Development of two processes for the synthesis of bridged azabicyclic systems: Intermolecular radical addition-homoallylic rearrangements leading to 2-azanorborn-5-enes and neophyl-type radical rearrangements to 2-azabenzonorbornanes

Article abstract of DOI:10.1039/b306717n

Radical thiol additions to 7-azanorbornadienes give 7-thio-substituted 2-azanorbornenes and Barton deoxygenations of 7-azabenzonorbornanols give 2-azabenzonorbornanes. The processes both involve novel nitrogen-directed radical rearrangements. The kinetics and mechanisms of the reactions are also discussed.

Full text of DOI:10.1039/b306717n

View Article Online / Journal Homepage / Table of Contents for this issue
Development of two processes for the synthesis of bridged
azabicyclic systems: intermolecular radical addition–homoallylic
rearrangements leading to 2-azanorborn-5-enes and neophyl-type
radical rearrangements to 2-azabenzonorbornanes†‡
David M. Hodgson,*a Magnus W. P. Bebbingtona and Paul Willisb
a Dyson Perrins Laboratory, Department of Chemistry, University of Oxford,
South Parks Road, Oxford, UK OX1 3QY. E-mail: david.hodgson@chem.ox.ac.uk
b AstraZeneca, R&D Charnwood, Bakewell Road, Loughborough, UK LE11 5RH
Received 12th June 2003, Accepted 4th August 2003
First published as an Advance Article on the web 22nd August 2003
Radical thiol additions to 7-azanorbornadienes give 7-thio-substituted 2-azanorbornenes and Barton deoxygenations
of 7-azabenzonorbornanols give 2-azabenzonorbornanes. The processes both involve novel nitrogen-directed radical
rearrangements. The kinetics and mechanisms of the reactions are also discussed.
Introduction
Azabicyclic systems are ubiquitous in natural products and
pharmaceutical agents.1 New processes for their construction
are therefore of continuing synthetic interest. Radical cyclis-
ations and rearrangements are potentially among the most
powerful methods for the synthesis of such polycycles.2–4
Homoallylic radical rearrangements, such as in the cyclo-
propylmethyl-homoallyl system (Scheme 1) have been the sub-
ject of considerable mechanistic study,5 but their application to
synthesis has yet to be widely exploited.6
Scheme 2
Scheme 1
Following reports that these processes could be directed by
substitution α to the radical centre (e.g. group X, Scheme 1),7–9 we
decided to investigate the possibility that a nitrogen atom could
be used in this way,7 allowing access to otherwise synthetically
challenging azacycles. Radical reduction of norbornenyl bromide
1 or nortricyclyl bromide 4 is known to produce the same
(∼60 : 40) mixture of norbornene 5 and nortricyclene 6 (Scheme
2).10,11 However, during the development of novel analgesics
related to epibatidine by a strategy involving indirect skeletal
interconversion of 7-aza- to 2-azabicyclo[2.2.1]heptyl ring
systems, we reported that radical deoxygenation of azanor-
tricyclanol 7 delivers a single product 8, even when R is poten-
tially radical stabilising (e.g. Ar, Scheme 3).12,13
Scheme 3 Reagents and conditions: i, ClCOCO2Me, DMAP, MeCN,
25 ЊC, 30 min; ii, Bu3SnH, AIBN, toluene, 100 ЊC, 45 min.
direct skeletal interconversion might be possible by inter-
molecular radical addition–homoallylic radical rearrangement
(Scheme 4). This would provide convergent access to synthetic-
ally important 2-azabicyclo[2.2.1]heptyl ring systems contain-
ing substitution not easily available by other methods.1 We also
considered that the rearrangement could be applicable to the
synthesis of a variety of other azabicyclic ring systems. Herein
we report a full account of our results concerning the develop-
ment of nitrogen-directed free radical rearrangements.14,15
Given this demonstration of ‘nitrogen-directed’ radical
rearrangement, we considered that a synthetically attractive
† This is one of a number of contributions from the current members
of the Dyson Perrins Laboratory to mark the end of almost 90 years of
organic chemistry research in that building, as all its current academic
staff move across South Parks Road to a new purpose-built laboratory.
‡ Electronic supplementary information (ESI) available: the prepar-
ation and characterisation of compounds 20b–i, 28c–g, 30b–g, 30i,
32e–g, 32i, 33b–g. See http//www.rsc.org/suppdata/ob/b3/b306717n/
Results and discussion
Radical additions to norbornadiene 12 and its derivatives, using
thiols for example, have been well-studied and usually result in
substituted nortricyclanes as the major products.16 Consistent
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3787

View Article Online
Table 1 Relative proportions of 13 and 14 formed from addition of
p-thiocresol to norbornadiene 12 at varying concentrationsa
Table 2 Thiol additions to 7-azabicyclo[2.2.1]heptadiene 19a
Entry
R
T/ЊC
t/h
Yield (%)
syn-20 : anti-20
[Thiol]initial/mol dmϪ3
1H NMR ratiob 13c : 14
a
b
c
d
e
f
Tol a
Ph
20
20
20
20
20
80
80
80
24
24
24
72
24
6
92
92
90
50
59
59
48
56
3 : 1
4 : 1
4 : 1
7 : 1
2.5 : 1
2 : 1
5.0 (neat)
2.0
1.0
40 : 60
26 : 74
17 : 83
14 : 86
2,6-DiMeC6H3
4-NO2C6H4
4-MeOC6H4
Bun
0.1
a The appropriate concentration of p-thiocresol was added to a solution
of 12 in toluene in one portion at 25 ЊC. b Average of two runs.
c exo : endo ratio of sulfides 13 was 5 : 1.
g
h
But
24
14
5 : 1
4 : 1
HO(CH2)3
a TolSH with 19b (PG = CO2Me) at 20 ЊC for 4 h gave 20i (R = Tol,
PG = CO2Me) in 66% yield (syn : anti, 4 : 1).
an interaction between the SOMO and the adjacent C–S σ
or σ* orbitals confers stability on radical 16, which drives the
cyclopropane ring opening leading to its formation. Such an
interaction would clearly not be possible in radical 18, and
furthermore exo-approach of the H-atom donor would be
hindered by the presence of the 7-tolylthio group which would
lie predominantly syn to the radical centre.
If the rates of H-atom transfer to 16 and 17 were assumed to
be equal, the product ratio would be a direct reflection of the
16–17 equilibrium position, but there does not seem to be any
evidence to support this assumption. Indeed, it has been argued
that norbornenyl radicals react more rapidly with certain
H-atom donors than do nortricyclyl radicals, and that the radi-
cal equilibrium 2–3 favours 3, even though more product from
the reduction of 2 is actually isolated.21 Our results suggest a
difference in the rate of H-atom transfer from p-thiocresol to 16
and 17. If radical 16 is stabilised, as we have suggested above,
and the rates of H-atom transfer to 16 and 17 were equal, the
ratio of products 13 : 14 would be greater than the ratio 5 : 6,
since radical 2 cannot be subject to the same stabilising inter-
action as 16. However, this analysis contradicts the experi-
mental results. Attack of the thiyl radical on norbornadiene is
known to occur predominantly from the exo face,22 and given
the results of Davies et al.,23 approach of the H-atom donor
(another molecule of p-thiocresol) would be expected to occur
from the same face. For this reason, the rate of H-atom transfer
to radical 16 could conceivably be slower than that to 17, as the
approach of the thiol would be subject to steric interactions
with the β-thio substituent. It is therefore likely that pro-
portionately more of 17 would be trapped than 16, so that the
ratio 13 : 14 does not reflect the 16–17 equilibrium position.
This analysis thus provides a plausible explanation for both
the absence of sulfide 15 in the addition of p-thiocresol to
norbornadiene and the differing ratios of bicyclic and tricyclic
products obtained from equilibria of substituted and unsubsti-
tuted norbornenyl and nortricyclyl radicals.
Scheme 4
Scheme 5
with these earlier studies, we found that addition of p-thiocresol
to norbornadiene 12 at 25 ЊC led to a mixture of nortricyclyl
sulfide 14 and bicyclic sulfide 1317,18 (Scheme 5, Table 1).
Further experiments at increasing dilution indicated that
the product ratio reached a constant level (within experi-
mental error, 13 : 14, ∼ 1 : 6), once the concentration of each
reactant was <0.1 mol dmϪ3. These latter results indicate
attainment of equilibrium between the substituted norbornenyl
and nortricyclyl radicals 16 and 17.
The ratio of norbornenes : nortricyclane products in this
case differs significantly from that obtained from generation of
the unsubstituted radicals, where a 60 : 40 ratio in favour of
norbornene is obtained at various concentrations.10,11 This
difference in behaviour requires an explanation. Given that no
7-substituted norbornenes 15 derived from radical 18 are pro-
duced in the thiol additions, then, one possibility is that ring-
opening of nortricyclyl radical 17 produces only norbornenyl
radical 16. Wong and Griller, however, were able to detect
7-substituted radicals in EPR studies on the addition of tert-
butoxyl to norbornadiene,19 which suggests that ring opening
of nortricyclyl radicals such as 17 to 7-substituted norbornenyl
radicals is not prevented stereoelectronically. An alternative
explanation is that the ring opening of 17 leading to 16 is pro-
moted by a radical stabilising effect of the sulfur substituent
β to the radical centre. A sulfur-bridged intermediate is thought
to be unlikely in similar bicyclic systems,20 but it is possible that
By comparison to the reactions of norbornadiene with thiols,
the corresponding 7-aza systems such as 19a and 19b (available
in two steps from alkoxycarbonyl-protected pyrrole and
tosylethyne)24 behave in dramatically different fashion (Scheme
6, Table 2), but in accord with the earlier analysis (Scheme 4).
Pleasingly, both aromatic and aliphatic thiols were found to
react cleanly with only a slight excess (1.1 equiv.) of azadiene
19a or 19b in benzene or toluene (0.1 mol dmϪ3) to give
Scheme 6 Reagents and conditions: i, RSH (0.9 equiv.), benzene or
toluene, 20 ЊC (R = aryl) or 80 ЊC (R = alkyl), 6–72 h.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3788

View Article Online
rearranged sulfides 20a–i in 6–72 h, depending on the reactivity
of the thiol. In the case of the less reactive aliphatic thiols, heat
was required for completion of the reaction in a satisfactory
time. Aliphatic thiols are considerably poorer H-atom transfer
agents than their aromatic counterparts.25 This may explain the
lower yields in these cases, as polymerisation could compete
with atom transfer. The major diastereoisomer of 20a in the
reaction with p-thiocresol was shown by NOE analysis to be the
syn-isomer arising from initial exo-attack of the thiyl radical on
azadiene 19a, and the predominant isomer obtained with the
other thiols is assigned as syn by analogy. Selectivity for
exo-attack was modest, which is consistent with formation of a
long C–S bond via a transition state whose energy is not greatly
affected by the steric bulk of the reacting partners. However, a
more hindered thiol, such as tert-butylthiol, shows higher
selectivity than the straight chain examples (entries f–h). This
suggests exo-attack is indeed less hindered, but the highest
selectivity was observed in the case of p-nitrothiophenol. The
p-nitrophenylthiyl radical is subject to greater delocalisation
than the other aromatic thiyl radicals, thereby increasing its
relative stability. This suggests that the p-nitrophenylthiyl
radical addition to 19a is less exothermic than in the other
examples, resulting in a later transition state and greater
selectivity for the comparatively less hindered exo attack.
Evidence that the product sulfides 20 possess the rearranged
2-azabicyclo[2.2.1]hept-5-ene framework is that oxidation of
20a (R = Tol) using buffered peracetic acid gave in 91% yield the
epimeric sulfones 21 (Scheme 7) with spectral data distinctly
different from the sulfones 2326 that would be obtained from
oxidation of the product of simple thiol addition across one
double bond in azadiene 19a. Subsequent desulfonylation of
sulfones 21 with 6% sodium-amalgam gave the known 2-aza-
bicyclic alkene 2212,13,27 in 33% yield as the only identifiable
product (Scheme 7).
Scheme 8 Reagents and conditions: i, PhSeH (0.9 equiv.), benzene,
25 ЊC. 24 h.
Fig. 1
can estimate that the true ratio of exo/endo attack is approx-
imately 2 : 1, 28 lower than for the addition of thiophenol. This
may reflect greater d-orbital interactions between the selenium
atom and the π-orbital lobes which converge on the endo face of
the diene and therefore increase the propensity for endo-attack
compared to the thiols.16
The combined results of the thiol and selenol additions to
19a allow an estimation of the rate constant for the rearrange-
ment 9
11 (Scheme 4). No products other than the 2-aza-
bicyclic epimers 20 are observed in the addition of aromatic
thiols. This suggests that the rate constant for the rearrange-
ment is at least one order of magnitude faster than the initial
rate of H-atom transfer from such thiols. Therefore an initial
thiol concentration of 0.1 mol dmϪ3 (typical in our studies) and
an approximate second order rate constant for H-atom transfer
of 108 dm3 molϪ1 sϪ1 25 indicate a rate constant of at least 108 sϪ1
,
for the rearrangement at 25 ЊC. The isolation of small amounts
of unrearranged selenides 25 in the selenol addition suggests
that the rate of rearrangement cannot be more than one order
of magnitude faster than H-atom transfer from benzeneselenol.
A second order H-atom transfer rate constant of 109 dm3 molϪ1
sϪ1 25 and a selenol concentration of 0.1 mol dmϪ3 indicate an
,
upper limit for the rate constant of 109 sϪ1 at 25 ЊC.
The absence of tricyclic products in any of the reactions with
azadienes 19 suggests that the lifetime of the azatricyclic inter-
mediate 10 is much shorter than that for either of the bicyclic
intermediates 9 and 11 (Scheme 4). In kinetic studies of the
nortricyclyl–norbornenyl radical rearrangement [2 and 3,
Scheme 2], the rates of ring opening and ring closure have been
shown to be similar: ∼107 sϪ1 at 25 ЊC.19 In the azabicyclic
system, however, it seems probable that the ring opening step
(10
11) is significantly faster, with the transition state being
lowered in energy by the stabilisation of the developing radical
α to nitrogen.29,30 Rate constants for the ring opening of cyclo-
propylmethyl radicals to but-3-enyl radicals are known to be
strongly influenced by substitution.5
Scheme 7 Reagents and conditions: i, CH3CO3H, NaOAc, CH2Cl2,
0 ЊC, 6 h, then 20 ЊC, 14 h; ii, Na–Hg, NaH2PO4, Na2HPO4, MeOH,
25 ЊC, 12 h.
Having demonstrated that nitrogen-directed homoallylic
radical rearrangement was a viable method for the synthesis of
azacycles, we next considered extending its application to a
related process—radical aryl migration. Compared with
radical rearrangements in norbornenyl systems, neophyl-type
rearrangements (1,2-shift of an aryl group, Scheme 9, X = CH2)
in the corresponding benzofused systems are much rarer.31 This
applies both to free radical additions to benzonorbornadiene
and to H-atom abstraction from benzonorbornane.32–34 Indeed,
many additions to benzonorbornadiene were found to proceed
Benzeneselenol shows similar behaviour with unsaturated
systems to aromatic thiols, but is an order of magnitude
superior as an H-atom transfer agent.25 Under otherwise identi-
cal conditions to the earlier reactions of aromatic thiols (Table
2), it was found that the reaction of benzeneselenol with aza-
diene 19a gives a small proportion (12%) of the selenide 25
arising from simple addition across one double bond, as well as
the expected rearranged selenide 24 (81%, Scheme 8).
Interestingly, the epimeric ratio of selenides 25 (exo : endo
1 : 6) was found to be different from the epimeric ratio of
rearranged selenides 24 (syn : anti 3 : 1), as determined by a
NOESY experiment. This could be explained if the rate of
H-atom transfer to endo-phenylselenyl radical 27 is faster than
that to exo-phenylselenyl radical 26 for steric reasons (Fig. 1).
From the yields and epimeric ratios of products 24 and 25 we
Scheme 9
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3789

View Article Online
without any rearrangement. This is presumably due to the
comparatively unfavourable disruption of aromaticity.35
These early observations, together with our more recent
studies (vide supra), suggested that extending the ‘aza homo-
allylic rearrangement’ process to 7-azabenzonorbornyl systems
(Scheme 9, X = NR) would be challenging, but could provide an
attractive entry to 2-azabenzonorbornanes.36 The latter systems
are of interest, for example as conformationally defined
adrenergic agents.37
7-Azabenzonorbornadienes (e.g. 28a, Scheme 10) were con-
sidered to be excellent substrates to examine the chemistry in
Scheme 9 (X = NR), since they are readily available via aryne
cycloadditions38 to pyrroles.39 However, radical addition of
p-thiocresol to 28a40 gave only simple addition to the double
bond, without any rearrangement, to give 29 (Scheme 10),
solely as the exo-isomer.
31a then proceeded smoothly in excellent yield (95%). We
were pleased to find that treatment of xanthate 31a with tris-
(trimethylsilyl)silane (TTMSS) (1.5 equiv., 0.12 mol dmϪ3
initial concentration) in boiling toluene with AIBN as initiator
for 2 h gave a 1 : 1 mixture of 32a,43 the product of direct
reduction, and 33a, which was identified as the desired
rearranged product by NMR studies and by comparison with
known alkene 20 (Scheme 7).27
The skeletal structure of the rearranged product was con-
firmed by hydrolysis of the N-methoxycarbonyl-protected
rearranged product 33b (prepared in analogous fashion to 33a)
to give the known amine 34.37 We supposed that increasing the
lifetime of the first-formed radical would lead to a higher
proportion of the rearranged product 33a. By increasing
the dilution by a factor of 3, and adding a mixture of TTMSS
and AIBN to a pre-heated solution of xanthate 31a over
100 minutes, the ratio increased to ∼20 : 1, and gave a 90%
isolated yield of 33a.
Encouraged by these results, we undertook a more detailed
study of the effect on the rearrangement of bicyclic core substi-
tution and also of variation of electronics in the aromatic ring.
3-Ethylpyrrole 3644 was prepared by hydrolysis of the known
1-phenylsulfonyl-3-ethylpyrrole 35.45 Both 3-ethylpyrrole and
commercially available 2,4-dimethylpyrrole were Boc-protected
to give substituted Boc-pyrroles 37 and 38 which were used
in benzyne cycloadditions (Scheme 12)40 to give adducts 28c
and 28d. Hydroboration then occurred with complete facial
and regioselectivity to produce alcohols 30c and 30d as single
diastereomers (Scheme 13).
Scheme 10 Reagents and conditions: i, TolSH, toluene, 25 ЊC, 24 h.
Therefore, hydroboration of 28a followed by oxidation to the
alcohol and then Barton deoxygenation was thought an
attractive alternative method for radical generation, since
hydroboration is known to proceed with good facial and regio-
selectivity in substituted alkenes,41 and the rate of H-atom
transfer would be slowed relative to a thiol.25 Hydroboration–
oxidation of 28a gave alcohol 30a as a single isomer (Scheme 11)
in satisfactory yield (68%), assigned as the exo-product
in accordance with previous work.42 Formation of xanthate
Radical deoxygenation of the xanthates 31c and 31d in the
manner previously described for 31a gave good yields of the
rearranged products (Scheme 13). Only traces of directly-
reduced products were observed in the crude 1H NMR spectra
1
[by comparison with the H NMR data of 32a (vide supra)].
Noteworthy is that NOESY experiments revealed that H-atom
transfer to form 33d occurs exclusively from the exo face, lead-
ing to a single diastereomer. These studies also showed that
the 7-alkyl substituent in 33c and 33d resides anti to nitrogen,
confirming that the earlier hydroborations were completely
exo-selective.
To determine the effect of arene ring electronics on the
neophyl-like migration, we prepared adducts 28e–g (Scheme 12)
by diazotization of the appropriate aromatic amino acid deriv-
atives in the presence of N-Boc-pyrrole.46 Hydrogenation43 of
these adducts was conducted to give reference samples of the
products of direct reduction 32e–g, as substitution of the aro-
matic ring was expected to affect the ratio of directly-reduced :
rearranged products.
Hydroboration–oxidation and then deoxygenation of the
derived xanthates 31e–g gave the expected products 33e–g in
good yields (Scheme 13). Dimethoxy-substitution in 31e was
found to have a retarding effect on the rearrangement, as shown
by a 6 : 1 ratio of rearranged : directly-reduced products—13%
of 32e was isolated in addition to 33e. This is consistent with
the initial formation of a nucleophilic secondary alkyl radical
and a slower cyclisation onto the more electron-rich aromatic
ring than in the unsubstituted system. This ratio in the reaction
of the difluorinated xanthate 31f did not appear to differ greatly
from that of 33a : 32a, which suggests only a mild overall elec-
tron-withdrawing effect due to the fluorine substituents. None
of the directly-reduced product was detected in the reaction of
31g, which is in agreement with rate-enhancements observed
(compared to phenyl) in studies of other naphthyl group radical
migrations.31
At this point, we chose to investigate the possible driving
force(s) for the nitrogen-directed rearrangements we had
observed. It was considered that there are potentially two
explanations for the selectivity of the rearrangements: first, that
interaction of the nitrogen lone pair with the adjacent radical
centre gives rise to a stabilising interaction (Fig. 2) in the
Scheme 11 Reagents and conditions: i, 9-BBN, THF, 25 ЊC, 24 h;
ii, KH, CS2, MeI, THF, 0–25 ЊC, 90 min; iii, PhOC(S)Cl, DMAP,
CH2Cl2, 25 ЊC, 18 h; iv, TTMSS, AIBN, toluene, reflux, 2 h; v, KOH,
ethylene glycol–water, reflux, 14 h.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3790

View Article Online
Scheme 12 Reagents and conditions: i, NaOH, MeOH, reflux, 2.5 h; ii, Boc2O, DMAP, MeCN, 25 ЊC, 16–24 h; iii, o-bromofluorobenzene, Mg, THF,
reflux, 2 h; iv, 3,4-dimethoxyanthranilic acid, isoamyl nitrite, MeCN, 25 ЊC, 30 min, then reflux, 2 h; v, 3,4-difluoroanthranilic acid, isoamyl nitrite,
THF, reflux, 90 min; vi, 3-amino-2-naphthoic acid, isoamyl nitrite, dioxan, reflux, 4 h; vii, H2 (1 atm), 5% Pd/C, MeOH, 25 ЊC, 10 min.
Table 3 Effect of the nitrogen protecting group R on rearrangement
product profile
R
Y
Yield 32 (%)
Yield 33 (%)
Boc
CS2Me
C(S)OPh
CS2Me
10
0
14
72
23
61
Me
C(O)tBu
Using the original N-Boc protected system, the ratio of
the directly-reduced : rearranged products was 1 : 7 when the
silane was added as one portion at the same concentration
([silane]initial = 0.04 mol dm Ϫ3) which suggests that the change to
N-methyl had accelerated the rearrangement. This result con-
firmed that a change in the amide-type resonance was not
essential for the rearrangement to occur, but given the low iso-
lated yield of 33h, we felt that it was not possible to confirm
that resonance was not a contributing factor in this chemistry.
For this reason, we elected to prepare a substrate with nitrogen
protected as an amide. In this case the lone pair is less available
for interaction with a radical centre, but resonance is more
significant than for the original carbamate. 1-tert-Butylcarb-
onylpyrrole 3948 was readily available from pyrrole and pivaloyl
chloride in the presence of DMAP and Et3N. Benzyne cyclo-
addition and hydroboration led to the alcohol 30i (Scheme 14).
Reaction of the derived xanthate 31i then gave the rearranged
product 33i and the directly reduced 32i (Scheme 15, Table 3).
Fig. 2
rearranged radical (e.g. Scheme 9, X = NR) that is not available
to the first formed radical. Second, that the C–N–C bond angle
is larger for the rearranged than for the first formed radical,
which could allow more significant and stabilising amide-type
resonance in the rearranged radical when R = Boc.
In order to distinguish between these two possibilities, the
protecting group on nitrogen was altered. The N-methyl pre-
cursor 31h was available from reduction of alcohol 30a (Scheme
14). Deoxygenation of 31h using TTMSS–AIBN added as one
portion gave the rearranged amine 33h as a single product,
albeit in low yield (23%), with no 32h detected (Scheme 15,
Table 3). A sample of 32h47 prepared independently from 32a
was subjected to the original deoxygenation conditions and was
found to be stable—the amine was fully recovered.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3791

View Article Online
Scheme 13 Reagents and conditions: i, 9-BBN, THF, 25 ЊC, 24 h; ii, KH, CS2, MeI, THF, 0–25 ЊC, 90 min; iii, TTMSS, AIBN, toluene, reflux, 2 h.
Significantly, the fraction of rearranged product was less than
in the N-Boc case. These three results suggest that SOMO-lone
pair interactions are responsible for the rearrangements. The
closer in energy the orbitals are, the greater the stabilising inter-
action (Fig. 2). There is considerable precedent for such an
interaction, as it is thought to be responsible for the low α-C–H
bond dissociation enthalpies of amines.30,49 In accordance with
this analysis, in the current chemistry the N-methyl substrate
showed the highest propensity for rearrangement, and the
N-tert-butylcarbonyl the lowest. The idea that changes in the
amide-type resonance were responsible for the rearrangements
can therefore be discounted.
The fact that the rate of the neophyl-like rearrangement
appears to depend on the substitution at nitrogen has impli-
cations for the mechanism of the rearrangement. Any rate-
determining step must be affected by the proposed orbital
interaction, in the absence of significant structural variation in
the bicyclic core with N-substitution. There has been much
debate over whether a species of type 41 (Scheme 16) is a dis-
crete intermediate or a transition state in the original neophyl
rearrangement.31 Ingold50 was able to generate radical 43 by
H-atom abstraction and measure its decay spectroscopically to
44, a process which occurred on the nanosecond timescale.
An estimation of the rate constant for H-atom transfer
from TTMSS to a secondary alkyl radical can be obtained from
the Arrhenius function derived by Chatgilialoglou and co-
workers.51 Using T = 384 K (the boiling point of toluene) gives a
Ϫ1
second order rate constant of ∼7 × 105 dm3 mol
s
Ϫ1. This in
turn allows an estimate of the rate constant for rearrangement:
reduction of xanthate 31a using TTMSS at an initial concen-
tration of 0.12 M gave approximately equal amounts of
rearranged and directly reduced products. This suggests that
the rate of rearrangement is, within an order of magnitude,
equal to the average rate of H-atom transfer to give the directly
reduced product. The first-order rate constant for the
rearrangement is thus approximately 105 sϪ1
.
It seems most unlikely that ring opening of 41 to give 42
could be rate-determining, as it would occur too quickly to be
the slowest step of this reaction.50 Instead, formation of 41,
with the accompanying disruption of aromaticity, is more likely
to be rate-determining. Under these circumstances it is difficult
to see how the product profile would show such a marked
dependence upon substitution at nitrogen, since the orbital
interaction that drives the rearrangement could not affect the
reaction until the radical began to develop α to nitrogen. How-
ever, the product profile can be explained if 41 is a transition
state, and the rearrangement of 40 to 42 occurs in one step. The
energy of the transition state would then be lowered by stabilis-
ation of the radical developing α to nitrogen and depend
directly on the N-protecting group.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3792

View Article Online
distilled from benzophenone ketyl, hydrocarbons from CaH2,
and alcohols from their magnesium alkoxides. All reactions
were monitored by TLC using commercially available (Merck
or Camlab) plates, pre-coated with a 0.25 mm layer of silica
containing a fluorescent indicator. Visualisation of reaction
components was achieved with 254 nm light, and with vanillin
or KMnO4 dips. Organic layers were dried using MgSO4, evap-
orated on a Buchi rotary evaporator, followed by drying on a
high vacuum oil pump (∼1 mmHg). Column chromatography
was carried out on Kieselgel 60 (40–63 µm). Melting points
were determined using a Gallenkamp hot stage apparatus
and are uncorrected. Elemental analysis was performed by
Elemental Microanalysis Limited, Okehampton, Devon, UK.
IR spectra were recorded as either KBr discs or thin films, using
a Perkin-Elmer 1750 FTIR spectrometer. Peak intensities are
specified as strong (s), medium (m) or weak (w). Only selected
absorbancies are reported. 1H NMR spectra of compounds
were recorded in CDCl3 unless otherwise stated, using a Varian
Gemini 200 (200 MHz), Bruker DPX400 (400 MHz) or
AMX500 spectrometer (500 MHz). Chemical shifts (δ) are
reported relative to CHCl3 (δ 7.27). Coupling constants (J) are
given in Hz, multiplicities are given as multiplet (m), doublet
(d), triplet (t) and quartet (q). 13C NMR spectra were recorded
on the Bruker DPX400 (100 MHz) or AMX500 (125 MHz)
spectrometer. Chemical shifts are reported relative to CDCl3
(central line of triplet δ 77.0) unless stated otherwise. 19F NMR
spectra were recorded on a Bruker AMX 500 (235 MHz)
spectrometer and chemical shifts are reported relative to CFCl3
(δ = 0.0). Mass spectra were obtained from the EPSRC Mass
Spectrometry Service Centre, Swansea with a Micromass,
ZAB-E instrument or 900 XLT high resolution double focusing
mass spectrometer with tandem ion trap. Alternatively they
were recorded in-house using a VG mass Lab. TRIO1 (GCMS)
or Micromass platform APCI spectrometer. All organolithiums
were titrated against 2,2,2Ј-trimethylpropionanilide before use.
A known amount of the amide was dissolved in THF (2 cm3)
and the organolithium was then added to this stirring solution
until a colour change was observed (clear colourless to bright
orange).‘Petroleum ether’ refers to the fraction boiling in the
range 30–40 ЊC.
Scheme 14 Reagents and conditions: i, LiAlH4, Et2O, reflux, 16 h;
ii, PhOC(S)Cl, DMAP, CH2Cl2, 25 ЊC, 16 h; iii, ButCOCl, DMAP,
Et3N, CH2Cl2, 25 ЊC, 24 h; iv, anthranilic acid, isoamyl nitrite, THF,
reflux, 3.5 h; v, rac-BINAP, [Rh(COD)Cl]2, catecholborane, THF,
25 ЊC, 3 h, then NaOH, H2O2, 0–25 ЊC, 16 h.
Scheme 15 Reagents and conditions: i, TTMSS, AIBN, toluene, reflux,
2 h.
7-(tert-Butoxycarbonyl)-7-azabicyclo[2.2.1]hepta-2,5-diene
19a52
2-p-Toluenesulfonyl-7-(tert-butoxycarbonyl)-7-aza-
bicyclo[2.2.1]hepta-2,5-diene26 (5.20 g, 14.9 mmol) was added
to a 250 cm3 beaker containing a stirrer bar, methanol (80 cm3),
Na2HPO4 (8.0 g, 56 mmol) and 6% Na–Hg (5.0 g) at Ϫ10 ЊC.
The mixture was stirred at 0 ЊC for 6 h. Water (100 cm3) was
then added and the mixture filtered. The filtrate was extracted
with CH2Cl2 (3 × 100 cm3), washed with brine (100 cm3), dried
(MgSO4) and the solvent removed at reduced pressure. Column
chromatography (SiO2, 20% Et2O : petroleum ether) gave 19a
(690 mg, 23%) as a colourless oil: Rf (25% Et2O : petroleum
ether) 0.33; νmax(thin film)/cmϪ1 2978m, 1708s, 1480w, 1458w,
Scheme 16
In summary, we present our work so far in the field of nitro-
gen-directed radical rearrangements. Two synthetically useful
and conceptually interesting radical rearrangement processes
have been established, leading to azabicycles that are not readily
accessible by other means. The nature of the orbital interaction
that drives the rearrangement has been confirmed. Extensions
of the processes to provide enantio-enriched products, as well
as manipulation of the products to targets of biological interest
are under investigation.
1345s, 1172s; δ (200 MHz) 6.99 (4 H, d, J 7.0, 4 × HC᎐C), 5.19
᎐
H
(2 H, s, C(1)H and C(4)H), 1.41 (9 H, s, But); δC(100 MHz)
154.6 (C᎐O), 144.5 (C᎐C), 143.5 (C᎐C), 80.3 (CMe ), 66.7
᎐
᎐
᎐
3
(CH), 66.1 (CH), 28.0 and 28.2 (CMe3); m/z (CIϩ, NH3) 194
(M ϩ Hϩ, 82%), 110 (10), 94 (M Ϫ Boc, 100) (Found: M ϩ Hϩ,
194.1181. C11H15NO2 requires 194.1181).
7-(Methoxycarbonyl)-7-azabicyclo[2.2.1]hepta-2,5-diene 19b24
Trifluoroacetic acid (1.5 cm3, 19.5 mmol) was added to a solu-
tion of diene 19a (186 mg, 0.96 mmol) in CH2Cl2 (15 cm3) and
stirred at room temperature for 1 h. The crude TFA salt was
azeotroped with toluene (3 × 40 cm3) and the resulting brown
solid (310 mg, >100%) used without further purification.
Methoxycarbonylation was achieved using the procedure of
Corey et al.53 as follows. To a solution of the TFA salt prepared
Experimental
All reactions requiring anhydrous conditions were conducted in
flame dried apparatus under an atmosphere of argon. Syringes
and needles for the transfer of reagents were oven dried and
allowed to cool in a desiccator over P2O5 before use. Ethers were
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3793

View Article Online
in the previous step in acetone (5 cm3) was added K2CO3
(830 mg, 6.0 mmol), and methyl chloroformate (310 µL, 4.0
mmol) and the mixture heated to reflux for 18 h. Water (10 cm3)
was added and the aqueous layer was extracted with Et2O (3 ×
5 cm3) and the combined extracts washed with 1 M NaOH
(10 cm3), brine (10 cm3), dried (MgSO4) and the solvent
removed at reduced pressure. Chromatography (SiO2, 50% Et2O
: petroleum ether) gave the carbamate 19b as a yellow oil
(92 mg, 63% from 18a): Rf (50% Et2O : petroleum ether) 0.35;
chromatography (SiO2, Et2O) gave 21 as a white solid (500 mg,
91%): Rf (50% Et2O : petroleum ether) 0.28; mp (from Et2O)
110–112 ЊC (Found: C, 61.7; H, 6.7; N, 4.0. C18H23NO4S
requires C, 61.7; H, 6.6; N, 4.0%); νmax(KBr)/cmϪ1 2963s,
2892m, 1688s, 1597s, 1408s, 1364s, 1260s, 1120s, 1068s; δH (500
MHz, 90 ЊC, DMSO-d6, syn : anti, 3 : 1) (syn-epimer) 7.77 (2 H,
d, J 8.0, 2 × CH of aromatic), 7.47 (2 H, d, J 8.0, 2 × CH of
aromatic), 6.50–6.47 (1 H, m, 1 × HC᎐C), 6.41–6.38 (1 H, m,
᎐
1 × HC᎐C), 4.66 (1 H, br s, C(1)H), 3.72 (1 H, dd, J 9.0 and 3.0,
᎐
δH (200 MHz) 7.02 and 7.00 (4 H, 2 × s, 4 × HC᎐C, split by
C(3)H exo), 3.39 (1 H, s, C(4)H), 3.31 (1 H, s, C(7)H), 2.52
(1 H, d, J 9.0, C(3)H endo), 2.45 (3 H, s, Me), 1.42 (9 H, s, But),
(anti-epimer) 7.69 (2 H, d, J 8.0, 2 × CH of aromatic), 7.46
᎐
carbamate rotamers), 5.25 (2 H, s, C(1)H and C(4)H), 3.62
(3 H, s, Me).
(2 H, d, J 8.0, 2 × CH of aromatic), 6.20–6.17 (2 H, m, HC᎐
᎐
CH), 4.67 (1 H, s, C(1)H), 3.79 (1 H, s, C(7)H), 3.39 (1 H, s,
C(4)H), 3.37 (1 H, dd, J 9.5 and 3.0, C(3)H exo), 2.52 (1 H, d,
J 9.0, C(3)H endo), 2.44 (3 H, s, Me), 1.38 (9 H, s, But); δC (125
General procedure for thiol/selenol additions
The thiol or benzeneselenol (0.9 equiv.) was added as one
portion to a solution of diene 19a or diene 19b (∼0.1 mol dmϪ3
)
MHz, 90 ЊC, DMSO-d ) (both isomers) 155.8 (C᎐O), 145.4,
᎐
6
in benzene or toluene and stirred under argon for the specified
time at the specified temperature. The solvent was then removed
at reduced pressure. In all cases the products were obtained as
an inseparable mixture of epimers. 1D NOE experiments estab-
lished that the relative stereochemistry of the major epimer was
with the thio group syn to nitrogen for 20a: irradiation of the
proton at C7 showed enhancement of the olefinic protons. The
stereochemistries of the other products were assigned by
analogy.
145.2, 139.2, 138.3, 137.3, 134.9, 134.4, 130.9, 130.7, 128.8,
128.7 (all C᎐C or C of aromatic), 80.6 (C7), 80.1, 79.7 (both
᎐
CMe3), 75.4 (C7), 62.2, 60.8 (both C1), 47.7 (C3), 45.9, 45.1
(both C4), 44.1 (C3), 29.1, 29.0 (both CMe3), 21.8 (Me–Ar); m/z
(CIϩ, NH3) 367 (M ϩ NH4ϩ, 18%), 350 (M ϩ Hϩ, 7), 311 (100),
250 (10), 201 (23), 189 (23), 157 (14), 96 (17) (Found: M ϩ Hϩ,
350.1429. C18H24NO4S requires 350.1426).
2-(tert-Butoxycarbonyl)-2-azabicyclo[2.2.1]hept-5-ene 2227 by
desulfonylation
2-(tert-Butoxycarbonyl)-7-syn/anti-(p-tolylthio)-2-azabicyclo-
[2.2.1]hept-5-ene 20a
Freshly prepared 6% Na–Hg (3.76 g, 9.8 mmol Na), and
Na2HPO4 (1.5 g, 10.5 mmol) were added to a stirred solution
of sulfones 21 (376 mg, 1.07 mmol) in anhydrous methanol
(10 cm3) under argon at Ϫ10 ЊC. The reaction was warmed to
25 ЊC over 3 h and left for a further 9 h at 25 ЊC. Water (10 cm3)
was added and the mixture filtered. The filtrate was extracted
with CH2Cl2 (3 × 20 cm3), the combined extracts dried (MgSO4)
and the solvent removed at reduced pressure to give an oil.
Chromatography (SiO2, 15% Et2O : cyclohexane) gave the carb-
amate 22 as a clear colourless oil (73 mg, 33%): Rf (50% Et2O:
petroleum ether) 0.63; δH (400 MHz) (3 : 2 mixture of rotamers)
Reaction of diene 19a (100 mg, 0.52 mmol) and p-thiocresol
(56 mg, 0.45 mmol) as described above for 24 h at 20 ЊC gave a
yellow oil after removal of solvent. Column chromatography
(SiO2, 15% Et2O : petroleum ether) gave 20a as a colourless oil
(132 mg, 92%): Rf (25% Et2O : petroleum ether) 0.24; νmax(thin
film)/cmϪ1 2975m, 2929s, 2889m, 1699s, 1493s, 1394s, 1363s,
1250s, 1177m, 1168s, 1100s; δH (500 MHz, 90 ЊC, DMSO-d6, syn
: anti, 3 : 1), (syn-epimer) 7.31 (2 H, d, J 8.0, 2 × CH of aro-
matic), 7.16 (2 H, d, J 8.0, 2 × CH of aromatic), 6.43 (2 H, m,
HC᎐CH), 4.45 (1 H, s, C(1)H), 3.54 (1 H, dd, J 9.5 and 3.0,
᎐
6.35 and 6.27 (2H, 2 × s, 2 × HC᎐C), 4.71 and 4.57 (1 H, 2 s,
᎐
C(3)H exo), 3.25 (1 H, s, C(7)H), 3.06 (1 H, s, C(4)H), 2.59 (1 H,
d, J 9.5, C(3)H endo), 2.30 (3 H, s, Me), 1.42 (9 H, s, But), (anti-
epimer) 7.26 (2 H, d, J 8.0, 2 × CH of aromatic), 7.15 (2 H, d,
C(1)H), 3.30 (1 H, dd, J 9.0 and 3.0, C(3)H exo), 3.16 (1 H, s,
C(4)H), 2.68–2.48 (1 H, m, C(3)H endo), 1.68–1.50 (2 H, m,
CH2), 1.44 (9 H, s, But); δC (125 MHz, mixture of rotamers)
J 8.0, 2 × CH of aromatic), 6.33 (2 H, m, HC᎐CH), 4.56 (1 H, s,
᎐
155.8, 155.6 (C᎐O), 136.5 (C5), 134.4, 133.7 (C6), 78.9 (CMe ),
᎐
3
C(1)H), 3.50 (1 H, s, C(7)H), 3.41 (1 H, dd, J 9.5 and 3.0, C(3)H
exo), 3.25 (1 H, s, C(4)H), 2.57 (1 H, d, J 9.5, C(3)H endo), 2.30
(3 H, s, Me), 1.40 (9 H, s, But); δC (125 MHz, 90 ЊC, DMSO-d6)
61.1, 59.9 (C1), 48.0 (C7), 46.2, 45.8 (C7), 43.4, 42.9 (C4), 28.4
(CMe3).
(both epimers) 155.8, 155.4 (both C᎐O), 138.8, 137.4, 137.2,
᎐
2-(tert-Butoxycarbonyl)-7-syn/anti-(phenylselenyl)-2-aza-
bicyclo[2.2.1]hept-5-ene 24 and 7-(tert-butoxycarbonyl)-2-exo/
endo-(phenylselenyl)-7-azabicyclo[2.2.1]hept-5-ene 25
136.0, 135.2, 132.9, 132.8, 131.9, 131.4, 130.6, 130.4 (all C᎐C or
᎐
C–Ar), 79.7, 79.5 (both CMe3), 65.9 (C7), 64.9, 63.8 (Both C1),
63.6 (C7), 49.2, 47.8 (Both C4), 47.2, 44.2 (Both C3), 29.0
(CMe3), 21.3 (Me–Ar); m/z (CIϩ, NH3) 318 (M ϩ Hϩ, 29%), 279
(33), 262 (100), 218 (48), 201 (23), 189 (23), 157 (14), 140 (13),
124 (14), 94 (78) (Found: M ϩ Hϩ, 318.1531. C18H24NO2S
requires 318.1528).
The experimental procedures and characterisation data for
thiol addition products 20b–i are included in the accompanying
electronic supplementary information.‡
Benzeneselenol (71 mg, 0.45 mmol) was added to a solution of
diene 19a (99 mg, 0.51 mmol) in toluene under argon and the
reaction was stirred at 20 ЊC for 24 h. Removal of solvent at
reduced pressure gave a yellow oil. Column chromatography
(SiO2, 15% Et2O : 85% petroleum ether) first gave selenide 25 as
a colourless oil which crystallised on standing to a white solid
(19 mg, 12 %): Rf (30% Et2O : 70% petroleum ether) 0.40; mp
(from light petroleum) 70–72 ЊC; νmax(thin film)/cmϪ1 3045w,
2792s, 2926s, 1688s, 1578m, 1477m, 1436m, 1368s, 1285m,
1233w, 1169m, 1097m, 846m, 737m; δH (400 MHz, exo : endo,
1 : 6: NOESY analysis suggested that the major isomer had the
selenyl substituent on the endo face: NOE enhancement was
observed between the olefinic protons and the C3 endo proton
trans to the proton attached to C2), (endo-epimer) 7.55–7.51
(2 H, m, 2 × H of Ph), 7.31–7.23 (3 H, m, 3 × H of Ph), 6.40
2-(tert-Butoxycarbonyl)-7-syn/anti-(p-tolylsulfonyl)-2-aza-
bicyclo[2.2.1]hept-5-ene 21
Peracetic acid (1.77 cm3 of 36–40% solution in water, 9.5 mmol)
was added to a mixture of sulfide 20a (500 mg, 1.57 mmol),
NaOAc (290 mg, 3.53 mmol) and Na2 HPO4 (4.5 g, 31 mmol) in
CH2Cl2 (50 cm3) at 0 ЊC. The reaction was stirred at 0 ЊC for 6 h
and then at 20 ЊC for 14 h. Water (50 cm3) was added and the
aqueous layer extracted with CH2Cl2 (3 × 50 cm3). The com-
bined organic extracts were washed with sodium bisulfite solu-
tion (50 cm3), brine (50 cm3), dried (MgSO4) and the solvent
was removed at reduced pressure to give a white solid. Column
(2 H, br s, 2 × HC᎐C), 4.81–4.60 (2 H, m, C(1)H and C(4)H),
᎐
3.64 (1 H, br s, C(2)H), 2.49 (1 H, br s, C(3)H exo), 1.44 (9 H, s,
But), 1.10 (1 H, dd, J 12.0 and 4.0, C(3)H endo), (exo-epimer)
7.55–7.51 (2 H, m, 2 × H of Ph), 7.31–7.23 (3 H, m, 3 × H of
Ph), 6.22 (2 H, br s, 2 × HC᎐C), 4.81–4.60 (2 H, m, C(1)H and
᎐
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3794

View Article Online
C(4)H), 3.48 and 3.40 (1 H, 2 × br s, C(2)H), 1.82 (1 H, br s,
C(3)H exo), 1.44 (9 H, s, But), 1.10 (1 H, dd, J 12.0 and 4.0,
C(3)H endo); δC (100 MHz) (endo-epimer observed only—exo
(50 cm3) was added and the mixture extracted with Et2O (3 ×
100 cm3). The combined organic extracts were washed with sat-
urated NaHCO3 (100 cm3), water (100 cm3) and dried (MgSO4).
The solvent was then removed at reduced pressure to give an
oil. Chromatography (SiO2, 20% Et2O : petroleum ether) gave
the amide 28i as a white solid (2.6 g, 43%): Rf (30% Et2O :
petroleum ether) 0.36; mp (from EtOAc–petroleum ether) 133–
134 ЊC (Found: C, 79.3; H, 7.6; N, 6.2. C15H17NO requires C,
79.3; H, 7.5; N, 6.2%); νmax(KBr)/cmϪ1 3042m, 2985s, 2970m,
1622s, 1474m, 1447m, 1403m, 1366s, 1292m, 1196m, 759s;
δH (400 MHz) 7.26 (2 H, s, 2 × CH of aromatic), 7.02 (2 H, br s,
too weak to appear in 13C spectrum) 154.5 (C᎐O), 135.8, 133.9,
᎐
132.8, 130.1, 129.2, 127.1 (all C᎐C or C of Ph), 80.3 (CMe ),
᎐
3
62.2, 60.5 (C1 and C4), 38.0 (C2), 34.5 (C3), 29.0 (But); m/z
(FABϩ, NOBA Matrix) 374 (M ϩ Naϩ, 30%), 352 (M ϩ Hϩ,
65%), 296 (100), 194 (62), 167 (68), 149 (77), 121 (56), 105 (75)
(Found: M ϩ Hϩ, 352.0817. C17H22NO2Se requires 352.0815).
Second to elute was 24 as a colourless oil which crystallised
on standing to a white solid (128 mg, 81%): Rf (30% Et2O :
petroleum ether) 0.27; mp (from petroleum ether) 58–59 ЊC
(Found: C, 57.9; H, 6.1; N, 4.0. C17H21NO3Se requires C, 58.3;
H, 6.0; N, 4.0%); νmax(thin film)/cmϪ1 3064m, 2975m, 2937m,
1700s, 1578m, 1487s, 1392m, 1293s, 1253s, 1153s, 1096m;
δH (500 MHz, 90 ЊC, DMSO-d6, syn : anti, 3 : 1) (syn-epimer)
7.57–7.49 (2 H, m, 2 × CH of Ph), 7.35–7.28 (3 H, m, 3 × CH of
HC᎐CH), 6.96–6.92 (2 H, m, 2 × CH of aromatic), 5.89 (2 H,
᎐
s, C(1)H and C(4)H), 1.18 (9 H, s, But); δC (100 MHz) 173.7
(C᎐O), 148.2 (2 × C(quat) of aromatic), 144.8 and 142.0
᎐
(2 × C᎐C), 125.0 (2 × C of aromatic), 121.4 and 120.2 (2 × C of
᎐
aromatic), 66.3 and 64.8 (C1 and C4), 38.4 (CMe3), 27.3
(CMe3); m/z (EIϩ) 227 (Mϩ, 5%), 170 (14), 142 (41), 128 (38),
115 (52), 89 (16), 57 (100), 41 (52) (Found: M ϩ Hϩ, 228.1381.
C15H18NO requires 228.1388).
Ph), 6.48–6.40 (2 H, m, HC᎐CH), 4.52 (1 H, s, C(1)H), 3.57
᎐
(1 H, dd, J 9.5 and 3.0, C(3) exo), 3.40 (1 H, s, C(7)H), 3.14 (1
H, s, C(4)H), 2.63 (1 H, d, J 9.5, C(3)H endo), 1.43 (9 H, s, But),
(anti-epimer) 7.57–7.49 (2 H, m, 2 × CH of Ph), 7.35–7.28 (3 H,
N-(tert-Butoxycarbonyl)-exo-2-(p-tolylthio)-1,2,3,4-tetrahydro-
m, 3 × CH of Ph), 6.40–6.33 (2 H, m, HC᎐CH), 4.68 (1 H, s,
1,4-iminonaphthalene 29
᎐
C(1)H), 3.43 (1 H, s, C(7)H), 3.40 (1 H, dd, J 9.5 and 3.0, C(3)H
exo), 3.35 (1 H, s, C(4)H), 2.57 (1 H, d, J 9.0, C(3)H endo), 1.41
(9 H, s, But); δC (125 MHz, 90 ЊC, DMSO-d6) (both isomers)
To a solution of 28a (100 mg, 0.41 mmol) in toluene (5 cm3) was
added p-thiocresol (77 mg, 0.62 mmol). The mixture was stirred
at 25 ЊC for 24 h, after which time the solvent was removed at
reduced pressure. Chromatography (10% Et2O : petroleum
ether) gave the sulfide 29 as a white solid (135 mg, 89%): Rf
(10% Et2O : 90% petroleum ether) 0.25; mp (from petroleum
ether) 88–89 ЊC; νmax(KBr)/cmϪ1 3020s, 2977s, 2931s, 1698s,
1493s, 1460s, 1366s, 1279s, 1256s, 1169s, 1090s, 1017m, 973m,
908s, 812s, 753s; δH (400 MHz) 7.43–7.34 (2 H, m, 2 × CH of
aromatic), 7.30–7.05 (6 H, 6 × CH of aromatic), 5.25 (1 H, br s,
C(1)H), 5.05 (1 H, br s, C(4)H), 3.21 (1 H, s, C(2)H), 2.36 (3 H,
s, Me), 2.02–1.95 (1 H, m, H of CH2), 1.93–1.86 (1 H, m, H of
155.8, 155.3 (C᎐O), 139.0, 137.4, 136.9, 135.5, 134.0, 133.9,
᎐
133.7, 133.6, 130.1, 129.9, 128.1, 127.9 (all C᎐C or C of Ph),
᎐
80.1, 79.6 (CMe3), 65.6, 64.8 (C1), 61.2, 59.3 (C7), 50.2, 48.4
(C4), 47.2, 44.8 (C3), 29.0, 28.9 (CMe3); m/z (EIϩ) 351 (Mϩ,
2%), 184 (10), 157 (18), 138 (69), 120 (10), 94 (83), 84 (27), 77
(33), 65 (28), 57 (100), 49 (24), 41(41) (Found: Mϩ, 351.0736.
C17H21NO3Se requires 351.0737).
N-(tert-Butoxycarbonyl)-1,4-dihydro-1,4-iminonaphthalene
28a40
CH2), 1.44 (9 H, s, But); δC (100 MHz) 154.6 (C᎐O), 145.1
᎐
(C(quat) of aromatic), 143.6 (C(quat) of aromatic), 137.0
(C(quat) of aromatic), 132.2 (C(quat) of aromatic) 131.3, 129.8,
126.9, 126.6, 120.3, 119.8 (all C of aromatic), 80.2 (CMe3), 65.2
(C4), 60.3 (C1), 48.4 (C2), 35.1 (C3), 28.3 (CMe3), 21.1 (Me);
m/z (EIϩ) 367 (Mϩ, 1%), 217 (35), 161 (93), 117 (100), 91 (62),
84 (55), 57 (95) (Found: M ϩ Hϩ, 368.1678. C22H26NO2S
requires 368.1684).
Prepared according to the method of Carpino et al. from
o-bromofluorobenzene and N-Boc-pyrrole in the presence of
magnesium in 58% yield (lit.40 58%): mp 71–72 ЊC (lit.40 72–
73 ЊC); Rf 0.30 (30% Et2O : petroleum ether); δH (200 MHz)
7.30–7.21 (2 H, m, 2 × CH of aromatic), 7.02–6.92 (4 H, m, 2 ×
CH of aromatic and HC᎐CH), 5.50 (2 H, s, C(1)H and C(4)H),
᎐
1.40 (9 H, s, But). Compounds 28c and 28d were prepared
analogously from pyrroles 37 and 38 respectively (vide infra).
Experimental details and characterisation data can be found in
the electronic supplementary information.‡
General procedure for alkene hydroboration: N-(tert-butoxy-
carbonyl)-exo-1,2,3,4-tetrahydro-1,4-iminonaphthalen-2-ol 30a
9-BBN (0.5 M in THF, 2.40 cm3, 1.20 mmol) was added drop-
wise to a stirred solution of 28a (250 mg, 1.02 mmol) in THF
(3 cm3) at 25 ЊC. The reaction was left to stir for 24 h. The flask
was then cooled to 0 ЊC and H2O2 (1.25 cm3 of a 30% aq.
solution) was added, followed by NaOH (1.7 cm3 of a 2 M aq.
solution). The reaction vessel was removed from the ice bath
and the reaction mixture was then stirred at 25 ЊC for 5 h. The
mixture was washed with saturated K2CO3 (10 cm3) and
extracted with Et2O (3 × 20 cm3). The combined extracts were
dried (MgSO4) and the solvent removed at reduced pressure to
give an oil. Column chromatography (50% Et2O : petroleum
ether) gave the alcohol 30a as an oil which crystallised on stand-
ing to a white solid (180 mg, 68%): Rf (50% Et2O : petroleum
ether) 0.23; mp (from Et2O) 90–91 ЊC; νmax(thin film)/cmϪ1
3254br s, 3024m, 2978s, 2928m, 2893s, 1704s, 1690s, 1456s,
1366s, 1348s, 1294m, 1251s, 1158s, 1141m, 1102s, 1088s, 1057s,
903m, 746s; δH (400 MHz) 7.27–7.23 (1 H, m, CH of aromatic),
7.19–7.17 (1 H, m, CH of aromatic), 7.14–7.06 (2 H, m, 2 × H
of aromatic), 5.11 (1 H, br s, C(1)H), 5.01 (1 H, br s, C(4)H),
4.04–4.00 (1 H, m, CH–OH), 3.20 (1 H, br. s, OH), 1.87–
1.84 (2 H, m, CH2), 1.38 (9 H, s, But); δC (100 MHz) 156.5
N-Methoxycarbonyl-1,4-dihydro-1,4-iminonaphthalene 28b54
Prepared according to the method of Cragg et al. from
anthranilic acid and commercially available N-methoxy-
carbonylpyrrole: Rf 0.23 (20% Et2O : petroleum ether); δH (200
MHz) 7.33–7.20 (2 H, m, 2 × CH of aromatic), 7.05–6.90 (4 H,
m, 2 × CH of aromatic and HC᎐CH), 5.58 (2 H, s, C(1)H and
᎐
C(4)H), 3.62 (3 H, s, MeO). Compound 28e was prepared by
adaptation of the procedure of Cragg et al.54 who used 3,6-di-
methoxyanthranilic acid. Compound 28f was prepared by
adaptation of the procedure of Vernon et al.55 who used
anthranilic acid in a benzyne cycloaddition with 1-Boc-2,5-di-
methylpyrrole. Compound 28g was prepared according to the
procedure of Remy and Bornstein and their co-workers.46
Details can be found in the electronic supplementary
information.‡
N-(tert-Butylcarbonyl)-1,4-dihydro-1,4-iminonaphthalene 28i
Isoamyl nitrite (3.10 g, 26.5 mmol) in THF (10 cm3) and
anthranilic acid (3.63 g, 26.5 mmol) in THF (10 cm3) were
added simultaneously to a solution of pyrrole 39 (4.0 g,
26.5 mmol) in THF (40 cm3) at reflux over 2 h. The reaction
was left at reflux for a further 1.5 h, then allowed to cool. Water
(C᎐O), 146.2, 141.5, (both C(quat) of aromatic) 127.0, 126.4,
᎐
121.3, 119.6 (all C of aromatic), 80.4 (CMe3), 72.0 (CH–O),
68.4 (C1), 60.8 (C4), 39.4 (C3), 28.2 (But); m/z (CIϩ, NH3), 279
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3795

View Article Online
(M ϩ NH4ϩ, 5%), 262 (M ϩ Hϩ, 70), 162 (100), 118 (25)
(Found: M ϩ Hϩ, 262.1440. C15H20NO3 requires 262.1443).
Details of the preparation and characterisation of alcohols
30b–g and 30i are to be found in the electronic supplementary
information.‡
white solid (95 mg, 94%, lit.43 86%); mp 40–41 ЊC (lit. 41–42 ЊC);
δH (400 MHz) 7.26–7.18 (2 H, m, 2 × CH of aromatic), 7.14–
7.09 (2 H, m, 2 × CH of aromatic), 5.12 (2 H, s, C(1)H and
C(4)H), 2.11 (2 H, d, J 8.0, 2 × CH exo), 1.41 (9 H, s, But), 1.29
(2 H, d, J 8.0, 2 × CH endo); δC (100 MHz), 155.2 (C᎐O),
᎐
144.8 (2 × C(quat) of aromatic), 126.3 (2 × C of aromatic),
119.4 (2 × C of aromatic), 79.9 (CMe3), 61.0 (C1 and C4), 26.7
(C2 and C3), 28.2 (CMe3).
Experimental details and characterisation data for reduced
compounds 32e–g and 32i are to be found in the electronic
supplementary information.‡
N-(Methyl)-exo-1,2,3,4-tetrahydro-1,4-iminonaphthalen-2-ol
30h
A solution of LiAlH4 (1.5 cm3 of a 1.0 M solution in Et2O,
1.5 mmol) was added to a stirred solution of alcohol 30a
(100 mg, 0.38 mmol) in Et2O (3.5 cm3) at 25 ЊC and the mixture
heated to reflux overnight. The mixture was then allowed to
cool and aqueous NaOH (2.5 cm3 of a 2 M aq. solution) was
added, followed by water (10 cm3) and the aqueous layer was
extracted with Et2O (3 × 20 cm3). The solvent was removed at
reduced pressure. Chromatography (EtOAc containing 2%
Et3N) gave alcohol 30h as a white solid (43 mg, 61%): Rf
(EtOAc containing 2% Et3N) 0.13; mp 123–124 ЊC; νmax(thin
film)/cmϪ1 3352s, 3249s, 2974s, 2946s, 2860m, 1446s, 1415s,
1337m, 1294m, 1167s, 1105s, 1066s, 1003m, 748s; δH (400 MHz)
7.30–7.27 (1 H, m, CH of aromatic), 7.21–7.12 (3 H, m, 3 × CH
of aromatic), 4.06 (1 H, s, C(1)H), 3.98 (1 H, s, C(4)H), 3.86–
3.83 (1 H, m, CH-O), 3.47 (1 H, br s, OH), 2.13 (3 H, s, Me),
1.89–1.84 (2 H, m, CH2); δC (100 MHz), 145.8, 140.6 (both
C(quat)), 127.0, 126.5, 123.5, 122.0 (4 × C of aromatic), 75.4,
72.6 (C1 and C4), 66.7 (C2), 40.6 (C3), 35.2 (Me); m/z (CIϩ,
NH3) 176 (M ϩ Hϩ, 100%), 143 (29) (Found: M ϩ Hϩ,
176.1072. C11H14NO requires 176.1075).
N-Methyl-1,2,3,4-tetrahydro-1,4-iminonaphthalene 32h47
A solution of LiAlH4 (1.0 M in Et2O, 1.0 cm3, 1.0 mmol) was
added to a stirred solution of carbamate 32a (100 mg, 0.41
mmol) in Et2O (4 cm3) at 25 ЊC and the mixture heated to reflux
overnight. The mixture was then allowed to cool and NaOH
(2.5 cm3 of a 2 M aq. solution) was added, followed by water
(10 cm3) and the aqueous layer was extracted with Et2O (3 ×
20 cm3). The solvent was removed at reduced pressure. Column
chromatography (EtOAc containing 2% Et3N) gave amine 32h
as a white solid (35 mg, 54%): Rf (EtOAc containing 2% Et3N)
0.20; δH (400 MHz) 7.22–7.17 (2 H, m, 2 × CH of aromatic),
7.13–7.07 (2 H, m, 2 × CH of aromatic), 4.06 (2 H, s, C(1)H and
C(4)H), 2.11 (2 H, d, J 8.0, 2 × CH exo), 2.03 (3 H, s, Me),
1.20 (2 H, dd, J 8.0 and 2.0, 2 × CH endo); δC (100 MHz) 144.3
(2 × C (quat) of aromatic), 126.2 (2 × C of aromatic), 121.7
(2 × C of aromatic), 67.3 (C1 and C4), 35.4 (Me), 27.1 (C2
and C3).
General procedure for xanthate preparation: N-(tert-butoxycarb-
onyl)-exo-2-(methylsulfanylthiocarbonyloxy)-1,2,3,4-tetrahydro-
1,4-iminonaphthalene 31a
General procedure for neophyl-type rearrangement: N-(tert-
butoxycarbonyl)-1,2,3,4-tetrahydro-1,4-methanoisoquinoline 33a
Alcohol 30a (200 mg, 0.76 mmol) was added to KH (400 mg of
a 30% suspension, 3 mmol) in THF (10 cm3) at 0 ЊC and the
mixture was then stirred at room temperature for 20 min. The
mixture was cooled to 0 ЊC and CS2 (228 mg, 3 mmol) was
added and the mixture stirred for 10 min at 0 ЊC. Finally MeI
(425 mg, 3 mmol) was added and the solution was stirred at
room temperature for 20 min. The reaction was quenched with
dropwise addition of water until effervescence had ceased and
the aqueous layer extracted with Et2O (2 × 10 cm3), the com-
bined extracts dried (MgSO4), and the solvent removed at
reduced pressure. Column chromatography (SiO2, 20% Et2O :
petroleum ether) gave 31a as a yellow solid (220 mg, 83%): Rf
(50% Et2O : petroleum ether) 0.59; νmax(KBr)/cmϪ1 3019m,
2975s, 2924s, 1696s, 1372s, 1294s, 1269s, 1222s, 1160s, 1092s,
1060s, 1002m, 771s; δH (400 MHz) 7.37 (1 H, s, CH of
aromatic), 7.30–7.15 (3 H, m, 3 × CH of aromatic), 5.42 (2 H, s,
C(1)H and C(4)H), 5.21 (1 H, s, CH-O), 2.59 (3 H, s, SMe),
2.23–2.15 (1 H, m, H of CH2), 2.09–1.99 (1 H, m, H of CH2),
1.43 (9 H, s, But); m/z (CIϩ, NH3) 369 (M ϩ NH4ϩ, 2%), 352
(M ϩ Hϩ, 15), 262 (43), 252 (50), 218 (16), 207 (38), 189 (10),
162 (69), 146 (100), 118 (62), 90 (19) (Found: M ϩ Hϩ,
352.1041. C17H21NO3S2 requires 352.1041).
A mixture of AIBN (60 mg, 0.37 mmol) and TTMSS (266 mg,
1.07 mmol), in toluene (2 cm3), was added over 100 min to a
preheated solution of xanthate 31a (250 mg, 0.71 mmol) in
toluene (21 cm3). The reaction was left at reflux for a further
30 min after completion of the addition, allowed to cool, and
the solvent removed at reduced pressure. Column chromato-
graphy (SiO2, 10% Et2O : petroleum ether) gave 33a as a colour-
less oil that crystallised on standing to a white solid (158 mg,
90%); Rf (50 Et2O : petroleum ether) 0.54; mp (from petroleum
ether) 65–66 ЊC; νmax(thin film)/cmϪ1 2977s, 2889m, 1698s,
1462m, 1392s, 1251m, 1181m, 1152s, 1093s, 838m, 755m;
δH (500 MHz, DMSO-d6, 90 ЊC) 7.34 (1 H, d, J 7.0, CH of
aromatic), 7.24 (1 H, d, J 7.0, CH of aromatic), 7.21–7.11 (2 H,
m, 2 × CH of aromatic), 4.94 (1 H, br s, C(1)H), 3.66 (1 H, s,
C(4)H), 3.48 (1 H, dd, J 9.0 and 3.0, C(3)H exo), 2.64 (1 H, dd,
J 9.0 and 2.0, C(3)H endo), 1.96 (1 H, d, J 9.0, C(9)H), 1.79
(1 H, d, J 9.0, C(9)H), 1.37 (9 H, s, But); δC (125 MHz, DMSO-
d , 90 ЊC) 155.4 (C᎐O), 146.4, 145.1 (2 × C(quat) of aromatic),
᎐
6
127.8, 126.7, 122.2, 121.0 (4 × CH of aromatic), 79.4 (CMe3),
62.2 (C1), 49.5 (C3), 49.1 (C9), 44.5 (C4), 28.2 (But); m/z (CIϩ,
NH3) 263 (M ϩ NH4ϩ, 21%), 246 (M ϩ Hϩ, 25), 207 (100), 146
(M ϩ H Ϫ Bocϩ, 96) (Found: M ϩ Hϩ, 246.1494. C15H20NO2
requires 246.1494).
Experimental procedures and characterisation data for
rearranged products 33b–g are to be found in the electronic
supplementary information.‡
The other xanthates were not fully characterised, but were
identified by the appearance of a 3 H singlet at ∼2.6 ppm and a
shift of the CH–O proton from ∼4.0 ppm to ∼5.3 ppm. Prepar-
ation of the other radical precursors is included with the
subsequent reduction–rearrangements.
N-Methyl-1,2,3,4-tetrahydro-1,4-methanoisoquinoline 33h
General procedure for hydrogenation of bicyclic alkenes: N-(tert-
a) From radical deoxygenation. To a solution of alcohol 30h
(122 mg, 0.70 mmol) in CH2Cl2 (5 cm3) was added DMAP
(342 mg, 2.8 mmol) and phenylchlorothionocarbonate (144 mg,
0.83 mmol). The reaction was stirred for 16 h and the solvent
was then removed at reduced pressure. Column chromato-
graphy (SiO2, 20% Et2O : petroleum ether containing 2% Et3N)
gave thiocarboxylate 31h as a yellow oil (140 mg, 65%): Rf
(EtOAc containing 2% Et3N) 0.55. The product was not fully
characterised but instead was identified by a change in chemical
butoxycarbonyl)-1,2,3,4-tetrahydro-1,4-iminonaphthalene 32a43
Alkene 28a (100 mg, 0.41 mmol) was dissolved in MeOH
(3 cm3), and 10% Pd/C (42 mg, 0.04 mmol Pd) was added. The
mixture was placed under 1 atm of hydrogen for 10 min at 25 ЊC
using a balloon. The flask was then opened to the atmosphere
and the mixture filtered through a Celite pad. The solvent was
removed at reduced pressure to give an oil. Chromatography
(SiO2, 15% Et2O : petroleum ether) gave compound 32a as a
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3796

View Article Online
shift of the CH–O proton from 3.8 in the starting alcohol to
5.1 ppm in the product and an increase in the aromatic proton
integral from 4 H to 9 H. AIBN (18 mg, 0.10 mmol) and
TTMSS (166 mg, 0.67 mmol) were added to a solution of thio-
carboxylate 31h (140 mg, 0.44 mmol) in toluene (17.5 cm3) and
the flask was then lowered into an oil bath preheated to 125 ЊC.
The mixture was stirred at reflux for 2 h and then allowed to
cool. The solvent was removed at reduced pressure. Column
chromatography (SiO2, EtOAc containing 2% Et3N) gave amine
33h as a yellow oil (16 mg, 23%): Rf (EtOAc containing 2%
Et3N) 0.17; νmax(thin film)/cmϪ1 3045m, 2974s, 2855s, 2780s,
1459m, 1275m, 1194m, 1174m, 1151m, 1128m, 1091m, 1032m,
1013m, 954w, 909w, 754m; δH (400 MHz) 7.29–7.23 (1 H, m,
CH of aromatic), 7.21–7.13 (3 H, m, 3 × CH of aromatic), 4.08
(1 H, s, C(1)H), 3.45 (1 H, dd, J 8.0 and 2.0, C(3)H exo), 3.41
(1 H, s, C(4)H), 2.04 (1 H, dd, J 9.5 and 3.0, C(9)H), 1.95 (3 H,
s, Me), 1.82 (1 H, dd, J 9.5 and 3.0, C(9)H), 1.44 (1 H, dd, J 8.0
and 2.0, C(3)H endo); δC (100 MHz), 146.5, 140.8 (both
C(quat)), 126.7, 125.2, 122.7, 120.5 (4 × CH of aromatic), 67.3
(C1), 56.0 (C3), 49.1 (C9), 45.0 (C4), 40.8 (Me); m/z (TOF ESϩ)
160 (M ϩ Hϩ, 100%) (Found: M ϩ Hϩ, 160.1129. C11 H14N
requires 160.1126).
was brought to reflux and left for 14 h. The mixture was allowed
to cool, extracted with Et2O (5 × 100 cm3), the organic extracts
washed with water (2 × 100 cm3), dried (K2CO3) and the solvent
removed at reduced pressure. Chromatography (Et2O, 1%
MeOH, 2% Et3N) gave the known amine 34 (50 mg, 47%): Rf
(Et2O, 2% Et3N) 0.12; νmax(thin film)/cmϪ1 3327br m, 3239br m,
3044m, 2977s, 2871m, 1698s, 1460s, 1374w, 1256w, 1167w,
1049w, 966w, 754s; δH (400 MHz) 7.30–7.05 (4 H, m, 4 × CH of
aromatic), 4.38 (1 H, s, C(1)H), 3.56 (1 H, s, C(4)H), 3.18 (1 H,
dd, J 9.0 and 3.0, C(3)H exo), 2.27 (1 H, dd, J 9.0 and 1.0,
C(3)H endo), 1.95–1.65 (1 H, br s, NH), 1.86 (1 H, d, J 7.0,
C(9)H), 1.76 (1 H, d, J 7.0, C(9)H); δC (100 MHz) 146.5, 145.2
(both C(quat) of aromatic), 144.0 and 143.9 (2 × C(quat) of
aromatic), 126.4, 125.9, 121.0, 119.1 (4 × CH of aromatic), 60.8
(C1), 48.3 (C9), 46.6 (C3), 44.4(C1).
3-Ethylpyrrole 3644
A solution of 3-ethyl-1-phenylsulfonylpyrrole 3545 (1.60 g,
6.8 mmol) in methanol (26 cm3) was added to NaOH (13 cm3 of
a 5 M aq. solution) and the mixture heated to reflux for 2.5 h
under argon. The mixture was allowed to cool and extracted
with EtOAc (3 × 50 cm3). The combined organic extracts were
washed with brine (100 cm3), dried (MgSO4) and the solvent
removed at reduced pressure to give spectroscopically pure 36
(520 mg, 80%): δH (200 MHz) 8.20–7.80 (1 H, br s, NH), 6.79–
6.77 (1 H, m, C(5)H), 6.65 (1 H, s, C(2)H), 6.20 (1 H, s, C(4)H),
2.61 (2 H, q, J 8.0, CH2), 1.31 (3 H, t, J 8.0, Me). The material
was Boc-protected56 immediately.
b) From hydride reduction. LiAlH4 (1.0 M in THF, 0.92 cm3,
0.92 mmol) was added to a solution of amine 33a (150 mg,
0.61 mmol) in THF (4 cm3) and the mixture heated at reflux for
2 h. The mixture was then allowed to cool and aqueous NaOH
(1.5 cm3of a 2 M aq. solution) was added, followed by water
(6 cm3). The aqueous layer was extracted with Et2O (3 × 5 cm3),
the combined organic extracts dried (MgSO4), and the solvent
removed at reduced pressure. Column chromatography (SiO2,
EtOAc containing 2% Et3N) gave amine 33h as a yellow oil
(44 mg, 45%): Rf (EtOAc containing 2% Et3N) 0.17. Other data
as above.
1-(tert-Butoxycarbonyl)-3-ethylpyrrole 37
DMAP (256 mg, 2.1 mmol) was added to a stirred solution of
Boc2O (5.45 g, 25 mmol) and 3-ethylpyrrole 36 (1.98 g, 21
mmol) in acetonitrile (20 cm3) and the reaction left to stir for 24
h under argon at 25 ЊC. Imidazole (3.0 g, 44 mmol) was added
and the solution stirred for 15 min according to the procedure
of Basel and Hassner57 for removal of Boc2O. CH2Cl2 (25 cm3)
was added and the solution washed with 0. % aq. HCl (3 × 50
cm3). The organic layer was separated, dried (MgSO4) and the
solvent was removed at reduced pressure. Column chromato-
graphy (SiO2, 3% Et2O : petroleum ether) gave the pyrrole 37 as
a yellow oil (2.58 g, 54%): Rf (10% Et2O : petroleum ether) 0.67.
An analytical sample was prepared by Kugelrohr distillation
(bp 80 ЊC at 10 mmHg). νmax(thin film)/cmϪ1 3148m, 3101m,
2969s, 2934s, 2874s, 1741s 1561m, 1489s, 1461s, 1404s, 1371s,
1313s, 1243s, 1162s, 1122s, 1071s, 1050m, 971s, 933m, 854s,
829m, 772s; δH (400 MHz) 7.17 (1 H, s, C(5)H), 7.00 (1 H, s,
C(2)H), 6.18 (1 H, s, C(4)H), 2.47 (2 H, q, J 7.5, CH2), 1.60
(9 H, s, But), 1.20 (3 H, t, J 7.5, Me); δ (100 MHz) 149.0 (C᎐O),
N-(tert-Butylcarbonyl)-1,2,3,4-tetrahydro-1,4-iminonaphthalene
32i and N-(tert-butylcarbonyl)-1,2,3,4-tetrahydro-1,4-methano-
isoquinoline 33i
Reaction of alcohol 30i (150 mg, 0.61 mmol) according to the
general procedure gave an oil. Chromatography (SiO2, 10%
Et2O : petroleum ether to 30% Et2O : petroleum ether, gradient
elution) gave xanthate 31i (176 mg, 86%) as an oil: Rf (15% Et2O
: petroleum ether) 0.14. AIBN (21 mg, 0.13 mmol) and TTMSS
(197 mg, 0.79 mmol) were added to a solution of xanthate 31i
(176 mg, 0.52 mmol) in toluene (21 cm3). The mixture was
brought to reflux and left for 2 h. The mixture was then allowed
to cool, and the solvent removed at reduced pressure. Column
chromatography (SiO2, 30% Et2O : petroleum ether) first gave
32i as a colourless oil that crystallised on standing to a white
solid (17 mg, 14%); data as in the electronic supplementary
information.‡
Second to elute was 33i as a white solid (73 mg, 61%): Rf
(30% Et2O : petroleum ether) 0.19; mp (from petroleum ether)
102–103 ЊC; νmax(KBr)/cmϪ1 2974s, 2882m, 1602s, 1475m,
1416s, 1386m, 1364m, 1309m, 1262m, 1212m, 1113m, 1003s,
764s; δH (500 MHz, toluene-d8, 90 ЊC) 7.15 (1 H, d, J 6.5, CH of
aromatic), 7.02–6.85 (3 H, m, 3 × CH of aromatic), 5.33 (1 H,
br s, C(1)H), 3.45 (1 H, dd, J 9.5 and 4.0, C(3)H exo), 3.14 (1 H,
s, C(4)H), 2.77 (1 H, d, J 9.5, C(3)H endo), 1.57–1.49 (2 H, m,
2 × C(9)H), 1.07 (9 H, s, But); δC (125 MHz, toluene-d8, 90 ЊC)
᎐
C
129.6, 120.0, 115.9, 112.6 (all pyrrole C), 83.1 (CMe3), 28.0
(CMe3), 20.0 (CH2), 14.5 (Me); m/z (CIϩ, NH3) 196 (M ϩ Hϩ,
100%), 152 (13), 149 (19), 96 (M Ϫ Bocϩ, 100) (Found: M ϩ
Hϩ, 196.1329. C11H18NO2 requires 196.1338).
1-(tert-Butoxycarbonyl)-2,4-dimethylpyrrole 38
DMAP (180 mg, 1.5 mmol) was added to a stirred solution of
Boc2O (650 mg, 3.00 mmol) and 2,4-dimethylpyrrole (500 mg,
5.25 mmol) in MeCN (5 cm3) and the reaction stirred under
argon at 25 ЊC for 24 h. The solvent was then removed at
reduced pressure. Column chromatography (SiO2, 3% Et2O :
petroleum ether) gave the pyrrole 38 as a colourless oil (325 mg,
55%): Rf (10% Et2O : petroleum ether) 0.70; νmax(thin film)/cmϪ1
3153m, 3123m, 3089m, 2978s, 2929s, 2750m,1736s, 1607m,
1535s, 1479s, 1456s, 1432s, 1391s, 1340s, 1257s, 1169s, 1095s,
1039m, 1005m, 989m, 968m, 859s, 848s, 802s; δH (400 MHz)
6.93 (1 H, s, C(5)H), 5.79 (1 H, s, C(3)H), 2.40 (3 H, s, Me–
C(2)), 2.01 (3 H, s, Me–C(4)), 1.59 (9 H, s, But); δC (100 MHz,
175.4 (C᎐O), 146.0, 144.8 (2 × C(quat) of aromatic), 127.1,
᎐
126.4, 121.2, 121.0 (4 × CH of aromatic), 62.4 (C1), 50.5 (C3),
47.8 (C9), 44.7 (C4), 38.9 (CMe3), 27.8 (CMe3); m/z (EIϩ) 229
(Mϩ, 21%), 144 (11), 129 (82), 116 (63), 84 (15), 57 (100)
(Found: M ϩ Hϩ, 230.1547. C15H20NO requires 230.1545).
1,2,3,4-Tetrahydro-1,4-methanoisoquinoline 3437
To a solution of carbamate 33b (150 mg, 0.74 mmol) in ethylene
CDCl ), 149.6 (C᎐O), 131.6, 120.4, 117.5, 114.2 (all pyrrole C),
82.7 (CMe3), 28.0 (CMe3), 15.4 (H3C–C(2)), 11.7 (H3C–C(4));
᎐
3
glycol (8 cm3) was added 10% aq. KOH (8 cm3). The mixture
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3797

View Article Online
m/z (CIϩ, NH3) 196 (M ϩ Hϩ, 52%), 149 (17), 96 (M Ϫ Bocϩ,
52) (Found: M ϩ Hϩ, 196.1328. C11H18NO2 requires 196.1338).
19 P. C. Wong and D. Griller, J. Org. Chem., 1981, 46, 2327–2329.
20 S. J. Cristol and R. P. Arganbright, J. Am. Chem. Soc., 1957, 79,
6039–6041.
21 Kuivila argues persuasively for a 2.7 : 1 equilibrium in favour of
nortricyclyl radical 3 over norbornenyl radical 2 based on trapping
experiments with dicyclohexylphosphine, see: M. S. Alnajjar and
H. G. Kuivila, J. Org. Chem., 1981, 46, 1053–1057.
22 T. V. V. Auken and E. A. Rick, Tetrahedron Lett., 1968, 9, 2709–
2712.
23 D. I. Davies, L. T. Parfitt, C. K. Alden and J. A. Claisse, J. Chem.
Soc. (C), 1969, 1585–1590.
24 H. J. Altenbach, B. Blech, J. A. Marco and E. Vogel, Angew. Chem.,
Int. Ed. Engl., 1982, 21, 778.
25 M. Newcomb, Tetrahedron, 1993, 49, 1151–1176.
26 R. Leung-Toung, Y. Liu, J. M. Muchowski and Y.-L. Wu, J. Org.
Chem., 1998, 63, 3235–3250.
27 A. Kasyan, C. Wagner and M. E. Maier, Tetrahedron, 1998, 54,
8047–8055.
28 Addition of the yields of syn-24 and exo-25, and of anti-24 and
endo-25 (as determined by NMR spectroscopy) gives a true ratio of
62 : 31, ∼2 : 1, for exo : endo attack by the phenylselenyl radical.
29 D. J. Pasto, R. Krasnansky and D. Zercher, J. Org. Chem., 1987, 52,
3062–3072.
1-(tert-Butylcarbonyl)pyrrole 3948
Et3N (5.66 g, 56 mmol) was added dropwise to a solution of
trimethylacetyl chloride (6.75 g, 56 mmol) and pyrrole (5.0 g,
76 mmol) in dry CH2Cl2 (30 cm3) at 25 ЊC. DMAP (0.68 g, 5.6
mmol) was then added as one portion and the mixture
stirred for 24 h at 25 ЊC. The solution was then diluted with
Et2O (30 cm3), washed with saturated KHSO4 (20 cm3), satur-
ated NaHCO3 (20 cm3) and water (40 cm3). The organic layer
was dried (MgSO4) and the solvent removed at reduced pres-
sure to give an oil. Column chromatography (SiO2, 3% Et2O :
petroleum ether) gave 39 as a colourless oil (4.9 g, 58%): Rf
(10% Et2O : petroleum ether) 0.80; νmax(thin film)/cmϪ1 2978s,
2935m, 2908w, 2877w, 1705s, 1460s, 1406m, 1371m, 1296s,
1223s, 1101s, 1078m, 1052m, 903s, 714s; δH (200 MHz), 7.48–
7.40 (2 H, m, C(2)H and C(5)H), 6.28–6.22 (2 H, m, C(3)H and
C(4)H), 1.45 (9 H, s, But); δ (100 MHz) 175.8 (C᎐O), 120.5 (C2
᎐
C
and C5), 111.9 (C3 and C4), 40.6 (CMe3), 28.5 (CMe3); m/z
(TOF CIϩ, NH3) 152 (M ϩ Hϩ, 100%), 131 (63), 87 (24), 70 (29)
(Found: M ϩ Hϩ, 152.078. C9H13NO requires 152.075).
30 D. D. M. Wayner, K. B. Clark, A. Rauk, D. Yu and D. A.
Armstrong, J. Am. Chem. Soc., 1997, 119, 8925–8932.
31 A. Studer and M. Bossart, Tetrahedron, 2002, 57, 9649–9667.
32 S. J. Cristol and G. W. Nachtigall, J. Org. Chem., 1967, 32, 3727–
3737.
33 S. J. Cristol and J. M. Sullivan, J. Am. Chem. Soc., 1971, 93, 1967–
Acknowledgements
1970.
34 H. R. Sonawane, B. S. Nanjundiah and R. G. Kelkar, Tetrahedron,
1986, 42, 6673–6682.
35 J. W. Wilt, in Free Radicals, ed. J. K. Kochi, Wiley, New York, 1973,
Vol. 1, pp. 333–501.
36 R. Pedrosa, C. Andres, J. M. Iglesias and A. Perez-Encabo, J. Am.
Chem. Soc., 2001, 123, 1817–1821.
37 G. L. Grunewald, D. J. Sall and J. A. Monn, J. Med. Chem., 1988,
31, 433–444.
We thank AstraZeneca Pharmaceuticals and the EPSRC for an
Industrial CASE award (to M. W. P. B.) and the EPSRC
National Mass Spectrometry Service Centre for mass spectra.
We also thank Professors S. J. Cristol and B. Giese for useful
discussions and Dr B. Odell for expert assistance with structure
determination using NMR.
38 H. Pellissier and M. Santelli, Tetrahedron, 2003, 59, 701–730.
39 Z. Chen and M. L. Trudell, Chem. Rev., 1996, 96, 1179–1193.
40 L. A. Carpino, R. E. Padykula, D. E. Barr, F. H. Hall, J. G. Krause,
R. F. Dufresne and C. J. Thoman, J. Org. Chem., 1988, 53, 2565–
2572.
References
1 D. Blondet and C. Morin, Heterocycles, 1982, 19, 2155–2181.
2 A. L. J. Beckwith and K. U. Ingold, in Rearrangements in Ground
and Excited States, ed. P. d. Mayo, AcademicPress, New York, 1980,
Vol. 1, pp. 161–310.
41 H. C. Brown and J. V. N. V. Prasad, Heterocycles, 1987, 25, 641–
657.
3 B. Giese, B. Koepping, T. Goebel, J. Dickhaut, G. Thoma, K. J.
Kulicke and F. Trach, Org. React., 1996, 48, 301–856.
4 Radicals in Organic Synthesis, eds. P. Renaud and M. P. Sibi, Wiley-
VCH, Weinheim, 2001.
5 D. C. Nonhebel, Chem. Soc. Rev., 1993, 22, 347–359.
6 M. Toyota, T. Wada and M. Ihara, J. Org. Chem., 2000, 65, 4565–
4570.
7 J. H. Rigby and F. C. Pigge, Tetrahedron Lett., 1996, 37, 2201–2204.
8 A. Srikrishna, R. Viswajanani, T. J. Reddy, D. Vijaykumar and
P. P. Kumar, J. Org. Chem., 1997, 62, 5232–5234.
9 I. E. Marko, S. L. Warriner and B. Augustyns, Org. Lett., 2000, 2,
3123–3125.
10 H. G. Kuivila, R. J. Strunk and C. R. Warner, J. Org. Chem., 1966,
31, 3381–3384.
11 T. A. Halgren, J. L. Firkins, T. A. Fujimoto, H. H. Suzukawa and
J. D. Roberts, Proc. Natl. Acad. Sci. U. S. A., 1971, 68, 3216–3218.
12 D. M. Hodgson, C. R. Maxwell and I. R. Matthews, Synlett, 1998,
1349–1350.
13 D. M. Hodgson, C. R. Maxwell, R. Wisedale, I. R. Matthews,
K. J. Carpenter, A. H. Dickenson and S. J. Wonnacott, J. Chem.
Soc., Perkin Trans. 1, 2001, 3150–3158.
14 D. M. Hodgson, M. W. P. Bebbington and P. Willis, Chem.
Commun., 2001, 889–890.
15 D. M. Hodgson, M. W. P. Bebbington and P. Willis, Org. Lett., 2002,
4, 4353–4356.
42 P. S. Anderson, Tetrahedron Lett., 1976, 1141–1144.
43 T. Ohwada, M. Miura, H. Tanaka, S. Sakamoto, K. Yamaguchi,
H. Ikeda and S. Inagaki, J. Am. Chem. Soc., 2001, 123, 10164–10172.
44 D. O. A. Garrido, B. Frydman, G. Buldain and M. I. Ojea, J. Org.
Chem., 1988, 53, 403–407.
45 D. M. Ketcha, K. P. Carpenter, S. T. Atkinson and H. R.
Rajagopolan, Synth. Commun., 1990, 20, 1647–1655.
46 D. E. Remy, F. H. Bisset and J. Bornstein, J. Org. Chem., 1978, 43,
4469–4472.
47 M. J. O. Arteunis, F. A. M. Borremans, J. Gelan, A. P. Marchand
and R. W. Allen, J. Am. Chem. Soc., 1978, 100, 4050–4054.
48 K. Nishide, S. Ichihashi, H. Kimura, T. Katoh and M. Node,
Tetrahedron Lett., 2001, 42, 9237–9240.
49 J. Lalevee, X. Allonas and J.-P. Fouassier, J. Am. Chem. Soc., 2002,
124, 9613–9621.
50 A. Effio, D. Griller, K. U. Ingold, J. C. Scaiano and S. J. Sheng,
J. Am. Chem. Soc., 1980, 102, 6063–6068.
51 C. Chatgilialoglou, J. Dickhaut and B. Giese, J. Org. Chem., 1991,
56, 6399–6403.
52 C. R. Maxwell, D. Phil. Thesis, University of Oxford, 1999.
53 E. J. Corey, M. G. Bock, A. P. Kozikowski, A. V. R. Rao, D. Floyd
and B. Lipshutz, Tetrahedron Lett., 1978, 19, 1051–1054.
54 G. M. L. Cragg, R. G. F. Giles and G. H. P. Roos, J. Chem. Soc.,
Perkin Trans. 1, 1975, 1339–1342.
55 J. M. Vernon, M. Ahmed and J. M. Moran, J. Chem. Soc.,
Perkin Trans. 1, 1977, 1084–1087.
16 D. I. Davies, J. Chem. Soc., Spec. Publ., 1970, 24, 201–237.
17 C. Maignan and R. A. Raphael, Tetrahedron, 1983, 39, 3245–3249.
18 S. J. Cristol, G. D. Brindell and J. A. Reeder, J. Am. Chem. Soc.,
1958, 80, 635–640.
56 L. Grehn and U. Ragnarsson, Angew. Chem., Int. Ed. Engl., 1984,
23, 296–301.
57 Y. Basel and A. Hassner, Synthesis, 2001, 550–552.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 7 8 7 – 3 7 9 8
3798