New reductive rearrangement of: N-arylindoles triggered by the grubbs-stoltz reagent et3sih/kotbu

Article abstract of DOI:10.1039/d0sc00361a

N-Arylindoles are transformed into dihydroacridines in a new type of rearrangement, through heating with triethylsilane and potassium tert-butoxide. Studies indicate that the pathway involves (i) the formation of indole radical anions followed by fragmentation of the indole C2-N bond, and (ii) a ring-closing reaction that follows a potassium-ion dependent hydrogen atom transfer step. Unexpected behaviors of 'radical-trap' substrates prove very helpful in framing the proposed mechanism.

Full text of DOI:10.1039/d0sc00361a

View Article Online
View Journal
Chemical
Science
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: J. A. Parkinson, A.
J. Smith, D. Dimitrova, J. N. Arokianathar, K. Kolodziejczak, A. Young, M. Allison, D. Poole, S. G. Leach, T.
Tuttle and J. Murphy, Chem. Sci., 2020, DOI: 10.1039/D0SC00361A.
Volume
Number
9
1
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
7
January 2018
Pages 1-268
Chemical
Science
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
rsc.li/chemical-science
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2041-6539
EDGE ARTICLE
Xinjing Tang et al.
Caged circular siRNAs for photomodulation of gene
expression in cells and mice
rsc.li/chemical-science

Please do not adjust margins
Page 1 of 8
Chemical Science
View Article Online
DOI: 10.1039/D0SC00361A
ARTICLE
New Reductive Rearrangement of N-Arylindoles Triggered by the
Grubbs-Stoltz Reagent Et3SiH/KOtBu
Andrew J. Smith,a Daniela Dimitrova,a Jude N. Arokianathar,a Krystian Kolodziejczak,a Allan Young,a
Mark Allison,a Darren L. Poole,b Stuart G. Leach,b John A. Parkinson,a Tell Tuttle*a and John A.
Murphy*a
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
N-Arylindoles are transformed into dihydroacridines in a new type of rearrangement, through heating with triethylsilane
and potassium tert-butoxide. Studies indicate that the pathway involves (i) the formation of indole radical anions followed
by fragmentation of the indole C2-N bond, and (ii) a ring-closing reaction that follows a potassium-ion dependent
hydrogen atom transfer step. Unexpected behaviors of ‘radical-trap’ substrates prove very helpful in framing the proposed
mechanism.
Introduction,
Et3SiH (5 equiv.)
KOtBu (2 equiv.)
100 °C, 20 h, PhMe
Me
Me
Me
O
1
OH
2
, 78 %
Me
Me
Me
Me
Indoles are an important class of compounds that feature
widely in medicinal chemistry. Accordingly, the synthesis and
reactivity of indoles have both been well explored. The indole
nucleus is electron-rich, and indoles routinely act as aromatic
nucleophiles, principally undergoing substitution by
electrophiles or addition of electrophiles at the 2,3-double-
bond.
In view of their position as electron-rich substrates, reductive
transformations of indoles are very rare,1 but we now report a
new radical-based rearrangement of N-arylindoles that arises
from reductive activation. The reaction is brought about by the
Grubbs-Stoltz reagent,2-9 resulting from heating triethylsilane
with potassium tert-butoxide. Since its discovery in 2013, this
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
Me
S
165 °C, 40 h
mesitylene
3
4
, 83 %
SiEt3
Et3SiH (3 equiv.)
KOtBu (0.2 equiv.)
45 °C, 96 h
N
Me
N
Me
, 78 %
5
6
Ph
Ph
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
130 °C, 18 h
N
Bn
N
H
reagent has been shown to drive
a range of useful
transformations, including cleavage of Ar-O (Scheme 1, 12)2,5
and Ar-S bonds (34),5 regioselective C-silylation of indoles
(56),3,4,6,7 reductive debenzylation of N-benzylindoles (78)8
and reduction of unsaturated hydrocarbons (910).8 A wide
range of mechanisms has been proposed for these
7
9
8
, 80 %
Et3SiH (30 equiv.)
KOtBu (30 equiv.)
130 °C, 18 h
10
, 85 %
SiEt2H
Me
transformations featuring diverse intermediates, including (i)
Et2SiH2 (3.5 equiv.)
KOtBu (0.2 equiv.)
80 °C, 24 h
Ar = p-MeOC6H4
6,8,9
silyl radicals 14,2,5,8,9 (ii) silyl radical anions, e.g. 15
and (iii)
silanate anions, e.g. 16;2,6,7,9 leading to extensive discussion in
the literature. This is illustrated by both open-shell and closed-
shell mechanisms being proposed by the same authors in back-
to-back papers, as possible pathways for converting 56.[6,7]
+
n
MeO
MeO
Ar
12
13
, 0%
, 97 %
11
OtBu
OtBu
K
Et3Si
Et3Si
Et3Si H K
a. Department of Pure and Applied Chemistry, University of Strathclyde, 295
Cathedral Street, Glasgow, G1 1XL, United Kingdom.
14
15
16
b. GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage,
SG1 2NY, United Kingdom.
Scheme 1. Selected transformations with (i) triethylsilane or
Electronic Supplementary Information (ESI) available: Experimental procedures
and spectroscopic data are provided as well as details of computational
investigations. See DOI: 10.1039/x0xx00000x
diethylsilane and (ii) potassium tert-butoxide.
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 2 of 8
ARTICLE
Journal Name
Me
Me
Me
Me
View Article Online
DOI: 10.1039/D0SC00361A
H
K
Et3SiOtBu
15
electron transfer
K
Et3SiH, KOtBu
130 °C, 18 h
N
Ph
N
N
N
N
H
K
NH2
 e
Ph
Ph
38
37
22
17
20
18
19
, 35 %
, 18 %
H
H-atom
abstraction
R
(i) Li, (5 equiv.)
H
R
THF, 30 o
C
R
+
B
(ii) filtration
(iii) PhB(pin)
N
Ar
R = SiEt3 or H
N
Ph
N
N
Ph
Ar
Ph
39
40
21
, 82%
H
H
Scheme 2. (a) Reductive ring-opening of N-allylindole 178 and
N-arylindole 20.1
Me
N
Me
H
K+-ion
dependent
H-atom
transfer
N
N
Most recently, in 2019, a unique role for the closely related
KOtBu + Et2SiH2 (as well as KOtBu + Et3SiH) in hydrosilylation of
styrenes (1112) was proposed by Jeon et al., and supported
with good evidence in a reaction that critically depends on K+-
arene interactions. (In the absence of potassium ions, polymer
13 is formed instead of 12).10
K
K
K
41
42
43
Base
H-atom
transfer
Me
Me
Me
K
workup
electron transfer
N
R
N
R
N
H
 e
R3
R2
46
30
R3
44
45
, R = K
, R = SiEt3
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
130 °C, 18 h
R2
R1
N
Scheme 3.
N
H
38-92 %
R1
paper, N-allyl-3-methylindole 17 (Scheme 2) was converted to 3-
methylindole 19 (35 %); unexpectedly, the reaction also afforded 2-
iso-propylaniline 18 (18 %). Reductive ring-opening reactions of
indoles, as seen in the formation of 18¸ were unknown at the time
and inspired the investigation reported in this paper; however, very
recently, elegant studies by Yorimitsu et al.1 have reductively
opened indoles 20 with excess Li powder, trapping the intermediate
vinyl and amidyl dianions with boron electrophiles to afford
benzazaborin products 21 in high yield.1,12
Table 1. Conversion of indoles to dihydroacridines.*
Entry
Substrate
Product
Yield
%
58
1
2
3
4
5
22, R1=R2=R3=H
30, R1=R2=R3=H
23, R1=R2=H, R3=Me
24, R1=R2=H, R3=Et
25, R1=H, R2=R3=Me
26, R1=R2=H, R3=Ph
31, R1=R2=H, R3=Me
32, R1=H, R2=R3=Me
32, R1=H, R2=R3=Me
33, R1=R2=H, R3=Ph
77
66
53
92
A selection of N-arylindoles 22–29 was prepared (see S.I.) and
tested under the conditions shown in Table 1. Indole heterocyclic
rings were cleaved, as the substrates underwent unexpected and
facile conversion to 9,10-dihydroacridines 30-36 in moderate to
excellent yields. Substituents were tolerated in the 2- and 3-
positions of the indole (entries 1-7), as well as in the N-aryl group
(entry 8). To rationalise the rearrangements, we adopt a working
6
7
8
27, R1=R2=H, R3=C8H17 34, R1=R2=H, R3=C8H17 55
28, R1=R2=H, R3=C4H9 35, R1=R2=H, R3=C4H9
71
29, R1=4-Me, R2=R3=H 36, R1=2-Me, R2=R3=H 38
* Reactions were carried out neat (i.e. without added solvent)
hypothesis for the mechanism, shown in Scheme 3.
A key
Thus, this apparently simple mixture of base and silane
provides great diversity in its reactions. Its applications already
include methods for the desulfurisation of fuels5 and for the
silylation of amines,11 together with proposed applications to
the controlled cleavage of lignins.2,5 To harness its chemistry,
further understanding is clearly needed of the mechanisms of
its reactions. In this paper, we report a new rearrangement
that transforms N-arylindoles to dihydroacridines, and we
provide evidence supporting (i) an initial fragmentation of the
C2-N bond of indole radical anions, followed by (ii) K+ ion-
assisted hydrogen atom transfer chemistry for the conversion
of the ring-opened intermediates to dihydroacridines.
intermediate is radical anion 37. This could form by electron
transfer from donor 15 to the indole or, alternatively, by H-atom
addition (H-atom addition is discussed later in this paper) to the
indole ring system, to form 39 or 40 (R = H) or an isomer, followed
by deprotonation under the basic conditions of the reaction. This
radical anion then fragments to afford the distal radical anion 38.
Abstraction of a hydrogen atom by the vinyl radical (likely donors
would include triethylsilane or silanate 16 or intermediates 39 or
40) affords the styrene/amide anion 41. This styrene now
undergoes regioselective hydrogen atom addition as proposed by
Jeon et al.10 to give benzyl radical anion 42, and instead of forming a
C-Si bond as in Jeon’s example, a C-C bond is formed as cyclisation
gives radical anion 43. This could function directly as an H-atom
donor to form 46 or, alternatively, deprotonation of the
cyclohexadienyl radical then affords the electron-rich species 44 (or,
following silylation, 4511) which behaves as an electron donor.
Our recent report[8] used Et3SiH/KOtBu to transform N-
benzylindoles e.g. 7 to indoles (8), and we proposed that
electron transfer from radical anion 15 to the substrate could
induce the fragmentation of the N-benzyl bond. In the same
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 8
Chemical Science
Please do not adjust margins
Journal Name
ARTICLE
a)
Computational studies performed on the 3-metVhieywl AartnicaleloOgnluinee
23 are consistent with this (Figure 1b). We rationalise that the
greater delocalisation associated with the bicyclic -system of
the indole results in this preferential localisation of the excess
electron density of the radical anion.
We investigated further substrates by computation and
experiment, starting with the N-naphthyl substrate 47. In this
case, the naphthyl -system is the preferred site for the radical
anion (Figure 1c). In the laboratory, that should result in
cleavage of the N-naphthyl bond to form an indolyl anion and
a naphthyl radical, rather than in cleavage of the indole
nucleus (Figure 1a, Case B). Accordingly, the -naphthyl
substrate 47 (Figure 1 d) was prepared. Under the standard
reaction conditions, the predicted cleavage occurred to form
3-methylindole 19 (55 %), and none of the corresponding
dihydroacridine was detected.
DOI: 10.1039/D0SC00361A
Case A
R
Me
R
R


N
N
N
H
Case B
R
R
R
N
N
N
+
+
H


b)


Two additional substrates, 48 and 49, were prepared,
featuring pyridine rings. Computation (see S.I.) suggested that,
due to the electron-deficient nature of these rings, the radical
anions of these substrates would house the added electron in
the pyridine ring, leading again to [indole N]pyridine bond
cleavage, rather than cleavage of the indole nucleus. This was
indeed the case, with indole being isolated in 48 % and 39 %
from 48 and 49 respectively following workup, and no
evidence of cleavage of the indole nucleus. Thus, the cleavages
of substrates 22-29, and 47-49 follow the regioselectivity that
is predicted from a study of the respective radical anions,
providing strong evidence that radical anions are
intermediates in these reactions.
Me
N
23ra
c)
Me
N
Table 2. Mechanistic Probes
Me
conditions
47ra
130 oC, 18 h
N
N
H
30
Ph
22
d)
Me
Me
Entry
Reagents
(3 equiv. of each)
Yield 30
N
N
N
N
(%)
58
33
0
0
0
1
2
3
4
5
Et3SiH, KOtBu
KDTBB, Me6Si2, KOtBu
KDTBB, KOtBu
Me6Si2, KOtBu
KOtBu
N
N
49
23
48
47
Figure 1. (a) Case A, when the unpaired electron in radical anion of N-
arylindoles localises on the indole group in the radical anion of 23 (i.e.
23ra), then fragmentation of the indole nucleus results, while with the
excess electron is housed in the N-aryl group in 47ra, then cleavage of
the indole N-aryl bond results (Case B); (b) radical anion 23ra showing
localisation of spin in the indole nucleus and (c) radical anion 47ra,
showing localisation of spin in the naphthyl group; (d) substrates
4749 undergo indole N-aryl bond cleavage. (for computational
methods, see S.I.)
To investigate the proposed mechanism involving SET, several other
reaction conditions were investigated (Table 2) with substrate 22.
The standard conditions for Et3SiH+KOtBu afforded 30 in 58 % yield
(entry 1). With a view to forming the radical anion of Me3SiOtBu,
analogous to 15, without using triethylsilane or any other
compound that contains a Si-H group, hexamethyldisilane was
treated with the radical anion, potassium di-tert-butylbiphenylide
(KDTBB) as reducing agent and KOtBu as base. This system afforded
product 30 (33%, entry 2). This successful outcome would result if
the disilane were reduced to a silyl radical and a silyl anion by
KDTBB. Addition of a silyl radical to N-phenylindole7 could give 39,
40 (R = SiEt3, Scheme 3) or an isomeric radical. Both 39 and 40 have
Considering the behavior of radical anion 37, fragmentation of
the 5-membered ring could depend on the excess electron of
the radical anion being localised extensively in the indole ring
system, rather than in the N-phenyl group (Figure 1a, Case A).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 4 of 8
ARTICLE
Journal Name
a labile hydrogen atom (circled) that can be removed either as a must be severed during the reaction. Substrate 51 VaielwsoArtaicflfeoOrdnleinde
proton, e.g. by trimethylsilyl anion, to form Me3SiH, or as a dihydroacridine 30 as well as the diastereoDmOeIr:i1c0.s1p0i3r9o/iDnd0SoCli0n0e3s651A2
hydrogen atom, e.g. by trimethylsilyl radical to again form Me3SiH, and 53. The loss of the side-chains, seen in the formation of 30, is
putting in place all the components for Scheme 3.
clearly associated with the alkene group in the side-chain, since
Table 1 shows that substrates with saturated side-chains, 27 and
28, suffered no loss of their saturated alkyl side-chains.
In the presence of KTDBB, but in the absence of any silane or
disilane, no product 30 was formed (entry 3). When KTDBB was
omitted (entry 4), no cleavage of the Si-Si bond could occur and so
the reaction gave no rearranged product. Entry 5 shows that KOtBu
alone could not bring about the rearrangement. From these results,
it was clear that (i) an appropriate silyl compound, (ii) KOtBu and (iii)
reducing power were all needed to bring about the rearrangement
efficiently.
HOtBu
KOtBu
KOtBu
HOtBu
N
N
N
Ph
Ph
Ph
50
54
55
Me
R
Me
Me
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
air
N
Ph
N
Ph
N
H
14 days
N
130 oC, 18 h
N
N
H
30
57
56
30
58
, R = K
22
, R = H
Ph
30ox
50
)
50
, 48 % (from
Me
Me
Me
Me
Me
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
+
+
+
130 oC, 18 h
N
Ph
, 2 %
N
N
Ph
N
H
N
Ph
N
N
Ph
N
Ph
Ph
53
, 12 %
Ph
51
30
52
51
52
53
59
air
14 days
Me
Me
N
N
H
30
N
Ph
60
30ox
, 12 %
51
(from
)
Scheme 4
Scheme 5
The requirement for all three components above was also shown
when potassium metal was tested as a reducing agent in the
presence of KOtBu (see S.I.). This led to consumption of the indole
substrates. Acridine products, rather than dihydroacridines, were
produced, reinforcing that reductive activation1 of the indole
nucleus leads to fragmentation of the indole 5-membered ring, but
the absence of H-atom sources meant that the reaction of Scheme
3 was not supported; the acridine products were formed in very low
yield (5%, see SI file). However, from this result and from the results
of Yorimitsu,1 it is clear that reductive activation of the indole
nucleus of N-arylindoles does lead to fragmentation of the 5-
membered ring.13
Scheme 5 proposes an explanation for both substrates 50 and 51.
Substrate 50 can suffer reversible deprotonation by tert-butoxide in
the allylic position to afford anion 54. Reprotonation of the anion
can afford 55 which is converted to radical anion 56. (We draw the
formation of the radical anion occurring following the
regioisomerisation of the alkene, but the opposite sequence may
well occur). Radical anion 56 then fragments to the 2-butenyl
radical 57, leaving indolylpotassium 58; following proton
abstraction, this gives N-phenylindole 22, which is then converted,
as in Scheme 3, into 9-methyldihydroacridine 30. (For NMR
evidence in favor of the alkene migration in the side-chain of 50,
see S.I. file).
We now looked for evidence of reactive intermediates on the
reaction pathway shown in Scheme 3, using substrates that
incorporated radical clocks. These substrates provided most
surprising outcomes and important mechanistic information.
Substrates 50, 51, (Scheme 4), 61 (Scheme 6) and 73 (Scheme 7)
were prepared and reacted under the standard conditions; the
products were then analyzed to provide evidence of reaction
intermediates.
The outcome of reaction with substrate 51 can be explained in like
manner. In this case, conversion to its radical anion 59 leads to
cyclisation to diastereomeric radicals that abstract hydrogen atoms
to afford 52 and 53. On the other hand, if KOtBu-induced
regioisomerisation migrates the alkene double-bond sequentially
along the side-chain, the substrate then is converted to radical
anion 60. This species can expel a delocalised pent-2-enyl radical,
again forming indolylpotassium 58, which, following proton
abstraction, is ultimately transformed to 9-methyldihydroacridine
30.
Substrate 50 gave an unexpected result when it afforded
dihydroacridine 30 in good yield, confirmed by NMR. For ease of
isolation, the product was exposed to oxidation in air, affording 300x
in 48% yield from 50. For this to happen, the butenyl side-chain
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 8
Chemical Science
Please do not adjust margins
Journal Name
ARTICLE
type A15b
Starting with substrate 61, reaction with KOtBu and EVt3ieSwiHArtaicflfeoOrdnleinde
a mixture of cyclopropyl product 62 (12 %)DaOnId: 10n.-1p0r3o9p/Dyl0SaCn0a0lo3g6u1Ae
63 (30 %). Following formation of radical anion 64,8 fragmentation
of the cyclopropyl ring would afford radical anion 65. As radical
anion 64 incorporates a benzylic cyclopropylcarbinyl radical, this
type B reaction will likely be reversible and thus both 64 and 65 can
be considered as likely intermediates. For 65, abstraction of (i) a
hydrogen e.g. from Et3SiH and (ii) a proton leads to 3-propyl-N-
phenylindole 67. This compound can then be processed, according
to Scheme 3, to afford dihydroacridine 63 following workup.
(a)
type B14,15
type C16
Ar
C
Ar
(b)
Me
Me nPr
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
130 oC, 18 h
N
N
N
H
H
Ph
[a]
[a]
61
62
63
, 30 %
, 12 %
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
130 oC, 18 h
e
N
N
Pr
Ph
Ph
78
73
N
Ph
N
N
N
N
79
Ph
Ph
n-Bu
Ph
Ph
65
66
64
67
Pr
N
N
Ph
N
R
Ph
75
81
80
N
Ph
N
N
air
69
64
68 Ph
Ph
Pr
O
Pr
O
O
n-Bu
Me
Me
Me
air
H
N
N
N
N
R
77
N
82
83
N
H
72
70
71a
71b
, R = K
, R = SiEt3
Scheme 8
Scheme 6
On the other hand, fragmentation of the 5-membered ring in 64
affords radical anion 68. Hydrogen atom abstraction leads to 69.
Regioselective hydrogen atom transfer10 to styrene 69 gives
benzylic radical 70. This is a cyclopropylbenzyl radical (type B) and
so can open reversibly. Cyclisation affords radical anion 71a.
Aromaticity is restored as 72 is formed; this could proceed with the
amide anion still in place, as in 71a, or following silylation of the
nitrogen as in 71b, as mentioned in Scheme 3).11
Studies were next carried out with substrate 61 and 73 bearing a
cyclopropyl substituent as a radical probe6,10 (Scheme 6).
Cyclopropyl groups are routinely used as indicators of radical
intermediates. Simple cyclopropylcarbinyl radicals (type A, Scheme
6a) ring-open very rapidly to form butenyl radicals.14,15 The rate of
the reverse reaction is comparatively very slow, so that these
reactions routinely afford ring-opened products as the sole
products from reactions. However, when the initial radical carbon is
benzylic (type B)14,15 or is a vinyl radical (type C),16 then the
cyclopropyl ring-opening is readily reversible and cyclic and/or
open-chain products can result. As seen below, these subtleties of
cyclopropanes as probes of radicals can be important in
understanding mechanisms.
The 2-cyclopropyl substrate 73 (Scheme 7) was transformed into
products from which 77 (trace amount) was separated and
characterised. Exposure of the residue to air allowed isolation of
acridine 76 (18 %).
The formation of 77 follows pathways that are familiar from other
examples in this paper, starting with the fragmentation of the 3-
membered ring in radical anion 78; (Scheme 8) the oxidation of 82
by air likely involves abstraction of a benzylic hydrogen atom, with
the peroxyl radical 83 as key intermediate on the pathway to
ketone 77. The formation of compound 76 (Scheme 7) was also
informative. Fragmentation of the 5-membered ring in 78 (see
nBu
Et3SiH (3 equiv.)
KOtBu (3 equiv.)
130 oC, 18 h
N
N
H
N
H
Ph
73
74
75
Scheme 9) leads to radical anion 84 which is
a type C
O2
cyclopropylvinyl radical, likely to be in equilibrium with its ring-
opened form 85. No product derived from 85 was detected, but 84
did lead to observed product.
nPr
O
O
Hydrogen abstraction by radical 84 leads to vinyl cyclopropane 86.
Following Jeon,10 regioselective hydrogen atom addition gives
radical 87¸which then proceeds to the cyclopropylmethyl-
N
N
76
77
, 18 %
, trace
Scheme 7
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 6 of 8
ARTICLE
Journal Name
R'
R'
View Article Online
DOI: 10.1039/D0SC00361A
Et3SiH (3 equiv.)
KOtBu (4 equiv.)
(a)
C
130 oC, 18 h
N
H
N
R
N
N
NPh
Ph
30
95, R = K or SiEt3,
Ph
, R' = CH3, 36 %
85
78
84
N
H
92
H
R'
H
N
R
87
N
R
N
H
N
R
93
N
R
86
Et3SiD (3 equiv.)
KOtBu (4 equiv.)
130 oC, 18 h
74
, R=K or H or SiEt3
94
air
R = K or SiEt3
R = K or SiEt3
CH2D
R'
D
CH3
+
+
N
N
N
N
H
30'
N
88
N
H
30
89
90
H
O2
30"
O
O
O
Scheme 10
N
91
spectrum showed two resonances for the methyl carbon, one a
1:1:1 triplet, due to a CH2D group in 30’, and the other a singlet,
due to an undeuterated methyl group in 30.
N
76
(b)
Additionally, we detected 30”. No signals were seen corresponding
to more than one deuterium atom in the methyl group. Deuteration
was also seen in the aryl groups. This process will have exchanged
original Ar-H hydrogen atoms for D atoms (and correspondingly
produced SiH bonds in place of SiD bonds, and these H atoms will
have led to some unlabeled methyl groups seen in the product.
N
Ph
N
N
84'
B
A
To check that this conversion of 92 to 30 was dependent on
potassium ions,10 the reaction was repeated with Et3SiH + NaOtBu;
this gave no reaction.
X
N
H
N
C
74
We also subjected 3-methyl-N-phenylindole 23 to reaction with
Et3SiD/KOtBu, focusing on deuterium incorporation into the methyl
group of 31 (Scheme 11). In this case, we observed CH3, CH2D and
CHD2 groups from the 13C{1H} and 13C{1H,2H} NMR spectra. This is in
line with the proposed mechanism. The formation of some
unlabelled methyl group and some monodeuterated CH2D group
supports the exchange reactions mentioned above, where Ar-H
hydrogen atoms on the substrate undergo exchange with R3Si-D
bonds, and the resulting R3Si-H allows the formation of some
unlabeled and monodeuterated methyl group in the
dihydroacridine product. Consistent with the exchange reactions,
deuteration of Ar-H positions was indeed seen in these reactions.
Scheme 9
substituted dihydroacridine 74. On deliberate exposure to air,
aromatisation slowly occurs to yield 88, followed by activation of
the benzylic C-H, ultimately leading to ketone 76.
Substrate 73 played an important role in our understanding of
mechanism. An alternative route to the dihydroacridine products
was initially considered [Scheme 9(b)] where fragmentation of the
indole radical anion 78 would afford distal radical anion 84’. Proton
abstraction, giving A and radical cyclisation would yield radical B.
However, this cyclopropylcarbinyl radical of type A would open to
give radical C, and no cyclopropyl product would be detected from
this intermediate. Accordingly, H-atom addition to the styrene 86 is
the favored route.
Me
Me
R
Et3SiD, (3 equiv.)
KOtBu, (3 equiv.)
130 °C, 18 h
N
N
D
D
H
Ph
104 mg
Scheme 3 proposed styrene 41 as a possible intermediate in the
rearrangement reaction. To check its viability, substrate 92 was
prepared (Scheme 10). On treatment with triethylsilane and KOtBu,
this afforded 9-methyldihydroacridine 30 (36 %). The upper part of
Scheme 10 shows the proposed pathway. Hydrogen atom addition
would form radical 93; notably, just a single H atom in 30 is
proposed to derive from Et3SiH in this mechanism, and to be
located in the methyl group of the product. To check this, the
63 mg of
23
31,
R = CH3
isotopomeric
mixture
31'
31"
, R = CH2D
, R = CHD2
Me Me
Me Me
Et3SiD, (3 equiv.)
KOtBu, (3 equiv.)
130 °C, 18 h
N
H
N
H
D
D
31
31
Scheme 11
2
reaction was repeated using Et3SiD instead of Et3SiH. The H-NMR
spectrum clearly showed labelling in the methyl group. The 13C {1H}-
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 8
Chemical Science
Please do not adjust margins
Journal Name
ARTICLE
In a further experiment, unlabeled product 31 was subjected to the
reaction conditions, to check whether the methyl groups were
stable to the conditions or would undergo exchange. No exchange
was seen in the methyl groups, although deuteration was seen on
the aryl rings,
Notes and references
View Article Online
DOI: 10.1039/D0SC00361A
1
(a) S. Tsuchiya, H. Saito, K. Nogi and H. Yorimitsu, Org. Lett.
2019, 21, 38553860. (b) H. Saito and H.Yorimitsu, H. Chem.
Lett. 2019, doi:10.1246/cl.190393
2
3
4
5
A. Fedorov, A. A. Toutov, N. A. Swisher and R. H. Grubbs,
Chem. Sci. 2013, 4, 1640–1645.
A. A. Toutov, W.-B. Liu, K. N. Betz, A. Fedorov, B. M. Stoltz,
and R. H. Grubbs, Nature 2015, 518, 80–84.
. A. Toutov, W.-B. Liu, K. N. Betz, B. M. Stoltz and R. H.
Grubbs Nat. Protoc., 2016, 10, 1897–1903.
A. A. Toutov, M. Salata, A. Fedorov, Y.-F. Yang, Y. Liang, R.
Cariou, K. N. Betz, E. P. A. Couzijn, J. W. Shabaker, K. N. Houk,
Additionally, and mindful of reports that H2 gas is produced on
reaction of Et3SiH with KOtBu,2-7 those two reagents were heated
for 1 hour, and then the headspace in the vessel, containing
hydrogen, was replaced by deuterium gas and substrate 23 was
introduced. This led to formation of the dihydroacridine 31.
Examination of the 2H NMR spectrum showed deuterium
incorporation into the methyl group of the product [and a much
lesser degree of deuterium labelling of the Ar-H positions ortho to
the NH group]. The methyl group carbon appeared as a singlet plus
a 1:1:1 triplet, resulting from the presence of CH3 and CH2D (the
combined yield of the products was 72 %). Therefore, the gas in the
headspace is not inert to the reaction. We further examined what
happened when the hydrogen gas was removed and replaced by
inert gas (argon). In this case, the transformation from indole
substrate to dihydroacridine products was completely suppressed
to < 1 % yield.
and
R.
H.
Grubbs,
Nat.
Energy
2017,
doi.org/10.1038/nenergy.2017.8
6
7
W.-B. Liu, D. P. Schuman, Y.-F. Yang, A. A. Toutov, Y. Liang, H.
F. T. Klare, N. Nesnas, M. Oestreich, D. G. Blackmond, S. C.
Virgil, S. Banerjee, R. N. Zare, R. H. Grubbs, K. N. Houk and B.
M. Stoltz, J. Am. Chem. Soc. 2017, 139, 6867–6879.
S. Banerjee, Y.-F. Yang, I. D. Jenkins, Y. Liang, A. A. Toutov,
W.-B.; Liu, D. P. Schuman, R. H. Grubbs, B. M. Stoltz, E. H.
Krenske, K. N. Houk and R N.; Zare, J. Am. Chem. Soc. 2017,
139, 6880–6887.
8
9
A. J. Smith, A. Young, S. Rohrbach, E. F. O’Connor, M. Allison,
H.-S. Wang, D. L. Poole, T. Tuttle and J. A. Murphy, Angew.
Chem. Int. Ed. 2017, 56, 13747–13751.
W. Xie, S.-W. Park, H. Jung, D. Kim, M.-H. Baik and S.Chang, J.
Am. Chem. Soc., 2018, 140, 9659–9668.
As a final point, we investigated whether the reaction was specific
to triethylsilane and to potassium tert-butoxide. Other silanes,
notably Me2PhSiH, MePh2SiH and Ph3SiH were also successful in the
conversion of 23 to 31. However, upon replacement of KOtBu with
NaOtBu (Table S1, entry 5), no reaction took place.
10 P. Asgari, Y. Hua, C. Thiamsiri, W. Prasitwatcharakorn, A.
Karedath, X. Chen, S. Sardar, K. Yum, G. Leem, B. S. Pierce, K.
Nam, J. Gao and J. Jeon, Nature Catalysis, 2019, 2, 164–173.
11 (a) A. A. Toutov, K. N. Betz, A. M. Romine, R. H. Grubbs, US
Patent Application US 2019/0218232A1. (b) F. Palumbo, S.
Rohrbach, T. Tuttle and J. A. Murphy, Helv. Chim. Acta, 2019,
in press DOI:10.1002/hlca.20190023.
12 In a different approach, that also adds electron density to
the indole, but not involving radicals or radical anions,
nucleophilic ring-opening of N-arylindoles by silyl anions,
R3SiLi, was reported: P. Xu, E.-U. Würthwein, C. G. Daniliuc
and A. Studer, Angew. Chem. Int. Ed., 2017, 56, 13872–
13875.
13 We did consider the possibility of the indole fragmentation
in Scheme 3 arising, instead, from fragmentation of radical
40 rather than a radical anion 37. Radical 40 could arise by
hydrogen atom addition to the 2-position of the indole 22.
However, the literature features no reports of
fragmentations of the indole nucleus arising from radicals
like 40, in the presence of the Grubbs-Stoltz reagent or
under other conditions.
Conclusions
In summary, the Stoltz-Grubbs reducing system transforms N-
arylindoles into 9,10-dihydroacridines in moderate to excellent
yields. The ring-opening of the indole is triggered by fragmentation
of intermediate indole radical anions; the styrenes formed in this
fragmentation are then activated by potassium-ion-dependent
hydrogen atom transfer to afford the dihydroacridine products. The
transformation provides important information about the nature of
chemistry that is undertaken by the Et3SiH+KOtBu reagent.
Conflicts of interest
There are no conflicts to declare.
14 V. W. Bowry, J. Lusztyk, and K. U. Ingold, J. Chem. Soc. Chem.
Commun., 1990, 923–925.
15 (a) T. A. Halgren, J. D. Roberts, J. H. Horner, F. N. Martinez, C.
Tronche and M. Newcomb, J. Am. Chem. Soc. 2000, 122,
2988–2994. (b) M. Newcomb, Vol.1, Ch. 5 in Encyclopedia of
Radicals in Chemistry, Biology and Materials Eds: A. Studer
and C. Chatgilialoglu, Wiley, 2012.
16 J. K. Crandall, G. L. Tindell and A. Manmade, Tetrahedron
Lett. 1982, 23, 3769–3772.
Acknowledgements
We thank the following sources for financial support: (i) GSK
and EPSRC for i–CASE awards (to AJS and DD) (ii) University of
Strathclyde Centre for studentship funding (ii) the EPSRC-
funded
ARCHIE-WeSt
High
Performance
Computer
(www.archie-west.ac.uk) for computational resource via EPSRC
grant no. EP/K000586/1
.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Chemical Science
Page 8 of 8
View Article Online
DOI: 10.1039/D0SC00361A
A New Rearrangement of N-Arylindoles Triggered by the Grubbs-Stoltz Reagent Et3SiH/KOtBu
A. J. Smith, D. Dimitrova, J. N. Arokianathar, K. Kolodziejczak, A. Young, M. Allison, D. L. Poole, S.G.
Leach, J. A. Parkinson, T. Tuttle and J. A. Murphy
TOC graphic
R3
R2
R3
Et3SiH, KOtBu
130 °C, 18 h
R2
R1
N
N
H
38-92 %
R1