Kinetic and thermodynamic control of axial chirality in biaryl-derived fused oxazolidine lactams exploiting a centre-axis relay of unit efficiency

Article abstract of DOI:10.1039/b706336a

The condensation of a 2-substituted-2-aminoethanol with methyl 2′-formylbiphenyl-2-carboxylate produces only two of the four possible axially chiral 6,7-dihydrodibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-ones (fused oxazolidine lactams), with kinetically controlled diastereoisomer ratios of up to 96: 4. Within each lactam product the central chirality of the oxazolidine-fused benzylic position C(4b) is relayed to the biaryl axis with unit efficiency, the mis-matching of these stereogenic elements being prohibited due to strain, as predicted by molecular mechanics calculations. Diastereoisomeric lactam pairs can be equilibrated by heating with acid, and under these thermodynamic conditions reversed diastereoisomer ratios of up to 26: 74 are observed. The Royal Society of Chemistry.

Full text of DOI:10.1039/b706336a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Kinetic and thermodynamic control of axial chirality in biaryl-derived fused
oxazolidine lactams exploiting a centre-axis relay of unit efficiency
David J. Edwards,a David House,b Helen M. Sheldrake,a Susan J. Stonea and Timothy W. Wallace*a
Received 27th April 2007, Accepted 18th June 2007
First published as an Advance Article on the web 3rd July 2007
DOI: 10.1039/b706336a
The condensation of a 2-substituted-2-aminoethanol with methyl 2ꢀ-formylbiphenyl-2-carboxylate
produces only two of the four possible axially chiral 6,7-dihydrodibenz[c,e]oxazolo[3,2-a]azepin-
9(4bH)-ones (fused oxazolidine lactams), with kinetically controlled diastereoisomer ratios of up to 96 :
4. Within each lactam product the central chirality of the oxazolidine-fused benzylic position C(4b) is
relayed to the biaryl axis with unit efficiency, the mis-matching of these stereogenic elements being
prohibited due to strain, as predicted by molecular mechanics calculations. Diastereoisomeric lactam
pairs can be equilibrated by heating with acid, and under these thermodynamic conditions reversed
diastereoisomer ratios of up to 26 : 74 are observed.
Introduction
The control of axial chirality in biaryls has been a focus of much
attention in recent years as these materials became prominent in
synthetic, materials and supramolecular chemistry, and prompted
the development of new methods for their enantioselective (at-
roposelective) synthesis.1 The key component of chiral biaryls, a
stereogenic axis, is also biologically significant, there being a large
number of compounds whose physiological properties depend on
their axial configuration. These include not only fixed-axis biaryls
such as (aR)-gossypol, an oral anti-spermatogenic agent,2 but also
many flexible systems exemplified by N-acetylcolchinol methyl
ether (NCME) 1, an antimitotic agent whose (aR)-form binds
strongly to tubulin3 and whose (aR)- and (aS)-forms equilibrate
in solution (ratio 3 : 1 in chloroform-d).4 These properties of
1 serve to illustrate the phenomenon of dynamic axial chirality
which, in appropriate circumstances, can be manipulated (e.g.
inverted) to useful effect without breaking bonds.5 The sense of
the axial chirality in NCME 1 depends on the configuration
of the benzylic stereocentre, where the acetamido substituent
prefers a pseudo-equatorial orientation, and the three-atom bridge
effectively operates as a mechanical centre-axis chirality relay.
The degree of axial chirality in biaryls, i.e. the aryl–aryl dihedral
angle, can also be controlled using suitable designs of multi-atom
bridge.6,7
type. However, the heterocyclic nucleus was already known in
the form of 310 and 4,11 and another series of lactams based on
2 was subsequently described by Levacher and co-workers.12,13
We report herein the details of our initial study, in which we
identify complementary kinetic and thermodynamic routes to
biaryl lactams of the form 2, and from which we conclude that
a chirality relay of unit efficiency links the biaryl axis and C(4b)
in such systems.
The interplay between benzylic central chirality and axis
configuration, as illustrated above, is a recurrent theme in atro-
poselective biaryl synthesis,1 and our interest in this area led us
Results and discussion
to investigate the potential of [7,5]-fused lactams based on 6,7-
dihydrodibenz[c,e]oxazolo[3,2-a]azepin-9-one 2 as conformation-
ally restrained biaryls. Although the ring system in 2 is analogous
to the [5,5]- and [6,5]-fused lactams introduced and exploited
so productively by Meyers and co-workers,8 the biaryl lactams
described in our preliminary publication9 were the first of their
The starting materials for our study were prepared from com-
mercial diphenic anhydride 5 using the two sequences shown in
Scheme 1. Heating the anhydride 5 with hydrazine hydrate gave
the hydrazide 6,14 which could be cleaved to the acid-aldehyde 7
using periodic acid;15 the ozonolysis of phenanthrene offers an
alternative route to 7.16 The corresponding ester 9 was prepared
by opening the anhydride 5 with methanol, followed by reduction
to the known ester-alcohol 8 using borane-methyl sulfide.17 The
oxidation of 8 to 9 with manganese(IV) oxide has been described,18
but in our hands this reaction sometimes gave significant amounts
aSchool of Chemistry, The University of Manchester, Oxford Road, Manch-
ester, M13 9PL, UK. E-mail: tim.wallace@manchester.ac.uk; Fax: +44 (0)
161 275 4939
bSynthetic Chemistry, GlaxoSmithKline, Medicines Research Centre, Gun-
nels Wood Road, Stevenage, SG1 2NY, UK
2658 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Scheme 2 Condensation of the biaryls 7 and 9 with aminoalcohols 11.
13a (12%), which were identified as the respective (aS,4bR,7S) and
(aR,4bS,7S) diastereoisomers by X-ray crystallography.9 These
lactams have highly twisted frameworks, with the aromatic rings
of each biaryl unit 40–45◦ from coplanarity and the respective
carbonyls 36◦ and 33◦ out of the plane of the adjacent aromatic
ring. The reactions of 7 with other commercial amino alcohols
11b–f also gave two lactams 12 and 13, each pair distinguishable
through the characteristic H(4b) signals in their 1H NMR spectra
(Table 2 and Table 3), but the yields and diastereoselectivity of the
process were at best moderate, leading to some isolation problems.
The situation was improved using the ester-aldehyde 9, whose
condensation with valinol 11a in toluene gave the lactam 12a
with a d.e. of 92% (entry 7). To shorten reaction times, other
condensations of 9 were carried out under more forcing conditions
(155 ◦C, 3 d) which provided convenient access to the lactams 12
except in the case of tert-leucinol 11f (entry 12). An alternative
route to the phenylglycinol-derived lactam 12b which proceeds at
lower temperature, based on the activation of 7 with Mukaiyama’s
salt, was developed by the Levacher group.13
Scheme 1 Preparation of biaryl starting materials 7 and 9.
of the lactone 10, a result which we ascribe to the variable activity
of the manganese reagent. Fortunately the oxidation of 8 under
Swern conditions19 afforded the ester-aldehyde 9 cleanly and in
good yield.
The polycyclic nucleus of 2 was first assembled by heating the
acid-aldehyde 7 with (S)-valinol 11a under dehydrating conditions,
1
which gave a mixture of two products as indicated by H NMR
spectroscopy and TLC (Scheme 2 and Table 1). The characterising
features of the NMR spectrum of the mixture were two singlets
at d 5.91 and 5.73 ppm, which were assigned to the respective
H(4b) signals of the major and minor products (ratio >5 : 1).
Chromatography afforded the crystalline lactams 12a (84%) and
In principle, the condensation of 7 or 9 with an amino alcohol 11
could provide up to four diastereoisomeric lactams, and in order to
rationalise the results in Table 1, in particular the absence of 14 and
15 from the product mixtures, the core structures 16 and 17 and the
related systems 18–21 were analysed using molecular mechanics
Table 1 Condensation of the biaryls 7 and 9 with aminoalcohols 11
Entry
Biaryl
Series
R
Methoda
Products (yield (%))b
Ratioc
1
2
3
4
5
6
7
8
9
7
7
7
7
7
7
9
9
9
9
9
9
a
b
c
d
e
f
a
b
c
d
e
f
Pri
Ph
Bn
Me
A
A
A
A
A
A
Ad
B
B
B
B
B
12a (84), 13a (12)
12b + 13b (75)
12c + 13c (54)
12d + 13d (59)
12e + 13e (61)
12f + 13f (55)
12a + 13a (71)
12b (88)
84 : 16
70 : 30
62 : 38
56 : 44
50 : 50
64 : 36
96 : 4
91 : 9
92 : 8
85 : 15
87 : 13
—
CH2OBn
But
Pri
Ph
Bn
Me
12c (83), 13c (6)
12d (72), 13d (10e)
12e (74), 13e (23e)
10
11
12
CH2OBn
But
–
f
a Method A: toluene, Dean–Stark, reflux, 36 h. Method B: xylene, bath temp. 155 ◦C, 72 h. b Isolated yields after chromatography. c Estimated from the
1H NMR spectrum of the product mixture. d Reflux maintained for 7 d. e Not obtained pure. f No products observed.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2659
©

View Article Online
2660 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
techniques (Fig. 1). Minimum energy conformations were located
for both 16 and 17 (Fig. 2), and the steric energy of 17 was
calculated to exceed that of 16 by more than 100 kJ mol−1. This is a
compelling difference, especially when considered in conjunction
with the estimated 80 kJ mol−1 by which the axial torsion barrier
in a simple three-atom bridged biaryl can be raised by o,oꢀ-
dimethylation.20 It can be concluded that the diastereoisomers
14 and 15 based on the mis-matched system 17 are effectively
disallowed by strain, with major contributions from excessive
bending and torsion (eclipsing). The structure 17 is also subject
to a loss of amide resonance due to the pyramidalisation of the
nitrogen,21 which is manifested inter alia by the O(9)–C(9)–N(8)–
C(7) dihedral angle (39.6◦) and the reduced sum of the bond angles
at the nitrogen atom (352.6◦). Some insight into the uniqueness
of the relationship between 16 and 17, which feature a three-
atom bridge containing two trigonal atoms, is provided by the
analyses of 18–21. In 18 and 19 the carbon atoms of the bridge are
tetrahedral while the nitrogen is pyramidal, and the two structures
would not appear to be differentiated on the basis of strain.
However, in 20 and 21 the presence of one trigonal atom in the
bridge is sufficient to bring about a matched/mis-matched centre-
axis relationship in favour of 20, based on steric strain. Examples
of this relay are discussed later.
A proposed mechanism for the formation of the lactams 12
and 13 is shown in Scheme 3. The initial step is the formation
of an imine species 22 which can equilibrate with the respective
trans- and cis-oxazolidines 23 and 24. In analogous systems such
equilibria are dominated by the ring-opened forms (cf. 22),22 but
intermolecular N-acylation is kinetically selective for the cis-2,4-
disubstituted oxazolidine form (cf. 24) in which one face of the
heterocyclic ring is unhindered.23 In contrast, the intramolecular
N-acylation which gives lactam 12 can proceed most favourably
through the arrangement depicted in (aS)-23, in which the groups
R and X are apart, whereas in the alternative (aR)-24 leading to
13 these groups are close. The higher selectivity observed using the
ester-aldehyde 9 and its failure to react with the bulky tert-leucinol
11f (Table 1, entry 12) are consistent with this mechanistic picture.
A second series of molecular mechanics calculations was carried
out on the lactams 12 and 13, and the results (Table 4) indicate that
the minor product 13 is the thermodynamically more stable isomer
of each pair. On the basis of precedents24 it was anticipated that
the interconversion of 12 and 13 might be induced by treatment
with acid, and the results of a study of this equilibration process
are shown in Table 5.
Table 4 Minimised steric energiesa and equilibrium compositions for
lactams 12 and 13
Minimised steric
energies/kJ mol−1
Predicted ratio
Series
12
13
DG/kJ mol−1
12:13 at 25 ◦C
a
b
c
d
e
f
127.624
129.012
136.075
108.918
183.423
146.698
124.729
127.606
132.697
107.626
176.621
142.445
2.894
1.406
3.378
1.292
6.802
4.253
23.7 : 76.3
36.2 : 63.8
20.4 : 79.6
37.3 : 62.7
6.0 : 94.0
15.2 : 84.8
a Calculated using MacroModel 8.0 (MM3 force field).
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2661
©

View Article Online
Fig. 1 Energy-minimised conformations of the fused dibenzazepines 16–21 and their steric energies, calculated using MacroModel 8.0 (MM3 force
field).
expected first-order kinetics (Fig. 3). The reaction was stopped
after 33 d, after which time the ratio 12a:13a was converging on a
value of 1 : 2, thus confirming the greater thermodynamic stability
of the diastereoisomer 13a and that the overall lactam-forming
sequence (Scheme 3) was kinetically controlled.
For convenience, many of the equilibrations shown in Table 5
were carried out at higher temperatures using an excess of trifluo-
roacetic acid (TFA), and two blanks (entries 3 and 6) confirmed the
requirement for acid. It is noteworthy that in all cases the isomer
distribution favoured the lactam 13, as predicted by calculation.
In some cases the calculated and observed isomer ratios were quite
close, but the largest discrepancy (entry 10) serves as a reminder of
the approximations made in assembling the results in Table 4: the
steric energy-minimised structure of 13e incorporated p-stacking
between the benzyl and biaryl components that would be negated
by aromatic solvents.
The mechanism for acid-induced equilibration is presumed to
Fig. 2 Potential energy plots of the C(1)–C(13b)–C(13a)–C(13) biaryl
dihedral angle (modulus) in lactams 16 and 17 in the region of their steric
energy minima, calculated using MacroModel 8.0 (MM3 force field).
involve the protonation of the lactam 12 followed by ring-opening
to the acyliminium species 26, which can isomerise to 27 and
recyclise to 28, providing access to 13 (Scheme 4). The rate-
determining step is likely to be the ring-opening of 25, in which
the C(4b)–O(5) bond is fixed with poor alignment for elimination.
In an initial experiment (entry 1), the valinol-derived lactam 12a
in acetonitrile-d3 was heated to 55 ◦C in the presence of a catalytic
amount of p-toluenesulfonic acid, and the solution was monitored
by H NMR spectroscopy. The equilibration process was slow
under these conditions, but the equilibration proceeded with the
1
Table 5 Acid-induced equilibration of the biaryl lactams 12 and 13
Entry Starting lactam
R
Solvent
Added acid Temperature T/◦Ca Time/d Products Observed ratiob Calculated ratioc
1
2
3
4
5
6
7
8
9
12a
Pri
Pri
Pri
Pri
Ph
Ph
Bn
Bn
Me
MeCN-d3 p-TsOH
Toluene-d8 p-TsOH
Toluene-d8 None
55
140
140
100
100
150
100
100
100
150
100
33d
3
7
7
7
12a + 13a 36 : 64
12a + 13a 27 : 73
26 : 74
30 : 70
30 : 70
28 : 72
39 : 61
40 : 60
27 : 73
27 : 73
40 : 60
13 : 87
21 : 79
12a
12a
12a
—
12a
Toluene-d8 CF3CO2H
Toluene-d8 CF3CO2H
12a + 13a 29 : 71
12b
12b + 13b 45 : 55
12b
Xylene
Toluene-d8 CF3CO2H
HCO2H None
Toluene-d8 CF3CO2H
None
2
12b
—
12c
7
12c + 13c 36 : 64
12c + 13c 43 : 57
12d + 13d 34 : 66
12e + 13e 30 : 70
12f + 13f 26 : 74
12c
3d
7
12d/13de
12e/13ef
12f/13fg
10
11
CH2OBn Xylene
CF3CO2H
CF3CO2H
3
7
But
Toluene
a Bath temperature. b Estimated from the 1H NMR spectrum of the product mixture. c Predicted equilibrium composition at T ◦C based on calculated
steric energies (Table 4). d Stopped before completion. e Starting ratio 56 : 44. f Starting ratio 50 : 50. g Starting ratio 64 : 36.
2662 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Scheme 3 Proposed mechanism for the formation of 12 and 13.
Scheme 4 Proposed mechanism for the acid-catalysed equilibration of 12
and 13.
Fig. 3 The progress of the equilibration of lactams 12a and 13a in CD3CN
at 55 ◦C. The fitted curve for [12a] corresponds to an observed rate constant
(k1 + k−1) of 1.1 × 10−6 s−1 and an equilibrium ratio of 1 : 2.
The results of this study confirm that a mobile biaryl axis
can be restrained in a predictable and fixed configuration by
incorporation within a fused oxazolidine lactam of the type
developed and used extensively in synthesis by Meyers et al.8
The most prominent feature of such lactams, illustrated by
the properties of 12a–f and 13a–f, is the effectiveness of the
mechanical link between the benzylic position C(4b) and the
biaryl axis, through which stereochemical information is relayed
with complete fidelity. Related chirality relays are present in the
chemical literature, one such example being the cyclisation of the
iminium species 29 which, on the basis of analogy with 20/21,
presumably gave ( )-30, although this was not specified.25 Other
reactions of this type were reported by Hilt et al.,26 who confirmed
the cyclisation of ( )-31 to ( )-32 using X-ray crystallography.
Only one other, non-crystalline, diastereoisomer was isolated from
this reaction (d.r. 8 : 1) but its configuration remained unspecified.
Again, by analogy with 20/21 the minor diastereoisomer was
probably ( )-33, as the alternative axis-centre permutation would
be disfavoured due to strain.
This can be seen in the crystal structure of 12a, in which the
dihedral angles C(4)–C(4a)–C(4b)–O(5) and C(9)–N(8)–C(4b)–
O(5) are 3.7◦ and 177.2◦ respectively, i.e. the oxazolidine C–O
bond is essentially orthogonal to both the aromatic and amide
p-systems (Fig. 4). The protonated oxygen therefore receives little
assistance from either when operating as a leaving group. The
isomer 13a is similarly situated, the corresponding dihedral angles
being 2.0◦ and 165.4◦.
Fig. 4 Dihedral angles affecting O(5) from the X-ray crystal structure
of 12a.9
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2663
©

View Article Online
and 13 could be useful in a variety roles that require robust, con-
formationally defined chiral frameworks. Experiments designed to
explore the chemistry of these lactams, with particular emphasis
on processes that allow the retention of the induced axial chirality
after disconnection of the oxazolidine ring, are in progress and
will be reported in due course.
Experimental
Melting points were determined using a Kofler hot-stage or an
Electrothermal apparatus and are uncorrected. IR spectra were
recorded for neat thin films on NaCl plates, using Perkin-Elmer
1710FT or Nicolet Nexus 670/870 spectrometers. NMR spectra
were measured on Bruker DPX200, AC300 and Advance 400
instruments for solutions in deuteriochloroform, unless otherwise
indicated; resonances were assigned with the aid of COSY, HMBC,
HMQC and DEPT spectra where appropriate. Low-resolution
mass spectra were measured on a Micromass LCT instrument
using a Waters 2790 separations module with electrospray (ES+)
ionisation and TOF fragment detection. High-resolution mass
spectra were recorded on ThermoFinnigan MAT95XP or Kratos
Concept S1 instruments. Data for most of the peaks of intensity
<20% of that of the base peak are omitted. Elemental analyses
were carried out by the UMIST and University of Manchester
microanalytical services. Optical rotations were measured at
589 nm using an AA-100 polarimeter (Optical Activity Ltd). The
sealed tubes used in equilibration experiments were screw-capped
and made from thick-walled glass (Aldrich Z18106-4).
Starting materials and solvents were routinely purified by
conventional techniques28 and most reactions were carried out
under nitrogen or argon. Organic solutions were dried using anhy-
drous magnesium sulfate and concentrated by rotary evaporation.
Analytical thin layer chromatography (TLC) was carried out on
Merck silica gel 60 on aluminium plates containing a 254 nm
fluorescent indicator. The chromatograms were visualised by the
use of UV light or the following developing agents: ethanolic
vanillin or potassium permanganate. Unless otherwise indicated,
preparative (column) chromatography was carried out using the
flash technique29 on 60H silica gel (Merck 9385). Compositions
of solvent mixtures are quoted as ratios of volume. ‘Petroleum’
refers to a light petroleum fraction, b.p. 40–60 ◦C, unless otherwise
stated. ‘Ether’ refers to diethyl ether. ‘Xylene’ refers to anhydrous
o-xylene (Aldrich 294780).
The most intriguing illustration of benzylic centre-axis comple-
mentarity is the observation by Zavada and co-workers that the
zinc bromide-induced elimination of cyanide ion from the fixed-
axis biaryl 34 produces only 35, indicating that only the axial cyano
group is expelled.27 Moreover, the reverse reaction, in which 35
undergoes the addition of cyanide ion, is also completely selective,
regenerating 34 in the absence of stereoisomers. The centre-axis
stereochemical relay thus operates in both directions, with the
configuration of the axis dictating the orbital alignment of the
imine moiety in 35.
Conclusions
The formation of biaryl-fused lactams 12 via the mechanism
shown in Scheme 3 amounts to the dynamic kinetic resolution
of 7 or 9, whose flexible biaryl axis is converged to a single
configuration in the form of the derived lactam 12. The overall
sequence involves the relay of chirality from the auxiliary amino
alcohol 11 to the benzylic fusion position, C(4b), which proceeds
with kinetic selectivity (up to 92% d.e.) originating in the faster
cyclisation of (aS)-23 as compared to (aR)-24. The alternative
cyclisations of 23 through an (R)-configured axis to give 14, and
of 24 through an (S)-configured axis to give 15, are prohibited by
strain, which is manifested in the lactams 12 and 13 as a chirality
relay of unit efficiency between C(4b) and the biaryl axis. The
diastereoisomeric pairs in each lactam series can be interconverted
by heating with acid, leading to equilibrium mixtures in which 13
generally predominates by ca. 3 : 1.
2ꢀ-(Hydrazinocarbonyl)biphenyl-2-carboxylic acid 614
To a solution of hydrazine monohydrate (4.0 mL, 82.5 mmol) in
water (4 mL) at 0 ◦C was added diphenic anhydride 5 (8.00 g,
35.7 mmol) over a period of 5 min with vigorous stirring and
◦
then the mixture was heated to 90 C for 1 h. The reaction was
cooled and poured into water (50 mL). The solution was acidified
with conc. HCl to pH 2 with vigorous stirring until a gummy
brown solid began to form. Chilling and scratching gave a fine
white precipitate. Filtration and crystallisation (EtOH) gave the
monohydrazide 6 (8.70 g, 95%) as a colourless solid, m.p. 182.5–
◦
Given the established advantages of amino alcohols as chiral
auxiliaries, based on their ready availability in enantiopure form
and the methods by which they can be installed, exploited and
disconnected,8 biaryl-fused oxazolidine lactams of the forms 12
183 C (Found: C, 65.4; H, 5.0; N, 10.6. C14H12N2O3 requires C,
65.62; H, 4.72; N, 10.93%); mmax/cm−1 3269, 3165; dH (300 MHz,
CDCl3) 7.76 (1 H, dd, J 1.5, 7.5 Hz, ArH), 7.52–7.38 (5 H, m,
ArH), 7.14–7.09 (3 H, m, ArH); dC (100 MHz, CDCl3) 127.4,
2664 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
127.6, 129.3, 129.6, 130.3, 130.6, 130.9, 132.3, 134.3, 140.4, 140.9,
168.2, 169.1; m/z 279 ([M + Na]+ , 25%), 257 ([M + H]+, 18), 239
(100); Rf 0.20 (EtOAc–EtOH, 3 : 2).
oil. Chromatography (EtOAc–hexane, 15 : 85) gave a sample of the
ester-aldehyde 9 (484 mg, 19%) as a colourless oil; dH (300 MHz,
CDCl3) 9.81 (1 H, s, CHO), 8.06 (1 H, d, J 7.0 Hz, ArH), 8.02 (1 H,
d, J 7.9 Hz, ArH), 7.64–7.50 (4 H, m, ArH), 7.30 (1 H, d, J 7.5 Hz,
ArH), 7.26 (1 H, d, J 8.7 Hz, ArH), 3.63 (3 H, s, OMe); mmax/cm−1
3062, 2951, 2842, 2751, 1728, 1696, 1596, 1435, 1258, 1196, 1129,
1086, 761, 668; Rf 0.52 (EtOAc–hexane, 3 : 7). Further elution of
the column yielded a mixed fraction (1.45 g, 61%) followed by 7H-
dibenzo[c,e]oxepin-5-one 10 (425 mg, 19%) as white crystals, m.p.
132–135 ◦C [lit.30 132–134 ◦C (sublimed)]; dH (300 MHz, CDCl3)
7.98 (1 H, d, J 7.9 Hz, ArH), 7.70–7.40 (7 H, m, ArH), 5.03 (2 H,
br d, J 15.5 Hz, 7-H2) (in accord with published data31); mmax/cm−1
2ꢀ-Formylbiphenyl-2-carboxylic acid 715
To a cooled solution (0 ◦C) of periodic acid (as H5IO6, 1.54 g,
6.8 mmol) in water (36 mL) and aq. ammonium hydroxide (8%
w/w, 14 mL) was added the hydrazide 6 (1.0 g, 3.9 mmol) dissolved
in aq. ammonium hydroxide (8% w/w, 7 mL) with vigorous
stirring, keeping the temperature below 5 ◦C. The mixture was
stirred at 5 ◦C for 5 min and then at room temperature for 15 min.
Conc. sulfuric acid (ca. 6 mL) was added, producing a gummy
brown solid, which was dissolved in dichloromethane (30 mL).
The layers were separated, the aqueous layer was extracted with
dichloromethane (3 × 30 mL), and the combined organic phase
was dried over MgSO4 and evaporated. Chromatography of the
residual brown oil (80 g SiO2, EtOAc–dichloromethane 1 : 4) and
crystallisation from EtOAc gave the title compound 7 (440 mg,
50%) as a white solid, m.p. 134–136 ◦C [lit.16 134–135 ◦C (MeOH–
water)] (Found: C, 74.1; H, 4.6. C14H10O3 requires C, 74.33; H,
4.46%); mmax/cm−1 3064, 2843, 1728, 1689, 1596, 1402, 1274, 1200,
757; dH (300 MHz, CDCl3) 10.30 (1 H, s, OH), 9.78 (1 H, s, CHO),
8.11 (1 H, dd, J 1.4, 7.9 Hz, ArH), 7.99 (1 H, dd, J 1.5, 7.5 Hz,
ArH), 7.62–7.48 (4 H, m, ArH), 7.30 (1 H, dd, J 1.2, 7.5 Hz, ArH),
7.23 (1 H, dd, J 1.1, 7.5 Hz, ArH); dC (75 MHz, CDCl3) 128.0,
128.3, 128.6, 129.6, 130.4, 131.5, 132.3, 132.7, 133.6, 134.1, 140.6,
145.2, 172.2, 192.1; m/z 290 ([M + 2Na + H2O]+, 100%), 249 ([M +
Na]+, 30), 209 (6), 181 (50); Rf 0.20 (EtOAc–dichloromethane,
1 : 4).
=
3069, 1716 (C O), 1600, 1449, 1379, 1277, 1226, 1111, 1091, 1048,
1011, 765, 740; Rf 0.42 (EtOAc–hexane, 3 : 7).
Method B. A solution of DMSO (0.51 mL, 561 mg, 7.2 mmol,
1.3 eq.) in dichloromethane (3 mL) was added to a 2 M solution of
oxalyl dichloride (3 mL, 6.0 mmol) in dichloromethane (20 mL)
at −78 ◦C and stirred for 30 min. The ester-alcohol 8 (1.33 g,
5.5 mmol) in dichloromethane (5 mL) was added and the reaction
mixture stirred for a further 30 min at −78 ◦C. Triethylamine
(2 mL, 1.45 g, 14.3 mmol) was added and the reaction mixture
was allowed to warm to room temperature while stirring for
3 h. The mixture was quenched with aq. ammonium chloride
(2 M, 20 mL), extracted with dichloromethane (3 × 30 ml), dried
(Na2SO4) and concentrated, giving the ester-aldehyde 9 as a pale
1
yellow oil (1.20 g, 91%), identical (TLC, H NMR) to material
obtained using Method A above.
Formation of biaryl lactams (Table 1)
Methyl 2ꢀ-hydroxymethylbiphenyl-2-carboxylate 8
Method A. From the acid-aldehyde 7 in toluene: A solution
of 2ꢀ-formylbiphenyl-2-carboxylic acid 7 and the appropriate
aminoalcohol 11 (1.4–1.5 mol. equiv.) in toluene was heated under
reflux in a Dean–Stark apparatus for 36 h. The solution was
A stirred suspension of diphenic anhydride 5 (3.52 g, 15.7 mmol)
in MeOH (50 mL) was heated to reflux for 3.3 h. The resulting
solution was concentrated in vacuo to yield a dark oil. To a stirred
solution of this oil in THF (100 mL) at 0 ◦C was added dropwise
a solution of borane–dimethylsulfide in THF (2 M, 14.1 mL,
28.2 mmol) and the resulting solution was allowed to warm
to room temperature over 23.5 h. The mixture was cautiously
quenched with MeOH (10 mL) and the volatiles removed in
vacuo. The residue was diluted with EtOAc (100 mL) and washed
with saturated aqueous NaHCO3 (2 × 50 mL). The organic
phase was dried, filtered, concentrated in vacuo and purified
by chromatography (EtOAc–hexane, 1 : 3), which gave the title
compound 8 (2.56 g, 67% over two steps) as a colourless oil; dH
(300 MHz, CDCl3) 7.94 (1 H, dd, J 0.8, 7.5 Hz, 6-H), 7.58–7.22
(6 H, m, ArH), 7.06 (1 H, d, J 6.8 Hz, 3ꢀ-H), 4.44 (1 H, d, J
12.1 Hz, CHHꢀOH), 4.38 (1 H, d, J 12.1 Hz, CHHꢀOH), 3.67 (3 H,
s, OCH3), 1.94 (1 H, s, OH) (in accord with published data17); Rf
0.23 (EtOAc–hexane, 3 : 7).
1
cooled, evaporated in vacuo and the residue was analysed by H
NMR spectroscopy to determine the diastereoisomer ratio (see
Table 2 and Table 3 for 1H NMR data). The products were isolated
by chromatography over silica gel.
Method B. From the ester-aldehyde 9 in xylene: A solution
of methyl 2ꢀ-formylbiphenyl-2-carboxylate 9 and the appropriate
aminoalcohol 11 (1.4–1.5 mol. equiv.) in xylene was heated at
155 ◦C (bath temp.) for 3 d. The solution was cooled, evaporated
in vacuo and the residue was analysed by 1H NMR spectroscopy
to determine the diastereoisomer ratio. The products were isolated
by chromatography over silica gel.
(aS,4bR,7S)-6,7-Dihydro-7-isopropyldibenz[c,e]oxazolo[3,2-
a]azepin-9(4bH)-one 12a and (aR,4bS,7S)-6,7-dihydro-7-
isopropyldibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 13a
Method A. The acid-aldehyde 7 (280 mg, 1.24 mmol) and (S)-
valinol 11a (180 mg, 1.74 mmol) in toluene (16 mL) was heated
under reflux in a Dean–Stark apparatus under Ar for 36 h. TLC
indicated complete conversion, the reaction was cooled and the
solvent removed in vacuo. The residue was partially purified by
chromatography over silica gel (80 g SiO2, EtOAc–hexane 1 : 5
as eluant), which gave the title compound 12a (210 mg, 58%) as
colourless crystals, m.p. 121–123 ◦C (EtOAc) (Found: C, 77.7; H,
Methyl 2ꢀ-formylbiphenyl-2-carboxylate 9
Method A (ref. 18). To a stirred solution of the ester-alcohol
8 (2.56 g, 10.6 mmol) in dichloromethane (200 mL) was added
MnO2 (80% pure; 10.82 g, 0.10 mol) and the resulting suspension
was heated under reflux for 3 h. The reaction mixture was cooled
to room temperature, filtered through Celite and concentrated in
vacuo to yield a mixture (ca. 1 : 1) of 9 and 10 (2.36 g) as a colourless
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2665
©

View Article Online
lactams 12c and 13c (total 210 mg, 54%), ratio 62 : 38 by 1H NMR
spectroscopy.
6.8; N, 5.0. C19H19NO2 requires C, 77.79; H, 6.53; N, 4.77%); [a]D23
−183 7 (c 1.9, CHCl3); mmax/cm−1 3068, 2959, 2885, 1639, 1600,
1449, 1406, 1289, 1216, 1173, 1080, 854; dC (75 MHz, CDCl3) 16.5
(MeA), 19.8 (MeB), 28.0 (CHMe2), 61.9 (7-C), 67.2 (6-C), 86.6
(4b-C), 122.3, 128.4 (two), 129.2, 129.4, 130.0, 130.2 and 131.4
(ArCH), 135.7, 136.2, 136.6 and 139.0 (quaternary ArC), 164.5
(8-C); m/z 609 ([2M + Na]+, 100%), 357 (79), 294 ([M + H]+, 82);
Rf 0.30 (EtOAc–hexane, 1 : 4). Mixed fractions were condensed to
give an oil (139 mg, 38%) which was purified by preparative TLC
[20 × 20 × 0.02 cm (Merck 1.05583); multiple elution with EtOAc–
hexane 1 : 6], giving 12a (95 mg, 26%) and the title compound
13a (42 mg, 11.5%), m.p. 114–115 ◦C (petroleum, b.p. 60–80 ◦C)
(Found: C, 78.0; H, 6.8; N, 5.0. C19H19NO2 requires C, 77.79; H,
6.53; N, 4.77%); [a]D27 +148 5 (c 2.25, CHCl3); mmax/cm−1 3064,
2959, 2874, 1639, 1596, 1445, 1402, 1344, 1169, 1080, 994, 963,
858, 742, 703; dC (75 MHz, CDCl3) 18.7 (Me), 19.7 (Me), 30.7
(CHMe2), 63.0 (7-C), 69.5 (6-C), 86.1 (4b-C), 122.5, 128.3, 128.4,
129.1, 129.7, 130.3, 130.7 and 131.8 (ArCH), 134.1, 136.2, 137.6
and 139.3 (quaternary ArC), 165.4 (8-C); m/z 609 ([2M + Na]+,
100%), 357 (88), 294 ([M + H]+, 81); Rf 0.28 (EtOAc–hexane,
1 : 4).
Method B. Heating the ester-aldehyde 9 (110 mg, 0.46 mmol)
and (S)-phenylalaninol 11c (104 mg, 0.69 mmol) in xylene (5 mL)
at 155 ◦C (bath temp.) for 3 d followed by removal of the
solvent in vacuo gave a mixture of isomeric lactams (ratio 92 :
8). Chromatography (EtOAc–hexane, 15 : 85) gave the title
compound 12c (129 mg, 83%) as a colourless solid, m.p. 84–
84.5 ◦C (dichloromethane) (Found: C, 80.75; H, 5.67; N, 4.03.
C23H19NO2 requires C, 80.92; H, 5.61; N, 4.10%); [a]2D6 −226
9
(c 1.0, CHCl3); mmax/cm−1 3069, 2927, 1638, 1447, 1406, 750; dC
(100 MHz, CDCl3) 37.0 (PhCH2), 58.2 (7-C), 69.4 (6-C), 86.0 (4b-
C), 121.7, 126.7, 128.13, 128.15, 128.7 (two), 128.9, 129.4, 129.6
(two), 129.9, 130.0 and 131.3 (ArCH), 134.8, 136.1, 136.2, 137.9
and 138.4 (quaternary ArC), 163.9 (8-C); m/z 705 ([2M + Na]+,
46%), 405 ([M + 2Na + H2O]+, 69), 364 ([M + Na]+, 100), 342 ([M +
H]+, 82) (Found: M, 341.1416; C23H19O2N requires 341.1410); Rf
0.38 (EtOAc–hexane, 3 : 7). Further elution of the column yielded
a mixed fraction (11 mg, 7%) and the title compound 13c (9 mg,
6%) as a colourless oil; mmax/cm−1 2913, 2844, 1642, 1448, 1081;
dC (100 MHz, CDCl3) 38.5 (CH2Ph), 58.5 (7-C), 70.5 (6-C), 85.8
(4b-C), 121.9, 126.7, 128.2, 128.4, 128.7 (two), 128.9, 129.5 (two),
129.6, 129.8, 130.2 and 131.5 (ArCH), 133.9, 135.9, 137.2, 137.8
and 139.4 (quaternary ArC), 164.2 (8-C); m/z 705 ([2M + Na]+,
11%), 405 ([M + 2Na + H2O]+, 47), 364 ([M + Na]+, 96), 342 ([M +
H]+, 100) (Found: M, 341.1412; C23H19O2N requires 341.1410); Rf
0.31 (EtOAc–hexane, 3 : 7).
From ester 9. Heating the ester-aldehyde
9 (230 mg,
0.96 mmol) and (S)-valinol 11a (148 mg, 1.43 mmol) in toluene
(15 mL) under reflux (bath temp. 120 ◦C) for 7 d gave the mixed
lactams 12a and 13a (total 200 mg, 71%, ratio 96 : 4).
(aS,4bR,7S)-6,7-Dihydro-7-phenyldibenz[c,e]oxazolo[3,2-
a]azepin-9(4bH)-one 12b and (aR,4bS,7S)-6,7-dihydro-7-
phenyldibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 13b
(aS,4bR,7S)-6,7-Dihydro-7-methyldibenz[c,e]oxazolo[3,2-
a]azepin-9(4bH)-one 12d and (aR,4bS,7S)-6,7-dihydro-7-
methyldibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 13d
Method A. Heating the acid-aldehyde 7 (296 mg, 1.31 mmol)
and the amino alcohol 11b (270 mg, 1.97 mmol) in toluene (15 mL)
under reflux (bath temp. 120 ◦C) for 36 h gave the mixed lactams
12b and 13b (total 320 mg, 75%, ratio 70 : 30).
Method A. Heating the acid-aldehyde 7 (290 mg, 1.28 mmol)
and (S)-2-amino-1-propanol 11d (144 mg, 1.92 mmol) in toluene
(15 mL) under reflux (bath temp. 120 ◦C) for 36 h gave the mixed
lactams 12d and 13d (total 200 mg, 59%), ratio 56 : 44 by 1H NMR
spectroscopy.
Method B. Heating the ester-aldehyde 9 (100 mg, 0.42 mmol)
and (S)-phenylglycinol 11b (86 mg, 0.63 mmol) in xylene (5 mL)
at 155 ◦C (bath temp.) for 3 d followed by removal of the
solvent in vacuo gave a mixture of isomeric lactams (ratio 91 : 9).
Chromatography (EtOAc–hexane, 1 : 4) gave the title compound
12b (120 mg, 88%) as colourless needles, m.p. 143–145 ◦C (MeOH)
(lit.13 138–140 ◦C); [a]D28 −86 8 (c 0.26, CHCl3); mmax/cm−1 3662,
3471, 3269, 3067, 3021, 2948, 2889, 1638, 1452, 1398, 1339, 1277,
1211, 1157, 904, 737; dC (100 MHz, CDCl3) 60.2 (6-C), 74.0 (7-C),
86.6 (4b-C), 121.8, 126.3 (two), 127.8, 128.2 (two), 128.8 (two),
129.1, 129.4, 130.0, 130.2 and 131.4 (ArCH), 134.8, 136.2, 136.4,
138.3 and 140.8 (quaternary ArC), 163.7 (8-C); m/z 677 ([2M +
Na]+, 49), 510 (53), 391 ([M + 2Na + H2O]+, 58), 355 (100),
350 ([M + Na]+, 82), 329 (76), 235 (73), 207 (38) (Found: M +
Na, 350.1145; C22H17O2NNa requires 350.1152); Rf 0.25 (EtOAc–
hexane, 3 : 7).
Method B. Heating the ester-aldehyde 9 (95 mg, 0.40 mmol)
and (S)-2-amino-1-propanol 11d (45 mg, 0.60 mmol) in xylene
(5 mL) at 155 ◦C (bath temp.) for 3 d followed by removal of the
solvent in vacuo gave a mixture of isomeric lactams (ratio 85 : 15).
Chromatography (EtOAc–hexane, 15 : 85) gave the title compound
12d (76 mg, 72%) as a colourless oil; [a]2D9 −143 7 (c 3.6, CHCl3);
mmax/cm−1 3065, 2975, 2887, 1638, 1451, 1410, 1217, 940, 912, 748,
730; dC (100 MHz, CDCl3) 18.3 (Me), 52.7 (7-C), 72.8 (6-C), 85.7
(4b-C), 121.7, 128.1 (two), 128.9, 129.4, 129.9, 130.0 and 131.1
(ArCH), 134.9, 136.17, 136.20 and 138.5 (quaternary ArC), 163.7
(8-C); m/z 266 ([M + H]+, 100%) (Found: M + H, 266.1165;
C17H16O2N requires 266.1176); Rf 0.20 (EtOAc–hexane, 3 : 7).
Further elution of the column yielded a mixed fraction (17 mg,
16%) and a fraction enriched in the minor isomer 13d (10 mg,
1
10%, ca. 80% d.e. by H NMR spectroscopy) as a colourless oil;
(aS,4bR,7S)-6,7-Dihydro-7-(phenylmethyl)dibenz[c,e]oxazolo[3,2-
a]azepin-9(4bH)-one 12c and (aR,4bS,7S)-6,7-dihydro-7-
(phenylmethyl)dibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 13c
mmax/cm−1 2964, 2923, 2912, 2841, 1632, 1463, 1407, 1094, 741; dC
(75 MHz, CDCl3) 19.1 (Me), 52.9 (7-C), 73.7 (6-C), 85.7 (4b-C),
121.7, 128.2, 128.3, 128.9, 129.7, 129.8, 130.2 and 131.5 (ArCH),
134.1, 135.9, 137.1 and 139.8 (quaternary ArC), 163.6 (8-C); m/z
266 ([M + H]+, 100%) (Found: M + H, 266.1181; C17H16O2N
requires 266.1176); Rf 0.14 (EtOAc–hexane, 3 : 7).
Method A. Heating the acid-aldehyde 7 (260 mg, 1.15 mmol)
and (S)-phenylalaninol 11c (260 mg, 1.72 mmol) in toluene
(15 mL) under reflux (bath temp. 120 ◦C) for 36 h gave the mixed
2666 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
(aS,4bR,7S)-7-Phenylmethoxymethyl-6,7-dihydrodibenz-
[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 12e and (aR,4bS,7S)-7-
phenylmethoxymethyl-6,7-dihydrodibenz[c,e]oxazolo-
[3,2-a]azepin-9(4bH)-one 13e
above, and then disconnecting the oxazolidine ring at C(4b). The
identities of H(4b) and O(5) were exchanged and the oxazolidine
ring was then reconnected. A constraint (force constant 100, range
10◦) was imposed on the (S)-configured biaryl torsion angle
(42.5◦) and the structure was subjected to a new csearch operation.
The torsional constraint was removed from the resulting structure,
ent-17, whose steric energy was then minimised (mini) once more.
MSE values for 18–21 were calculated similarly. The structures 16
and ent-17 were each subjected to a drive analysis of the dihedral
angle C(1)–C(13b)–C(13a)–C(13) in the region of the respective
minima (interval 0.1◦), which provided the data for the plots in
Fig. 2. The calculated MSE values (kJ mol−1) were: 16, 103.95; 17,
214.47; 18, 118.87; 19, 118.28; 20, 27.67; 21, 55.44.
Method A. Heating the acid-aldehyde 7 (100 mg, 0.44 mmol)
and (R)-2-amino-3-benzyloxy-1-propanol 11e (120 mg,
0.66 mmol) in toluene (15 mL) under reflux (bath temp.
120 ◦C) for 36 h gave the mixed lactams 12e and 13e (total 100 mg,
61%), ratio 1 : 1 by 1H NMR spectroscopy.
Method B. Heating the ester-aldehyde 9 (79 mg, 0.33 mmol)
and (R)-2-amino-3-benzyloxy-1-propanol 11e (89 mg, 0.49 mmol)
in xylene (5 mL) at 155 ◦C (bath temp.) for 3 d followed by removal
of the solvent in vacuo gave a mixture of isomeric lactams (ratio
87 : 13). Chromatography (EtOAc–hexane, 15 : 85) gave the title
compound 12e (90 mg, 74%) as a colourless gum; [a]2D9 −96
7 (c 4.1, CHCl3); mmax/cm−1 3063, 2950, 2885, 1640, 1451, 1409,
1214, 1097, 920, 748, 737; dC (100 MHz, CDCl3) 56.0 (7-C), 67.7
(CH2), 69.2 (CH2), 73.5 (CH2), 86.0 (4b-C), 121.9, 127.8 (three),
128.1, 128.2, 128.5 (two), 129.0, 129.4, 130.0 and 130.1 (ArCH),
134.5, 136.2 (two), 138.1 and 138.4 (quaternary ArC), 164.2 (8-
C); m/z 394 ([M + Na]+, 100%), 372 ([M + H]+, 47) (Found: M +
Na, 394.1421; C24H21O3NNa requires 394.1414); Rf 0.29 (EtOAc–
hexane, 3 : 7). Further elution of the column yielded a mixed
fraction (28 mg, 23%, ratio 12e : 13e 1 : 1.6) Rf 0.29 + 0.21 (EtOAc–
hexane, 3 : 7).
Acid-induced equilibration of lactams (Table 5)
Except for entry 1, all temperatures are those of oil baths and are
approximate ( 5 ◦C).
Entry 1. A solution of the lactam 12a (11.2 mg, 0.038 mmol)
and p-toluenesulfonic acid hydrate (2.6 mg, 0.014 mmol) in
CD3CN (6 mL) was heated at 55 2 ◦C (bath) for a total of 33 d,
during which time samples were periodically removed for analysis
by 1H NMR spectroscopy. The integrals for the respective H(4b)
signals were used to monitor the change in the diastereoisomer
ratio [13a]/[12a]. The data used to plot Fig. 3 are listed in Table 6.
Table 6 Data used to plot Fig. 3
(aS,4bR,7S)-7-tert-Butyl-6,7-dihydrodibenz[c,e]oxazolo[3,2-
a]azepin-9(4bH)-one 12f and (aR,4bS,7S)-7-tert-Butyl-6,7-
dihydrodibenz[c,e]oxazolo[3,2-a]azepin-9(4bH)-one 13f
12a (%)
Time/h
100.0
98.1
96.8
77.6
65.2
58.6
46.4
43.8
40.2
38.2
36.7
36.0
0.0
9.0
15.0
Method A. Heating the acid-aldehyde 7 (280 mg, 1.24 mmol)
and (S)-tert-leucinol 11f (216 mg, 1.84 mmol) in toluene (15 mL)
under reflux (bath temp. 120 ◦C) for 36 h gave the mixed lactams
12f and 13f (total 210 mg, 55%), ratio 64 : 36 by 1H NMR
spectroscopy, as a colourless gum; mmax/cm−1 3067, 2966, 2873,
2244, 1650, 1452, 1394, 1219, 1153, 1083, 912, 741; m/z 308 ([M +
H]+, 100%).
107.0
184.5
250.5
419.5
493.0
611.0
688.0
758.5
805.5
Method B. Heating the ester-aldehyde 9 (230 mg, 0.96 mmol)
and the amino alcohol 11f (168 mg, 1.43 mmol) in xylene (15 mL)
at 155 ◦C (bath temp.) for 3 d gave no lactams by 1H NMR
spectroscopy.
For the equilibration of A and B via first-order processes, the
change in concentration of A is related to the respective forward
and reverse rate constants k1 and k−1 using the expression
Molecular mechanics calculations (Table 4, Fig. 1 and Fig. 2)
d[A]
dt
= −k1[A] + k−1[B]
(1)
Minimised steric energy (MSE) values for the lactams 12, 13
(Table 4), and 16–21 (Fig. 1) were calculated on a PC running
Linux (Fedora Core 3) with MacroModel v. 8.0 (Maestro v.
5.0 interface) using the MM3 force field and Monte Carlo
conformational search (csearch) method (1000 iterations). The
standard minimisation parameters were: no solvent; PRCG
method; convergence on gradient; max. number of iterations 3000;
convergence threshold 0.0200. The procedure returned multiple
repeats of the same global MSE in each case. The difference,
DG, for each pair of structures 12 and 13 was converted into the
equilibrium populations indicated in Table 4 using the equation
DG = −RT lnK (where R = 8.31451 J K−1 mol−1). The dihedral
driving (drive) plots shown in Fig. 2 were acquired by locating
the energy-minimised structure for 16 using csearch as described
and the equilibration process is represented by the integrated rate
equation, eqn (2).
[A]t = [A]∞ + ([A]0 − [A]∞) e−(k
The value of k−1 corresponds to k1/K, where K is the value of
[B]/[A] at equilibrium.
(2)
+k )t
−1
1
Fig. 3 was plotted using the observed isomer ratios listed in
Table 6. The calculated exponential decay curve for [12a] was
generated using eqn (2) with the variables [12a]∞, {[12a]0 − [12a]∞}
and (k1 + k−1) optimised by minimising the sum of the R2 values
for each data point. The data yielded the following parameters
(errors approximate, based on 2% accuracy in NMR integrals):
k1 = 7.3 × 10−7 s−1
(
0.3 × 10−7)
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2667
©

View Article Online
k–1 = 3.7 × 10−7 s−1
[A]∞ = 34% ( 2%)
(
0.3 × 10−7)
heated at 150 ◦C for 3 d. The equilibration, monitored by 1H NMR
spectroscopy, stabilised with an estimated 12e:13e ratio of 30 : 70.
[A]0–[A]∞ = 67% ( 2%)
Entry 11. A mixture of the lactams 12f and 13f (ratio ca.
64 : 36; 20 mg, 0.065 mmol) in toluene-d8 (0.5 mL) in an NMR
tube was treated with a solution of TFA in toluene-d8 (6.7 M,
0.1 mL, 0.67 mmol). The stoppered tube was sealed in a screw-
capped thick-walled glass tube and heated at 100 ◦C for 7 d. The
equilibration, monitored by 1H NMR spectroscopy, stabilised with
an estimated 12f:13f ratio of 26 : 74.
Entry 2. A solutionofthe lactam 12a (8.2mg, 0.028mmol) and
p-toluenesulfonic acid hydrate (2.0 mg, 0.011 mmol) in toluene-d8
(0.6 mL) was heated in a sealed tube at 140 ◦C for 3 d, during
which time the mixture was periodically analysed by H NMR
spectroscopy. The estimated equilibrium ratio 12a : 13a was 27 :
1
73.
Entry 3. A solution of the lactam 12a (11.2 mg, 0.038 mmol)
◦
in toluene-d8 (0.6 mL) was heated in a sealed tube at 140 C for
Acknowledgements
1
7 d, after which time the H NMR spectrum indicated that 12a
We are grateful to the Association for International Cancer
Research, the EPSRC and GlaxoSmithKline for their financial
support of this work. We also thank Val Boote and Steve Kelly for
assistance with MS and NMR measurements, and Drs Andrew
Regan and Ian Watt for valuable discussions. We also wish to
acknowledge the use of the EPSRC’s Chemical Database Service
at Daresbury.32
remained intact. No other products were evident.
Entry 4. A sample of the lactam 12a (20 mg, 0.07 mmol) in
toluene-d8 (0.5 mL) in an NMR tube was treated with a solution
of TFA in toluene-d8 (6.7 M, 0.1 mL, 0.67 mmol). The stoppered
tube was sealed in a screw-capped thick-walled glass tube and
◦
1
heated at 100 C for 7 d. The equilibration, monitored by H
NMR spectroscopy, stabilised with an estimated 12a:13a ratio of
29 : 71.
References
Entry 5. A sample of the lactam 12b (20 mg, 0.06 mmol) in
toluene-d8 (0.5 mL) in an NMR tube was treated with a solution
of TFA in toluene-d8 (6.7 M, 0.1 mL, 0.67 mmol). The stoppered
tube was sealed in a screw-capped thick-walled glass tube and
1 Reviews: (a) T. W. Wallace, Org. Biomol. Chem., 2006, 4, 3197–3210;
(b) G. Bringmann, A. J. Price Mortimer, P. A. Keller, M. J. Gresser, J.
Garner and M. Breuning, Angew. Chem., Int. Ed., 2005, 44, 5384–5427.
2 A. I. Meyers and J. J. Willemsen, Chem. Commun., 1997, 1573–1574
and references cited therein.
3 A. Brossi, J. Med. Chem., 1990, 33, 2311–2319. See also: U. Berg and
H. Bladh, Helv. Chim. Acta, 1999, 82, 323–325.
◦
1
heated at 100 C for 7 d. The equilibration, monitored by H
NMR spectroscopy, stabilised with an estimated 12b:13b ratio of
45 : 55.
4 A. Brossi, O. Boye´, A. Muzaffar, H. J. C. Yeh, V. Toome, B. Wegrzynski
and C. George, FEBS Lett., 1990, 262, 5–7.
Entry 6. A sample of the lactam 12b (50 mg) in xylene (10 mL)
was sealed in a screw-capped thick-walled glass tube and heated
at 150 ◦C for 2 d. Evaporation of the solvent and analysis of the
residue by 1H NMR spectroscopy indicated that no equilibration
had occurred.
5 For examples, see: K. Mikami, K. Aikawa, Y. Yusa, J. J. Jodry and M.
Yamanaka, Synlett, 2002, 1561–1578; A. Alexakis and C. Benhaim,
Eur. J. Org. Chem., 2002, 3221–3236; T. Ooi, Y. Uematsu, M. Kameda
and K. Maruoka, Angew. Chem., Int. Ed., 2002, 41, 1551–1554; B. Lygo,
B. Allbutt and S. R. James, Tetrahedron Lett., 2003, 44, 5629–5632; K.
Morioka, N. Tamagawa, K. Maeda and E. Yashima, Chem. Lett., 2006,
35, 110–111.
6 D. J. Edwards, R. G. Pritchard and T. W. Wallace, Acta Crystallogr.,
Sect. B: Struct. Sci., 2005, 61, 335–345.
7 Z. Zhang, H. Qian, J. Longmire and X. Zhang, J. Org. Chem., 2000,
65, 6223–6226.
Entry 7. A sample of the lactam 12c (20 mg, 0.06 mmol) in
toluene-d8 (0.5 mL) in an NMR tube was treated with a solution
of TFA in toluene-d8 (6.7 M, 0.1 mL, 0.67 mmol). The stoppered
tube was sealed in a screw-capped thick-walled glass tube and
heated at 100 C for 7 d. The equilibration, monitored by H
NMR spectroscopy, stabilised with an estimated 12c:13c ratio of
36 : 64.
8 D. Romo and A. I. Meyers, Tetrahedron, 1991, 47, 9503–9569; M. D.
Groaning and A. I. Meyers, Tetrahedron, 2000, 56, 9843–9873.
9 D. J. Edwards, R. G. Pritchard and T. W. Wallace, Tetrahedron Lett.,
2003, 44, 4665–4668.
◦
1
10 Y. Kubo, T. Araki and K. Maruyama, Bull. Chem. Soc. Jpn., 1985, 58,
2863–2869.
Entry 8. A sample of the lactam 12c (20 mg, 0.06 mmol) in
formic acid (4 mL) was sealed in a screw-capped thick-walled glass
tube and heated at 100 ◦C for 3 d. Evaporation of the solvent and
analysis of the residue by 1H NMR spectroscopy gave an estimated
12c:13c ratio of 43 : 57.
11 L. Giraud, E. Lacoˆte and P. Renaud, Helv. Chim. Acta, 1997, 80, 2148–
2156.
12 M. Penhoat, V. Levacher and G. Dupas, J. Org. Chem., 2003, 68, 9517–
9520.
13 M. Penhoat, S. Leleu, G. Dupas, C. Papamicae¨l, F. Marsais and V.
Levacher, Tetrahedron Lett., 2005, 46, 8385–8389.
14 R. A. Labriola and A. Felitte, J. Org. Chem., 1943, 8, 536–539.
15 E. F. M. Stephenson, J. Chem. Soc., 1954, 2354–2357.
16 P. S. Bailey and R. E. Erickson, Org. Synth., 1973, coll. vol. 5,
pp. 493–495.
Entry 9. A mixture of the lactams 12d and 13d (ratio ca.
56 : 44; 20 mg, 0.075 mmol) in toluene-d8 (0.5 mL) in an NMR
tube was treated with a solution of TFA in toluene-d8 (6.7 M,
0.1 mL, 0.67 mmol). The stoppered tube was sealed in a screw-
capped thick-walled glass tube and heated at 100 ◦C for 7 d. The
equilibration, monitored by 1H NMR spectroscopy, stabilised with
an estimated 12d:13d ratio of 34 : 66.
17 K. Schlo¨gl and R. Scho¨lm, Monatsh. Chem., 1980, 111, 259–274.
18 K. Schlo¨gl and R. Scho¨lm, Liebigs Ann. Chem., 1980, 1877–1888.
19 A. J. Mancuso and D. Swern, Synthesis, 1981, 165–185.
20 H. Kalchhauser, K. Schlo¨gl, W. Weissensteiner and A. Werner, J. Chem.
Soc., Perkin Trans. 1, 1983, 1723–1729.
21 The rotation barrier in formamide has been estimated at 72 kJ mol−1
using ab initio methods. E. D. Glendening and J. A. Hrabal, J. Am.
Chem. Soc., 1997, 119, 12940–12946. See also: X. Lopez, J. I. Mujika,
G. M. Blackburn and M. Karplus, J. Phys. Chem. A, 2003, 107, 2304–
2315.
Entry 10. A mixture of the lactams 12e and 13e (ratio ca. 50 :
50; 50 mg, 0.16 mmol) and TFA (0.2 mL, 2.7 mmol) in xylene
(10 mL) was sealed in a screw-capped thick-walled glass tube and
2668 | Org. Biomol. Chem., 2007, 5, 2658–2669
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
22 K. Higashiyama, H. Inoue and H. Takahashi, Tetrahedron Lett., 1992,
33, 235–238.
27 M. Tichy, M. Budesinsky, J. Gu¨nterova, J. Zavada, J. Podlaha and I.
Cisarova, Tetrahedron, 1999, 55, 7893–7906.
23 W. Ando, Y. Igarashi and L. Huang, Chem. Lett., 1987, 1361–1364. See
also: D. Seebach and J. D. Aebi, Tetrahedron Lett., 1984, 25, 2545–2548.
24 M. Amat, N. Llor and J. Bosch, Tetrahedron Lett., 1994, 35, 2223–
2226; M. Amat, J. Bosch, J. Hidalgo, M. Canto´, M. Pe´rez, N. Llor, E.
Molins, C. Miravitlles, M. Orozco and J. Luque, J. Org. Chem., 2000,
65, 3074–3084.
28 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Purification of
Laboratory Chemicals, Pergamon, Oxford, 2nd edn, 1980.
29 W. C. Still, M. Kahn and A. Mitra, J. Org. Chem., 1978, 43, 2923–2925.
30 V. Brandmeier and M. Feigel, Tetrahedron, 1989, 45, 1365–1376.
31 J. M. Schomaker, B. R. Travis and B. Borhan, Org. Lett., 2003, 5,
3089–3092.
25 P. Pigeon and B. Decroix, Tetrahedron Lett., 1997, 38, 1041–1042.
26 G. Hilt, F. Galbiati and K. Harms, Synthesis, 2006, 3575–3584.
32 D. A. Fletcher, R. F. McMeeking and D. Parkin, J. Chem. Inf. Comput.
Sci., 1996, 36, 746–749.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 2658–2669 | 2669
©