Asymmetric control in Diels-Alder cycloadditions of chiral 9-aminoanthracenes by relay of stereochemical information

Article abstract of DOI:10.1016/j.tetasy.2007.04.012

Two approaches to the synthesis of chiral 9-amino anthracenes are described. The first, by nucleophilic addition of organolithium reagents to imines promoted by BF₃·OEt₂, unexpectedly provided stable aminoboranes as products. The second approach, using palladium catalysed cross coupling, was more successful for primary amines, and the key 9-(α-methylbenzylamino)anthracene subjected to cycloadditions with N-methyl maleimide and maleic anhydride. Excellent reactivity was achieved with good levels of diastereoselectivity, through a favourable combination of electrostatic and hydrogen bonding effects. Trial studies of the retro Diels-Alder reaction of these cycloadducts were also performed.

Full text of DOI:10.1016/j.tetasy.2007.04.012

Tetrahedron: Asymmetry 18 (2007) 1003–1012
Asymmetric control in Diels–Alder cycloadditions of chiral
9-aminoanthracenes by relay of stereochemical information
Harry Adams, Ramadan A. Bawa, Keith G. McMillan and Simon Jones*
Department of Chemistry, University of Sheffield, Dainton Building, Brook Hill, Sheffield S3 7HF, UK
Received 27 March 2007; accepted 4 April 2007
Abstract—Two approaches to the synthesis of chiral 9-amino anthracenes are described. The first, by nucleophilic addition of organo-
lithium reagents to imines promoted by BF3ÆOEt2, unexpectedly provided stable aminoboranes as products. The second approach, using
palladium catalysed cross coupling, was more successful for primary amines, and the key 9-(a-methylbenzylamino)anthracene subjected
to cycloadditions with N-methyl maleimide and maleic anhydride. Excellent reactivity was achieved with good levels of diastereoselec-
tivity, through a favourable combination of electrostatic and hydrogen bonding effects. Trial studies of the retro Diels–Alder reaction of
these cycloadducts were also performed.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
asymmetric synthesis requires a key crystallisation to
enhance levels of enantioselectivity. Thus, we recently
began a programme aimed at investigating the stereocontrol
element and have prepared oxazolines, in which the stereo-
genic centre is too far from the reaction centre to control
the selectivity of the cycloaddition,3 and oxazolidines,4
whereby suitable choice of nitrogen protecting groups can
lead to a change in the sense of diastereoselection. Earlier
we had noted that 9-(1-hydroxyethyl)-anthracene 2 led to
an enhanced reaction rate and a different sense of diaste-
reoselection, presumably through hydrogen bonding
effects.5 We thus envisioned that 9-amino anthracenes
could participate in hydrogen bonding and lead to
increased reactivity of the diene through the interaction
of the lone pair with the aromatic system.6
The introduction of stereocontrol elements into reagents to
allow subsequent stereoselective transformations is still one
of the most powerful methods for introducing new stereo-
genic centres into target substrates, despite intense compe-
tition from catalytic methods. We have previously
demonstrated that excellent levels of diastereoselectivity
are observed in the Diels–Alder reaction of 9-(methoxy-
ethyl)-anthracene 1 (Fig. 1) with dienophiles, such as
maleic anhydride, maleimides and p-benzoquinone.1
1
2
MeO
HO
Figure 1. Chiral 9-substituted anthracenes.
2. Results and discussion
Research from this group and others has also demon-
strated that these and similar compounds can function as
chiral auxiliaries.1b,2 The two key issues in using these com-
pounds as auxiliaries is obtaining high diastereoselectivities
in the cycloaddition step, coupled with the ease of synthesis
of the chiral anthracene. Although parent ether 1 has
always given exceptionally high diastereoselectivity, its
2.1. Synthesis of rac-9-(N-a-methylbenzylamino)anthracene
10
There are a number of existing ways to construct the target
molecule, however, we initially chose to access racemic
material in order to assess the viability of the reaction by
a four-step synthetic route via key imine 6. Thus, nitration
of anthracene using concd HNO3/HCl in glacial acetic
acid7 afforded a pale yellow precipitate of 9-nitro-10-
chloro-9,10-dihydroanthracene 3 that was converted to
*
Corresponding author. Tel.: +44 0114 222 9483; fax: +44 0114 222
9346; e-mail: simon.jones@sheffield.ac.uk
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.04.012

1004
H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
9-nitroanthracene 4 after treatment with aqueous NaOH.
This was reduced with a suspension of SnCl2 in concd
HCl to give a yellow precipitate of the desired 9-amino-
anthracene 5,8 which was condensed with benzaldehyde
in toluene to afford target compound 6 as an orange solid
in 78% yield (Scheme 1).
mal decomposition of aminoboranes in sealed tubes under
high vacuum,9 by the reaction of metalloborohydrides and
ammonium halides,10 and by the reaction of N-disubsti-
tuted aminodichloroboranes with an excess of Grignard
reagents.11 However, this remains the first example of
anthrylaminoboranes prepared to date.
2.2. Preparation of (R)-9-(N-a-methylbenzylamino)anthra-
cene 10
NO2
NO2
(i)
(ii)
67%
over 2 steps
Since preparation via organometallic addition appeared to
be unviable, transition metal-catalysed formation of the
key carbon–nitrogen bond was considered as an alternative
strategy.12 This had the added advantage of using
commercial, enantiomerically pure chiral amines, negating
a resolution or asymmetric synthesis step in our route. A
cross-coupling reaction of (R)-a-methylbenzylamine 9 with
9-bromoanthracene was carried out using Pd(OAc)2
(4.8 mol % Pd) and (ꢀ)-BINAP (16 mol %) as a catalyst
system to form 9-N-a-methylbenzylaminoanthracene 10
in 100% conversion within 21 h and 84% yield with 93%
ee. The enantiomeric excess of the product was slightly
lower than the precursor primary amine 9 (98% ee), the
partial racemisation in such reactions being previously
noted by Buchwald et al.13 (Scheme 3).
Cl
3
4
Ph
N
NH2
(iv)
(iii)
83%
78%
5
6
Scheme 1. Reagents and conditions: (i) concd HNO3/HCl, <30 ꢁC; (ii)
NaOH (10%), 60–70 ꢁC; (iii) SnCl2/concd HCl, glacial HOAc, 70–80 ꢁC;
(iv) PhCHO, toluene, reflux.
In an initial attempt to add a methyl group across the C@N
bond, methylmagnesium bromide and methyl lithium were
added to imine 6 leading to the recovery of the starting
material. In order to increase the reactivity, BF3ÆOEt2
was added prior to the addition of methyl lithium at
ꢀ78 ꢁC. The resulting mixture was stirred for 21 h at room
temperature giving complete conversion to a single compo-
nent, which unexpectedly contained three methyl groups
clearly visible in the 1H NMR spectrum. Surprisingly,
two of these methyl groups could not be detected by
13C NMR spectroscopy. Similar results were obtained when
using n-BuLi, giving complete conversion to a single com-
ponent containing extra two butyl groups from the 1H
NMR spectrum, with two methylene signals missing from
the 13C NMR spectrum. A 13C NMR experiment at
400 MHz revealed the two extra methyl and methylene
groups as broad signals with very low intensity, suggesting
the interaction with a proximal boron atom through quad-
rupole relaxation. This implied that the compounds
obtained were anthrylaminodialkylboranes 7 and 8 formed
by nucleophilic attack on the intermediate aminoborane
complex, which was supported by 11B NMR spectroscopy
and high resolution mass spectroscopy (Scheme 2). This
class of compounds has previously been prepared by ther-
Ph
Br
HN
Me
Me
NH2
(i)
+
Ph
84%
93% ee
9
10
Scheme 3. Reagents and conditions: (i) 4.8% Pd(OAc)2, 16% (ꢀ)-BINAP,
t-BuONa, toluene, 90–95 ꢁC.
2.3. Preparation of (R)-9-(N-methyl-a-
methylbenzylamino)anthracene
Since our previous studies have shown that the proximity
of the stereodirecting group has important consequences
on the level of diastereoselectivity observed, analogues
bearing alkyl groups at the nitrogen centre were also con-
sidered. Attempts to prepare the N-methyl analogue by a
similar palladium catalysed cross coupling with (R)-N-
methyl-a-methylbenzylamine using Pd(OAc)2/(ꢀ)-BINAP
as a catalyst combination in the presence of t-BuONa with
toluene as the solvent led to recovery of the starting mate-
rials. This reaction was repeated using a variety of mono-
and bidentate phosphine ligands [Pd2(dba)3/P(o-tolyl)3,
Pd(OAc)2/(ꢀ)-BINAP, Pd2(dba)3/(ꢀ)-BINAP, Pd2(dba)3/
dppe, Pd2(dba)3/(o-tolyl)3P, Pd(Ph3)4/PPh3], with starting
materials being recovered in all cases. However, a combina-
tion of Pd2(dba)3 and Buchwald’s phosphine ligand 1314
produced 20% conversion of the starting materials to the
target amino anthracene 11. Surprisingly, unlike the
ligands previously used, ligand 13 gave rise to both imine
12 and anthracene via a b-hydride elimination pathway
Ph
Ph
F
B
N
N
R
F
(i)
(ii)
6
R= Me or n-Bu
Ph
R
B
N
R
R
1
(Scheme 4). The H NMR data of the crude material also
7 R= Me 32% yield
8 R= -Bu 70% yield
revealed that imine 12 was formed as the E-isomer.15
n
Since this aminophosphine ligand appeared to give promis-
ing results, other related phosphine–nitrogen containing
Scheme 2. Reagents and conditions: (i) MeLi or n-BuLi, toluene, ꢀ78 ꢁC,
BF3ÆOEt2, 24 h; (ii) aq NaHCO3, rt.

H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
1005
Br
NMR data of the crude material also revealed that the
imine was formed as the E-isomer.
Me
Ph
Me
(i)
+
N
H
Although compound 11 was made in 16% yield, it was
impractical to carry out any further transformations with
such a low yield. Attempts to improve the yield by employ-
ing a deptrotonation/alkylation strategy with compound
10 failed, as did use of alternative alkylation reagents such
as benzyl bromide. In view of this, only auxiliary 10 was
taken forward for evaluation.18
Ph
Me
Me
N
Me
Me
N
+
+
Ph
12
11
O
2.4. Diels–Alder cycloaddition of (R)-9-(N-a-methylbenz-
ylamino) anthracene 10 with N-methylmaleimide and maleic
anhydride
PCy2
Me2N
N
PPh2
Fe
14
13
As the reactivity and diastereoselectivity of methyl ether 1
in Diels–Alder cycloadditions with N-methylmaleimide
have been previously reported,1b this was used as a bench-
mark for evaluating the chiral anthracene prepared. Amine
10 was reacted with 1 equiv of N-methylmaleimide and
maleic anhydride for 2 h at 80–85 ꢁC, in each case, produc-
ing a mixture of two diastereoisomers in excellent conver-
sion and good selectivity (Scheme 6, Table 1).
Scheme 4. Reagents and conditions: (i) 6% Pd2(dba)3/6% 13 or 14,
t-BuONa, toluene, 80–85 ꢁC.
ligands were considered. Thus, oxazolinylphosphine ligand
14 was prepared via a three-step sequence following a liter-
ature procedure16 and used with Pd2(dba)3 to give target
aminoanthracene 11 in ca. 20% conversion, but with no
formation of anthracene or imine as a side product.
Although not pursued, optimisation of this class of ferrocen-
yl ligands may be useful in the development of catalyst sys-
tems that do not result in the decomposition of valuable
starting materials by b-hydride elimination.
Ph
HN
Me
O
X
+
O
Since b-hydride elimination tends to be exasperated by
electron rich aryl halides, it was thought that the electronic
density around the anthracene ring may be reduced by hav-
ing another withdrawing group, such as a bromine atom,
attached to C10 of the anthracene ring.17 Theoretically,
this could lead to the formation of 9,10-diaminoanthracene
17. When the cross-coupling reaction of 9,10-dibromoan-
thracene and secondary amine 14 was carried out using
Pd2(dba)3/ligand 13 in the presence of NaOt-Bu using tol-
uene as a solvent, aminoanthracene 11 was formed (17%
conversion). No formation of 9,10-di-(N-methyl-a-meth-
ylbenzylamino)anthracene 17 was observed which could
be attributed to the fact that intermediate 16 is electron-
rich with b-hydride elimination taking place to form amine
10
X=O
X=NMe
(i)
O
O
X
X
O
O
+
HN
HN
Me
Ph
Me
Ph
18a,b
19a,b
Scheme 6. Reagents and conditions: (i) toluene, 80–85 ꢁC, 2 h.
Table 1. Reactions of amine 10 with maleic anhydride and N-methyl
1
11 along with the imine (Scheme 5). Once again, the H
maleimides
Conversiona Diastereomeric Yield (major isomer)b
(%)
Ph
ratioa
Br
Br
Me
Me
Ph
N
Me
(i)
X = NMe 100
18a:19a = 92:8 89%
18b:19b = 91:9 Not determined
Me
+
N
X = O
100
H
a Calculated from the integrals of appropriate signals in the 1H NMR
spectrum.
b Based on isolated product.
Br
16
15
Ph
Me
Ph
Me
Me
Me
N
N
N
Me
The reactivity of this amine was found to be higher than
the reactivity of ether 1 and neither temperature nor reac-
tion time had an effect on the diastereoselectivity. In each
case, scalemic mixtures of 18a, 19a and 18b, 19b (76:24
and 86:14, respectively) obtained after partial diastereo-
meric enhancement by recrystallisation were stirred for
2 h in toluene at 80–85 ꢁC. No change in the diastereomeric
ratio was observed indicating that these reactions were
Me
Me
+
N
+
Ph
11
Me
17
Ph
Scheme 5. Reagents and conditions: (i) 4% Pd2(dba)3/8% 13, t-BuONa,
toluene, 85 ꢁC.

1006
H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
irreversible and hence kinetically controlled. In the case of
the maleimide reaction, effective recrystallisation allowed
the identity of the major diastereoisomer 18a to be deter-
mined by single crystal X-ray diffraction.19 Although sepa-
ration was more difficult and not preparatively useful,
repeated crystallisation of the maleic anhydride adducts
allowed isolation of the major and minor isomers, the
identity of which was again confirmed by single crystal
X-ray diffraction (Figs. 2–4).
specific rotation with that in the literature, however this
was not conducted in the same solvent, which may have
led to the discrepancy reported.20
The origin of the stereoselectivity is very unlikely to result
from the direct interaction of the stereogenic centre with
the dienophile, since this group is located too far away
from the approaching dienophile, essentially occupying
space beneath the anthracene ring system, exemplified by
the single crystal X-ray structure (Figs. 5 and 6).
It should be noted that the stereochemical outcome is
opposite to that which has previously been reported.6 In
earlier work, the absolute sense of diastereroselection was
confirmed by conversion of the anhydride cycloadduct into
a lactone, whose absolute configuration was known. The
stereochemistry was then assigned by comparison of the
However, the reaction can be rationalised via a relay of ste-
reochemical information from the fixed stereogenic centre
via the stereochemically labile amine atom. The dienophile
approaches from the least hindered face of aminoanthra-
cene 10 opposite to the a-methylbenzyl group (models A
Figure 2. Crystal structure of (3aS,9aS)-3a,4,9,9a-tetrahydro-4-[(R)-a-methylbenzylamino]-2-methyl-4,9-[10,20]benzeno-1H-benzo[f]isoindole-1,3-(2H)-
dione 18a. Thermal ellipsoids shown at 50% probability.
Figure 3. Crystal structure of (11S,15S)-9,10,11,15-tetrahydro-9-[(R)-a-
methylbenzylamino]-9,10[30,40]-furanoanthracene-12,14-dione 18b. Ther-
mal ellipsoids shown at 50% probability.
Figure 4. Crystal structure of (11R,15R)-9,10,11,15-tetrahydro-9-[(R)-a-
methylbenzylamino]-9,10[30,40]-furanoanthracene-12,14-dione 19b. Ther-
mal ellipsoids shown at 50% probability.

H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
1007
importance of a combination of electrostatic repulsion
and attractive hydrogen bonding in establishing high levels
of stereoselection. This is further exemplified by the
Diels–Alder reactions that have been reported with
N-methyl analogue 11. Here there is no hydrogen bond,
instead relying on repulsive effects from the lone pair and
methyl group, resulting in a very poor diastereoselectivity
(1.2:1).2
2.5. Retro Diels–Alder reaction of adduct 18a
Attempts at a retro Diels–Alder reaction were carried out
on adduct 18a under basic and acidic conditions. This reac-
tion was first conducted under basic conditions using a
number of strong bases (NaH, LDA and t-BuLi) in an
attempt to deprotonate the amino group, thus increasing
the electron density and effecting an efficient retro Diels–
Alder reaction.21 However, all these bases returned unre-
acted starting material, indicating that the amine proton
might be too sterically hindered to be removed. Promoting
the retro Diels–Alder reaction by simply heating at reflux
was thought to be hopeless, as amine 10 showed high reac-
tivity in the competing forward Diels–Alder reaction.
Therefore, adduct 18a was refluxed in toluene in the pres-
ence of trifluoroacetic acid in order to trap any amine 10
that may form from a retro Diels–Alder reaction as an
ammonium salt, leading to an electron-deficient anthracene
derivative, which would in turn slow the forward Diels–
Alder reaction and allow the isolation of the N-methylma-
leimide and ammonium salt of the auxiliary which could be
recovered.
Figure 5. Crystal structure of 9-[(R)-a-methylbenzylamino]anthracene 10.
Thermal ellipsoids shown at 50% probability.
Figure 6. Space-filling representation of the crystal structure of 9-[(R)-a-
methylbenzylamino]anthracene 10.
However, the product of this reaction anthrone 20 was
formed after elimination of the stereodirecting group and
partial cleavage of the N-methylsuccinamide unit. This
probably results from protonation of the carbonyl group
leading to cleavage of the bond between C10 and C11.
Hydrolysis of the resulting iminium ion 21 by water would
give anthrone 20 (Scheme 7).
and B) (Fig. 7). The diastereoselection can then be ex-
plained by the dienophile approaching the anthracene ring
with the carbonyl oxygen orientated towards the hydrogen
of the amino group, thus forming a hydrogen bond, and
additionally avoiding possible electrostatic repulsion with
the nitrogen lone pair (model A). This leads to the forma-
tion of isomers 18a and 18b as the major diastereoisomers.
The minor diastereoisomers 19a and 19b are formed when
the dienophile attacks the anthracene ring with the car-
bonyl oxygen orientated towards the nitrogen lone pair
(model B).
In an attempt to confirm this, a 1H NMR experiment
involving heating adduct 18a at 90 ꢁC in deuterated toluene
in the presence of trifluoroacetic acid was carried out. After
3 h, the formation of anthrone 20 along with (R)-a-meth-
ylbenzyl ammonium trifluoroacetate was seen in the 1H
NMR spectrum. This finding supports the proposed
mechanism, with the water coming from the deuterated
chloroform.
Given the relatively small size difference between the lone
pair and the hydrogen atom, this example highlights the
O
O
O
X
O
O
X
O
X
X
O
O
H
H
N
N
HN
HN
Me
R
R
Me
Ph
Ph
model B
model A
Me
19a, b (minor)
18a, b (major)
R =
Ph
X=O, NMe
Figure 7. Possible models to explain the diastereoselection.

1008
H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
OH
Me
O
N
O
O
Me
NMe
O
N
H
H
O
H
HN
Me
HN
Me
CF3COO
Me
NH
Ph
Ph
18a
Ph
O
O
Me
O
Me
O
N
N
H
H
Me
Ph
H2O
CF3COO
H3N
+
CF3COO
Me
NH
O
20
21
Ph
Scheme 7. Possible mechanism for the formation of anthrone 20.
3. Conclusion
red spectra were recorded on a Perkin–Elmer Spectrum
RX/FT-IR system with a DuraSamplIR II ATR accessory.
250 MHz and 300 MHz 1H NMR were carried out on a AC-
250 or Avance 300 instrument, respectively, supported by
an Aspect 200 or Aspect 300 data system. 13C NMR
(100 MHz) was carried out on a Bruker AMX-400 spec-
trometer. Residual proton signals from the deuterated sol-
We have shown that 9-anthryl amines can be readily pre-
pared via palladium catalysed coupling reactions, and in
doing so we have discovered new routes for preparing an-
thryl aminoboranes and new ligands for the N-arylation
reaction that appear to inhibit competitive b-hydride elim-
ination. The 9-anthryl amine produced displayed excellent
reactivity and very good diastereoselectivity in trial Diels–
Alder reactions, the reaction being controlled by a through
space relay from the adjacent fixed stereodirecting group.
Our results on the sense of diastereoselection contradict
those previously reported by Snyder, although this is prob-
ably due to an error made in comparison of specific rota-
tions with known compounds. The main issue with the
methodology is the problem encountered in performing
the retro Diels–Alder reaction required for use as an effec-
tive auxiliary.
vents were used as references [chloroform (1H 7.25 ppm, 13
C
77 ppm)]. Coupling constants were measured in Hz. Mass
spectra were recorded on a Micromass Autospec M spec-
trometer. Elemental microanalysis was performed using a
Perkin–Elmer 2400 CHNS/0 Series II elemental analyser.
Data collected were measured on a Bruker Smart CCD
area detector with Oxford Cryosystems low temperature
system. Single crystals of compounds 10, 18a, 18b and
19b were grown by liquid diffusion from dichlorometh-
ane/petroleum ether (40–60 ꢁC), mounted in inert oil and
transferred to the cold gas stream of the diffractometer.
4.3. Preparation of 9-nitroanthracene 47
4. Experimental
4.1. Materials
Concentrated nitric acid (4 cm3) was added dropwise to a
suspension of anthracene (10.0 g, 56.0 mmol) in glacial ace-
tic acid (40 cm3) maintaining the temperature below 30 ꢁC.
This was stirred vigorously for 1 h to form a clear solution.
A mixture of concentrated HCl (50 cm3) and glacial acetic
acid (50 cm3) was added slowly resulting in a pale yellow
precipitate of 9-nitro-10-chloro-9,10-dihydroanthracene 3.
This was filtered, washed with glacial acetic acid
(3 · 25 cm3) and thoroughly with water until the washings
were neutral. The resulting yellow solid was treated with a
warm solution (60–70 ꢁC) of 10% NaOH (200 cm3), fil-
tered, washed with warm water until the washings were
neutral, air-dried and recrystallised from glacial acetic acid
affording a fluffy yellow solid (8.31 g, 67% yield), mp 153–
157 ꢁC (acetic acid) (lit.7 148–149 ꢁC acetic acid); dH
(250 MHz; CDCl3) 8.59 (1H, s, ArCH), 8.03 (2H, d, J
7.6, ArCH), 7.92 (2H, d, J 7.6, ArCH), 7.68–7.52 (4H, m,
ArCH); dC (62.5 MHz; CDCl3) 130.7 (ArCNO2), 130.4
The solvents utilised were dried as appropriate, unless
otherwise stated. THF, toluene and ether were dried over
sodium, whereas CH2Cl2 was dried over calcium hydride.
Glassware was flame-dried and dried under vacuum before
use. TLC was carried out using Fluorochem Limited Silica
Gel 40–63u 60A. Visualisation of the TLC plates was
achieved by employing UV lamps at 254 nm and 356 nm
or KMnO4 and then heating.
4.2. Instrumentation
Melting points were measured on a Gallenkamp apparatus
and are uncorrected. Specific rotations were performed on
an Optical Activity LTD. AA-10 automatic polarimeter at
589 nm (Na D-line) and measured at 20 ꢁC unless otherwise
stated. [a]D values are given in 10ꢀ1 deg cm2 gꢀ1. All infra-

H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
1009
(2 · ArC), 128.9 (2 · ArC), 128.4 (Ar-C), 126.2 (2 · ArC),
122.6 (2 · ArC), 121.3 (4 · ArC). NMR data were in accor-
dance with the literature.22
were combined, washed with brine (2 · 30 cm3), dried over
Na2SO4, filtered and the solvent was removed in vacuo.
The crude product was purified by column chromatogra-
phy (100% petroleum ether 40–60 ꢁC) to give the title com-
pound as a yellow thick oil (0.27 g, 32% yield), mmax ATR/
cmꢀ1 1670; dH (250 MHz; CDCl3) 8.29 (1H, s, ArCH),
8.22–8.16 (1H, m, ArCH), 8.03–7.94 (1H, m, ArCH),
7.79 (1H, d, J 8.6, ArCH), 7.49–7.40 (2H, m, ArCH),
7.25–7.14 (2H, m, ArCH), 6.96 (1H, m, ArCH), 6.89–
6.80 (3H, m, ArCH), 6.67–6.64 (2H, m, ArCH), 5.60 (1H,
q, J 7.3, NCHCH3), 1.39 (3H, d, J 7.3, CHCH3), 0.75
(3H, s, BCH3), ꢀ0.65 (3H, s, BCH3); dC (100 MHz; CDCl3)
142.5 (ArC), 138.2 (ArC), 131.7 (ArC), 131.6 (ArC), 130.4
(ArC), 130.0 (ArC), 128.6 (2 · ArCH), 128.4 (ArCH),
127.5 (2 · ArCH), 127.4 (ArCH), 126.8 (ArCH), 125.6
(ArCH), 125.1 (ArCH), 125.0 (2 · ArCH), 124.8 (ArCH),
124.5 (ArCH), 123.9 (ArCH), 60.5 (CH), 21.7 (CH3), 6.8
(br s, BCH3), 5.8 (br s, BCH3); dB (128.4 MHz; CDCl3)
48.1 (br s, NBMe2); m/z (EI+) 338 (12%), 337.2008 (46,
C24H2411BN requires 337.2002); 336 (10), 233 (46), 232
(100), 231 (50), 218 (22), 217 (78), 216 (34), 105 (65).
4.4. Preparation of 9-aminoanthracene 58
A suspension of 9-nitroanthracene 4 (7.24 g, 32.5 mmol) in
glacial acetic acid (145 cm3) was heated to 70–80 ꢁC for
1.5 h. To the resulting clear solution was added a slurry of
SnCl2 (31.0 g, 163.2 mmol) in concentrated HCl (110 cm3)
via a dropping funnel. The resulting yellow precipitate was
stirred at 80 ꢁC for a further 30 min, cooled to room temper-
ature, filtered, washed with concentrated HCl (3 · 10 cm3),
treated with a solution of 5% NaOH for approximately
15 min with manual stirring from time to time, filtered,
washed thoroughly with water until the washings were neu-
tral and vacuum-dried at 50 ꢁC for 6 h to afford a yellow
powder (5.18 g, 83% yield). No further purification was
required, mp 165–170 ꢁC (lit.8 153–154 ꢁC, benzene); dH
(250 MHz; CDCl3) 7.90 (4H, m, ArCH), 7.85 (1H, s,
ArCH), 7.43 (4H, m, ArCH), 4.85 (2H, br s, NH2); dC
(62.5 MHz; CDCl3) 129.0 (2 · ArC), 125.2 (4 · ArC),
123.8 (2 · ArC), 121.1 (2 · ArC), 116.3 (4 · ArC). NMR
data were in accordance with the literature.8
4.7. Preparation of rac-N-(9-anthryl)-a-butylbenzylamino-
N-dibutylborane 8
4.5. Preparation of 9-N-anthrylbenzylidene 6
9-N-Anthrylbenzylidene 6 (0.20 g, 0.71 mmol) was dis-
solved in dry toluene (10 cm3), cooled to ꢀ20 ꢁC and
BF3ÆOEt2 (0.2 cm3, 1.64 mmol) was added. The reaction
mixture was stirred for 5 min at ꢀ20 ꢁC, cooled to ꢀ78 ꢁC
and n-BuLi (1.2 cm3, 3.00 mmol) was added dropwise.
The resulting mixture was stirred for 6 h, allowed to warm
to room temperature and left stirring for a further 12 h. The
reaction was quenched with satd aq NaHCO3 (10 cm3). The
organic layer was separated and the aqueous layer extracted
with EtOAc (3 · 10 cm3). The organic phases were com-
bined, washed with brine (2 · 20 cm3), dried over Na2SO4,
filtered and the solvent was removed in vacuo. The crude
product was purified by column chromatography (100%
petroleum ether 40–60 ꢁC) to give the title compound as a
yellow thick oil (0.23 g, 70% yield), mmax ATR/cmꢀ1 1520;
dH (250 MHz; CDCl3) 8.32–8.27 (2H, m, ArCH), 8.02–
7.98 (1H, m, ArCH), 7.74 (1H, d, J 8.4, ArCH), 7.49–7.43
(2H, m, ArCH), 7.28 (1H, d, J 8.4, ArCH), 7.14 (1H, app.
t, J 8.4, ArCH), 6.86–6.80 (2H, m, ArCH), 6.70 (2H, app.
t, J 8.4, ArCH), 6.40 (2H, d, J 7.0, ArCH), 5.37 (1H, dd,
J 11.5, 4.4, CH), 2.15 (1H, m, CHH), 1.91–1.59 (5H, m,
2 · CH2 and CHH), 1.55–1.49 (2H, m, CH2), 1.35–0.93
[8H, m, 4 · CH2], 0.91–0.74 [6H, m, 3 · CH2], 0.48 (3H, t,
J 7.2, CH3), 0.19 (2H, dd, J 9.3, 6.9, BCH2); dC
(100 MHz; CDCl3) 139.6 (ArC), 138.3 (ArC), 131.6
(ArC), 131.4 (ArC), 130.4 (ArC), 130.3 (ArC), 129.6
(2 · ArCH), 128.3 (ArCH), 127.2 (ArCH), 127.0
(2 · ArCH), 126.6 (ArCH), 125.7 (ArCH), 125.2 (ArCH),
125.1 (ArCH), 124.9 (ArCH), 124.8 (ArCH), 124.5 (ArCH),
124.4 (ArCH), 123.8 (ArCH), 66.0 (CH), 36.4 (CH2), 31.9
(CH2), 29.7 (CH2), 29.4 (CH2), 28.4 (CH2), 28.0 (CH2),
26.8 (CH2), 25.8 (CH2), 22.5 (CH2), 20.7 (br s, BCH2),
18.8 (br s, BCH2), 14.3 (CH3), 14.0 (CH3), 13.8 (CH3); dB
(128.4 MHz; CDCl3) 49.2 [br s, NB(n-Bu)2]; m/z (EI+) 464
(16%), 463.3420 (47, C33H4211BN requires 463.3410); 462
(13), 407 (12), 406 (16), 317 (44), 316 (82), 315 (26), 260
(67), 259 (74), 204 (42), 203 (64), 147 (68), 91 (100).
A mixture of benzaldehyde (4 cm3, 39.5 mmol) and
9-aminoanthracene
5 (3.85 g, 20.0 mmol) in toluene
(40 cm3) was refluxed for 24 h. The reaction mixture was
allowed to cool to room temperature and the solvent was
removed under reduced pressure to give a yellow-brown
solid. The crude product was purified by column chroma-
tography (5% EtOAc/petroleum ether 40–60 ꢁC) to afford
the target compound as an orange solid (4.36 g, 78% yield),
mp 138–142 ꢁC; (Found: C, 89.8; H, 5.4; N, 4.9. C21H15N
requires C, 89.7; H, 5.4; N, 5.0); mmax (ATR)/cmꢀ1 1630;
dH (250 MHz; CDCl3) 8.52 (1H, s, PhCH@N), 8.35 (1H,
s, ArCH), 8.13–8.07 (2H, m, ArCH), 8.05–7.95 (4H, m,
ArCH), 7.64–7.53 (3H, m, ArCH), 7.53–7.60 (4H, m,
ArCH); dC (62.5 MHz; CDCl3) 165.3 (C@N), 146.0
(ArC), 135.9 (Ar-C), 132.0 (2 · ArCH), 131.9 (2 · ArC),
129.0 (2 · ArCH), 128.2 (2 · ArCH), 125.5 (2 · ArCH),
124.9 (2 · ArCH), 124.0 (2 · ArCH), 122.2 (2 · ArCH),
122.0 (2 · ArC); m/z (ES+) 282.1281 (100%, C21H16N
requires 282.1283); (EI+) 282 (28), 281 (100), 280 (36),
204 (16), 203 (12), 178 (25), 177 (22), 176 (24), 151 (15),
140 (14), 126 (6), 103 (8).
4.6. Preparation of rac-N-(9-anthryl)-a-methylbenzylamino-
N-dimethylborane 7
9-N-Anthrylbenzylidene 6 (0.70 g, 2.50 mmol) was dis-
solved in dry toluene (30 cm3), cooled to ꢀ20 ꢁC and
BF3ÆOEt2 (0.7 cm3, 5.60 mmol) was then added. The reac-
tion mixture was stirred for 5 min at ꢀ20 ꢁC, cooled to
ꢀ78 ꢁC and MeLi (7 cm3, 9.80 mmol) was added dropwise.
The resulting mixture was stirred for 6 h, allowed to warm
to room temperature and stirred for a further 12 h. The
reaction was quenched with satd aq NaHCO3 (20 cm3),
the organic layer was separated and the aqueous layer
was extracted with EtOAc (3 · 10 cm3). The organic phases

1010
H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
4.8. Preparation of (R)-9-(N-a-methylbenzylamino)anthra-
5 h, and then allowed to stir overnight at room tempera-
ture. The excess of LiAlH4 was destroyed with water
(10 cm3), followed by adding diethyl ether (15 cm3), after
which the reaction mixture was filtered through Celite.
The organic layer was separated, dried over Na2SO4, fil-
tered and the solvent removed to afford an oil which was
dried in vacuo to give N-methyl-a-methylbenzylamine
(0.61 g, 61% yield). No further purification was required,
[a]D = +77.7 (c 1, CHCl3) [lit.24 +78.3 (c 2.18, CHCl3)];
dH (300 MHz; CDCl3) 7.27–7.17 (5H, m, ArCH), 3.57
(1H, q, J 6.1, CH), 2.24 (3H, s, NCH3), 1.37 (3H, d, J
6.1, CH3); dC (125 MHz; CDCl3) 145.4 (ArC), 128.4
(2 · ArCH), 126.9 (ArCH), 126.6 (2 · ArCH), 60.2 (CH),
34.5 (NCH3), 22.8 (CH3). NMR data were in accordance
with the literature.24
cene 1023
A mixture of (ꢀ)-BINAP (0.01 g, 0.016 mmol, 16 mol %)
and Pd(OAc)2 (0.0012 g, 0.0048 mmol, 4.8 mol %) was stir-
red in dry toluene (2 cm3) for 5 min at room temperature.
9-Bromoanthracene (0.51 g, 2.00 mmol), (R)-a-methyl-
benzylamine (0.4 cm3, 3.00 mmol), NaOt-Bu (0.27 g,
2.80 mmol) and dry toluene (2 cm3) were added. The reac-
tion mixture was stirred for 21 h at 90–95 ꢁC and cooled to
room temperature. The reaction was then diluted with
diethyl ether (10 cm3), filtered though Celite, concentrated
and dried in vacuo. The crude product was purified by
column chromatography (10% EtOAc/petroleum ether
40–60 ꢁC) to afford 0.50 g (84% yield) of the dried product,
mp 134–136 ꢁC; [a]D +198 (c 0.5, CHCl3); mmax (ATR)/
cmꢀ1 1556, 1411, 1354; dH (250 MHz; CDCl3) 8.16–8.12
(2H, m, ArCH), 8.13 (1H, s, ArCH), 7.98–7.94 (2H, m,
ArCH), 7.46–7.23 (9H, m, ArCH), 4.73 (1H, q, J 6.7,
CH), 4.24 (1H, br s, NH), 1.57 (3H, d, J 6.7, CH3); dC
(62.5 MHz; CDCl3) 144.9 (ArC), 140.3 (ArC), 132.2
(2 · ArC), 128.8 (2 · ArCH), 128.6 (2 · ArCH), 127.1
(ArCH), 126.2 (2 · ArCH), 126.0 (ArC), 125.1 (2 · ArCH),
124.6 (2 · ArCH), 123.3 (ArCH), 121.5 (ArCH), 60.3 (CH),
23.1 (CH3); m/z (ES+) 298.1602 (100%, C22H20N requires
298.1596); m/z (EI+) 298 (12), 297 (45), 204 (9), 193
(100), 192 (60), 177 (9), 165 (52), 151 (6), 139 (6), 105
(59), 91 (9).
4.11. Preparation of 9-[N-methyl-N-(R)-a-methylbenzyl-
amino]anthracene 1123
A mixture of Pd2(dba)3 (0.06 g, 0.06 mmol, Pd 6 mol %)
and (S)-2-[(S)-2-(diphenylphosphino)ferrocenyl]-4-(1-meth-
ylethyl)oxazoline 14 (0.03 g, 0.06 mmol, 6 mol %) in dry
toluene (3 cm3) was stirred for 5 min at room temperature.
9-Bromoanthracene (0.51 g, 2.00 mmol), N-methyl-N-a-
methylbenzylamine (0.41 g, 3.00 mmol), NaOt-Bu (0.27 g,
2.80 mmol) and dry toluene (3 cm3) were added and the
reaction mixture was refluxed for 24 h. The reaction was
allowed to cool to room temperature, after which diethyl
ether (10 cm3) was added, and the diluted mixture passed
through a short plug of Celite. The solvent was removed
under reduced pressure to give the crude material (25%
conversion). This material was purified by column chroma-
tography (neat petroleum ether 40–60 ꢁC) to afford a yel-
low solid of the title compound (0.10 g, 16% yield), mp
4.9. Preparation of (R)-(+)-a-methylbenzyl formamide24
(R)-(+)-a-Methylbenzylamine (1.0 cm3, 7.83 mmol) was
dissolved in THF (15 cm3) at 50 ꢁC in the presence of
Amberlyst (ca. 0.05 g) as
a catalyst. Ethylformate
(4.8 cm3, 59.5 mmol) was added to the reaction mixture
and then heated at reflux for 21 h. This was allowed to cool
to room temperature, diethyl ether (15 cm3) added, the cat-
alyst filtered and the solvent removed in vacuo. (R)-(+)-a-
Methylbenzylformamide was obtained as an oily residue,
which did not require purification (1.15 g, 99% yield),
[a]D = +168.6 (c 1, CHCl3), [lit.24 +171.5 (c 6.07, CHCl3)];
dH (300 MHz; CDCl3) 8.06 (1H, s, CHO), 7.26–7.22 (5H,
m, ArCH), 6.32–5.80 (1H, br s, NH), 5.12 (0.8H, quintet,
J 7.2, CH major rotamer), 4.63 (0.2H, quintet, J 7.2, CH
minor rotamer), 1.49 (0.6H, d, J 7.2, CH3 minor rotamer),
1.45 (2.4H, d, J 7.2, CH3 major rotamer); dC (125 MHz;
CDCl3) 164.1 (CHO minor rotamer), 160.3 (CHO, major
rotamer), 142.8 (ArC, minor rotamer), 142.6 (ArC, major
rotamer), 128.8 (ArCH, minor rotamer), 128.6 (ArCH, ma-
jor rotamer), 127.7 (ArCH, minor rotamer), 127.4, 126.0
(ArCH, major rotamer), 125.7 (ArCH, minor rotamer),
51.6 (CH, minor rotamer), 47.5 (CH, major rotamer),
23.5 (CH3, minor rotamer), 21.7 (CH3, major rotamer).
NMR data were in accordance with the literature.24
120–121 ꢁC; [a]D = +70 (c 1, CHCl3); mmax (ATR/cmꢀ1
)
2920, 1672, 1590, 1492, 1450; dH (250 MHz; CDCl3) 9.00
(1H, d, J 8.6, ArCH), 8.40 (1H, s, ArCH), 8.30 (1H, app.
d, J 9.5, ArCH), 8.12–8.05 (2H, m, ArCH), 7.77–7.73
(2H, m, ArCH), 7.67–7.58 (1H, m, ArCH), 7.58–7.47
(5H, m, ArCH), 7.41–7.35 (1H, m, ArCH), 5.03 (1H, q, J
6.7, CHCH3), 3.04 (3H, s, NCH3), 1.09 (3H, d, J 6.7,
CH3CH); dC (62.5 MHz; CDCl3) 146.4 (ArC), 143.6
(ArC), 132.9 (ArC), 132.6 (ArC), 132.5 (ArC), 131.4
(ArC), 129.7 (ArCH), 128.6 (ArCH), 128.5 (ArCH), 127.7
(2 · ArCH), 127.2 (2 · ArCH), 125.6 (2 · ArCH), 125.4
(ArCH), 125.3 (ArCH), 125.0 (ArCH), 124.8 (ArCH),
124.6 (ArCH), 64.2 (CH), 41.6 (NCH3), 22.7 (CH3); m/z
(EI+) 311.1661 (22%, C23H21N requires 311.1674), 268
(100), 267 (25), 265 (17), 208 (36), 206 (43), 191 (32), 180
(34), 178 (32), 152 (34), 105 (45).
4.12. Preparation of 9-[N-methyl-N-(R)-a-methylbenzyl-
amino]anthracene 11 (from 9,10-dibromoanthracene)
A mixture of Pd2(dba)3 (0.04 g, 0.04 mmol, Pd 4 mol %)
and 2,20-dimethylaminodicyclohexylphosphinobiphenyl 13
(0.03 g, 0.08 mmol, 8 mol %) in toluene (4 cm3) was stirred
for approximately 5 min at room temperature, and 9,10-
dibromoanthracene (0.17 g, 0.50 mmol), N-methyl-N-a-
methylbenzylamine (0.20 g, 1.50 mmol), NaOt-Bu (0.13 g,
1.40 mmol) and toluene (2 cm3) were added and the reac-
tion mixture was stirred for 52 h at 85–90 ꢁC. The reaction
4.10. Preparation of (R)-(+)-N-methyl-a-methylbenzyl-
amine24
(R)-(+)-a-Methylbenzylformamide (1.10 g, 7.38 mmol) was
dissolved in THF (4 cm3), and then added dropwise to a
suspension of LiAlH4 (1.00 g, 26.3 mmol) in THF
(12 cm3) at 65 ꢁC. The reaction mixture was refluxed for

H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
1011
was allowed to cool to room temperature. Diethyl ether
(10 cm3) was added and the diluted mixture passed through
a short pad of Celite. The solvent was removed under
reduced pressure to afford the crude material (0.21 g, 17%
conversion). Spectroscopic data were as described
previously.
C26H21NO3 requires C, 78.97; H, 5.35; N, 3.54.);
[a]D = +26 (c 1, CHCl3); mmax (ATR/cmꢀ1) 1777; dH
(500 MHz; CDCl3) 7.70–7.60 (2H, m, ArCH), 7.59–7.51
(2H, m, ArCH), 7.49–7.18 (6H, m, ArCH), 7.15–7.09
(2H, m, ArCH), 6.95–6.82 (1H, m, ArCH), 4.74 (1H, d, J
3.1, CH), 4.52 (1H, app. q, J 6.3, NCHCH3), 3.92 (1H,
d, J 9.9 CH), 3.71 (1H, dd, J 9.9, 3.1, CH), 3.08 (1H, br
s, NH), 1.79 (3H, d, J 6.3, CHCH3); dC (100 MHz; CDCl3)
170.5 (CO), 170.0 (CO), 148.4 (ArC), 141.5 (ArC), 141.3
(ArC), 141.1 (ArC), 136.7 (ArC), 128.7 (2 · ArCH), 127.7
(ArCH), 127.5 (ArCH), 126.6 (ArCH), 126.5 (2 · ArCH),
126.2 (2 · ArCH), 125.2 (ArCH), 124.7 (ArCH), 123.8
(ArCH), 121.3 (ArCH), 66.3 (C), 53.5 (CH), 48.5 (CH),
47.7 (CH), 45.4 (CH), 27.7 (CH3); m/z (ES+) 414 (10%),
396.1587 (75, MH+. C26H22NO3 requires 396.1600), 292
(100).
4.13. Preparation of (3aS,9aS)-3a,4,9,9a-tetrahydro-4-[(R)-
a-methylbenzylamino]-2-methyl-4,9-[10,20]benzeno-1H-
benzo[f]isoindole-1,3-(2H)-dione 18a23
9-(N-a-Methylbenzylamino)anthracene
10
(0.08 g,
0.27 mmol) was dissolved in dry toluene (10 cm3). The
resulting solution was heated to 80–85 ꢁC. N-Methylmalei-
mide (0.03 g, 0.27 mmol) was added as a solid and the reac-
tion mixture was stirred for 2 h. The resulting mixture was
allowed to cool to room temperature, and the solvent re-
moved under reduced pressure to afford the title cycloaddi-
tion adduct (0.10 g, 89% yield) as two diastereoisomers in
the ratio of 92:8. A sample was purified by diffusion recrys-
tallisation using CH2Cl2/petroleum ether 40–60 ꢁC to
afford the target cycloadduct as a white solid, mp 158–
160 ꢁC (CH2Cl2/petroleum ether 40–60 ꢁC); [a]D = +50 (c
0.5, CHCl3); mmax (ATR)/cmꢀ1 1691; dH (250 MHz;
CDCl3) 7.68–7.65 (2H, m, ArCH), 7.50–7.41 (3H, m,
ArCH), 7.33–7.27 (2H, m, ArCH), 7.25–7.17 (2H, m,
ArCH), 7.12–7.01 (3H, m, ArCH), 6.86 (1H, app. td, J
7.6, 1.3, ArCH), 4.64 (1H, d, J 3.1, CH), 4.48–4.59 (1H,
m, CH3CH), 3.58 (1H, d, J 8.2, COCH), 3.37 (1H, dd, J
8.2, 3.1, CHCHCH), 3.14 (1H, br d, J 6.4, NH), 2.52
(3H, s, NCH3), 1.78 (3H, d, J 6.7, CH3); dC (62.5 MHz;
CDCl3) 176.9 (CO), 176.7 (CO), 149.1 (ArC), 142.1
(ArC), 141.8 (ArC), 141.7 (ArC), 137.3 (ArC), 128.6
(2 · ArCH), 127.0 (ArCH), 126.5 (2 · ArCH), 126.4
(2 · ArCH), 126.3 (ArCH), 126.2 (ArCH), 124.8 (ArCH),
124.7 (ArCH), 123.7 (ArCH), 121.1 (ArCH), 66.4 (C),
53.6 (CH), 47.8 (CH), 46.3 (CH), 45.7 (CH), 27.9 (CH3),
24.2 (CH3); m/z (EI+) 409.1904 (1%, C27H25N2O2 requires
409.1916), 408 (1), 297 (64), 280 (4), 243 (2), 204 (6), 194
(15), 193 (100), 178 (8), 165 (45).
1
Selected H NMR signals for the minor diastereoisomer
19a (11R,15R)-9,10,11,15-tetrahydro-9-[(R)-a-methylbenz-
ylamino]-9,10[30,40]-furanoanthracene-12,14-dione 18b dH
(500 MHz; CDCl3) 4.70 (1H, s, CH), 4.68–4.62 (1H, m,
CH), 3.70 (1H, d, J 10.1, CH), 3.52 (1H, dd, J 10.1, 2.8,
CH), 1.79 (3H, d, J 5.7, CH3).
4.15. Preparation of 1-methyl-(3R)-(10-oxo-9,10-dihydro-
anthracen-9-yl)pyrrolidine-2,5-dione 11421
The major maleimide diastereoisomer 18a (0.50 g,
1.23 mmol) was dissolved in dry toluene (10 cm3). Trifluo-
roacetic acid (0.5 cm3, 7.00 mmol) was added and the
resulting mixture was stirred at reflux for 4 h, cooled to
room temperature and the solvent removed under reduced
pressure. The crude was purified by column chromatogra-
phy (5% MeOH/CH2Cl2) to give a yellow solid (0.26 g,
69% yield) of the title compound, mp 85–88 ꢁC [(lit.
rac)21 122–122.5 ꢁC]; [a]D = +125 (c 0.08, CHCl3); mmax
(ATR)/cmꢀ1 1694, 1663, 1598; dH (500 MHz; CDCl3)
8.32–8.28 (2H, m, ArCH), 7.66–7.62 (1H, m, ArCH),
7.61–7.59 (1H, m, ArCH), 7.53–7.47 (3H, m, ArCH),
7.35–7.38 (1H, m, ArCH), 5.16 (1H, d, J 2.9, CH), 3.48–
3.43 (1H, m, COCH), 2.87 (3H, s, NCH3), 2.21 (1H, dd,
J 18.5, 9.2, COCH), 1.87 (1H, dd, J 18.5, 4.9, COCH);
dC (125 MHz; CDCl3) 183.8 (CO), 178.0 (CO), 175.0
(CO), 142.4 (ArC), 138.2 (ArC), 133.8 (ArCH), 133.4
(ArC), 133.2 (ArCH), 132.4 (ArC), 128.6 (ArCH), 128.0
(ArCH), 127.9 (2 · ArCH), 127.8 (2 · ArCH), 50.1 (CH),
41.6 (CH), 28.9 (CH2), 24.8 (CH3); m/z (EI+) 306.1122
(17%, C19H16NO3 requires 306.1130), 305 (17), 236 (22),
221 (16), 205 (13), 194 (19), 193 (100), 178 (7), 165 (25).
NMR data were in accordance with the literature.21
1
Selected H NMR signals for the minor diastereoisomer
19a (3aR,9aR)-3a,4,9,9a-tetrahydro-4-[(R)-a-methylbenzyl-
amino]-2-methyl-4,9-[10,20]benzeno-1H-benzo[f]isoindole-
1,3-(2H)-dione dH (250 MHz; CDCl3) 4.63 (1H, d, J 7.0,
CH), 3.18 (1H, dd, J 8.6, 3.1, CH), 2.34 (3H, s, NCH3).
4.14. Preparation of (11S,15S)-9,10,11,15-tetrahydro-9-
[(R)-a-methylbenzylamino]-9,10[30,40]-furanoanthracene-
12,14-dione 18b
Maleic anhydride (0.10 g, 1.01 mmol) was added as a solid
to a stirred solution of 9-a-methylbenzylaminoanthracene
10 (0.30 g, 1.01 mmol) in toluene (4 cm3) at 80–85 ꢁC.
The reaction mixture was stirred for 2 h, cooled to room
temperature, and the solvent was evaporated to give a yel-
low solid of the title compound (0.40 g, 100%) as a mixture
of diastereoisomers in the ratio of 91:9. The major diaste-
reoisomer 18b was purified by recrystallisation from petro-
Acknowledgements
We thank the Committee of Higher Education in Libya
and The Libyan Arab Peoples Bureau for a scholarship
(R.A.B.).
References
leum ether (40–60)/EtOAc, followed by
a
second
recrystallisation from petroleum ether (40–60)/CH2Cl2,
mp 178–183 ꢁC; (Found: C, 78.74; H, 5.16; N, 3.41.
1. (a) Jones, S.; Atherton, J. C. C. Tetrahedron: Asymmetry
2001, 12, 1117–1119; (b) Atherton, J. C. C.; Jones, S. J. Chem.

1012
H. Adams et al. / Tetrahedron: Asymmetry 18 (2007) 1003–1012
Soc., Perkin Trans 1 2002, 2166–2173; (c) Adams, H.; Jones, 13. Wagaw, S.; Rennels, R. A.; Buchwald, S. L. J. Am. Chem.
S.; Ojea-Jimenez, I. Org. Biomol. Chem. 2006, 4, 2296–
Soc. 1997, 119, 8451–8458.
2303.
14. Old, D. W.; Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc.
1998, 120, 9722–9723.
15. Kyba, E. P. Tetrahedron Lett. 1973, 51, 5117–5120.
16. Richards, C. J.; Mulvaney, A. W. Tetrahedron: Asymmetry
1996, 7, 1419–1430.
2. Burgess, K. L.; Corbett, M. S.; Eugenio, P.; Lajkiewicz, N. J.;
Liu, X.; Sanyal, A.; Yan, W.; Yuan, Q.; Snyder, J. K. Bioorg.
Med. Chem. 2005, 13, 5299–5309.
3. Jones, S.; Bawa, R. A. Tetrahedron 2004, 60, 2765–2770.
4. Adams, H.; Bawa, R. A.; Jones, S. Org. Biol. Chem. 2006, 4,
4206–4213.
17. Beletskaya, I. P.; Bessmertny, A. G.; Guilard, R. Tetrahedron
Lett. 1999, 40, 6393–6397.
5. Atherton, J. C. C.; Jones, S. Tetrahedron Lett. 2001, 42, 8239–
8241.
6. During the course of this work, similar chemistry was
reported; see: Sanyal, A.; Yuan, Q.; Snyder, J. K. Tetrahedron
Lett. 2005, 46, 2475–2478.
7. Braun, C. E.; Cook, C. D.; Merritt, C.; Rousseau, J. E. Org.
Synth. 1951, 31, 77–79.
8. Janovec, L.; Suchar, G.; Imrich, J.; Kristian, P.; Sasinkova,
V.; Alfoldi, J.; Sedlak, E. Coll. Czech. Chem. Comm. 2002, 67,
665–678.
9. Niedenzu, K.; Dawson, J. W. J. Am. Chem. Soc. 1959, 81,
5553–5555.
10. Schaeffer, G. W.; Anderson, E. R. J. Am. Chem. Soc. 1949,
71, 2143–2145.
11. (a) Brown, J. F. J. Am. Chem. Soc. 1952, 74, 1219–1221; (b)
Kawate, Y. H.; Nishida, T. A.; Nakagawa, M. J. Org. Chem.
1999, 64, 8821–8828.
18. N-Methyl analogue 11 was prepared during the course of this
work by another group and gave low diastereoselectivites.
Further work on this compound was therefore abandoned.
See Ref. 2.
19. Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication numbers CCDC 632904–632907. Copies of the
data can be obtained, free of charge, on application to
CCDC, 12 Union Road, Cambridge CB2 lEZ, UK [fax:
+44(0)-1223-336033 or e-mail: deposit@ccdc.cam.ac.uk].
20. We thank Professor J. K. Snyder for his help in trying to
establish the origin of this discrepancy.
21. Koerner, M.; Rickborn, B. J. Org. Chem. 1990, 55, 2662–2672.
22. Schuster, I. I. J. Org. Chem. 1981, 46, 5110–5118.
23. Although this compound has been previously reported, no
analytical data were presented. See Ref. 6.
12. Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576,
125–146.
24. Chantrapromma, K.; Ollis, W. D.; Sutherland, I. O. J. Chem.
Soc., Perkin Trans. 1 1983, 1049–1061.