Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations

Article abstract of DOI:10.1016/j.tet.2019.130663

Conditions were found for the [1,5]-hydride shift nitro-Mannich reaction that led to the synthesis of 2,3-disubstituted tetrahydroquinolines. Two simple cyclic amine substrates gave diastereomerically pure rearranged products in 65 and 90% yields by refluxing in HFIP. A more general procedure used Gd(OTf)3 as a catalyst and successfully rearranged other cyclic and acyclic amines in 42–84% yield with diastereomeric ratios of 75:25 to >95:5 in favour of the anti-diastereoisomer (9 examples). Two examples of sulphur containing heterocycles gave lower yields of 9 and 25%. Electron withdrawing substituents were shown to have a deleterious effect on the success of the reaction. The results indicated the limitation of the [1,5]-hydride shift nitro-Mannich reaction with respect to the stability of the intermediate iminium ion.

Full text of DOI:10.1016/j.tet.2019.130663

Tetrahedron xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Investigation of the [1,5]-hydride shift as a route to nitro-Mannich
cyclisations
,
*
a
a,
James C. Anderson a , Chia-Hao Chang , Merina K. Corpinot 1, Michael Nunn b,
ꢀ
Oliver J. Ware a
a Department of Chemistry, University College London, 20 Gordon Street, London, WC1H 0AJ, UK
b Early Chemical Development, Pharmaceutical Sciences, R&D, AstraZeneca, Macclesfield, SK10 2NA, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Conditions were found for the [1,5]-hydride shift nitro-Mannich reaction that led to the synthesis of 2,3-
disubstituted tetrahydroquinolines. Two simple cyclic amine substrates gave diastereomerically pure
rearranged products in 65 and 90% yields by refluxing in HFIP. A more general procedure used Gd(OTf)3
as a catalyst and successfully rearranged other cyclic and acyclic amines in 42e84% yield with diaste-
reomeric ratios of 75:25 to >95:5 in favour of the anti-diastereoisomer (9 examples). Two examples of
sulphur containing heterocycles gave lower yields of 9 and 25%. Electron withdrawing substituents were
shown to have a deleterious effect on the success of the reaction. The results indicated the limitation of
the [1,5]-hydride shift nitro-Mannich reaction with respect to the stability of the intermediate iminium
ion.
Received 30 August 2019
Received in revised form
27 September 2019
Accepted 28 September 2019
Available online xxx
Keywords:
[1,5]-hydride shift
Nitro-Mannich
Tetrahydroquinolines
HFIP promoted
Gd(OTf)3 catalysis
© 2019 Elsevier Ltd. All rights reserved.
1. Introduction
During our studies of this reaction Kim reported thermal Lewis acid
catalysed and thiourea catalysed examples of cyclic amines
The functionalisation of Csp3-H bonds adjacent to an amine via
1,5-hydride transfer and subsequent nucleophilic cyclisation rep-
resents a redox-neutral reaction cascade that has proven useful for
the synthesis of structurally diverse amino heterocycles (for
example Scheme 1) [1]. Amongst the examples of hydride acceptors
(Scheme 2c) [5].
We set out to investigate the use of nitro alkenes as hydride
acceptors in these sorts of processes more fully as we recognised it
could simplify our previous synthesis of 2,3-functionalised [6] and
2,3,4-functionalised [7] terahydroquinolines by a nitro-Mannich
type cyclisation (Scheme 3). The nitro-Mannich reaction [8] is
(C ¼ X, Scheme 1), the use of carbonyls, imines, and especially
a,b-
unsaturated carbonyl and malonic acid derivatives are the most
prevalent [2].
useful for the synthesis of stereodefined b-nitroamines that can be
used in heterocycle synthesis [9]. Tetrahydroquinolines are an
important motif in many biologically active natural products and
medicinal compounds [10], often requiring the de novo construc-
tions of the heterocycle. A particularly efficient and versatile
approach has made use of the nitro-Mannich cyclisation of nitro
nucleophiles onto a pendant aniline imine to give stereodefined
nitro-substituted tetrahydroquinolines [6,7,11]. Our previous work
relied upon the formation of an imine by condensation of an aniline
with an aldehyde and then cyclisation with either a nitronic acid
nucleophilic formed by a weak acid or nitronate anion formation
from an intermolecular nucleophilic conjugate addition to a nitro
styrene (Scheme 3a). We report here our studies into the scope and
limitations of the [1,5]-H shift nitro-Mannich reaction for the syn-
thesis of 2,3-disubstiuted tetrahydroquinolines that expands the
use of anilines in this process (Scheme 3b). Our results compliment
The use of the nitro alkene function as a hydride acceptor in
these reactions would enable a nitro-Mannich cyclisation to occur
(Scheme 2a), but reports of this are scarce. An isolated report de-
tailing thermal conditions by Rabong et al. (Scheme 2b) indicated
the probable difficulty of this process in comparison to the use of
a
,b
-unsaturated carbonyl derivatives [3]. The addition of an ester
stabilising group revealed that -nitroacrylate esters were
a
competent substrates for the [1,5]-H shift, but required the use of a
catalytic Lewis acid (Mg(OTf)2 10 mol%) in addition to heat [4].
* Corresponding author.
E-mail address: j.c.anderson@ucl.ac.uk (J.C. Anderson).
Corresponding author for single crystal X-ray structure determination.
1
https://doi.org/10.1016/j.tet.2019.130663
0040-4020/© 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

2
J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
2. Results and discussion
Repeat of the literature conditions to rearrange 1 (nBuOH,
reflux, 80 h) gave only 5% of desired product 2 (Lit 39%, Scheme 2b)
[3]. The anti-stereochemistry was assigned from the 1H NMR with a
key trans diaxial coupling of 9.7 Hz across the nitroamine [3], which
was later supported by the single crystal X-ray determination of a
similar analogue (vide infra). Repeating the reaction in PhMe at
reflux saw only degradation over 80 h. To investigate this reaction
further we studied the effect of hydrogen bond activation on this
process. We speculated that a more acidic solvent might lead to a
higher yielding reaction due to enhanced hydrogen bonding to the
nitro alkene. This would lead to a better hydride acceptor and faster
reaction if we assumed hydride migration was the rate determining
step. Refluxing 1 in CF3CH2OH was complete by tlc in 9.5 h and led
to a 55% isolated yield of 2 (Table 1) as a single diastereoisomer. Use
of the more acidic alcohol heaxafluoroisopropanol (HFIP) led to a
90% yield in 4 h. In retrospect HFIP is also known to stabilise cations
due to its high dielectric constant and may also help lower the
activation energy of the reaction by stabilising the developing
iminium cation after hydride migration (Scheme 2a) [12]. While
preparing this manuscript reports appeared using the unique
properties of HFIP to promote 1,5-hydride transfer from anilines to
a quinone methides. We decided to explore the scope of the reac-
tion with HFIP [13]. The use of HFIP in a stoichiometric amount in
refluxing PhMe or CH2Cl2 led to recovered starting material. The
piperidinyl analogue (E)-1-(2-(2-nitrovinyl)phenyl)piperidine (3)
[2b,3] had only given unquantifiable traces of product in the orig-
inal nBuOH method [3] and in the thiourea catalysed examples by
Kim [5a]. Refluxing in HFIP for a longer time period of 18 h we were
able to isolate the desired cyclised product 4 in 65%, as a single
diastereoisomer. The coupling constant across the nitro amine is
only 6.2 Hz, which is some way off the magnitude expected for a
trans-diaxial coupling if we assumed the substituents occupied a
pseudo equatorial conformation. The anti-stereochemistry is
depicted by Kim [5] for this and related ring fused products, but no
structural proof was published. We have tentatively assigned the
anti-stereochemistry to 4 based on comparison of its 1H NMR data
with other analogues (vide infra). As an aside we did attempt the
use of Schreiner's thio-urea catalyst [14] (5 mol%) for the rear-
rangement of 3 in refluxing toluene for 4 days after no reaction at
room temperature. Analysis by tlc revealed mainly starting mate-
rial, with some other spots which did not correspond to the desired
product 4. This was in line with the observations of Kim who found
that thio-urea catalysis or thermal conditions did not rearrange
substrate 3 [5].
Scheme 1. [1,5]-H shift cyclisation reaction.
Scheme 2. The [1,5]-H shift nitro-Mannich reaction.
A known ethereal substrate 5 [15] was prepared from salicyl
aldehyde [16] and subjected to refluxing conditions in HFIP, but
gave no product. It has been noted in the literature that acyclic
benzyl ethers are harder to isomerise than cyclic ethers [2n]. A
corresponding acyclic amino analogue 6 was prepared by optimi-
sation of the published route to the cyclisation precursor 1 and 3
(Scheme 4) [3]. An SNAr reaction between 2-fluorobenzaldehyde 7
and N-methylbenzylamine gave amino aldehyde 8 in 74% yield
which was subjected to a Henry reaction to give the cyclisation
precursor 6 in 24% yield. Heating in HFIP unfortunately only gave
starting material. With the limited substrate window shown by
HFIP, we decided to explore the original literature rearrangement 1
to 2 (Scheme 2b) with Lewis acids.
Scheme 3. Synthesis of isoquinolines by the nitro-Mannich reaction.
those of Rabong and Kim. We show that piperidines as well as
pyrrolidines can be rearranged in high yields, that the rearrange-
ment is very substrate dependent, but that certain acyclic benzylic
and alkyl amines will rearrange. Our results expands the substrate
scope of the rearrangement, definitively confirm the sense of dia-
stereoselection and give some rationale to guide the use of this
rearrangement in target synthesis.
Lewis acids BF3(OEt)2, TiCl4, Ti(OiPr)4 and Sc(OTf)3 had all shown
to be useful for the corresponding [1,5]-H shift and cyclisation with
ethers and a,b-unsaturated carbonyls. Treatment of 1 with catalytic
and stoichiometric quantities of these in CH2Cl2 at rt and reflux
gave colour changes and precipitates, but starting material was
recovered after aqueous work up. Presumably these particular
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
3
Table 1
Screening of fluorinated alcohols for the [1,5]-H shift nitro-Mannich reaction.
Substrate
Solvent
Temp. (ꢀC)
Time (h)
Product
Yield (%)a
CF3CH2OH
HFIP
78
58
9.5
4
55
90
HFIP
58
18
65
b
HFIP
HFIP
58
58
162
18
-
0
0
b
-
a
Isolated Yield.
Recovered starting material.
b
one solvent was picked from each octant to screen. Each reaction
was heated to 60 ꢀC for 12 h with aliquots taken at t ¼ 0, 0.33, 0.67,
1, 2, 3, 4, 6, 8 and 12 h and monitored by HPLC (Table 2). As
degradation was observed under some conditions the time chosen
for comparison was when the percentage area of the product peak
was at its greatest for each set of conditions. Whilst MeCN provided
the initial result and was the solvent used for a similar rearrange-
ment in the literature [4], it was observed to perform poorly as a
solvent in these conditions with the product peak accounting for
47% of the total peak area. Perhaps surprisingly PhMe appeared to
perform best, with a 94% conversion after 12 h and very little side
product formation or degradation occurring despite the reaction
proceeding more slowly than with MeCN and 2-MeTHF, for
example.
Two other lanthanide triflates were surveyed in this reaction;
Yb(OTf)3 gave 14% yield by HPLC, whereas Ce(OTf)3 gave no con-
version. A control reaction using 5 mol% TfOH also gave no con-
version confirming that the Lewis acid Gd(OTf)3 was responsible for
catalysing this reaction. The optimum reaction conditions were
found to be heating a solution of 1 in PhMe (10.0 mL per mmol) and
Gd(OTf)3 (30 mol%) to 100 ꢀC for 12 h to give 2 in 79% isolated yield.
A small library of nitroalkenes were prepared to investigate the
limitations of this reaction (Fig. 1). The synthesis of the nitroalkenes
used either the SNAr/Henry route from o-fluorobenzaldehyde
starting materials (Scheme 4) or if that failed the amine was
introduced using palladium catalysed amination [20] followed by a
Henry reaction from o-bromobenzaldehyde starting materials. The
substrate tolerance of the two amination reactions and the com-
mercial availability of either the o-fluoro- or o-bromobenzaldehyde
starting materials defined the substrate scope prepared.
Scheme 4. Synthesis of acyclic amino analogue.
Lewis acids coordinate to the amine of 1. We next screened Lewis
acids that had been used to catalyse the conjugate addition of nu-
cleophiles to nitroalkenes. However SnCl2 [17], and Cu(OTf)2 [18],
gave no product under a variety of conditions. The cyclisation of the
corresponding compound with an additional geminal vinyl ester
had been achieved with Mg(OTf)2 in refluxing MeCN [4]. While
repeat of these conditions on 1 gave no reaction the use of another
Lewis acid from this work Gd(OTf)3 (10 mol%) under the same
conditions (MeCN, reflux, 18 h) gave the desired product 2 in 33%
isolated yield.
A solvent screen was then conducted on a mixture of 1 with
50 mol% Gd(OTf)3 (Table 2). TIBCO Spotfire® was used to visualise
principle component analysis (PCA), whereby solvent data from an
internal AstraZeneca database was arranged by orthogonal prop-
erties [19]. The axes used to portray the solvents in 3D space were
not linked to specific properties (e.g. boiling point or dielectric
constant), but instead were functions generated from these prop-
erties. From this 3D visualisation it was possible to divide solvents
which should behave similarly into octants of chemical space and
Table 2
A series of cyclic amines 9e11 were synthesised to compare to
the successful rearrangement of 1 and 3. Acyclic variants 12e14
were also prepared to compare with 6. The synthesis of N-allyl
derivatives was attempted, but were unsuccessful by either of the
two routes. Although the desired SNAr and Palladium catalysed
allylations proceeded, albeit in low yield, the subsequent Henry
reaction gave complex mixtures of products. The electronics of the
Solvent screen with Gd(OTf)3 on 1 to give 2 (Scheme 2b).
Solvent
Time (h)
Area of HPLC product peak (%)
DMF
PhMe
12
12
4
3
4
12
6
12
0
94
47a
65b
41b
75c
46b
0
MeCN
2MeTHF
IPA
p-xylene
1-hexanol
Bu3N
aromatic template of
1 were investigated with a series of
substituted aromatic and heteroaromatic derivatives 15e19. The
nitroalkenes were each treated with 30 mol% Gd(OTf)3 and heated
to 100 ꢀC in PhMe for 18 h to give in most cases the desired rear-
ranged products (Table 3). The rearranged products were
a
Major degradation peak.
Several other products.
20% starting material.
b
c
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

4
J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
Fig. 1. Nitroalkene library.
characterised by 1H NMR. The products lost the vinyl protons
associated with the nitroalkene and gained two vicinally coupled
protons CHNO2 and NCH, diagnostic of the b-nitroamine formed
diastereoisomer of 4 was isolated with a nitroamine coupling
constant J ¼ 6.2 Hz. We therefore tentatively assign this the anti-
relative stereochemistry based upon the magnitude of the nitro
amine coupling constant and comparison to that of anti-20. The
major diastereoisomer of 21 exhibited a nitroamine coupling con-
stant J~5 Hz, which we think is not definitive enough to justify a
stereochemical assignment by comparison to 4 and 20. The
coupling constants for the nitroamine of the three compounds
22e24 were obscured and/or ambiguous so no structural assign-
ment could be made by 1H NMR.
To demonstrate the use of these compounds for the synthesis of
1,2-diamines, product 2 was subjected to a standard Zn/HCl
reduction and trifluoroacetylation which we have used routinely
(Scheme 5) [22]. The protected diamine was isolated in an unop-
timised 46% yield over the two steps. The coupling constant across
the 1,2-diamine was identical to that starting material (J ¼ 9.7)
indicating that the anti-stereochemistry had been retained.
The carbocyclic precursors 1, 3 and 9 gave good conversions (90, 57
and 67% respectively), but the diastereoselectivity was poor for the
7-membered ring substrate 9 (66:34). Morpholine 10 gave no re-
action and thiomorpholine 11 gave a poor 9% yield of rearranged
product 21 as a single diastereoisomer after heating for 7 days. The
rearrangement of acyclic precursors was successful with the
nitroalkenes derived from neutral 6 or electron rich 12 benzyl-
amines and alkylamine 14 giving rearranged products 22, 23 and 24
in 67, 84 and 74% yields respectively. The dr of 23 was 90:10 and 24
was isolated essentially as a single diastereoisomer. In contrast the
neutral phenyl product 22 had a crude dr of 75:25. The nitroalkene
derived from the electron withdrawn chloro-benzylamine 13 took
36 h for complete reaction and gave a major rearranged product
that contained a TfO group. We were unable to fully elucidate the
structure of this product. The electronic character of the central
aromatic template was investigated and the bromosubstituted
analogue 25 was formed in 42% yield. The pyridine nucleus 16 gave
degradation under the standard conditions and with 1.3 equiva-
lents of Gd(OTf)3 in an attempt to sequester the coordinating ability
of the pyridyl nitrogen. The more electron rich derivatives gave the
dimethoxy 26, methylenedioxy 27 and thiophene 28 products in
62, 72 and 25% yields respectively. All of the central aromatic de-
rivatives that rearranged gave products with excellent dr and
mostly as single diastereoisomers.
Characterising the relative stereochemistry was based in part on
a single crystal X-ray determination of 26 that confirmed the anti-
stereochemistry across the nitro amine [21]. The coupling constant
across the nitro amine of 26 was ~9.5 Hz, in line with a trans-diaxial
coupling and similar to the reported value for 2 [3]. The analogues
25, 27 and 28 have a similary large 9.5, 9.6 and 10.2 Hz J value
respectively across the nitro amine and were thus assigned the
anti-stereochemistry. The minor diastereoisomer syn-25 has an
expectedly lower corresponding J value of 2.7 Hz. The two di-
astereoisomers of 20 were separable. In the 1H NMR for the major
diastereoisomer it was not clear which J value of the NCH signal
3. Conclusion
We have developed two methods for the [1,5]-H shift nitro-
Mannich reaction for the synthesis of 2,3-disubstituted tetrahy-
droquinolines that allows a greater substrate scope for this reac-
tion. HFIP was found to cyclise cyclic amine substrates 1 and 3
under mild conditions (58 ꢀC) to give the tetrahydroquinoline
products 2 and 4 in 90% and 65% yields respectively as single di-
astereoisomers. While examples of the cyclisation of pyrrolidinyl
amine 1 have been published in the literature, the cyclisation of the
piperidinyl amine 3 had not been achieved. HFIP did not promote
the [1,5]-H shift nitro-Mannich reaction of acyclic amines, so an
alternative procedure was investigated that used catalytic Gd(OTf)3
in refluxing PhMe. This system was successful for 5-membered (1),
6-membered (3) and 7-membered (9) cyclic amine substrates.
Rearrangement of 9 gave 20 which had been prepared in a similar
manner by Kim. Our 67% yield and 2:1 dr is not as efficient as their
results of 98%, 4:1 dr with Sc(OTf)3 in MeCN and 68%, dr 4:1 with a
thiourea catalyst. In contrast and uniquely the Gd(OTf)3 catalysed
conditions were able to cyclise acyclic tertiary neutral (6) and
electron rich benzylic- (12) and alkyl- (14) amines in high yields
with diastereomeric ratios ranging from 75:25 to >95:5. Sub-
stituents on the aniline precursors (15, 17 and 18) were on the
whole well tolerated with yields ranging from 42 to 72% and dia-
stereomeric ratios of 90:10 to >95:5. Although the electron poor
aniline with a para-Br-substituent was the lowest yielding. This,
together with the results that the heteroyclic amine derived from
morpholine (10) and the electron withdrawn benzylic amine 13
possessing a Cl-substituent were both inert to the reaction condi-
tions, emphasised the extent of stabilisation required for the in-
termediate iminium ion to be formed. The sulphur containing
amino substituted thiophene 19 gave only a moderate 25%
d
4.12 (1H, ddd, J ¼ 7.6, 6.4, 3.1 Hz) was due to coupling from the
CHNO2 proton as it appeared as a multiplet ( 4.67), as did the two
NCHCH2 protons ( 1.82e1.87 and 1.58-1.50). However both the
nitroamine protons for the minor diastereoisomer were distinct at
4.90 (1H, ddd, J ¼ 12.4, 5.7, 4.3 Hz, CHNO2) and 4.08 (1H, dt,
J ¼ 8.5, 4.3 Hz, NCH). The large J ¼ 12.4 Hz coupling constant of
CHNO2 was due to coupling to one of the benzylic protons 3.53
d
d
d
d
d
(1H, dd, J ¼ 16.2, 12.4 Hz). The nitroamine coupling constant was
J ¼ 4.3 Hz. By comparison to the diastereoisomers of 25, we can
tentatively assign the minor diastereoisomer of 20, syn and the
major diastereoisomer anti, with a corresponding nitroamine
coupling constant for the latter of J ¼ 7.6 or 6.4 Hz. Only one
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
5
Table 3
(dr > 95:5) of rearranged product 28 and the heteroyclic amine
derived from thiomorpholine (11) gave a low yield (9%) of 21 after
prolonged heating (7 days). The sulphur atom in substrates 11 and
19 could be coordinating to the metal catalyst and leading to other
by-products. Preference for the cyclisation to favour the anti-dia-
stereoisomer was confirmed from a combination of X-ray and
Substrate scope for the Gd(OTf)3 catalysed [1,5]-H shift nitro-Mannich reaction.a.
Substrate
Product
Yield (%)b
79
drc
1
>95:5
coupling constant data for the product
b-nitro-amine function.
Overall, conditions for a wider substrate scope have been demon-
strated for the [1,5]-H shift nitro-Mannich reaction. The reaction is
substrate dependent with a fine balance for iminium ion stability
required for rearrangement. Optimisation of different metal cata-
lysts to suit particular substrates and the application of this
methodology to natural product synthesis is currently under
investigation.
3
9
57
45
95:5
>95:5
4. Experimental section
23
>95:5
4.1. General
All non-aqueous reactions were carried out in oven-dried
glassware under an inert atmosphere unless otherwise indicated.
All reaction temperatures refer to the values of the external heating
element and not that of the reaction mixture. Room temperature
implies a temperature range of 20e25 ꢀC. A temperature of 0 ꢀC was
achieved using an ice-water bath whereas cryogenic conditions
(ꢁ78 ꢀC or ꢁ68 ꢀC) were achieved using a dry ice and acetone or
CH2Cl2 bath respectively. All additions of reagent occurred as a
single portion or fast unless otherwise stated. Column chroma-
10
11
e
0d
9e
e
>95:5
6
major 50%
minor 17%
>95:5
>95:5
12
84
90:10
tography was carried out using BDH (40e60 mm) silica gel and
analytical thin layer chromatography was carried out using Merck
Keiselgel aluminium-backed plates coated with silica gel. Auto-
matic column chromatography was carried on a Biotage® Isolera™
Spektra or CombiFlash® EZ Prep fitted with RediSep® or Biotage®
SNAP Ultra cartridges. Components were visualised using ultra-
violet light (254 nm) and a basic potassium permanganate dip.
Removal of solvent in vacuo was achieved using Büchi rotary
evaporators and either the house vacuum or a Büchi Vac® V-500
pump. Where a compound has been prepared using a specific
literature procedure, this is referenced by the compound name.
All commercial chemicals and solvents were used as supplied
unless otherwise stated. The dry solvents toluene and THF were
obtained from a solvent tower, where degassed solvents were
13
14
e
0f
74
e
>95:5
15
38
4
>95:5
>95:5
passed through two columns of activated alumina and a 7 mm filter
under 4 bar pressure. Other anhydrous solvents were purchased
bottled from the Aldrich chemical company and used as provided.
Activation of 4 Å molecular sieves was achieved by heating under a
high vacuum.
Melting points are uncorrected and were obtained using a
Reichert Melting Point Apparatus. Infrared (IR) spectra were
recorded on a PerkinElmer spectrum 100 FT-IR (ATR mode). 1H
NMR spectra were recorded at 300 MHz on a Bruker Avance 300
spectrometer, at 400 MHz on a Bruker Avance 400 spectrometer, at
500 MHz on a Bruker Avance 500 spectrometer, at 600 MHz on a
Bruker Avance 600 spectrometer or at 700 MHz on a Bruker Avance
700 spectrometer in the stated solvent using the residual protic
16
17
e
0g
62
e
>95:5
18
19
a
72
25
>95:5
>95:5
solvent CHCl3
(
d
¼ 7.26 ppm, s) or DMSO ( ¼ 2.56 ppm, qn) as the
d
internal standard. Chemical shifts are quoted in ppm using the
following abbreviations: s, singlet d, doublet t, triplet q, quartet qn,
quintet m, multiplet, br, broad or a combination of these. The
coupling constants (J) are measured in Hertz.13C NMR spectra were
recorded at 125 MHz on a Bruker Avance 600 Spectrometer or at
175 MHz on a Bruker Avance 700 Spectrometer in the stated solvent
Nitroalkene (1 equivalent) and Gd(OTf)3 (30 mol%) were heated in PhMe
(10.0 mL per mmol) to 100 ꢀC until the reaction reached completion.
b
Isolated Yield.
Measured by 1H NMR.
No conversion.
Refluxed 7 days.
Complete consumption of starting material, unknown product.
Degradation.
c
d
using the internal reference of CHCl3 (
¼ 39.52 ppm, sept) as the internal standard.226 Chemical shifts
are reported to the nearest 0.1 ppm. Mass spectrometry data was
d
¼ 77.0 ppm, t) or DMSO
e
(d
f
g
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

6
J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
3.9 Hz), 3.23e3.17 (4H, m), 1.79 (8H, m); 13C NMR (150 MHz, CDCl3)
156.6 (C), 138.2 (CH), 136.2 (CH2), 132.6 (CH), 129.0 (CH), 123.8 (C),
d
122.0 (CH), 120.8 (CH), 56.8 (CH2), 29.3 (CH2), 27.3 (CH2); HRMS (ES)
[MþH]þ Calcd for C14H19N2O2 247.1447; Found 247.1454.
4.2.3. (E)-4-(2-(2-Nitrovinyl)phenyl)morpholine (10). The requisite
aldehyde [23] (368 mg, 1.88 mmol), prepared by General procedure
A, was subjected to General procedure C and gave after purification
by automated column chromatography (0e10% EtOAc/heptane), 10
(186 mg, 38%) as a dark red oil:: IR ymax (neat) 3099, 3033, 2890,
Scheme 5. Formation of 1,2-diamine.
collected on Thermo Finnigan Mat900xp (EI/CI) VG-70se (FAB) and
Waters LCT Premier XE (ES) instruments.
2849, 1625, 1596 cmꢁ1 1H NMR (400 MHz, CDCl3)
; d 8.40 (1H, d,
J ¼ 13.8 Hz), 7.67 (1H, d, J ¼ 13.8 Hz), 7.53e7.44 (2H, m), 7.13 (2H, dd,
J ¼ 7.7, 6.3 Hz), 3.94e3.88 (4H, m), 3.00e2.96 (4H, m); 13C NMR
4.2. Synthesis of nitroalkenes
(175 MHz, CDCl3)
d 153.8 (ArC), 137.3 (NCH2CH2), 136.5 (NCH2CH2),
General procedure A for the synthesis of 2-aminobenzaldehydes
The known 2-aminobenzaldehydes were synthesised by the
literature procedure based on the work of Rabong [3], involving the
SNAr reaction between the requisite fluorobenzaldehydes and
amines to give the known precursor benzaldehydes to nitroalkenes
1 [3], 3 [16], 6 [4], 9 [23],10 [24],12 [25],13 [25],14 [26],15 [27], and
16 [28].
133.1 (ArCH), 129.3 (ArCH), 124.6 (ArC), 123.9 (ArCH), 119.9 (ArCH),
67.2 (OCH2), 53.6 (NCH2); HRMS (ES) [MþH]þ Calcd for C12H14N2O3
235.1077; Found 235.1071.
4.2.4. (E)-4-(2-(2-Nitrovinyl)phenyl)thiomorpholine
(11). The
requisite aldehyde [24] (340 mg, 1.64 mmol), prepared by General
procedure B, was subjected to General procedure C and gave after
purification by column chromatography (10% EtOAc/hexane), 11
(260 mg, 63%) as a light red crystalline solid: mp 80e82 ꢀC; IR ymax
General procedure B for the Pd catalysed synthesis of
aminobenzaldehydes
(neat) 3098, 2952, 2912, 2878, 2831, 1624, 1596, 1505 cmꢁ1 1H
;
A 2-bromoaryl aldehyde (1 equivalent) was added to a sus-
pension/solution of Cs2CO3 (1.4 equivalents), Pd(OAc)2 (0.01
equivalents) and ( )BINAP (0.03 equivalents) in PhMe (10 mL per
mmol). The reaction mixture was stirred at rt for 5 min before a
secondary amine (1.1 equivalents) was added and the resulting
mixture heated to 90 ꢀC (heating block) for 18 h. Following cooling
to rt, the solvent was removed in vacuo and the crude product was
purified by column chromatography to give the known precursor
benzaldehydes to nitroalkenes 11 [24], 17 [29], and 19 [30].
NMR (600 MHz, CDCl3)
d
8.39 (1H, d, J ¼ 13.8 Hz), 7.64 (1H, d,
J ¼ 13.8 Hz), 7.51 (1H, d, J ¼ 7.8 Hz), 7.49e7.45 (1H, m), 7.16e7.12
(2H, m), 3.26e3.20 (4H, m), 2.91e2.86 (4H, m); 13C NMR (150 MHz,
CDCl3)
d 155.0 (C), 137.3 (CH), 136.3 (CH), 133.0 (CH), 128.9 (CH),
124.9 (C), 124.0 (CH), 120.8 (CH), 55.7 (CH2), 28.40 (CH2); HRMS (ES)
[M]þ Calcd for C12H14N2O2S 250.07705; Found 250.0771.
4.2.5. (E)-N-(4-Methoxybenzyl)-N-methyl-2-(2-nitrovinyl)aniline
(12). The requisite aldehyde [25] (616 mg, 2.40 mmol), prepared by
General procedure A, was subjected to General procedure C and
gave after purification by column chromatography (10% EtOAc/
hexane), 12 (228 mg, 32%) as a red oil: IR ymax (neat) 3097, 3030,
2950, 2832, 2798, 1623, 1609, 1594, 1508 cmꢁ1; 1H NMR (600 MHz,
General procedure C for the synthesis of nitroalkenes
The literature procedure by Rabong [3] was optimised. A solu-
tion of aldehyde (1 equivalent) in PhMe (5.0 mL per mmol) was
added KF (0.19 equivalents), Me2NH$HCl (1.5 equivalents), MeNO2
(3.5 mL per mmol) and the solution heated to 90 ꢀC (heating block)
for 18 h. The solvent was then removed in vacuo to give the crude
product which was purified by column chromatography.
CDCl3)
d
8.52 (1H, d, J ¼ 13.8 Hz), 7.65 (1H, d, J ¼ 13.8 Hz), 7.50 (1H,
dd, J ¼ 7.7, 1.4 Hz), 7.46e7.39 (1H, m), 7.13 (1H, d, J ¼ 8.2 Hz), 7.09
(1H, dd, J ¼ 11.4, 4.1 Hz), 6.88e6.84 (2H, m), 4.06 (2H, s), 3.80 (3H,
s), 2.67 (3H, s); 13C NMR (150 MHz, CDCl3)
d 159.1 (C), 154.4 (C),
137.2 (CH), 136.9 (CH), 132.8 (CH), 130.0 (CH), 129.4 (C), 129.3 (CH),
124.8 (C), 123.3 (CH), 121.2 (CH), 113.9 (CH), 61.7 (CH2), 55.4 (CH3),
41.6 (CH3); HRMS (ES) [MþH]þ Calcd for C17H19N2O3 299.13174;
Found 299.13181.
4.2.1. (E)-N-Benzyl-N-methyl-2-(2-nitrovinyl)aniline (6). The requi-
site aldehyde 8 [4] (2.25 g, 10.0 mmol), prepared by General pro-
cedure A, was subjected to General procedure C and gave after
purification by column chromatography (2% EtOAc/hexane), 6
(761 mg, 28%) as a bright red oil: IR ymax (neat) 3099, 3058, 3025,
4.2.6. (E)-N-(4-Chlorobenzyl)-N-methyl-2-(2-nitrovinyl)aniline (13).
The requisite aldehyde [26] (261 mg, 1.00 mmol), prepared by
General procedure A, was subjected to General procedure C and
gave after purification by column chromatography (10% EtOAc/
hexane), 13 (105 mg, 35%) as a red oil: IR ymax (neat) 3102, 3032,
2879, 2800,1625,1596,1553,1509 cmꢁ1; 1H NMR (700 MHz, CDCl3)
2945,1623,1593,1552,1507 cmꢁ1; 1H NMR (400 MHz, CDCl3)
d 8.52
(1H, d, J ¼ 13.8 Hz), 7.64 (1H, d, J ¼ 13.8 Hz), 7.52e7.48 (1H, m),
7.45e7.39 (1H, m), 7.36e7.30 (2H, m), 7.27 (3H, m), 7.16e7.12 (1H,
m), 7.12e7.06 (1H, m), 4.13 (2H, s), 2.69 (3H, s); 13C NMR (175 MHz,
CDCl3) d 154.4 (ArC), 137.4, (NCH2CH2), 137.1 (ArC), 137.0 (NCH2CH2),
132.8 (ArCH), 129.3 (ArCH), 128.7 (ArCH), 128.7 (ArCH), 127.7
(ArCH),124.7 (ArC),123.4 (ArCH),121.1 (ArCH), 62.2 (NCH2Ph), 41.98
(NCH3); HRMS (ES) [MþH]þ Calcd for C16H17N2O2 269.1290; Found
269.1298.
d
8.50 (1H, d, J ¼ 13.8 Hz), 7.64 (1H, d, J ¼ 13.8 Hz), 7.51 (1H, d,
J ¼ 7.8 Hz), 7.42 (1H, t, J ¼ 7.3 Hz), 7.30 (2H, d, J ¼ 8.3 Hz), 7.20 (2H, d,
J ¼ 8.2 Hz), 7.11 (2H, m), 4.09 (2H, s), 2.68 (3H, s); 13C NMR
(175 MHz, CDCl3) d 154.0 (C), 137.1 (CH), 136.8 (CH), 135.8 (C), 133.5
(C) 132.8 (CH), 130.0 (CH), 129.3 (CH), 128.8 (CH), 124.8 (C), 123.6
(CH), 121.2 (CH), 61.4 (CH2), 42.1 (CH2); HRMS (ES) [MþH]þ Calcd
for C16H16Cl35N2O2 303.0900; Found 303.0908.
4.2.2. (E)-1-(2-(2-Nitrovinyl)phenyl)azepane (9). The requisite
aldehyde [23] (690 mg, 3.35 mmol), prepared by General procedure
A, was subjected to General procedure C and gave after purification
by column chromatography (2% EtOAc/hexane), 9 (477 mg, 58%) as
a dark red oil: IR ymax (neat) 3098, 2952, 2912, 2878, 2831, 1624,
4.2.7. (E)-N,N-Dibutyl-2-(2-nitrovinyl)aniline (14). The requisite
aldehyde [26] (700 mg, 3.00 mmol), prepared by General procedure
A, was subjected to General procedure C and gave after purification
by column chromatography (0e2% EtOAc/hexane), 14 (398 mg,
48%) as a red oil: IR ymax (neat) 2953, 2927, 2858, 1623, 1593,
1596, 1505 cmꢁ1
; d 8.43 (1H, d,
1H NMR (400 MHz, CDCl3)
J ¼ 13.7 Hz), 7.58 (1H, d, J ¼ 13.7 Hz), 7.44 (1H, dd, J ¼ 7.8, 1.6 Hz),
7.41e7.36 (1H, m), 7.15 (1H, dd, J ¼ 8.3, 0.9 Hz), 7.00 (1H, dd, J ¼ 11.1,
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
7
1509 cmꢁ1
;
1H NMR (600 MHz, CDCl3)
d
8.41 (1H, d, J ¼ 13.8 Hz),
crystalline solid: mp 91e93 ꢀC; IR ymax (neat) 3128, 2950, 2917,
2835, 2678, 1596, 1502 cmꢁ1; 1H NMR (600 MHz, CDCl3)
8.41 (1H,
7.65 (1H, d, J ¼ 13.8 Hz), 7.49 (1H, dd, J ¼ 7.8, 1.3 Hz), 7.43e7.40 (1H,
m), 7.17 (1H, d, J ¼ 8.2 Hz), 7.07 (1H, t, J ¼ 7.7 Hz), 3.03e2.99 (4H, m),
1.44 (4H, tt, J ¼ 7.7, 6.5 Hz), 1.30e1.22 (4H, m), 0.86 (6H, t,
d
d, J ¼ 13.3 Hz), 7.40 (1H, d, J ¼ 13.3 Hz), 6.84 (1H, s), 6.55 (1H, s), 5.96
(2H, s), 3.27 (4H, m), 2.00e1.95 (4H, m); 13C NMR (150 MHz, CDCl3)
J ¼ 7.4 Hz); 13C NMR (150 MHz, CDCl3)
d
153.4 (C), 137.4 (CH), 136.7
d 152.5 (C),150.1 (C),142.1 (C),138.7 (CH),132.8 (CH),112.8 (C),106.9
(CH), 132.26 (CH), 129.3 (C), 126.1 (CH), 123.1 (CH), 122.5 (CH), 54.5
(CH2), 29.6 (CH2), 20.5 (CH2), 14.0 (CH3); HRMS (ES) [MþH]þ Calcd
for C16H25N2O2 277.1916; Found 277.1921.
(CH), 101.8 (CH2), 98.1 (CH), 53.9 (CH2), 25.5 (CH2). HRMS (ES)
[MþH]þ Calcd for C13H15N2O4 263.1032; Found 263.1040.
4.2.12. (E)-1-(2-(2-Nitrovinyl)thiophen-3-yl)pyrrolidine
(19).
4.2.8. (E)-1-(4-Bromo-2-(2-nitrovinyl)phenyl)pyrrolidine
(15).
The requisite aldehyde [30] (170 mg, 0.94 mmol), prepared by
General procedure B, was subjected to General procedure C and
gave after purification by column chromatography (10% EtOAc/
hexane), 20 (259 mg, 62%) as a yellow crystalline solid: mp
The requisite aldehyde [27] (530 mg, 2.09 mmol), prepared by
General procedure A, was subjected to General procedure C and
gave after purification by automated column chromatography
(0e10% EtOAc/heptane), 15 (242 mg, 39%) as a red crystalline solid:
mp 88e89 ꢀC; IR ymax (neat) 3112, 2987, 2950, 2922, 2851, 1607,
157e159 ꢀC; IR ymax (neat) 3087, 2959, 2922, 2852, 1534 cmꢁ1 1H
;
NMR (600 MHz, CDCl3)
d
8.56 (1H, dd, J ¼ 12.2, 1.0 Hz), 7.37 (1H, dd,
1588, 1551, 1530 cmꢁ1
;
1H NMR (600 MHz, CDCl3)
d
8.33 (1H, d,
J ¼ 5.7, 1.1 Hz), 7.22 (1H, d, J ¼ 12.2 Hz), 6.56 (1H, d, J ¼ 5.7 Hz),
J ¼ 13.4 Hz), 7.46 (1H, d, J ¼ 2.4 Hz), 7.42 (1H, d, J ¼ 13.4 Hz), 7.37
3.65e3.61 (4H, m), 2.12e2.05 (4H, m); 13C NMR (150 MHz, CDCl3)
(1H, dd, J ¼ 8.9, 2.4 Hz), 6.75 (1H, d, J ¼ 9.0 Hz), 3.35 (4H, m),
d
154.0 (C), 135.0 (CH), 132.5 (CH), 127.6 (CH), 119.9 (CH), 105.3 (C),
2.01e1.96 (4H, m); 13C NMR (150 MHz, CDCl3)
d
150.1 (C), 138.3
52.2 (CH2), 25.8 (CH2). HRMS (ES) [MþH]þ Calcd for C10H13N2O2S
(CH), 135.4 (CH), 135.0 (CH), 131.8 (CH), 119.9 (C), 117.4 (CH), 110.9
(C), 52.9 (CH2), 25.8 (CH2); HRMS (ES) [M]þ Calcd for C12H23BrN2O2
297.0239; Found 297.0248.
225.0698; Found 225.0115.
4.3. [1,5]-H shift cyclisations using HFIP
4.3.1. (3aS*,4R*)-4-nitro-1,2,3,3a,4,5-hexahydropyrrolo[1,2-a]quino-
lone (2). A solution of 1 (60 mg, 0.27 mmol) in HFIP (4 mL) was
heated to 58 ꢀC (heating block) for 4 h then cooled to rt and the
solvent removed in vacuo to give a crude residue which was puri-
fied by column chromatography (2% EtOAc in hexane) to give dia-
stereomerically pure 2 [3] (54 mg, 90%) as a yellow crystalline solid:
mp 94e95 ꢀC.
Alternatively 1 (1 mmol) could be rearranged using Gd(OTf)3
according to general procedure D below to give diastereomerically
pure 2 [3] (79%).
4.2.9. (E)-3-(2-Nitrovinyl)-2-(pyrrolidin-1-yl)pyridine
(16). The
requisite aldehyde [28] (584 mg, 3.35 mmol), prepared by General
procedure A, was subjected to General procedure C and gave after
purification by column chromatography (20% EtOAc/hexane), 16
(282 mg, 30%) as a bright red crystalline solid: mp 64e66 ꢀC; IR
ymax (neat) 3176, 3104, 2983, 2958, 2876, 2859, 1683, 1607, 1586,
1545 cmꢁ1
;
1H NMR (600 MHz, CDCl3)
d
8.36 (1H, d, J ¼ 13.3 Hz),
8.25 (1H, dd, J ¼ 4.6, 1.8 Hz), 7.59 (1H, dd, J ¼ 7.6, 1.6 Hz), 7.37 (1H, d,
J ¼ 13.3 Hz), 6.68 (1H, dd, J ¼ 7.6, 4.7 Hz), 3.65e3.60 (4H, m),
1.99e1.94 (4H, m); 13C NMR (150 MHz, CDCl3)
d 158.4 (C), 151.3
(CH), 138.8 (CH), 138.0 (CH), 135.1 (CH), 113.5 (CH), 110.6 (C), 51.3
(CH2), 25.9 (CH2); HRMS (ES) [MþH]þ Calcd for C11H14N3O2
220.1086; Found 220.1112.
4.3.2. (4aS*,5R*)-5-nitro-2,3,4,4a,5,6-hexahydro-1H-pyrido[1,2-a]
quinolone (4). In an identical procedure for the formation of 2,
nitroalkene 3 (92 mg, 0.40 mmol) was rearranged in HFIP over 18 h
to give diastereomerically pure 4 (53 mg, 58%) as a low melting
4.2.10. (E)-1-(4,5-Dimethoxy-2-(2-nitrovinyl)phenyl)pyrrolidine
(17). The requisite aldehyde [29] (353 mg, 1.50 mmol), prepared by
General procedure B, was subjected to General procedure C and
gave after purification by column chromatography (20% EtOAc/
hexane), 17 (259 mg, 62%) as a dark purple crystalline solid: mp
136e138 ꢀC; IR ymax (neat) 3118, 2964, 2918, 2862, 2848, 2826,
waxy solid: IR ymax (neat) 2938, 2850, 1597, 1576, 1540 cmꢁ1
;
1H
NMR (600 MHz, CDCl3)
d
7.18e7.15 (1H, td, J ¼ 8.3, 0.8 Hz), 7.05 (1H,
d, J ¼ 7.3 Hz), 6.84 (1H, d, J ¼ 8.3 Hz), 6.76 (1H, td, J ¼ 7.4, 0.8 Hz),
4.70 (1H, ddd, J ¼ 7.7, 6.2, 5.1 Hz), 3.95e3.89 (1H, m), 3.54 (1H, ddd,
J ¼ 10.8, 6.2, 2.5 Hz), 3.44 (1H, dd, J ¼ 15.5, 7.8 Hz), 3.16 (1H, dd,
J ¼ 15.5, 5.0 Hz), 2.83 (1H, td, J ¼ 12.6, 2.9 Hz), 1.95e1.88 (1H, m,H),
1.80e1.72 (1H, m), 1.69e1.44 (4H, m,H); 13C NMR (150 MHz, CDCl3)
1625, 1584, 1569, 1546, 1517 cmꢁ1; 1H NMR (600 MHz, CDCl3)
(1H, d, J ¼ 13.3 Hz), 7.45 (1H, d, J ¼ 13.2 Hz), 6.83 (1H, s), 6.45 (1H, s),
3.93 (3H, s), 3.86 (3H, s), 3.36e3.33 (4H, m), 2.04e1.96 (4H, m); 13
NMR (150 MHz, CDCl3) 153.9 (C), 148.5 (C), 143.3 (C), 138.8 (CH),
d 8.45
d
145.3 (C), 129.2 (CH), 128.3 (CH), 120.0 (C), 118.8 (CH), 113.1 (CH),
C
86.4 (CH), 59.0 (CH), 48.1 (CH2), 31.5 (CH2), 30.1 (CH2), 24.5 (CH2),
24.1 (CH2); HRMS (ES) [MþH]þ Calcd for C13H17N2O2 233.1290;
Found: 233.1305. Data in agreement with the literature [3].
Alternatively 3 (0.25 mmol) could be rearranged using Gd(OTf)3
according to general procedure D below to give 4 (79%) as an
inseparable 95:5 mixture of diastereoisomers.
d
132.4 (CH), 111.1 (C), 110.7 (CH), 99.9 (CH), 56.4 (CH3), 56.0 (CH3),
53.6 (CH2), 25.6 (CH2); HRMS (ES) [MþH]þ Calcd for C14H19N2O4
279.1345; Found 279.1347.
4.2.11. (E)-1-(6-(2-Nitrovinyl)benzo[d][1,3]dioxol-5-yl)pyrrolidine
(18). The requisite aldehyde was prepared by General procedure B:
6-bromopiperonal (460 mg, 2.02 mmol) gave the aldehyde
(375 mg, 86%) as a yellow crystalline solid after purification by
column chromatography (20% EtOAc/hexane): mp 71e73 ꢀC; IR
General procedure D for 1,5-hydride shift nitro-Mannich cyclisation
using Gd(OTf)3
Nitroalkene (1 equivalent) and Gd(OTf)3 (0.3 equivalents) were
heated in toluene (10.0 mL per mmol) to 100 ꢀC until the reaction
reached completion (18 h unless otherwise stated). The toluene
was then removed in vacuo and the resultant residue purified via
column chromatography to give the cyclised product.
ymax (neat) 2969, 2945, 2865, 1627, 1604, 1493 cmꢁ1
(600 MHz, CDCl3)
s), 3.40e3.35 (4H, m), 2.02e1.96 (4H, m). 13C NMR (150 MHz, CDCl3)
;
1H NMR
d
10.01 (1H, s), 7.22 (1H, s), 6.44 (1H, s), 5.94 (2H,
d
188.1 (CH), 153.7 (C), 150.6 (C) 140.6 (C), 117.4 (C), 108.6 (CH), 101.7
(CH2), 95.8 (CH), 54.0 (CH2), 26.0 (CH2); HRMS (ES) [MþH]þ Calcd
4.3.3. (20)-anti (6R*,6aS*)- and (20)-syn (6R*,6aR*)- 6-nitro-
for C12H14NO3 220.0974; Found. 220.0979.
5,6,6a,7,8,9,10,11-octahydroazepino[1,2-a]quinolone. The
nitro-
The requisite aldehyde (334 mg, 1.53 mmol) was subjected to
General procedure C and gave after purification by column chro-
alkene 9 (90 mg, 0.37 mmol) was subjected to General procedure D
and gave a crude 66:34 mixture of diastereoisomers. Purification by
column chromatography (5% EtOAc/hexane) gave 20-anti (6R*,6aS*)
matography (10% EtOAc/hexane), 18 (140 mg, 36%) as
a red
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

8
J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
(39 mg, 45%) as a yellow oil; IR ymax (neat) 3025, 2923, 2858, 1602,
1578, 1540, 1499 cmꢁ1 1H NMR (600 MHz, CDCl3)
7.10 (1H, t,
column chromatography (10% EtOAc/hexane) gave 23 (96 mg, 84%)
as an inseparable 90:10 mixture of diastereoisomers as a low
melting orange waxy solid: IR ymax (neat) 2927, 2833, 1603, 1575,
;
d
J ¼ 7.8 Hz), 7.06 (1H, d, J ¼ 7.4 Hz), 6.66 (1H, t, J ¼ 7.3 Hz), 6.60 (1H, d,
J ¼ 8.3 Hz), 4.69e4.67 (1H, m), 4.12 (1H, ddd, J ¼ 7.6, 6.4, 3.1 Hz),
3.85 (1H, ddd, J ¼ 15.3, 5.9, 2.9 Hz), 3.54 (1H, d, J ¼ 17.8 Hz), 3.21
(1H, dd, J ¼ 17.8, 5.4 Hz), 3.11 (1H, ddd, J ¼ 15.6, 10.6, 5.3 Hz),
2.11e2.03 (1H, m), 1.87e1.82 (1H, m), 1.72e1.59 (4H, m), 1.58e1.50
; d 7.20e7.16 (1H, m),
1508 cmꢁ1 1H NMR (600 MHz, CDCl3)
7.07e7.03 (2H, m), 6.87e6.84 (2H, m), 6.82 (1H, d, J ¼ 7.6 Hz), 6.72
(1H, d, J ¼ 8.3 Hz), 6.68 (1H, td, J ¼ 7.5, 1.0 Hz), 4.90 (1H, td, J ¼ 6.6,
3.4 Hz), 4.80 (1H, d, J ¼ 6.6 Hz), 3.79 (3H, s), 3.77 (1H, ddd, J ¼ 11.9,
6.7, 1.1 Hz), 3.58 (1H, dd, J ¼ 12.0, 3.4 Hz), 2.99 (3H, s); 13C NMR
(1H, m), 1.48e1.40 (1H, m); 13C NMR (150 MHz, CDCl3)
d 143.0 (C),
129.2 (CH), 128.0 (CH), 116.5 (CH), 116.0 (C), 111.0 (CH), 81.9 (CH),
60.6 (CH), 49.4 (CH2), 33.5 (CH2), 27.4 (CH2), 27.2 (CH2), 26.1 (CH2),
25.9 (CH2). HRMS (ES) [MþH]þ Calcd for C14H19N2O2 247.1447;
Found 247.1461; and 20-syn (6R*,6aR*) (21 mg, 23%) as a yellow oil;
(150 MHz, CDCl3) d 159.0 (C), 145.3 (C), 133.6 (C), 130.4 (CH), 130.1
(CH), 128.3 (CH), 121.7 (C), 118.1 (CH), 114.4 (CH), 111.7 (CH), 85.8
(CH), 55.4 (CH3), 51.2 (CH2), 46.4 (CH), 39.3 (CH3); HRMS (ES)
[Mþ2H-OCH3]þ Calcd for C16H17N2O2 269.1290 Found: 269.1294.
IR ymax (neat) 3021, 2930, 2903, 2855, 1600, 1574, 1530, 1494 cmꢁ1
;
1H NMR (600 MHz, CDCl3)
d
7.12 (1H, t, J ¼ 7.7 Hz), 7.08 (1H, t,
4.3.7. 1-Butyl-3-nitro-2-propyl-1,2,3,4-tetrahydroquinoline
(24).
J ¼ 6.8 Hz), 6.67 (1H, td, J ¼ 7.4, 0.7 Hz), 6.62 (1H, d, J ¼ 8.3 Hz), 4.90
(1H, ddd, J ¼ 12.4, 5.7, 4.3 Hz), 4.08 (1H, dt, J ¼ 8.5, 4.3 Hz), 3.92 (1H,
ddd, J ¼ 15.1, 6.1, 3.1 Hz), 3.53 (1H, dd, J ¼ 16.2, 12.4 Hz), 3.27e3.16
(2H, m), 2.15e2.06 (1H, m), 1.72e1.47 (6H, m), 1.43e1.33 (1H, m);
The nitroalkene 14 (83 mg, 0.30 mmol) was subjected to General
procedure D and after purification by column chromatography
(0e2% EtOAc/hexane) gave diastereomerically pure 24 (61 mg, 74%)
as a yellow oil: IR ymax (neat) 3036, 2954, 292, 2868, 1601, 1575,
13C NMR (150 MHz, CDCl3)
d
142.9 (C), 129.9 (CH), 128.3 (CH), 116.6
1544, 1498 cmꢁ1; 1H NMR (600 MHz, CDCl3)
d
7.10 (1H, t, J ¼ 7.8 Hz),
(CH), 116.1 (C), 110.8 (CH), 81.1 (CH), 60.0 (CH), 49.9 (CH2), 28.7
(CH2), 27.0 (CH2), 26.7 (CH2), 26.2 (CH2), 25.6 CH2).
7.07 (1H, d, J ¼ 7.3 Hz), 6.68 (1H, td, J ¼ 7.4, 0.8 Hz), 6.58 (1H, d,
J ¼ 8.3 Hz), 4.75 (1H, m), 4.04 (1H, m), 3.52 (1H, d, J ¼ 18.1 Hz), 3.43
(1H, ddd, J ¼ 14.6, 9.7, 5.0 Hz), 3.12 (1H, dd, J ¼ 18.1, 5.6 Hz), 2.99
(1H, ddd, J ¼ 14.7, 9.8, 6.4 Hz), 1.60 (2H, ddd, J ¼ 10.2, 6.8, 2.7 Hz),
1.55e1.34 (4H, m), 1.33e1.25 (2H, m), 0.97 (3H, td, J ¼ 7.2, 2.6 Hz),
4.3.4. 5-Nitro-1,2,4,4a,5,6-hexahydro-[1,4]thiazino[4,3-a]quinolone
(21). The nitroalkene 11 (100 mg, 0.40 mmol) was subjected to
General procedure D for 7 days and after purification by column
chromatography (10% EtOAc/hexane) gave diastereomerically pure
21 (9 mg, 9%) as a low melting waxy solid: IR ymax (neat) 2955, 2913,
0.93 (3H, t, J ¼ 7.4 Hz); 13C NMR (150 MHz, CDCl3)
d 142.5 (C), 129.2
(CH), 127.7 (CH),117.0 (C), 116.7 (CH), 111.9 (CH), 81.1 (CH), 60.8 (CH),
50.9 (CH2), 34.4 (CH2), 29.8 (CH2), 26.2 (CH2), 20.2 (CH2), 19.4 (CH2),
14.2 (CH3), 14.1 (CH3); Calcd for C16H25N2O2 277.19160 Found:
277.19169.
2890, 2850, 2825, 1596, 1563, 1507 cmꢁ1
CDCl3)
;
1H NMR (600 MHz,
7.17 (1H, td, J ¼ 8.3, 1.5 Hz), 7.09 (1H, d, J ¼ 7.2 Hz), 6.80 (1H,
d
td, J ¼ 7.4, 0.8 Hz), 6.76 (1H, d, J ¼ 8.3 Hz), 4.67 (1H, q, J ¼ 5.0 Hz),
4.34 (1H, dddd, J ¼ 11.0, 4.4, 2.0,1.1 Hz), 4.29 (1H dt, J ¼ 14.5, 2.8 Hz),
3.48e3.42 (2H, m, ArCH2), 3.15 (1H, dd, J ¼ 16.6, 5.2 Hz), 3.03 (1H,
ddd, J ¼ 13.5, 12.2, 2.7 Hz), 2.80 (1H, dd, J ¼ 13.0, 11.0 Hz), 2.39 (1H,
d, J ¼ 13.0 Hz), 2.31 (1H, ddd, J ¼ 13.4, 4.0, 2.5 Hz); 13C NMR
4.3.8. (25)-anti (3aS*,4R*) and (25)-syn (3aR*,4R*) -7-bromo-4-
nitro-1,2,3,3a,4,5-hexahydropyrrolo[1,2-a]quinolone. The
nitro-
alkene 15 (118 mg, 0.397 mmol) was subjected to General proced-
ure D and gave a crude 90:10 mixture of diastereoisomers.
Purification by column chromatography (5% EtOAc/hexane) gave
25-anti (3aS*,4R*) (45 mg, 38%) as an orange crystalline solid: mp
138 ꢀC (dec.); IR ymax (neat) 2971, 2946, 2856, 1595, 1532,
(150 MHz, CDCl3) d 142.9 (C), 129.4 (CH), 128.5 (CH), 119.6 (C), 119.1
(CH), 113.5 (CH), 83.9 (CH), 60.5 (CH), 50.1 (CH2), 28.8 (CH2), 28.2
(CH2), 23.4 (CH2); HRMS (ES) [MþH]þ Calcd for C12H15N2O2S
251.0854; Found 251.0858.
1495 cmꢁ1; 1H NMR (600 MHz, CDCl3)
d
7.22 (1H, dd, J ¼ 8.6,1.9 Hz),
7.16 (1H, s), 6.36 (1H, d, J ¼ 8.6 Hz), 4.42 (1H, ddd, J ¼ 12.0, 9.7,
4.7 Hz), 3.76 (1H, td, J ¼ 9.6, 5.5 Hz), 3.47e3.41 (2H, m), 3.27e3.21
(m, 2H), 2.25 (1H, ddd, J ¼ 12.3, 6.9, 1.4 Hz), 2.16 (1H, ddd, J ¼ 12.8,
9.7, 7.5 Hz), 2.06e1.96 (1H, m), 1.77 (1H, tdd, J ¼ 11.9, 9.5, 7.8 Hz);
4.3.5. 1-Methyl-3-nitro-2-phenyl-1,2,3,4-tetrahydroquinoline (22).
The nitroalkene 6 (109 mg, 0.41 mmol) was subjected to General
procedure D and gave a crude 75:25 mixture of diastereoisomers.
Purification by column chromatography (10% EtOAc/hexane) gave
22-major (55 mg, 50%) as a yellow solid: mp 93e94 ꢀC; IR ymax
(neat) 3060, 3026, 2895, 2827, 1601, 1576,1546,1497 cmꢁ1; 1H NMR
13C NMR (150 MHz, CDCl3)
d 142.1 (C), 131.4 (CH), 131.1 (CH), 119.4
(C), 112.7 (CH), 108.2 (C), 83.9 (CH), 60.4 (CH), 47.8 (CH2), 33.4 (CH2),
30.8 (CH2), 23.5 (CH2); HRMS (ES) [M]þ Calcd for C12H13BrN2O2
297.0239; and 25-syn (3aR*,4R*) (5 mg, 4%) as an orange crystalline
solid: mp 137 ꢀC (dec.); IR ymax (neat) 2970, 2946, 2855, 1594, 1532,
(500 MHz, CDCl3)
d 7.36e7.28 (3H, m), 7.23e7.18 (3H, m), 7.03 (1H,
d, J ¼ 7.3 Hz), 6.72 (2H, m), 5.19 (1H, apparent singlet), 4.84 (1H,
apparent dd, J ¼ 7.8, 3.9 Hz), 3.42 (1H, ddd, J ¼ 16.8, 3.7, 1.4 Hz), 3.01
1495 cmꢁ1; 1H NMR (600 MHz, CDCl3)
d
7.21 (dt, J ¼ 11.1, 5.6 Hz,1H),
(1H, dd, J ¼ 16.8, 4.4 Hz); 13C NMR (175 MHz, CDCl3)
d
144.4 (C),
7.16e7.14 (m,1H), 6.40 (d, J ¼ 8.6 Hz,1H), 5.11 (dt, J ¼ 5.1, 2.7 Hz,1H),
3.74 (ddd, J ¼ 9.2, 6.3, 2.7 Hz, 1H), 3.39 (td, J ¼ 8.8, 3.4 Hz, 1H),
3.33e3.26 (m, 3H, NCH2), 2.22 (dtd, J ¼ 9.5, 6.9, 2.8 Hz, 1H),
2.10e2.03 (m, 1H), 2.03e1.95 (m, 1H), 1.86e1.78 (m, 1H); 13C NMR
139.5 (C), 129.3 (CH), 129.0 (CH), 128.5 (CH), 128.4 (CH), 126.6 (CH),
117.3 (CH), 110.6 (CH), 83.8 (CH), 64.8 (CH), 37.8 (CH3), 27.6 (CH2);
HRMS (ES) [MþH]þ Calcd for C16H17N2O2 269.1290 Found:
269.1986; and 22-minor (18 mg, 17%) as a yellow oil: IR ymax (neat)
3025, 2895, 2833, 1602, 1577, 1542, 1499 cmꢁ1; 1H NMR (500 MHz,
(150 MHz, CDCl3)
d 142.8 (C), 131.1 (CH), 130.5 (CH), 118.7 (C), 112.9
(CH), 108.5 (C), 78.7 (CH), 58.5 (CH), 47.3 (CH2), 31.6 (CH2), 28.4
(CH2), 23.1 (CH2).
CDCl3)
d
7.36e7.29 (3H, m), 7.28 (1H, d, J ¼ 7.8 Hz), 7.12 (1H, d,
J ¼ 7.4 Hz), 7.03 (2H, dt, J ¼ 3.7, 2.1 Hz), 6.75 (1H, td, J ¼ 7.4, 0.9 Hz),
6.72(1H, d, J ¼ 8.2 Hz), 5.30 (1H, m), 5.18 (1H, m), 3.25e3.22 (1H,
m), 3.19e3.15 (1H, m), 2.92 (3H, s); 13C NMR (125 MHz, CDCl3)
4. 3. 9. (3aS*, 4R*)-7, 8-dimethoxy-4-nitro-1, 2, 3, 3a, 4, 5-
hexahydropyrrolo[1,2-a]quinolone (26). The nitroalkene 17 (111 mg,
0.40 mmol) was subjected to General procedure D and after puri-
fication by column chromatography (20% EtOAc/hexane) gave dia-
stereomerically pure 26 (69 mg, 62%) as an orange crystalline solid:
mp 163e164 ꢀC; IR ymax (neat) 2989, 2938, 2919, 2878, 1618, 1592,
d
144.3 (C), 135.7 (C), 129.3 (CH), 128.5 (CH), 128.5 (CH), 128.3 (CH),
127.1 (CH), 116.8 (CH), 116.3 (CH), 109.8 (CH), 81.5 (CH), 64.5 (CH),
37.6 (CH3), 26.4 (CH2).
4.3.6. (2S*,3R*)-2-(4-Methoxyphenyl)-1-methyl-3-nitro-1,2,3,4-
tetrahydroquinoline (23). The nitroalkene 12 (114 mg, 0.38 mmol)
was subjected to General procedure D and after purification by
1535, 1521 cmꢁ1; 1H NMR (600 MHz, CDCl3)
d 6.63 (1H, s), 6.14 (1H,
s), 4.45 (1H, ddd, J ¼ 11.8, 9.8, 4.9 Hz), 3.87 (3H, s), 3.81 (3H, s), 3.72
(1H, td, J ¼ 9.2, 5.7 Hz), 3.47 (1H, td, J ¼ 8.6, 3.0 Hz), 3.42 (1H, ddd,
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
9
J ¼ 14.9, 11.9, 0.7 Hz), 3.29 (1H, dd, J ¼ 16.0, 8.4 Hz), 3.20 (1H, dd,
J ¼ 15.0, 4.9 Hz), 2.25e2.20 (1H, m), 2.16e2.09 (1H, m), 2.06e1.96
dtd, J ¼ 15.6, 12.2, 8.9 Hz), 1.75 (1H, ddd, J ¼ 21.5, 11.7, 7.4 Hz); 13C
NMR (175 MHz, CDCl3) d 143.8 (C), 129.0 (CH), 128.2 (CH), 118.6 (C),
(1H, m), 1.82e1.75 (1H, m); 13C NMR (150 MHz, CDCl3)
d
149.4 (C),
116.2 (CH), 110.8 (CH), 62.2 (CH), 49.6 (CH), 47.6 (CH2), 34.9 (CH2),
31.2 (CH2), 23.6 (CH2); HRMS (ES) [MþH]þ Calcd for C14H16F3N2O
285.1209; Found 285.1215.
140.9 (C), 138.0 (C), 113.7 (CH), 108.5 (C), 97.3 (CH), 84.8 (CH), 60.4
(CH), 57.0 (CH3), 56.1 (CH3), 48.5 (CH2), 33.3 (CH2), 30.6 (CH2), 23.5
(CH2); HRMS (ES) [MþH]þ Calcd for C14H19N2O4 279.1345 Found:
279.1345 m.p. 163e164 ꢀC. Relative stereochemistry was from sin-
gle crystal X-ray diffraction, see ESI.
Declaration of competing interest
None.
4.3.10. (3aS*,4R*)-4-nitro-1,2,3,3a,4,5-hexahydro-[1,3]dioxolo[4,5-g]
pyrrolo[1,2-a]quinolone (27). The nitroalkene 18 (100 mg,
0.38 mmol) was subjected to General procedure D and after puri-
fication by column chromatography (10% EtOAc/hexane) gave dia-
stereomerically pure 27 (72 mg, 72%) as a yellow crystalline solid:
mp 166e168 ꢀC; IR ymax (neat) 2981, 2942, 2895, 2870, 2836, 2782,
Acknowledgments
We thank the EPSRC (DTA award) and AstraZeneca for funding
and Dr K. Karu for mass spectra.
1625, 1614, 1540, 1502 cmꢁ1 1H NMR (600 MHz, CDCl3)
d 6.57 (1H,
Appendix A. Supplementary data
s), 6.16 (1H, s), 5.85 (2H, dd, J ¼ 3.3, 1.3 Hz), 4.43 (1H, ddd, J ¼ 11.8,
9.7, 4.9 Hz), 3.70 (1H, td, J ¼ 9.3, 5.7 Hz), 3.40 (2H, ddd, J ¼ 11.9, 9.0,
4.6 Hz), 3.24 (1H, dd, J ¼ 16.1, 8.5 Hz), 3.17 (1H, dd, J ¼ 15.1, 4.9 Hz),
2.24e2.18 (1H, m), 2.14e2.08 (1H, m), 2.03e1.94 (1H, m), 1.77 (1H,
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2019.130663.
tt, J ¼ 11.1, 8.4 Hz); 13C NMR (150 MHz,CDCl3)
d 147.7 (C), 139.0 (C),
References
138.7 C), 109.0 (CH), 109.0 (C), 100.8 (CH2), 94.4 (CH), 84.7 (CH), 60.4
(CH), 48.7 (CH2), 33.8 (CH2), 30.6 (CH2), 23.4 (CH2); HRMS (ES)
[MþH]þ Calcd for C13H15N2O4 263.1032 Found: 263.1029.
[1] (For reviews see: ) (a) M. Tobisu, N.A. Chatani, Angew. Chem. Int. Ed. 45 (2006)
1683;
(b) S.C. Pan, J. Org. Chem. 8 (2012) 1374;
(c) B. Peng, N. Maulide, Chem. Eur J. 19 (2013) 13274;
(d) M.C. Haibach, D. Seidel, Angew. Chem. Int. Ed. 53 (2014) 5010;
(e) L. Wang, J. Xiao, Adv. Synth. Catal. 356 (2014) 1137;
(f) L. Wang, J. Xiao, Top. Curr. Chem. 374 (2016) 17.
[2] (For representative examples see: Carbonyl acceptors) (a) J.G. Schulz,
A. Onopchenko, J. Org. Chem. 43 (1978) 339;
4.3.11. 5-Nitro-4,5,5a,6,7,8-hexahydrothieno[3,2-e]indolizine
(28).
The nitroalkene 19 (36 mg, 0.15 mmol) was subjected to General
procedure D and after purification by column chromatography (10%
EtOAc/hexane) gave diastereomerically pure 28 (9 mg, 25%) as a
low melting white waxy solid: IR ymax (neat) 3324, 3295, 2912,
2865, 1715, 1669, 1602, 1547, 1506 cmꢁ1; 1H NMR (700 MHz, CDCl3)
(b) I.D. Jurberg, B. Peng, E. Wostefeld, M. Waserloos, N. Maulide, Angew. Chem.
Int. Ed. 51 (2012) 1950 (Imine acceptors);
(c) J.C. Duff, J. Chem. Soc. (1941) 547;
(d) W. Verboom, M.R.J. Hamzink, D.N. Reinhoudt, R. Visser, Tetrahedron Lett.
25 (1984) 4309;
d
7.64 (1H, d, J ¼ 5.6 Hz), 6.72 (1H, d, J ¼ 5.6 Hz), 4.52e4.48 (ddd,
J ¼ 10.6, 6.0, 3.2 Hz), 4.46 (1H, dd, J ¼ 10.2, 7.0 Hz), 3.96 (1H, d,
J ¼ 15.7 Hz), 3.60e3.55 (1H, m), 3.35e3.25 (1H, m), 3.21 (1H, dd,
J ¼ 15.7, 3.5 Hz), 2.12 (1H, dt, J ¼ 12.5, 6.4 Hz), 2.06 (1H, dt, J ¼ 9.7,
4.3 Hz), 2.03e1.98 (1H, m), 1.78 (1H, dd, J ¼ 12.2, 5.7 Hz); 13C NMR
too faint to analyse; HRMS (ES) [MþH]þ Calcd for C10H13N2O2S
225.0692 Found: 225.0701.
(e) C. Zhang, S. Murarka, D. Seidel, J. Org. Chem. 74 (2009) 419;
(f) K. Mori, Y. Ohshima, K. Ehara, T. Akiyama, Chem. Lett. 38 (2009) 524;
ꢀ
€
(g) J. Wofling, E. Frank, G. Schneider, L.F. Tietze, Angew. Chem. Int. Ed. 38
(1999) 200;
(h) P.A. Vadola, I. Carrera, D. Sames, J. Org. Chem. 77 (2012) 6689;
(i) K. Mori, T. Kawasaki, T. Akiyama, Org. Lett. 14 (2012) 1436;
(j) S. Wang, X.-D. An, S.-S. Li, X. Liu, Q. Liu, J. Xiao, Chem. Commun. 54 (2018)
13833;
(k) S.-S. Li, S. Zhu, C. Chen, K. Duan, Q. Liu, J. Xiao, Org. Lett. 21 (2019) 1058;
(l) S. Zhu, C. Chen, K. Duan, Y.-M. Sun, S.-S. Li, Q. Liu, J. Xiao, J. Org. Chem. 84
(2019) 8440 (a,b-unsaturated carbonyl acceptors);
(m) R.S. Atkinson, R.H. Green, J. Chem. Soc. Perkin Trans. 1 (1974) 394;
(n) S.J. Pastine, K.M. McQuaid, D. Sames, J. Am. Chem. Soc. 127 (2005) 12180;
(o) K.M. McQuad, J.Z. Long, D. Sames, Org. Lett. 11 (2009) 2972;
(p) J.C. Ruble, A.R. Hurd, T.A. Johnson, D.A. Sherry, M.R. Barbachyn,
P.L. Toogood, G.L. Bundy, D.R. Graber, G.M. Kamilar, J. Am. Chem. Soc. 131
(2009) 3991;
4.4. 2,2,2-Trifluoro-N-((anti)-1,2,3,3a,4,5-hexahydropyrrolo[1,2-a]
quinolin-4-yl)acetamide (29). A solution of HCl (6 M, 1.2 mL) was
added dropwise to a solution of 2 (75 mg, 0.34 mmol) in EtOAc/
EtOH (1:1, 10 mL) at 0 ꢀC. Zinc dust (223 mg) was added and the
yellow mixture was stirred for 30 min. The colourless reaction
mixture was then quenched with sat. aq. NaHCO3 (sat.) and the
resultant white precipitate filtered through a short pad of Celite and
washed with EtOAc (30 mL). The aqueous layer was separated and
extracted with EtOAc (30 mL). The combined organic layers were
dried (MgSO4) and the solvent removed in vacuo to give the crude
amine as a colourless oil which was used directly.
(q) K. Mori, T. Kawasaki, S. Sueoka, T. Akiyama, Org. Lett. 12 (2010) 1732;
(r) S. Zhu, C. Chen, M. Xiao, L. Yu, L. Wang, J. Xiao, Green Chem. 19 (2017) 653.
[3] C. Rabong, C. Hametner, K. Mereiter, V.G. Karstev, H. Jordis, Heterocycles 75
(2008) 799.
[4] J.F. Briones, G.S. Basarab, Chem. Commun. 52 (2016) 8541.
[5] (a) Y.J. Kim, Y. Kim, D.Y. Kim, Bull. Korean Chem. Soc. 38 (2017) 578;
(b) T.H. Youn, D.Y. Kim, Synth. Commun. 47 (2017) 2109.
[6] (a) J.C. Anderson, A. Noble, P. Ribelles Torres, Tetrahedron Lett. 53 (2012)
5707;
(b) J.C. Anderson, J.P. Barham, C.D. Rundell, Org. Lett. 17 (2015) 4090.
[7] J.C. Anderson, C.D. Rundell, Synlett 27 (2016) 41.
[8] A. Noble, J.C. Anderson, Chem. Rev. 113 (2013) 2887.
[9] (a) J.C. Anderson, H.A. Chapman, Org. Biomol. Chem. 5 (2007) 2413;
(b) J.C. Anderson, L.R. Horsfall, A.S. Kalogirou, M.R. Mills, G.J. Stepney,
G.J. Tizzard, J. Org. Chem. 77 (2012) 6186;
To the crude amine from above in CH2Cl2 (7.0 mL) at ꢁ78 ꢀC was
added DIPEA (0.18 mL, 1.0 mmol) and TFAA (0.14 mL, 1.0 mmol), the
mixture stirred for 10 min and then warmed to rt. The reaction
mixture was diluted with water (7.0 mL), separated and the
aqueous layer extracted with CH2Cl2 (2 x 15 mL). The combined
organic layers were dried (MgSO4), filtered and the solvent was
removed in vacuo. The resultant crude material was purified by
column chromatography (30% EtOAc in hexane) to give 29 as an off-
white low melting point waxy solid (45 mg, 46%): IR ymax (neat)
(c) J.C. Anderson, A. Noble, D.A. Tocher, J. Org. Chem. 77 (2012) 6703;
(d) J.C. Anderson, A.S. Kalogirou, M.J. Porter, G. Tizzard, J. Beil, J. Org. Chem. 9
(2013) 1737;
(e) J.C. Anderson, I.B. Campbell, S. Campos, J. Shannon, D.A. Tocher, Org. Bio-
mol. Chem. 13 (2015) 170;
(f) J.C. Anderson, I.B. Campbell, S. Campos, I.H. Reid, C.D. Rundell, J. Shannon,
G. Tizzard, J. Org. Biomol. Chem. 14 (2016) 8270;
(g) A. Abil, J.C. Anderson, M. Corpinot, E.S. Gascoigne, J. Tetrahedron 74 (2018)
5458.
3220, 3066, 2927, 2844, 171, 1602, 1555, 1502 cmꢁ1
(700 MHz, CDCl3)
6.61 (1H, t, J ¼ 7.3 Hz), 6.44 (1H, d, J ¼ 8.1 Hz), 6.12 (1H, s), 4.08 (1H,
ddd, J ¼ 11.9, 9.7, 5.1 Hz), 3.40 (1H, td, J ¼ 8.9, 2.1 Hz), 3.34 (1H, td,
J ¼ 9.9, 5.3 Hz), 3.29 (1H, dd, J ¼ 16.4, 8.8 Hz), 3.08 (1H, dd, J ¼ 15.2,
4.9 Hz), 2.84 (1H, dd, J ¼ 15.1, 11.9 Hz), 2.18e2.08 (2H, m), 1.92 (1H,
;
1H NMR
d
7.11 (1H, t, J ¼ 7.6 Hz), 6.99 (1H, d, J ¼ 7.4 Hz),
[10] (For reviews, see: ) (a) B. Nammalwar, R.A. Bunce, Molecules 19 (2014) 204;
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663

10
J.C. Anderson et al. / Tetrahedron xxx (xxxx) xxx
ꢀ
(b) V. Sridharan, P.A. Suryavanshi, J.C. Menendez, Chem. Rev. 111 (2011) 7157;
(c) Y.-G. Zhou, Acc. Chem. Res. 40 (2007) 1357;
[19] L.J. Diorazio, D.R.J. Hose, N.K. Adlington, Org. Process Res. Dev. 20 (2016) 760.
[20] A.R. Muci, S.L. Buchwald, Practical palladium catalysts for C-N and C-O bond
formation, in: N. Miyaura (Ed.), Topics in Current Chemistry, vol. 219,
Springer-Verlag, Berlin, 2002, p. 133.
[21] Crystallographic data (excluding structure factors) for the structures in this
paper have been deposited with the Cambridge Crystallographic Data Centre
as supplementary publication no. 1904994. Copies of the data can be ob-
tained, free of charge, on application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK, (fax: þ44-(0)1223e336033 or e-mail:deposit@ccdc.cam.ac.uk).
[22] J.C. Anderson, G.J. Stepney, M.R. Mills, L.R. Horsfall, A.J. Blake, W. Lewis, J. Org.
Chem. 76 (2011) 1961.
(d) A. Mitchinson, A. Nadin, J. Chem. Soc., Perkin Trans. 1 (2000) 2862;
(e) V. Kouznetsov, A. Palma, C. Ewert, A. Varlamov, J. Heterocycl. Chem. 35
(1998) 761;
(f) A.R. Katrizky, S. Rachwal, B. Rachwal, Tetrahedron 52 (1996) 15031.
[11] (a) R. Maity, S.C. Pan, Org. Biomol. Chem. 13 (2015) 6825;
(b) Z.-X. Jia, Y.-C. Luo, P.-F. Xu, Org. Lett. 13 (2011) 832.
[12] (HFIP has been recognised as being able to promote various reactions, see for
example and references there in: ) (a) J. Wencel-Delord, F.A. Colobert, Org.
Chem. Front. 3 (2016) 394;
(b) I. Colomer, A.E.R. Chamberlain, M.B. Haughey, T.J. Donohoe, Nat. Rev.
Chem. 1 (2017) 0088;
[23] N. Krogsgaard-Larsen, M. Begtrup, M. Herth, J. Kehler, Synthesis 2010 (2010)
4287.
(c) X.-D. An, J. Xiao, Chem. Rec. 19 (2019) 1.
[24] R.K. Bhaskarachar, V.G. Revanasiddappa, S. Hegde, J.P. Balakrishna, S.Y. Reddy,
Med. Chem. Res. 24 (2015) 3516.
[13] (a) S.-S. Li, X. Lv, D. Ren, C.-L. Shao, Q. Liu, J. Xiao, Chem. Sci. 9 (2018) 8253;
(b) Z. Lv, F. Hu, K. Duan, S.-S. Li, Q. Liu, J. Xiao, J. Org. Chem. 84 (2019) 1833.
[14] P.R. Schreiner, A. Wittkopp, Org. Lett. 4 (2002) 217.
[15] D. Seebach, H.F. Leitz, V. Ehrig, Chem. Ber. 108 (1975) 1924.
[16] S.P. Waters, M.W. Fennie, M.C. Kozlowski, Tetrahedron Lett. 47 (2006) 5409.
[17] D. Enders, T. Otten, Synlett (1999) 747.
[25] W. Cao, X. Liu, W. Wang, L. Lin, X. Feng, Org. Lett. 13 (2011) 600.
[26] F. Hinderer, U.H.F. Bunz, Chem. Eur J. 19 (2013) 8490.
[27] M. Shiraishi, M. Seto, K. Aikawa, N. Kanzaki, M. Baba, in: Acrylamide Deriva-
tive, Process for Producing the Same, and Use, EP1593673, 2005 (A1).
[28] S.J. Mahoney, E. Fillion, Chem. Eur J. 18 (2012) 68.
[18] (a) H. Choi, Z. Hua, I. Ojima, Org. Lett. 6 (2004) 2689;
(b) D.M. Mampreian, A.H. Hoveyda, Org. Lett. 6 (2004) 2829.
[29] A.C. Sather, O.B. Berryman, J. Rebek, Org. Lett. 14 (2012) 1600.
[30] D. Prim, G. Kirsch, Tetrahedron 55 (1999) 6511.
Please cite this article as: J.C. Anderson et al., Investigation of the [1,5]-hydride shift as a route to nitro-Mannich cyclisations, Tetrahedron,
https://doi.org/10.1016/j.tet.2019.130663