4, 515-519.
[3]
[4]
[13] For reviews see: a) W. C. Wertjes, E. H. Southgate, D. Sarlah, Chem.
Soc. Rev. 2018, 47, 7996-8017; b) M. Okumura, D. Sarlah, Eur. J. Org.
Chem. 2019, DOI: 10.1002/ejoc.201901229. For seminal reports see: c)
E. H. Southgate, J. Pospech, J. Fu, D. R. Holycross, D. Sarlah, Nat.
Chem. 2016, 8, 922-928; d) E. H. Southgate, D. R. Holycross, D. Sarlah,
a) D. Hager, D. W. C. MacMillan, J. Am. Chem. Soc. 2014, 136,
16986−16989; b) D. Uraguchi, N. Kinoshita, T. Kizu, T. Ooi, J. Am. Chem.
Soc. 2015, 137, 13768−13771; c) J. L. Jeffrey, F. R. Petronijevic, D. W.
C. MacMillan, J. Am. Chem. Soc. 2015, 137, 8404−8407; d) M. Nakajima,
E. Fava, S. Loescher, Z. Jiang, M. Rueping, Angew. Chem., Int. Ed. 2015,
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Angew. Chem. Int. Ed. 2017, 56, 15049-15052; e) L. W. Hernandez, J.
Pospech, U. Klöckner, T. W. Bingham, D. Sarlah, J. Am. Chem. Soc.
2017, 139, 15656-15659; f) L. W. Hernandez, U. Klöckner, J. Pospech,
L. Hauss, D. Sarlah, J. Am. Chem. Soc. 2018, 140, 4503-4507; g) W. C.
Wertjes, M. Okumura, D. Sarlah, J. Am. Chem. Soc. 2019, 141, 163-167.
[14] For reviews on dearomatization see a) Q. Ding, X. Zhou, R. Fan, Org.
Biomol. Chem. 2014, 12, 4807-4815; b) S. P. Roche, J. A. Porco, Angew.
Chem. Int. Ed. 2011, 50, 4068-4093; c) A. R. Pape, K. P. Kaliappan, E.
P. Kündig, Chem. Rev. 2000, 100, 2917-2940. For selected further
photocatalytic examples see: d) M. J. James, J. L. Schwarz, F. Strieth-
Kalthoff, B. Wibbeling, F. Glorius, J. Am. Chem. Soc. 2018, 140, 8624-
8628; e) D. Alpers, M. Gallhof, J. Witt, F. Hoffmann, M. A. Brasholz,
Angew. Chem. Int. Ed. 2017, 56, 1402-1406; f) W. Dong, Y. Yuan, X.
Gao, M. Keranmu, W. Li, X. Xie, Z. Zhang, Org. Lett. 2018, 20, 5762-
5765; g) A. Flynn, K. McDaniel, M. Hughes, D. B. Vogt, N. T. Jui,
ChemRxiv 2019, DOI: 10.26434/chemrxiv.9874454; h) R. McAtee, E. A.
Noten, C. R. J. Stephenson, Chem Rxiv 2019, DOI:
10.26434/chemrxiv.9864071. For selected non-photochemical examples
see: i) J. J. Farndon, X. Ma, J. F. Bower, J. Am. Chem. Soc. 2017, 139,
14005-14008; j) K. Douki, H. Ono, T. Taniguchi, J. Shimokawa, M.
Kitamura, T. Fukuyama, J. Am. Chem. Soc. 2016, 138, 14578-14581; k)
G. Zhao, G. Xu, C. Qian, W. Tang, J. Am. Chem. Soc. 2017, 139, 3360-
3363; l) E. R. Welin, A. Ngamnithiporn, M. Klatte, G. Lapointe, G. M.
Pototschnig, M. S. J. McDermott, D. Conklin, C. D. Gilmore, P. M.
Tadross, C. K. Haley, K. Negoro, E. Glibstrup, C. U. Grünanger, K. M.
Allan, S. C. Virgil, D. J. Slamon, B. M. Stoltz, Science 2019, 363, 270-
275.
[21] Only applicable when imine formation when is facile.
[22] In almost all cases, diastereomers are easily separable using silica gel
column chromatography.
[23] Imines derived from alkylaldehydes or alkylamines were not good
substrates under these reaction conditions. See supporting information
for more details.
[24] R. A. Garza-Sanchez, T. Patra, A. Tlahuext-Aca, F. Strieth-Kalthoff, F.
Glorius, Chem. Eur. J. 2018, 24, 10064-10068.
[25] This observation was further verified by control experiments whereby
under the reaction conditions lepidine remained unreacted without the
imine substrate (i.e. no reduction, or interaction with a radical trap, see
supporting information), whereas the imine – without the presence of
lepidine – shows full conversion to over-reduced amine and aza-pinacol
dimer.
[26] Calculated using Nicewicz’s method: H. G. Roth, N. A. Romero, D. A.
Nicewicz, Synlett 2016, 27, 714-723.
[27] C. P. Seath, D. B. Vogt, Z. Xu, A. J. Boyington, N. T. Jui, J. Am. Chem.
Soc. 2018, 140, 15525-15534.
[28] F. De Vleeschouwer, V. Van Speybroeck, M. Waroquier, P. Geerlings, F.
De Proft, Org. Lett. 2007, 9, 2721-2724.
[29] Preliminary investigations into protonated species showed negligible
differences in regioselectivity to the unprotonated form.
[30] For full computational details, see supporting information.
[31] A pre-print of this manuscript is available at: J. A. Leitch, T. Rogova, F.
Duarte,
D.
J.
Dixon,
Chem
Rxiv,
2019,
DOI:
10.26434/chemrxiv.10284104.
[32] Crystallographic data is available free of charge from The Cambridge
Crystallographic Data Centre (www.ccdc.cam.ac.uk) for the following
compounds: 3aa(endo) CCDC 1956748, 3aa(exo) CCDC 1956749, 5a
CCDC 1956750
[15] For reviews see: a) M. P. Wiesenfeldt, Z. Nairoukh, T. Dalton, F. Glorius,
Angew. Chem. Int. Ed. 2019, 58, 10460-10476; b) D. S. Wang, Q.-A.
Chen, S.-M. Lu, Y.-G. Zhou, Chem. Rev. 2012, 112, 2557-2590. For key
references see: c) M. P. Wiesenfeldt, Z. Nairoukh, W. Li, F. Glorius,
Science 2017, 357, 908-912; d) Z. Nairoukh, M. Wollenburg, C.
Schlepphorst, K. Bergander, F. Glorius, Nat. Chem. 2019, 11, 264-270;
e) W.-X. Huang, B. Wu, X. Gao, M.-W. Chen, B. Wang, Y.-G. Zhou, Org.
Lett. 2015, 17, 1640-1643; f) W.-B. Wang, S.-M. Lu, P.-Y. Yang, X.-W.
Han, Y.-G. Zhou, J. Am. Chem. Soc. 2003, 125, 10536-10537; g) A.
Iimuro, K. Yamaji, S. Kandula, T. Nagano, Y. Kita, K. Mashima, Angew.
Chem. Int. Ed. 2013, 52, 2046-2050; h) S. Chakraborty, W. W.
Brennessel, W. D. Jones, J. Am. Chem. Soc. 2014, 136, 8564-8567.
[16] For reading on potential benefits of sp3 rich heterocycles in
physicochemical modulation and bioisostere investigations, see a) V. V.
Levterov, O. Michurin, P. O. Borysko, S. Zozulya, I. V. Sadkova, A. A.
Tolmachev, P. K. Mykhailiuk, J. Org. Chem. 2018, 83, 14350-14361; b)
T. Ritchie, S. MacDonald, R. Young, S. Pickett, Drug Disc. Today 2011,
16, 164-171; c) D. Staveness, T. M. Sodano, K. Li, E. A. Burnham, K. D.
Jackson, C. R. J. Stephenson, Chem, 2019, 5, 215-226; d) D. Staveness,
J. L. Collins, R. C. McAtee, C. R. J. Stephenson, Angew. Chem. Int. Ed.
2019, DOI: 10.1002/anie.201909492.
[17] a) W. Hu, Z. Kang, D. Zhang, H. Xu, CN106588926, 2017-04-26; b) Z.
Kang, D. Zhang, W. Hu, Org. Lett. 2017, 19, 3783-3786.
[18] a) Y. Miura, N. Hayashi, S. Yokoshima, T. Fukuyama, J. Am. Chem. Soc.
2012, 1234, 11995-1997; b) Z. Xu, X. Bao, Q. Wang, J. Zhu, Angew.
Chem. Int. Ed. 2015, 54, 14937-14940; c) J. P. Marino, M. W. Kim, R.
Lawrence, J. Org. Chem. 1989, 54, 1782-1784.
[19] For quinoline dearomatization see: a) Y. Terada, I. Takeuchi, H. Hamada,
Chem. Pharm. Bull 1986, 34, 1917-1923. This tentative structure was
later confirmed via X-ray crystallography, see: b) K. Hatano, I. Takeuchi,
Y. Hamada, T. Yashiro, Y. Kurono, Bull. Chem. Soc. Jpn. 1990, 63, 3013-
3015; c) A. Grozavu, H. B. Hepburn, P. J. Smith, H. K. Potukuchi, P. J.
Lindsay-Scott, T. J. Donohoe, Nat. Chem. 2019, 11, 242-247; d) S.-G.
Wang, W. Zhang, S.-L. You, Org. Lett. 2013, 15, 1488-1491; e) S.-G.
Wang, S.-L. You, Angew. Chem. Int. Ed. 2014, 53, 2194-2197. For
review of dearomative silylation and borylation reactions of heteroarenes
see: f) S. Park, S. Chang, Angew. Chem. Int. Ed. 2017, 56, 7720-7738,
and references therein.
[20] T. Rossolini, B. Ferko, D. J. Dixon, Org. Lett. 2019, 21, 6668-6673.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914390
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents (Please choose one layout)
Layout 1:
RESEARCH ARTICLE
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here))
Layout 2:
RESEARCH ARTICLE
J. A. Leitch, T. Rogova, F. Duarte,* D. J.
Dixon*
1-9
Dearomative Photocatalytic
Construction of Bridged 1,3-
Diazepanes
Building bridges: The reductive photocatalytic dearomatization of quinoline
derivatives using N-arylimines is described. The interesting bridged 1,3-diazepane
architecture is constructed through the addition of an intermediary -amino radical to
the C4-position of a 4-substituted quinoline and subsequent reductive cyclization.
This article is protected by copyright. All rights reserved.