N-Aryl Groups Are Ubiquitous in Cross-Dehydrogenative Couplings Because They Stabilize Reactive Intermediates

Article abstract of DOI:10.1002/chem.201700430

The mechanism of cross-dehydrogenative coupling (CDC) reactions has been examined by experimental and computational methods. We provide a rationale for the ubiquity of the N-aryl group in these reactions. The aryl substituent stabilizes two intermediates

Full text of DOI:10.1002/chem.201700430

A Journal of
Accepted Article
Title: N-Aryl Groups are Ubiquitous in Cross Dehydrogenative
Couplings Because They Stabilize Reactive Intermediates
Authors: Althea S.-K. Tsang, A. Stephen K. Hashmi, Peter Comba,
Marion Kerscher, Bun Chan, and Matthew Todd
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: Chem. Eur. J. 10.1002/chem.201700430
Link to VoR: http://dx.doi.org/10.1002/chem.201700430
Supported by

10.1002/chem.201700430
Chemistry - A European Journal
N-Aryl Groups are Ubiquitous in Cross Dehydrogenative Couplings Because
They Stabilize Reactive Intermediates
Althea S.-K. Tsang,a A. Stephen K. Hashmi,b,c Peter Comba,d Marion Kerscher,d
Bun Chan*†a and Matthew H. Todd*a
[a] A. S.-K. Tsang, B. Chan, M. H. Todd, School of Chemistry, The University of
Sydney, NSW 2006, Australia.
E-mail: bun.chan@nagasaki-u.ac.jp; matthew.todd@sydney.edu.au
[b] A. S. K. Hashmi, Institute of Organic Chemistry, Im Neuenheimer Feld 270,
Heidelberg University, 69120 Heidelberg, Germany.
[c] A. S. K. Hashmi, Chemistry Department, Faculty of Science, King Abdulaziz
University (KAU), Jeddah 21589, Saudi Arabia
[d] Institute of Inorganic Chemistry, Im Neuenheimer Feld 270, Heidelberg
University, 69120 Heidelberg, Germany.
† Current address: Graduate School of Engineering, Nagasaki University,
Bunkyo 1-14, Nagasaki 852-8521, Japan
Abstract: The mechanism of cross-dehydrogenative coupling (CDC) reactions
has been examined by experimental and computational means. We have
provided a rationale for the ubiquity of the N-aryl group in these reactions. The
aryl substituent stabilizes two intermediates and the high-energy transition
state that connects them, which together represent the rate-determining step.
This knowledge has enabled us to predict new CDC substrates that react
either well or poorly.
Keywords: amines, C,C coupling, mechanistic studies, radicals
Introduction
There has been considerable interest in methodology for cross-dehydrogenative
coupling (CDC) reactions, given the operational simplicity of the transformations and
the high product yields often obtained (Figure 1).[1] These reactions are typically one-
pot procedures, where an oxidant is combined with an incoming nucleophile and an
unfunctionalised substrate. The reaction products are generally formed with high
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
regioselectivity and uncontrolled further reaction at the site of substitution is rarely
observed. The use of phenyl-substituted tetrahydroisoquinoline 1a has become
ubiquitous in the field, with the majority of methodologies describing transformations
for this motif, yet little explanation exists as to why this substrate performs so well, in
part because there are few detailed mechanistic studies of this reaction class.[2]
While conditions vary greatly, from the use of transition metals[3] as catalysts to
metal-free[4] or photoredox[5] conditions, the substrates subjected to CDC reactions
are generally similar, with variation predominantly on the aromatic rings. Elaboration
of the CDC products has been demonstrated, with the methodology incorporated into
the synthesis of an anthelminthic drug.[6] Through the use of synthetic and
computational techniques, in this work we provide an explanation for the successful
DDQ-mediated transformation of phenyltetrahydroisoquinolines (e.g. 1a), the
alternative reactivity of acyl-substituted tetrahydroisoquinolines and offer predictions
of reactivity for other substrates.
Figure 1. Schematic representation of a cross-dehydrogenative coupling reaction of
phenyltetrahydroisoquinoline 1a.
Results and Discussion
The DDQ-mediated CDC reaction of 1a[7] has previously been shown to proceed
through the corresponding iminium ion intermediate 3 (Figure 2; nucleophile is
nitromethane to give 2a),[8] with the same intermediate identified when CuCl2.H2O
was used.[2c] The isolation of 3 as a complex with the DDQH− byproduct confirmed
the product of oxidation resulting from an overall hydride step, but gave little insight
into the preceding steps and intermediates. The reaction between 1a and DDQ was
1
monitored by H NMR spectroscopy (Figure 3). The appearance of a signal at ~ 9
ppm, attributed to the iminium hydrogen of 3, was almost instantaneous. While the
broad nature of the signals suggested the presence of radical intermediates, the
spectra obtained did not give any further information towards the identity of these
species. It was also observed that the reaction proceeded too quickly to be monitored
meaningfully by NMR spectroscopy, with little difference in chemical shift or signal
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
intensity observed. The addition of a stoichiometric amount of iPr2EtN led to the
disappearance of the downfield signals representing the iminium ion intermediate 3
and appearance of the signals representing the nitromethyl product 2a.
Figure 2. DDQ-mediated CDC reaction of phenyl-substituted tetrahydroisoquinolines
1 and iminium intermediate 3 derived from 1a.
Figure 3. 1H NMR spectra of a) starting material 1a; b) the reaction of 1a and DDQ in
CD3NO2 after 5 minutes with solvent suppression; c) after addition of iPr2EtN; d)
nitromethyl product 2a.
The reaction of 1a with DDQ was monitored by UV-vis spectroscopy (Figure 4a), with
three local maxima observed immediately, corresponding to the DDQ radical anion
DDQ•−.[9] A gradual decrease in the absorbance was observed over a period of hours.
Mechanistically, this suggested the possibility of a single electron transfer (SET) step
as part of the reaction pathway. Given the speed of the reaction, with an
instantaneous colour change observed upon addition of the oxidant, monitoring
under stopped-flow conditions allowed for spectra to be recorded at intervals of
approximately 0.1 seconds (see Supplementary Information, Fig S1). Indeed,
maximum absorbance was recorded after only 30 seconds. When the number of
equivalents of DDQ was changed, the maximum absorbance varied accordingly
(Figure 4b), which suggested that an equilibrium between the substrate/substrate
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
radical cation and DDQ/DDQ radical anion existed in solution. Although this approach
towards monitoring is indirect, i.e. only the fate of the oxidant is observed, the DDQ-
mediated oxidation of other phenyltetrahydroisoquinolines was monitored in a similar
manner. We note that an initial hydrogen atom transfer (HAT) mechanism has also
been proposed for related processes (see below). However, based on the rapid color
change observed, we believe that if HAT does play a part here, it would only be a
minor contributing factor. While the same local maxima corresponding to the DDQ
radical anion were observed, the rate of the reaction differed according to the
substitution pattern at the para position of the aromatic ring (Figure 4c), with an
electron-donating substituent (R = OMe, 1b) giving a faster increase in absorbance.
The resulting Hammett plot (Figure 4d, see Supplementary Information for
experimental details) showed that a linear free-energy relationship existed for this
step of the reaction, with the negative value of ρ (−2.138) suggesting that an
electron-deficient centre is generated in the rate-limiting step.
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
Figure 4. UV-vis spectra (CH3NO2, 25 °C) of a) 1a and DDQ (0.40 mM each) after 1
minute, with maxima at 456, 517 and 588 nm representing the DDQ radical anion
DDQ•−, b) in reactions of 1a (0.40 mM) and DDQ (0.5 – 3.0 eq.), c) in reactions of
1a–e (R = H, OMe, tBu, Me, F respectively) and DDQ (0.40 mM each) and d)
Hammett plot of data obtained.
To gain further insight into the reaction, the reaction profile was modelled using
computational means (Figure 5, see Supplementary Information for all computational
methods). For the phenyl-substituted amine 1a and other analogues, the energy
difference between the starting reactants (0, phenyltetrahydroisoquinolines and DDQ)
and the proposed radical ions (RI, 4 and DDQ radical anion) is small (< 20 kJ mol-1),
thus permitting an equilibrium to be established. Between the different analogues, the
magnitude of energy difference between the starting materials and radical ions is in
agreement with the difference in initial rate as derived from the UV-vis
measurements. While the ionic intermediates (Int, iminium ion 3 and corresponding
hydroxyquinone) are in a significant thermodynamic well (~ 120 kJ mol-1), the barrier
for hydrogen abstraction via the transition state TS (~ 80 kJ mol-1, see Figure 6d for
the transition state structure of 1a) is considerable and is consistent with the slow
decrease in UV-vis absorbance and hence disappearance of the DDQ radical anion
DDQ•−. The subsequent reaction of the iminium ion intermediate 3 with nitromethane
is slightly endergonic, with the substituted products obtained in good yield. The
combined experimental and computational results suggest that the DDQ radical
anion DDQ•− (and hence the corresponding amine radical cation 4) exists. At this
point the results do not yet show whether this species on the reaction pathway itself
or is a result of a side reaction. The transition state located could be that for either
hydride or hydrogen atom transfer. Stability analysis on the closed-shell transition
state TS wavefunctions gave 〈S2〉 values of ~ 0.7 (versus 0.0 for a pure singlet, 0.75
for a pure doublet and 2.0 for a pure triplet). This suggests that the transition state
has a substantial biradical character derived from the proposed radical ions RI, which
in turn suggests that the single electron transfer step is in fact on the reaction
pathway. The biradical nature of the transition state is consistent with the proposed
mechanism and contrary to a recently published model[10a] for a similar oxidation, in
which hydride transfer was proposed.[10b,c]
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
Figure 5. Profile of the reaction between 1a-f and DDQ, calculated at B3LYP/6-
31G(2df,p), with condensed phase free energies at 298 K in kJ/mol relative to
reactants 1a and DDQ; transition state structure for 1a.
The sequence above that was experimentally observed and computationally
modeled is in agreement with some mechanisms proposed for other CDC
reactions (e.g., an iron-catalyzed, tert-butyl peroxy radical-[11a], O2-[11b,c] or
DDQ-[2c] mediated CDC)[11d] but distinct from mechanisms proposed for others
(e.g., metal-[11e,f] or tBuOO•/tBuO•-[11g] mediated production of C-based radicals
through initial hydrogen atom transfer) which may nevertheless also benefit
from the stabilisation of partially-formed radicals in the relevant transition
states.11h This highlights an important point: reactions grouped together in the
literature under the term “cross dehydrogenative couplings” most likely exhibit
a range of related reaction mechanisms.[2a,12]
Other synthetically more versatile protecting groups for the heteroatom, such
as acyl substituents have (to date) rarely been used in substrates for CDC
reactions.[13] The acyl-substituted
5 was prepared and subjected to the DDQ-
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
mediated CDC reaction with nitromethane (Figure 6a). The desired product 6a
was not obtained. Instead, imide 6b was observed, which may be associated
with the involvement of water or oxygen. Using the Boc-substituted
7, the
major product isolated from the reaction with DDQ was the oxygen-bridged
dimer 8a (Figure 6b). Monitoring the production of the DDQ radical anion in the
reactions of
5 and 7 as starting materials by UV-vis spectroscopy revealed only
weak absorbance (Figure 6c), which suggested that the reaction mechanism
for these acyl-substituted substrates was different to that for the aryl substrates
1
. Modeling of the reaction profile for these substrates showed that the
pathway proposed for the aryl-substituted amines (i.e. single electron
1
transfer as a first step) was higher in energy than a direct oxidation of the
substrate and thus likely to be unfavourable (Figure 6d). The computational
results also showed that the ground-state energy of the final acyl-substituted
product 6a is similar to that of 2a suggesting 6a is not unduly
thermodynamically unstable; indeed 6a has been synthesised and isolated by
us using alternative approaches.[14]
The computational results to this point revealed that an electron-rich
substituent at the heteroatom promoted oxidation via single electron transfer, with the
atom transfer step likely to be rate determining. This suggested that modifications at
the benzylic position would affect the rate (and possibly the outcome) of the reaction.
Accordingly, the 1-methyl substituted phenyltetrahydroisoquinoline 9 was synthesised
and subjected to a reaction with DDQ in nitromethane (Figure 7a). Even under
forcing conditions (reflux for 48 hours), the desired nitromethyl product 10 was not
obtained, with 9 reisolated. Monitoring of the reaction for production of the DDQ
radical anion by UV-vis spectroscopy showed that single electron transfer occurred
similarly to the reaction starting with 1a (Figure 7b). That the starting material was
reisolated from the reaction suggested that the single electron transfer step was
reversible. Modeling of the reaction pathway of 9 by computational means showed
that the barrier to electron transfer was similar to that of 1a, but the barrier to the
transition state for atom transfer was far higher (Figure 7c). The difference in the
structure of the transition states can be seen in Figure 7d. According to the
computational results, the atom abstraction step for 1a was rate-limiting and hence
likely to be sensitive to substitution at the benzylic position. Steric effects in 9 require
that the methyl group occupy an equatorial position (Figure 7d, right), which prevents
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
the coplanar arrangement of the relevant C–H bond and the heteroatom’s lone pair.
Subsequent resonance stabilisation in TS9-DDQ does not occur, thus raising the
energy of this transition state and preventing complete oxidation. By contrast, the
coplanar arrangement adopted between the reacting C–H bond in substrate 1a and
the heteroatom lone pair in the transition state TS1a-DDQ (Figure 7d, left) facilitates
stabilisation.[2a, 15]
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
Figure 6. DDQ-mediated reaction of a) acetyl-substituted 5 and b) Boc-substituted 7
tetrahydroisoquinolines with nitromethane; c) UV-vis absorbance at 588 nm of the
reactions of 1a, 5 and 7 with DDQ in nitromethane, monitored under stopped flow
conditions, d) free energy profiles of the reactions of 1a, 5 and 7 with DDQ in
nitromethane.
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
Figure 7. a) Attempted DDQ-mediated oxidation of 9 in nitromethane; b) UV-vis
absorbance at 588 nm of the reactions of 1a and 9 with DDQ in nitromethane under
stopped flow conditions; c) free energy profiles of the reactions of 1a and 9 with DDQ
in nitromethane; d) transition states for the reactions of 1a (top) and 9 (bottom) with
DDQ.
This substrate
9 presents an example where the use of computations can
rationalise the experimental observations and explain the overall reactivity. As
the presence of an aromatic ring at the heteroatom facilitated the reaction, it
was predicted that naphthyl-substituted tetrahydroisoquinolines (11, Figure 8a)
would also undergo the CDC reaction successfully. Subsequently, the reaction
profiles of the tetrahydroisoquinolines 11a and 11b were calculated (Figure
8d). It was predicted that the radical intermediates arising from the 1’-naphthyl
substituted 11a would be destabilised, when compared to its analogue 11b
and the phenyl-substituted 1a, due to the aplanar nature of the ring systems
(Figure 8c and Supplementary Information, Fig S3). Conversely, the
conformations and calculated energies for the 2’-naphthyl substituted 11b are
comparable to that of 1a. Overall, the energies of the transition state, iminium
ion intermediate and final products are similar for 1a, 11a and 11b, suggesting
that the reactions of 11a and 11b would proceed. Naphthyl-substituted amines
11a and 11b were then synthesised and subjected to a reaction with DDQ in
nitromethane, with the nitromethyl products 12a and 12b obtained in excellent
yield (Figure 8a). The reactions of 11a and 11b were monitored by UV-vis
spectroscopy, with the maxima representing the DDQ radical anion observed
in each case (Figure 8b). When comparing the rate of increase of absorbance,
the aplanar 11a showed a slower increase in absorbance than the analogue
11b. This experimental observation was in agreement with the computational
results, with the slower rate of increase corresponding to the higher energy of
the radical intermediates arising from single electron transfer. Studies into the
DDQ-mediated oxidative addition of nitromethane to naphthyl-substituted
amines 11a and 11b have provided a successful example where
computational predictions were successfully confirmed through experimental
and spectroscopic means.[16]
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
Figure 8. a) Reaction of 11a and 11b with DDQ in nitromethane; b) UV-vis
absorbance at 588 nm of the reactions of 11a and 11b with DDQ in nitromethane,
with 1a for comparisons; c) lowest energy conformations of 11a and 11b; d) Free
energy profile of the reactions
Conclusions
We have used a variety of experimental and computational techniques to
explore the mechanism of the DDQ-mediated CDC reaction of substituted
tetrahydroisoquinolines. Throughout these studies, the experimental results
have been in agreement with those derived computationally, with this
combined approach successful in explaining the reactivity of aryl- and acyl-
substituted substrates and also predicting the reaction outcome of other
tetrahydroisoquinolines. The use of a phenyl substituent at the heteroatom
lowers the energy of the radical intermediates (compared to when an acyl
group is used) and the energy of the transition state for subsequent atom
transfer, thus accounting for the extensive use of phenyltetrahydroisoquinoline
1a as a substrate for CDC reactions. This effect is presumably resonance-
based, explaining the modulating effect of substituents on the N-aryl ring and
the lesser reported ability of alkyl groups to participate in CDC reactions. In all
of the “successful” substrates discussed here, complete regioselectivity is
observed with the incoming nucleophile installed at the benzylic position only.
Overoxidation and further substitution of the products 2a-e
observed, as the presence of substituents at the benzylic position prevent
subsequent oxidation. This was also explored with the methyl-substituted
, 12a and 12b is not
9
,
with computational results showing that the steric environment at the benzylic
position prevented oxidation, thus explaining why overoxidation did not occur.
The mechanistic studies showed that evaluating the stability and the
geometries of the intermediate radicals could provide an initial prediction
regarding the feasibility of new substrates for DDQ-mediated CDC reaction. It
is common practice to suggest plausible mechanisms for CDC reactions based
on extrapolation from literature examples. This may be hazardous if this class
of reactions ultimately exhibits a range of reaction mechanisms. A comparison
between the results obtained here and those obtained by other researchers in
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
the few other investigations involving physical measurements of related CDC
reactions suggests that this may be the case.
Acknowledgements
We thank The University of Sydney and ARC Centre of Excellence for Free
Radical Chemistry and Biotechnology for funding and the DAAD, the Agnes
Campbell Foundation and the Lamberton Foundation for scholarships (to A. S.-
K. T.).
References
[1] a) From C-H to C-C Bonds: Cross-Dehydrogenative–Coupling, Royal Society
of Chemistry, Cambridge, 2015, Li, C.-J. (Ed); b) C. J. Scheuermann, Chem.
Asian J. 2010, 5, 436–451.
[2] a) Mechanisms of Cross-Dehydrogenative-Coupling Reactions, A. S.-K.
Tsang, S. J. Park, M. H. Todd in Ref 1a, Chapter 11, 254–294; b) H. H. Jung, P.
E. Floreancig, Tetrahedron 2009, 65, 10830–10836; c) E. Boess, C. Schmitz, M.
Klussmann, J. Am. Chem. Soc. 2012, 134, 5317–5325; d) E. Boess, D.
Sureshkumar, A. Sud, C. Wirtz, C. Farès, M. Klussmann, J. Am. Chem. Soc.
2011, 133, 8106–8109.
[3] a) Copper-Catalyzed Cross-Dehydrogenative-Coupling Reactions, X. Zheng,
Z. Li, in Ref 1a, Chapter 3, 55–66; b) Iron-Catalyzed Cross-Dehydrogenative-
Coupling Reactions, C. Darcel, J.-B. Sortais, S. Q. Duque, in Ref 1a, Chapter 4,
67–92; (c) Z. Li, C.-J. Li, J. Am. Chem. Soc., 2004, 126, 11810–11811; d) Z. Li,
C.-J. Li, J. Am. Chem. Soc. 2005, 127, 6968–6969; e) Z. Li, C.-J. Li, J. Am.
Chem. Soc. 2005, 127, 3672–3673; f) D. Sureshkumar, A. Sud, M. Klussmann,
Synlett 2009, 1558–1561; g) A. Sud, D. Sureshkumar, M. Klussmann, Chem.
Commun. 2009, 3169–3171; h) C. M. R. Volla, P. Vogel, Org. Lett. 2009, 11,
1701–1704.
[4] a) Cross-Dehydrogenative-Coupling Reactions without Metals, H. Ito, K.
Ueda, K. Itami, in Ref 1a, Chapter 8, 153–196; b) D. Cheng, W. Bao, Adv. Synth.
Catal. 2008, 350, 1263–1266; (c) D. Cheng, W. Bao, J. Org. Chem. 2008, 73,
6881–6883; d) X.-Z. Shu, X.-F. Xia, Y.-F. Yang, K.-G. Ji, X.-Y. Liu, Y.-M. Liang,
J. Org. Chem. 2009, 74, 7464–7469; (e) Y. Li, W. Bao, Adv. Synth. Catal. 2009
351, 865–868.
,
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
[5] a) Light-Assisted Cross-Dehydrogenative-Coupling Reactions, L. Furst, C. R.
J. Stephenson, in Ref 1a, Chapter 10, 219–253; b) A. G. Condie, J. C. González-
Gómez, C. R. J. Stephenson, J. Am. Chem. Soc. 2010, 132, 1464–1465; c) M.
Rueping, S. Zhu, R. M. Koenigs, Chem. Commun. 2011, 47, 8679–8681; d) D. P.
Hari, B. König, Org. Lett. 2011, 13, 3852–3855; e) D. B. Freeman, L. Furst, A. G.
Condie, C. R. J. Stephenson, Org. Lett. 2012, 14, 94–97.
[6] A. S.-K. Tsang, K. Ingram, J. Keiser, D. B. Hibbert, M. H. Todd, Org. Biomol.
Chem. 2013, 11, 4921–4924.
[7] A. S.-K. Tsang, M. H. Todd, Tetrahedron Lett. 2009, 50, 1199–1202.
[8] A. S.-K. Tsang, P. Jensen, J. M. Hook, A. S. K. Hashmi, M. H. Todd, Pure
Appl. Chem. 2011, 83, 655–665.
[9] a) J.-M. Lü, S. V. Rosokha, J. K. Kochi, J. Am. Chem. Soc. 2003, 125, 12161–
12171; b) K. M. Zaman, N. Nishimura, S. Yamamoto, Y. Sueishi, J. Phys. Org.
Chem. 1994, 7, 309–315; c) S. Fukuzumi, K. Ohkubo, Y. Tokuda, T. Suenobu, J.
Am. Chem. Soc. 2000, 122, 4286–4294.
[10] a) V. S. Batista, R. H. Crabtree, S. J. Konezny, O. R. Luca, J. M. Praetorius,
New J. Chem. 2012, 36, 1141-1144; b) C. Höfler, C. Rüchardt, Liebigs Ann./Recl.
1996, 183–188; c) S. Yamabe, S. Yamazaki, S. Sakaki, Int. J. Quant. Chem.
2015, 115, 1533–1542.
[11] a) M. O. Ratnikov, X. Xu, M. P. Doyle, J. Am. Chem. Soc. 2013, 135, 9475–
9479; b) M. Brzozowski, J. A. Forni, G. P. Savage, A. Polyzos, Chem. Commun.
2015, 51, 334–337; c) S. Murata, M. Miura and M. Nomura, J. Chem. Soc.,
Chem. Commun. 1989, 116–118; d) H. Bartling, A. Eisenhofer, B. König, R. M.
Gschwind, J. Am. Chem. Soc. 2016, 138, 11860–11871; e) E. Shirakawa, N.
Uchiyama, T. Hayashi, J. Org. Chem. 2011, 76, 25–34; f) M. Ghobrial, K.
Harhammer, M. D. Mihovilovic, M. Schnürch, Chem. Commun. 2010, 46, 8836–
8838; g) E. Boess, L. M. Wolf, S. Malakar, M. Salamone, M. Bietti, W. Thiel, M.
Klussmann, ACS Catal. 2016, 6, 3253–3261; h) M. O. Ratnikov, M. P. Doyle J.
Am. Chem. Soc. 2013, 135, 1549–1557
[12] a) A. E. Wendlandt, S. S. Stahl, Angew. Chem. Int. Ed. 2015, 54, 14638–
14658; b) F. Wurche, W. Sicking, R. Sustmann, F.-G. Klärner, C. Rüchardt,
Chem. Eur. J. 2004, 2707–2721; c) X. Guo, H. Zipse, H. Mayr, J. Am. Chem.
Soc. 2014, 136, 13863–13873; d) B. Chan, L. Radom, J. Phys. Chem. A 2007
,
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
111, 6456–6467; e) E. Baciocchi, T. D. Giacco, F. Elisei, O. Lanzalunga, J. Am.
Chem. Soc. 1998, 120, 11800–11801.
[13] a) M. Ghobrial, M. Schnürch, M. D. Mihovilovic, J. Org. Chem. 2011, 76,
8781–8793; b) H. Richter, O. García Mancheño, Eur. J. Org. Chem. 2010, 4460–
4467; c) C. Dubs, Y. Hamashima, N. Sasamoto, T. M. Seidel, S. Suzuki, D.
Hashizume, M. Sodeoka, J. Org. Chem. 2008, 73, 5859–5871; d) N. Sasamoto,
C. Dubs, Y. Hamashima, M. Sodeoka, J. Am. Chem. Soc. 2006, 128, 14010–
14011; e) B. Schweitzer-Chaput, M. Klussmann, Eur. J. Org. Chem. 2013, 666–
671.
[14] a) M. Ahamed, T. Thirukkumaran, W. Y. Leung, P. Jensen, J. Schroers, M.
H. Todd, Eur. J. Org. Chem. 2010, 5980–5988; b) N. R. Amarasinghe, P. Turner,
M. H. Todd, Adv. Synth. Catal. 2012, 354, 2954–2958.
[15] a) E. Baciocchi, M. Bietti, O. Lanzalunga, Acc. Chem. Res. 2000, 33, 243–
251; b) S. F. Nelsen, J. T. Ippoliti, J. Am. Chem. Soc. 1986, 108, 4879–4881.
[16] While this work was being written up, an iron-catalysed synthesis of 12a was
reported, where a slower rate of reaction was indeed noted as predicted, see Ref
[11b].
This article is protected by copyright. All rights reserved.

10.1002/chem.201700430
Chemistry - A European Journal
TOC Graphic
This article is protected by copyright. All rights reserved.