Photochemical [2 + 2] cycloaddition reactions of 6-alkenyl-3- phenylcyclohex-2-en-1-ones: Using biradical conformation control to account for exceptions to the "rule of five"

Article abstract of DOI:10.1039/c0ob01131b

A series of 6-alkenyl-3-phenylcyclohex-2-enones has been synthesised and the structures of the products obtained from them on irradiation have been determined. The 6-propenyl compounds afforded a tricyclic 'parallel' [2 + 2] cycloaddition product and a bi

Full text of DOI:10.1039/c0ob01131b

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 2959
www.rsc.org/obc
PAPER
Photochemical [2 + 2] cycloaddition reactions of
6-alkenyl-3-phenylcyclohex-2-en-1-ones: using biradical conformation control
to account for exceptions to the “rule of five”†‡
Stephen A. Bradley,a Brian J. Bresnan,b Sylvia M. Draper,c Niall W. A. Geraghty,*b Mark Jeffares,a
Thomas McCabe,c T. Brian H. McMurry*c and John E. O’Brienb
Received 7th December 2010, Accepted 3rd February 2011
DOI: 10.1039/c0ob01131b
A series of 6-alkenyl-3-phenylcyclohex-2-enones has been synthesised and the structures of the products
obtained from them on irradiation have been determined. The 6-propenyl compounds afforded a
tricyclic ‘parallel’ [2 + 2] cycloaddition product and a bicyclic enone resulting from hydrogen abstraction
in the biradical intermediate. The 6-butenyl and 6-pentenyl analogues gave ‘crossed’ cycloaddition
products only. Although the regiochemistry of these cycloaddition reactions cannot be explained in
terms of the ‘rule of five’, it is compatible with the concept of ‘biradical conformation control’ which is
based on a consideration of the energy and structure of the possible 1,4-biradical intermediates.
Introduction
In previous work we have examined the photochemistry of 5-
alkenyl-3-phenylcyclopentenones and have shown that 1 gave
the so-called parallel2 cycloaddition products 2. This mode of
addition was not changed by the presence of electron donating or
withdrawing substituents in the p-position of the 3-phenyl group,3
or by the presence of a phenyl group in the 2- or 3- positions of
the propenyl side chain.4,5 The last can also be substituted in the
p-position by electron donating or withdrawing groups with no
effect on the mode of cycloaddition.6
The related 5-(3¢-butenyl)-3-phenylcyclopentenone 3 undergoes
a similar photochemical reaction, but the major product 4 is the
result of cross cycloaddition. The minor product 5 is formed
through parallel cycloaddition. The homologue, 5-(4¢-pentenyl)-
3-phenylcyclopentenone 6, gives a product 7, which is formed by
parallel cycloaddition7 These results were rationalised using the
“rule of five” initially proposed by Srinivasan8 and Hammond,9
Not all alkenyl substituted cycloalkenones are as well behaved
and later explained by Gleiter10 in terms of “through space” and
in terms of the rule of five. The behaviour of 6-alkenylcyclohex-
“through bond” interactions.
2-enones is of particular interest in this respect as they form
the non-rule of five adduct, either exclusively or as the major
product, on irradiation (Scheme 1). This behaviour is general for
aSchool of Chemical and Pharmaceutical Sciences, Dublin Institute of
both simple 6-alkenylcyclohex-2-enones11 and fused ring systems
Technology, Kevin Street, Dublin 8, Ireland; Fax: 00 -353-1-4024989;
such as 1-alkenyl-2(1H)-naphthalenones.12 This paper describes
Tel: 00-353-1-4024572
bSchool of Chemistry, National University of Ireland, Galway, Ireland.
E-mail: niall.geraghty@nuigalway.ie; Fax: 00-353-91-525700; Tel: 00-353-
91-492474
an extension of our previous work to 6-alkenyl-3-phenylcyclohex-
2-enones; this work was carried out with a view to establishing
whether or not the regiochemical outcome of the [2 + 2]
cycloaddion reactions of these molecules would be consistent
with the general behaviour of 6-alkenylcyclohex-2-enones, and to
providing an explanation for the failure of such compounds to
comply with the rule of five.
cSchool of Chemistry, Trinity College, Dublin, Ireland. E-mail: tmcmurry@
tcd.ie; Fax: 00-353-1-6712826; Tel: 00-353-1-8961992
† Photochemistry of cyclic enones: Part 14. For Part 13, see ref. 1.
‡ CCDC reference numbers 735146–735148. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0ob01131b
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2959–2968 | 2959
©

View Article Online
Scheme 1
Synthesis
Our original intention was to synthesise compounds of the type
8, n = 1,2,3 (Scheme 2). We believed that the presence of the
ester group would drive the equilibrium between the two possible
conformations, 8¢ and 8¢¢, towards 8¢, where the alkenyl side
chain is “axial” and thus more favourably orientated for a [2 +
2] cycloaddition reaction. However the structure of 8, n = 1, as
determined by X-ray analysis (Fig. 1) shows that the molecule in
the solid state adopts the conformation 8¢¢, in which it is the ester
group which is “axial”. Although in hindsight, this conformation
may minimise repulsions between the carbonyl groups of the enone
and ester, the finding does not affect our reasoning about the
ratio of the two conformations in solution. While we were able to
make the diketo-esters 9, R = Me, Et from the appropriate alkyl 3-
oxobutanoate and phenyl vinyl ketone, ring closure using a variety
of bases and acids as catalysts resulted in concomitant loss of the
ester group giving 3-phenylcyclohexenone 10 (Scheme 2).
Scheme 2
Fig. 1 3-Phenyl-6-(2¢-propenyl)cyclohex-2-en-1-one, 8, n = 1 (CCDC No.
735148).
In an alternative approach, we first attached the alkenyl groups
to methyl 3-oxobutanoate to give 11, n = 1,2,3, and then condensed
these with phenyl vinyl ketone to afford 12, n = 1,2,3. In
a preliminary experiment with sodium methoxide, 12, n = 1,
was shown to undergo ring closure but again with loss of the
ester group to afford 13. However, using DBU as base, the
compounds 12, n = 1,2,3, all give the corresponding 6-alkenyl-6-
methoxycarbonyl-3-phenyl-2-cyclohexenone 8, n = 1,2,3. All the
compounds were characterised by their 1H- and 13C-NMR spectra,
using DEPT, nOe, and 2-D techniques.
Scheme 3
finding that in all cases the isolated product(s) was the only
low MW product(s) formed, and that no starting material
remained unreacted. The presence of a styrene chromophore in
these 3-phenylcyclopentenone systems results in polymerization
being competitive with cycloaddition as a reaction channel, and
accounts for the low yields observed. Polymer formation has
also been observed in earlier work.2 The 1H-NMR spectrum
Results
In the event, the presence of the ester group at the 6-position was
not required, as when we irradiated the enone 13 in acetonitrile,
a single product was obtained in 11% yield (Scheme 3). The
reaction was followed by GCMS, as were the other photo-
chemical reactions, and this confirmed the spectroscopic (NMR)
2960 | Org. Biomol. Chem., 2011, 9, 2959–2968
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
showed that the allyl and enone olefinic hydrogens were no longer
present, but that there was a triplet at d 6.31. The product must
therefore contain a new olefinic hydrogen coupled to a methylene
group. The presence of a double bond was confirmed by the
13C-NMR spectrum which contained signals at d 125.7(CH) and
125.6(q). This suggests that after initial bond formation between
the terminal carbon of the allylic group and the 2-position of
the enone, hydrogen abstraction from the original 4-position of
the enone, forming 14 with its styrene chromophore, occurs in
1
preferenc to ring closure. H- and 13C-NMR data support this
structure.
Photoreaction of the methyl ester 8, n = 1, was complete after
12 h, and gave the corresponding keto-olefin 15 as the main
product; this was never obtained in a pure state as it was always
contaminated with a small amount of a minor component which
was isolated from the mixture by low temperature crystallisation at
-50 ◦C from hexane-ether (9 : 1). There was no olefinic signal in its
1H-NMR spectrum. This spectrum and the 13C-NMR spectrum,
together with 2-D and nOe techniques, enabled us to identify
this compound as 16, the product of a parallel cycloaddition.
Spectroscopically, the major component 15 showed the presence of
a styrene residue and saturated carbonyl group. Irradiation of the
higher homologue 8, n = 2 (Scheme 3) was stopped after 14 h, and
gave a single product 17 whose structure was confirmed by single
crystal X-Ray analysis (Fig. 2). The NMR spectra were consistent
with this structure. Decoupling, nOe, and 2-D techniques were
used to identify the individual signals.
Fig. 3 Methyl 11-oxo-3-phenyltricyclo[4.3.2.03,10]undecane-6-carboxylate
18 [CCDC No. 735147].
allylcyclopentenones 17 can, for example, be considered to proceed
via a a 1,5-ring closure of alkene units which are in a ‘parallel’
orientation (19A; n = 1) (Scheme 4), resulting in the formation
of 2. Ring closure of alkene units in the corresponding ‘crossed’
orientation (19B; n = 1) would involve a 1,6 ring closure and is
thus disfavoured by the rule of five. The successful application of
the rule of five is however dependent on arbitrarily considering
the alkene units to be connected by a pathway (19A, 19B) which
does not include the carbonyl group; considering this pathway
results in an incorrect prediction of the regiochemical outcome of
the reaction (20A and 20B). Although a 1,5-closure is not possible
for the 5-butenyl compound 3, the regioselectivity observed in
the formation of the major cycloadduct 4 can be accommodated
by the extension to the rule of five which states that where
1,5-closure is not possible, the reaction should proceed by 1,6-
closure. Unsatisfactorily however, such an approach only results
in a correct regiochemical prediction if the pathway involving the
carbonyl group, deliberately excluded above, is considered (20B;
n = 2). The frontier orbital (FO) formulation of the rule of five due
to Gleiter10 suggests that alkenes connected by a bridge containing
an odd number of carbon atoms should involve preferential
parallel cycloaddition; if the connecting bridge contains an even
Fig. 2 Methyl 7-oxo-1-phenyltricyclo[4.2.2.03,8]decane-6-carboxylate 17
(CCDC No. 735146).
Similarly, the keto-ester 8, n = 3, on irradiation (17 h) gave a
single compound which was identified as the cross-product 18. The
structure was again determined by single crystal X-Ray analysis
(Fig. 3); the NMR spectra obtained for the photoadduct were
consistent with this structure.
Discussion
The “rule of five”8,9,10 states that intramolecular [2 + 2] cycloaddi-
tion reactions will, where possible, occur via a 1,5-ring closure
and the five-membered ring biradical thus formed. Although
the rule provides a basis for rationalising the outcome of many
[2 + 2] cycloaddition reactions, its application in some cases
is less than convincing. The photochemical ring closure of 5-
Scheme 4
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2959–2968 | 2961
©

View Article Online
number of carbon atoms, cross addition should predominate.
Assuming the pathway involving the carbonyl group is again
arbitrarily excluded, the regiochemical outcome of the reactions of
1 and 3 is in keeping with this FO based version of the rule of five.
It should be pointed out however that this approach is explicitly
based10 on the intermediacy of an exciplex, a species whose
involvement in photochemical [2 + 2] cycloaddition reactions has
been seriously questioned.13 The basic rule of five has nothing to
contribute in terms of rationalising the regioselectivity observed
in the photoaddition of 5-(4¢-pentenyl)-3-phenylcyclopenten-one
6. If the pathway involving the carbonyl group is again ignored,
the formation of 7 via a 1,7-ring closure (19A; n = 3) would be in
keeping with Gleiter’s FO analysis (odd, parallel).
Using the rule of five to rationalise the regiochemical outcome
of the intramolecular [2 + 2] cycloaddition reactions of 6-
alkenylcyclohexenones described in this paper is even less satisfac-
tory. The formation of the parallel cycloadduct 16 on irradiation
of 8, n = 1, can be interpreted as a rule of five allowed, 1,6-closure
(21A, n = 1), but only if one again arbitrarily ignores the 1,5-
closure possible via the pathway containing the carbonyl group
(22B, n = 1). It is worth noting that the regiochemical preference
demonstrated by 8, n = 1, in its cycloaddition reaction is shared
by the structurally related, but electronically quite different, 1-
methyl-1-allyl-2(1H)-naphthalenone.12 Applying the FO approach
is equally unsatisfactory as it suggests that irradiation of 8, n = 1,
should result in the formation of the incorrect regioisomer through
the involvement of either 21B, n = 1 (even, crossed) or 22B, n = 1
(odd, parallel). The enone 8, n = 2 has an extra carbon in the chain
connecting the alkene units, but irradiation results in a reaction
whose regiochemistry is inverted relative to that observed for 8,
n = 1. This means that the FO approach again fails to provide a
rationalisation of the reaction outcome. The addition of a further
methylene to the linking chain does not change the regiochemistry
of the cycloaddition with 8, n = 3, forming 18: a prediction based
on the FO approach would thus, almost by default, be correct. The
basic rule is successful in relation to the formation of 17 from 8, n =
2, as using the pathway containing the carbonyl group, 22B, it can
be considered to involve a 1,6-ring closure. It does not however
provide any basis for dealing with the formation of 18 from 8, n = 3.
As it is clear that the rule of five in its basic form, or as formulated in
terms of FO interactions, fails to provide an understanding of the
photoaddition reactions of 6-alkenylcyclohexenones, a different
mechanistic framework is thus required.
The idea that the product determining factor in these photo-
chemical reactions might be the conformation of the biradical
intermediates derives from the fact that these are almost invariably
of the triplet type. This means that they are sufficiently long-lived
to allow conformational relaxation to occur, and that intersystem
crossing (ISC) is required before the final bond forming step can
take place. The rate of ISC depends on the level of spin–orbit
coupling (SOC) within the biradical and this in turn is strongly de-
pendent on its geometry,22,23 specifically the inter-radical distance
(IRD) and the relative orientation of the singly occupied p-orbitals.
In the work described here the orientational potential for SOC in
a particular biradical was evaluated qualitatively in terms of the
level of interaction apparent in its spin density plot. Once ISC has
occurred and the biradical is in the singlet state, the formation
of closed shell products is very rapid and the opportunity for
further conformational change is thus extremely limited. In the
case of cycloaddition reactions this final step could result in the
formation of a four-membered ring through ring closure, or the
regeneration of the starting material(s) through bond cleavage.
It is in fact reasonable to suggest that those biradicals which
have undergone relatively efficient ISC would favour cyclobutane
formation as the geometric factors facilitating ISC (short IRD and
favourable orbital overlap) would also favour ring closure rather
than bond cleavage. Biradicals which do not enjoy these structural
features will not be product forming and will simply revert to
starting material. Thus predicting the outcome of a cycloaddition
involves identifying the biradical(s) which possess these structural
characteristics and hence the product(s) that should be formed.
An analysis of the photochemical behaviour described above for
6-alkenyl-3-phenylcyclohex-2-enones, in terms of this concept is
set out in detail below.
As mentioned above intramolecular [2 + 2] cycloaddition in
systems such as 8, n = 1, 2, 3, and 13 can result in the formation of
regioisomeric products as the interacting alkene units can orientate
themselves relative to each other in a parallel or crossed fashion. A
molecular modelling study of the biradicals derived from 8, n = 1, 2,
3, and 13 was carried out with a view to evaluating the potential of
the biradical conformational control concept as a tool for under-
standing the regiochemistry of these photochemical reactions. The
study considered all possible biradicals, including those involving
1◦ radicals, as trapping experiments23 have indicated that these can
be formed as intermediates in the course of these photochemical
[2 + 2] cycloaddition reactions. The approach involved initially
identifying the low energy conformation(s) of each of the four
possible biradicals using a semi-empirical (AM1) conformational
search procedure. A DFT calculation (UB3YLP/6-31G*) was
then used to refine the structure and energy of each low energy con-
formation. Potential product forming biradicals were identified on
the basis of IRDs, and whether the spin density plots indicated that
the singly occupied p-orbitals were favourably orientated relative
to each other. Biradicals of high relative energy were however
excluded, even if their structures were appropriate, on the basis
that their formation would not be competitive.
An alternative concept, biradical conformation control, has
been successfully used to account for both the regiochemistry
and stereochemistry of photochemical cycloaddition and cy-
clization reactions which involve biradical intermediates. Thus
the regioselectivity of Paterno–Bu¨chi14 and intramolecular [2 +
2] cycloaddition reactions15 has been rationalised in terms of
this concept. The stereoselectivity of cyclizations of the Nor-
rish/Yang type16 and of those involving 1,5-biradicals17,18 has
also been satisfactorily explained, as has the stereoselectivity of
the intramolecular addition of C–H bonds to cyclic enones19 and
that of Paterno–Bu¨chi cycloadditions.20 Thus, although a [2 + 2]
cycloaddition may be a relatively complicated reaction in terms of
the range of possible product determining intermediates, this body
of work, together with trapping21 and other data,13 strongly argues
that the outcome of the reaction is determined by the structure and
behaviour of an intermediate 1,4-biradical.
None of the 1,4-biradicals that can be formed from 13 has
a low energy conformation which would facilitate closure to a
˚
cyclobutane. The IRD in each case is greater than 3.00 A (Table 1)
and there is no overlap in any of the spin density plots at the
0.004 electrons/au3 level, a value chosen empirically to facilitate
a comparison of the spatial relationship of the singly occupied
2962 | Org. Biomol. Chem., 2011, 9, 2959–2968
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
the formation of 17. Although the corresponding parallel 2◦/3◦
biradical (Table 2, E) is lowest in energy (Table 1), its singly
occupied orbitals, as indicated in the spin density plot, are not
appropriately orientated for ring closure and so it is not product
forming. The observed formation of 17 is thus entirely in keeping
with the principle of biradical conformation control.
Table 1 Relative energies and inter-radical distances for biradicals
1◦/3◦
Crossed
2◦/2◦
Crossed
1◦/2◦
Parallel
2◦/3◦
Parallel
Enone
DEa IRDb
DEa
IRDb DEa
IRDb DEa IRDb
13
9.17 3.02
10.00 3.20
10.16 3.18
14.10 3.11
13.40 3.02
13.73 2.95
16.75 2.87
0.00 3.12
0.00 3.13
0.00 3.16
0.00 3.17
The cyclohexenone 8, n = 3, also forms a crossed cycloadduct,
18, regioselectively. A consideration of the spin-density plots of
the four possible biradical intermediates again indicates that the
2◦/2◦ crossed (Table 2, F) and the 1◦/2◦ parallel biradicals are
potentially product forming, but that the observed product should
be that derived from the former because of the significant energy
difference (DE = 4.95 kcal mol-1) between the two biradicals
(Table 1). In this case there is the additional factor that the level
of overlap in the spin density plot is also much greater for the
crossed 2◦/2◦ radical for which overlap occurs up to the 0.0063
electrons/au3 level; overlap disappears at the 0.0045 electrons/au3
level for the corresponding parallel 1◦/2◦ biradical. The parallel
2◦/3◦ biradical (Table 2, G), despite being lowest in energy, can
again be excluded from the analysis as its singly occupied orbitals
are not appropriately orientated for ring closure and therefore this
biradical would be expected to undergo bond cleavage, reverting
to starting material.
In view of the many exceptions,24–27 to which the results
presented here can be added, it is clear that the empirical ‘rule
of five’ and the theoretical models which have been developed to
support it, do not provide a basis for a comprehensive under-
standing of the outcome of intramolecular [2 + 2] cycloaddition
reactions. The failure to allow for the effect of substituents is
one obvious limitation. In contrast, the ‘biradical conformation
control’ concept provides an understanding of the photochemical
behaviour of these systems in terms of the energy and structure
of what are the well-established intermediates in their reactions,
with substituents playing a role in determining the structure
of the minimum energy conformation of the intermediate. The
results presented here suggest that the behaviour of other systems
undergoing intramolecular photochemical [2 + 2] cycloaddition
should be re-examined using this approach and this work is
currently underway.
8, n = 1 7.90 3.02
8, n = 2 3.03 3.03
8, n = 3 4.97 2.90
9.33
2.89
11.78 2.79
a Energy relative to 2◦/3◦ parallel biradical, kcal mol-1; b inter-radical
˚
distance, A.
p-orbitals in the various biradicals. Thus the expectation for
this system is that cyclobutane formation would not occur, the
biradicals instead undergoing bond cleavage to regenerate starting
material. However the conformation of the lowest energy biradical
(Table 1), the 2◦/3◦ parallel biradical (Table 2, A), is set up for
intramolecular hydrogen abstraction as it involves a very short
˚
carbon radical/hydrogen distance (2.67 A), with the same atoms
forming part of a six-membered chair arrangement (Fig. 4a).
The fact that the hydrogen abstraction product 14 is actually
formed on irradiation validates the biradical conformational
control approach as a means of understanding the photochemical
behaviour of these systems.
Fig. 4 Hydrogen abstraction forming 14 and 15.
The molecular modelling analysis indicates that the biradical
intermediates derived from the carbomethoxy substituted cy-
clohexenone 8, n = 1, have lowest energy conformations that
are almost identical to those of the biradicals derived from
13 (Table 1). Thus it is not surprising that the formation of
the hydrogen abstraction product 15 should be observed, again
produced via the biradical of lowest energy, the 2◦/3◦ parallel
biradical (Table 2, B; Fig. 4b). However in this case molecular
modelling analysis of the 1◦/2◦ parallel biradical obtained from
8, n = 1, indicates that it has two low energy conformations
of almost equal energy (DE = 0.12 kcal mol-1). In contrast to
the situation for 13, there is a small degree of overlap in the
spin density plot of the higher energy of these two biradicals
at the 0.004 electrons/au3 level (Table 2, C), thus accounting
for the competitive formation of the parallel cycloadduct 16 on
irradiation of this enone. Irradiation of the cyclohexenone 8, n =
2, results in the regioselective formation of the crossed cycloadduct
17. Molecular modelling indicates that overlap occurs in the spin
density plots of the minimum energy forms of the 1◦/2◦ parallel
and the 2◦/2◦ crossed biradicals. However the energy difference
between the two (DE = 4.40 kcal mol-1) (Table 1) suggests that
the reaction should involve the latter (Table 2, D) and result in
Experimental
Melting points are uncorrected. Solvents were dried and distilled.
Photochemical experiments were carried out using an Applied
Photophysics 400 W medium pressure mercury lamp and a Pyrex
filter. These reactions were monitored by GC-MS. IR spectra were
measured in cm-1 using Perkin Elmer Paragon 1000 or 1600 FT-
IR machines as neat liquids, or as Nujol mulls. NMR spectra
were recorded using a Varian Gemini 2000, Bruker MSL 300,
or a Bruker DPX machine, using deuteriochloroform as solvent
(unless otherwise stated). Peak positions are given in ppm from
TMS, and J values in Hz. In the assignment of NMR signals,
atoms in side-chains are indicated with an apostrophe (e.g. 2¢).
Calculations were performed using Spartan ’04 (Wavefunction,
Inc., Irvine, CA, 2004). An AM1 based conformational search
procedure, based on the default Monte Carlo facility, was used
to identify the low energy conformations for each biradical. The
geometry of conformations with an inter-radical distance of less
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2959–2968 | 2963
©

View Article Online
Table 2 UB3YLP/6-31G* derived biradical structures
Enone
Biradical
Minimum Energy Conformer
Spin Density Surface
13
Parallel, 2◦/3◦
8, n = 1
Parallel, 2◦/3◦
8, n = 1
Parallel, 1◦/2◦
8, n = 2
Crossed, 2◦/2◦
8, n = 2
Parallel, 2◦/3◦
8, n = 3
Crossed, 2◦/2◦
8, n = 3
Parallel, 2◦/3◦
2964 | Org. Biomol. Chem., 2011, 9, 2959–2968
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
˚
than 3.5 A were then refined using a DFT (UB3LYP/6-31G*)
at this temperature for 1h. The mixture was acidified with
aqueous hydrochloric acid (20 mL, 1 M) and extracted with
dichloromethane (3 ¥ 30 mL). The combined organic layers were
washed with water, and dried (MgSO4). The solvent was removed
and the residue distilled to give an oil (13.57 g, bp 172–176 ◦C/0.7
mbar). The oil was dissolved in hexane/ether (9 : 1) at -40 ◦C to
give a precipitate, which was washed with hexane/ether (9 : 1) at
-40 ◦C. The product was dried in a desiccator in vacuo to give
ethyl 2-acetyl-5-oxo-5-phenyl-2-(2¢-propenyl)pentanoate (10.99 g,
53%), mp 48–49 ◦C. IR nmax (nujol) 1737, 1712, and 1686 cm-1; 1H
NMR d (200 MHz) 1.25 (3H, t, J = 7.2, ester CH3), 2.17 (3H, s,
acetyl CH3), 2.28 (2H, m, 3-H), 2.65 (2H, d, J = 7.2, 1¢-CH2),
2.88 (2H, m, 4-H), 4.2 (2H, q, J = 7.2, ester CH2), 5.10 (2H, m,
3¢-CH2), 5.62 (1H, ddt, J = 17.0, 10.5, and 6.7, 5-H, 2¢-CH), 7.49
(3H, m, m- and p-PhH), and 7.92 (2H, m, o-PhH); 13C NMR d
13.9 (ester CH3) 25.9 (3-C), 26.8 (acetyl CH3), 33.1 (4-C), 36.8
(1¢-C), 61.4 (ester CH2), 62.5 (2-C), 119.2 (3¢-C), 128.0 (o-PhC),
128.6 (m-PhC), 132.0 (p-PhC), 133.1 (2¢-C), 136.6 (ipso-PhC),
171.7 (1-C), 198.8 (5-C), and 204.7 (acetyl C O). Found: C,
71.41; H, 7.29. C18H22O4 requires C, 71.50; H, 7.33%.
optimization.
Crystal data
3-Phenyl-6-(2¢-propenyl)cyclohex-2-en-1-one 13 (Fig. 1).
C17H18O3, M = 270.31, Triclinic, a = 6.121(6), b = 10.219(3), c =
˚
12.1028(18) A, a = 80.995(17), b = 83.01(3), g = 82.18(5), U =
3
¯
˚
736.9(8) A , T = 298 K, space group P1, Z = 2, m(Mo-Ka) =
0.08 mm-1, 2843 reflections collected, 1387 unique, (Rint = 0.0176),
aR1 = 0.0527, wR2 [I > 2s(I)] = 0.1294, Gof = 1.033, CCDC
deposition number 735148.
Methyl 7-oxo-1-phenyltricyclo[4.2.2.03,8]decane-6-carboxylate
17 (Fig.2). C18H20O3, M = 284.34, Triclinic, a = 6.6572(9), b =
9.8399(8), c = 12.3209(15) A, a = 102.795(8), b = 99.112(11), g =
105.356(9), U = 738.47(15) A , T = 298 K, space group P1, Z = 2,
m(Mo-Ka) = 0.09 mm-1, 2763 reflections collected, 1396 unique,
(Rint 0.0325), aR1 = 0.0507, wR2 [I > 2s(I)] = 0.1283, Gof = 1.001,
CCDC deposition number 735146.
˚
˚
3
¯
Methyl 11-oxo-3-phenyltricyclo[4.3.2.03,10]undecane-6-carboxy-
Methyl 2-oxo-4-phenyl-1-(2¢-propenyl)cyclohex-3-ene-1-carbo-
xylate 8, n = 1. A mixture of the diketo methyl ester 12, n = 1
(6.29 g, 21.8 mmol) and DBU (3.32 g, 21.8 mmol) in methanol
(50 mL) were refluxed for 30 min. The mixture was acidified
with aqueous hydrochloric acid (30 mL, 1 M), and extracted with
dichloromethane (3 ¥ 50 mL). The combined organic layers were
washed with dilute hydrochloric acid (50 mL) and water (50 mL),
and then dried (MgSO4). Removal of the solvent, and distillation
of the residue gave an oil (4.02 g, bp 162–165 ◦C/2 mbar). The
oil was dissolved in hexane/ether (8 : 2, 20 mL) and cooled to
-50 ◦C to give a white precipitate. The product was washed with
hexane/ether (8 : 2) at -50 ◦C, and dried in a desiccator in vacuo to
give the keto ester 8, n = 1, as needles, mp 65–66 ◦C, (2.95 g, 47%);
IR nmax (Nujol) 1728, 1671, and 1640 cm-1; 1H NMR d (400 MHz)
2.10 (1H, ddd, J = 14.0, 9.5, 5.0, 6ax-H), 2.60 (2H, m, 6eq-H and
1a¢-CH2), 2.77 (2H, m, 5ax-H and 1b¢-CH2), 2.94 (1H, dddd, J =
18.5, 9.3, 5.0, and 2.0, 5eq-H), 3.72 (3H, s, ester CH3), 5.15 (2H, m,
3¢-CH2), 5.80 (1H, ddt, J = 17.5, 10.0, and 7.2, 2¢-H), 6.47 (1H, br s,
late 18 (Fig.3). C19H22O3, M = 298.37, Triclinic, a = 6.6266(7),
˚
b = 10.1023(10), c = 12.3043(11) A, a = 90.769(8), b = 96.613(10),
3
¯
˚
g = 103.782(8), U = 793.91(14) A , T = 298 K, space group P1, Z =
2, m(Mo-Ka) = 0.08 mm-1, 2133 reflections collected, 1578 unique,
(Rint 0.0183), aR1 = 0.0373, wR2 [I > 2s(I)] = 0.1003, Gof = 1.086,
CCDC deposition number 735147.
aR1 = RꢀFo| - |Fcꢀ/R|Fo|,
wR2 = [Rw(Fo - Fc )2/Rw(Fo )2]1/2
2
2
2
Synthesis
Methyl 2-acetyl-5-oxo-5-phenyl-2-(2¢propenyl)pentanoate 12, n =
1. Sodium methoxide [from sodium (0.2 g, 12 mmol)] in
methanol (10 mL) was added slowly to a stirred mixture of methyl
2-acetylpent-4-enoate (4.00 g, 25.6 mmol) and phenyl vinyl ketone
(3.38 g, 25.6 mmol) at 0 ◦C. The mixture was stirred at this
temperature for 1 h, then acidified using aqueous hydrochloric acid
(20 mL, 1 M), and extracted with dichloromethane (3 ¥ 30 mL).
The combined organic layers were washed with water, and dried
(MgSO4). The solvent was removed in vacuo and the residue was
fractionally distilled to give 12, n = 1 as a viscous oil (3.91 g,
53%); bp 162–164 ◦C/0.3 mbar; IR nmax (neat) 1741, 1712, and
1686 cm-1; 1H NMR d (200 MHz) 2.11 (3H, s, acetyl CH3), 2.22–
2.29 (2H, m, 3-H), 2.61 (2H, br d, 1¢-CH2), 2.84 (2H, m, 4-H), 3.66
(3H, s, ester CH3), 5.04 (2H, m, 3¢-CH2), 5.60 (1H, ddt, J = 17.2,
9.9, and 7.3, 2¢-CH), 7.34- 7.48 (3H, m, m- and p-PhH), 7.86 (2H,
m, o-PhH); 13C NMR d (50 MHz) 25.7 (3-C), 26.6 (acetyl CH3),
32.9 (4-C), 36.7 (1¢-C), 52.1 (ester CH3), 62.3 (2-C), 119.0 (2¢-C),
127.8 (2¢-C and o-PhC), 128.4 (m-PhC), 131.9 (p-PhC), 136.4 (ipso-
PhC), 172.0 (1-C), 198.6 (5-C), and 202.4 (acetyl C O). Found:
C, 70.68; H, 7.15. C17H20O4 requires C, 70.81; H, 6.99%.
3-H), 7.41 (3H, m, m- and p-PhH) and 7.55 (2H, m, o-PhH); 13
C
NMR d (80 MHz) 25.3 (5-C), 29.8 (6-C), 38.4 (1¢-C), 52.3 (ester
CH3), 55.3 (1-C), 118.8 (3¢-C), 124.2 (3-C), 126.2 (o-PhC), 128.7
(m-PhC), 130.2 (p-PhC), 133.2 (2¢-C), 137.9 (ipso-PhC), 158.5 (4-
C), 171.5 (ester C) and 195.0 (2-C). Found: C, 75.58; H, 6.56%.
C17H18O3 requires C, 75.53; H, 6.71.
3-Phenyl-6-(2¢-propenyl)cyclohex-2-en-1-one 13. Sodium hy-
dride (850 mg, 35.4 mmol) and ethyl 2-acetyl-5-oxo- 5-phenyl-2-
(2¢-propenyl)-pentanoate (10.0 g, 33.1 mmol) in tetrahydrofuran
(50 mL) were refluxed for 1 h. The mixture was acidified with
aqueous hydrochloric acid (50 mL, 1 M) and extracted with
dichloromethane (3 ¥ 50 mL). The combined organic layers were
washed with hydrochloric acid and water, and then dried (MgSO4).
Removal of the solvent and distillation of the residue afforded an
oil, bp 165–170 ◦C. This oil was dissolved in acetone (30 mL)
and cooled to -75 ◦C, to afford a colourless precipitate which was
collected, and washed with acetone (10 mL) at -75 ◦C. The product
was dried in a desiccator in vacuo, to give the cyclohexenone 13
(4.96 g, 70%) mp 49 ◦C; IR nmax (neat) 1669, 1651, and 1649 cm-1;
1H NMR d (400 MHz) 1.89 (1H, dddd, J = 13.6, 11.5, 9.5, and
Ethyl
2-acetyl-5-oxo-5-phenyl-2-(2¢-propenyl)pentanoate.
Sodium ethoxide [from sodium (0.05 g, 2 mmol)] in ethanol
(10 mL) was added slowly to a stirred mixture of ethyl 2-
acetylpent-4-enoate (12.87 g, 75.7 mmol) and phenyl vinyl ketone
(10.00 g, 75.7 mmol) at 0 ◦C, and the reaction maintained
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2959–2968 | 2965
©

View Article Online
5.5, 5ax-H), 2.18–2.30 (2H, m, 5eq- and 1b¢-CH2), 2.42 (1H, ddt,
J = 11.5, 8.8, and 4.5, 6-H), 2.73 (2H, m, 4ax-H and 1a¢-CH2),
2.83 (1H, m, 4eq-H), 5.10 (2H, m, 3¢-CH2), 5.85 (1H, m, 2¢-H),
6.43 (1H, partially resolved d, J = 1.5, 2-H), 7.42, (3H, m, m- and
p-PhH), and 7.55 (2H, m, o-PhH); 13C NMR d (80 MHz) 27.4 (5-
C), 27.43 (4-C), 33.7 (1¢-C), 45.3 (6-C), 116.7 (3¢-C), 125.0 (2-C),
126.6 (o-PhC), 128.7 (m-PhC), 129.6 (p-PhC), 136.1 (2¢-C), 138.6
(ipso-PhC), 158.6 (6-C) and 200.7 (1-C). Found: C, 84.73; H, 7.56.
C15H16O requires C, 84.87; H, 7.60%.
(8 : 2, 20 mL) and cooled to -50 ◦C to give a white precipitate,
which was collected and washed at -50 ◦C with hexane/ether
(8 : 2). The solid was dried in air and then in a desiccator in vacuo.
The product (2.93 g, 52%) had mp 67–68 ◦C; IR nmax (neat) 1723,
1
1665, and 1638 cm-1; H NMR d (400 MHz) 2.10 (5H, m, 1¢-
and 2¢-CH2, and 6ax-H), 2.51 (1H, m, 6eq-H), 2.78 (1H, m, 5ax-H),
2.95 (1H, m, 5eq-H), 3.72 (3H, s, ester-CH3), 5.07 (2H, m, 4¢-H),
5.80 (1H, ddt, J = 17.1, 10.0, and 7.0, 5¢-H), 6.45 (1H, br s, 3-
H), 7.42 (3H, m, m- and p-PhH), and 7.55 (2H, m, o-PhH); 13C
NMR d (80 MHz) 25.3 (5-C), 29.9 (6-C), 33.1 (1¢-C), 36.4 (2¢-
C), 52.3 (ester CH3), 55.7 (1-C), 115.1 (4¢-C), 124.3 (3-C), 126.1
(o-PhC), 128.7 (m-PhC), 130.1 (p-PhC), 137.4 (3¢-C), 138.0 (ipso-
PhC), 158.2 (4-C), 171.0 (ester C O), and 195.4 (2-C). Found: C,
76.08; H, 7.15.%. C18H20O3 requires C, 76.03, H, 7.09.
Methyl 2-acetylhex-5-enoate 11, n = 2. Methyl 3-oxobutanoate
(8.59 g, 74.0 mmol) was added to a solution of sodium methoxide
[from sodium (1.75 g, 74.0 mmol)] in methanol (100 mL). 4-
Bromo-1-butene (10 g, 74.0 mmol) was added over 10 min, and
the mixture refluxed for 7 h. The solvent was removed, and the
residue acidified with dilute hydrochloric acid (30 mL, 1 M) and
extracted with diethyl ether (3 ¥ 100 mL). The combined ether
layers were dried (MgSO4), and the solvent removed. Fractional
distillation of the residue gave the ketoester 11, n = 2 as a colourless
Methyl 2-acetylhept-6-enoate 11,
n
=
3. Methyl 3-
oxobutanoate (9.24 g, 70.0 mmol) was added to a solution of
sodium methoxide [from sodium (1.66 g 72.0 mmol)] in methanol.
5-Bromo-1-pentene (10.0 g, 67.0 mmol) was added over 10 min,
and the mixture refluxed for 8 h. The solvent was removed and
the residue was acidified with hydrochloric acid (30 mL, 1 M)
and extracted with dichloromethane (3 ¥ 50 mL). The combined
organic layers were washed with water and dried (MgSO4). The
solvent was removed in vacuo and the residue distilled to give the
ketoester 11, n = 3 as a colourless oil, bp 60–62 ◦C/3.0 mbar; IR
◦
oil (6.92 g, 55%), bp 67–69 C/0.5 mmbar; IR nmax (neat) 1741,
1
1717, and 1641 cm-1; H NMR d (400 MHz) 1.78 (2H, m, 3-H),
1.90 (2H, 4-H), 2.06 (acetyl CH3), 3.33 (1H, t, J = 7, 2-H), 3.57
(3H, s, ester CH3), 4.86 (2H, m, 6-CH2), 5.70 (1H, ddt, J = 16.7,
10.0, and 6.6, 5-H); 13C NMR d (80 MHz) 26.7 (3-C), 28.5 (acetyl
CH3), 30.4 (4-C), 51.8 (ester CH3), 58.1 (2-C), 115.4 (6-C), 136.6
(5-C), 169.7 (1-C), and 202.3 (acetyl C O). Found: C, 63.53; H,
8.2%. C9H14O3 requires C, 63.51; H, 8.29.
1
nmax (neat) 1744, 1717, and 1641 cm-1; H NMR d (400 MHz)
1.36 (2H, m, 4-CH2), 1.83 (2H, m, 3-CH2), 2.05 (2H, q, J = 7-H,
5-CH2), 2.20 (3H, s, acetyl CH3), 3.41 (1H, t, J = 7.5, 2-H), 3.71
(ester CH3), 4.96 (2H, m, 7-CH2), 5.75 (1H, ddt, J = 17.0, 10.3, and
6.6, 6-H); 13C NMR d (80 MHz) 26.5 (4-C), 27.5 (3-C), 28.6 (acetyl
CH3), 33.2 (5-C), 52.2 (ester CH3), 59.4 (2-C), 114.9 (7-C), 137.6
(6-C), 170.1 (1-C), and 202.8 (acetyl C O). Found: C, 65.09; H,
8.66%. C10H16O3 requires C, 65.23; H, 8.75.
Methyl 2-acetyl-2-(3-oxo-3-phenylpropyl)hex-5-enoate 12, n = 2.
Sodium methoxide [from sodium (0.2 g, 12.0 mmol)] in methanol
(10 mL) was added slowly to a stirred mixture of methyl 2-
acetylhex-4-enoate 11, n = 2 (12.8 g, 75.7 mmol) and phenyl vinyl
ketone (10.0 g, 75.7 mmol) at 0 ◦C. The mixture was stirred
at this temperature for 1 h, acidified with dilute hydrochloric
acid (20 mL, 1 M), and extracted with dichloromethane (3 ¥
30 mL). The combined organic layers were dried (MgSO4), and
the solvent removed. The residue was fractionally distilled to give
the diketoester 12, n = 2 as a viscous oil (11.62 g, 51%), bp 170–
172 ◦C/0.5 mmbar; IR nmax (neat) 1741, 1712, and 1686 cm-1;
1H NMR d (200 MHz) 1.92 (2H, m, 3-CH2), 2.11 (3H, s, acetyl
CH3), 2.17 (2H, m, 4-CH2), 2.28 (2H, m, 1¢-CH2), 2.82 (2H, m,
2¢-CH2), 3.67 (3H, s, ester CH3), 5.06 (2H, m, 6-CH2), 5.60 (1H,
ddt, J = 17.0, 10.5, and 6.7, 5-H), 7.45 (3H, m, m- and p-PhH), and
7.87 (2H, m, o-H); 13C NMR d (50 MHz) 25.7 (1¢-C), 26.7 (acetyl
CH3), 34.4 (3-C), 30.4 (4-C), 32.4 (2¢-C), 52.2 (ester CH3), 62.3 (2-
C), 117.0 (6-C), 127.8 (o-PhC), 130.4 (5-C), 128.4 (m-PhC), 131.9
(p-PhC), 136.6 (ipso-PhC), 172.6 (1-C), 198.6 (3¢-C), and 202.7
(acetyl C O). Found: C, 71.38; H, 7.39%. C18H22O4 requires C,
71.5; H, 7.33.
Methyl 2-acetyl-2-(3-oxo-3-phenylpropyl)hept-6-enoate (12, n =
3). Sodium methoxide [from sodium (0.2 g, 12 mmol)] 1n
methanol (10 mL) was added slowly to a stirred mixture of methyl
2-acetylhept-6-enoate (4.0 g, 25.6 mmol) and phenyl vinyl ketone
(3.36 g, 25.6 mmol) at 0 ◦C. The resulting mixture was stirred
at this temperature for 1 h, acidified with dilute hydrochloric
acid (20 mL, 1 M), and extracted with dichloromethane (3 ¥
30 mL). The combined organic layers were washed with water
and dried (MgSO4). The solvent was removed in vacuo and the
residue distilled to afford the diketoester 12, n = 3 as a viscous
yellow oil (3.91 g, 53%), bp 162–164 ◦C/0.3 mbar. IR nmax (neat)
1
1741, 1712, and 1686 cm-1; H NMR d (200 MHz) 1.21 (2H, m,
4-H), 1.84 (2H, m, 3-H), 2.04 (2H, m, 5-H), 2.11 (acetyl CH3), 2.25
(2H, m, 1¢-CH2), 2.79 (2H, m, 2¢-CH2), 3.67 (ester CH3), 4.94 (2H,
m, 7-CH2), 5.60 (1H, ddt, J = 16.5, 10.5, and 6.6, 6-CH), 7.38 (2H,
m, m-PhH), 7.48 (1H, m, o-PhH), and 7.87 (2H, m, p-PhH); 13C
NMR d (50 MHz) 23.0 (4-C), 25.7 (1¢-C), 26.6 (acetyl CH3), 31.6
(3-C), 33.1 (2-¢C), 33.5 (5-C), 52.0 (ester CH3), 62.4 (2-C), 115.0
(7-C), 127.7 (o-PhC), 128.3 (m-PhC), 132.8 (p-PhC), 136.4 (ipso-
PhC), 172.4 (1-C), 198.5 (3¢-C), and 202.7 (acetyl C O). Found:
C, 72.26; H, 7.15%. C17H20O4 requires C, 72.13; H, 7.33.
Methyl 2-oxo-4-phenyl-1-(3¢-butenyl)cyclohex-3-ene-1-carboxy-
late 8, n = 2. Methyl 2-acetyl-(3-oxo-3-phenylpropyl)hex-5-
enoate 12, n = 2 (6.00 g, 19.8 mmol), DBU (3.01 g, 19.8 mmol)
and methanol (10 mL) were refluxed for 30 min. The mixture
was acidified with dilute hydrochloric acid (30 mL, 1 M), and
then extracted with dichloromethane (3 ¥ 50 mL). The combined
organic layers were washed with dilute hydrochloric acid and
water, and were dried (MgSO4). Removal of the solvent and
fractional distillation of the residue gave a yellow oil (3.75 g, bp
184–189 ◦C/0.6 mbar). The oil was dissolved in hexane/ether
Methyl 2-oxo-4-phenyl-1-(4¢-pentenyl)cyclohex-3-ene-1-carbo-
xylate 8, n = 3. Methyl 2-acetyl-2-(3-oxo-3-phenylpropyl)hept-
6-enoate 12, n = 3 (7.2 g, 22.7 mmol), DBU (3.46 g, 22.7 mmol)
and methanol (50 mL) were refluxed for 30 min. The mixture was
2966 | Org. Biomol. Chem., 2011, 9, 2959–2968
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
acidified with dilute hydrochloric acid (30 mL, 1 M), and extracted
with dichloromethane (3 ¥ 50 mL). The combined organic layers
were washed with dilute hydrochloric acid (50 mL) and water, and
then dried (MgSO4). The solvent was removed in vacuo, and the
residue distilled to give a yellow oil (3.75 g, bp 184–189 ◦C/0.6
mbar). ◦The oil was dissolved in hexane/ether (8 : 2) and cooled
to -50 C to afford a white precipitat◦e, which was collected and
washed with the same solvent at -50 C. The product was dried
(desicc◦ator) to give the cyclohexenone 8, n = 3 (3.39 g, 50%), mp
75.40; H, 6.63%. C17H18O3 requires C, 75.53; H, 6.71. MS (EI) m/z
270.
The filtrate from the isolation of 16 was concentrated to
give an oil (20 mg), which was shown spectrscopically to
be a mixture of the tricyclic ketone 16 and methyl 9-oxo-
2-phenylbicyclo[3.3.1]non-2-ene-5-carboxylate 15: 1H NMR d
(400 MHz), 1.64–2.45 (7H, ms, 4endo-H, 6exo-H, 6endo-H, 7exo-H, 7endo
-
H, 8exo-H, 8endo-H), 2.68 (1H, dd, J = 19.2, and 3.8, 4exo-H), 3.49
(1H, s, 1-H), 3.80 (3H, s, ester CH3), 6.29 (1H, t, J = 3.7, 3-H), and
7.30 (5H, m, o-, m-, and p-PhH). MS (EI) m/z 270.
1
70–72 C; IR nmax (Nujol) 1723, 1665, and 1638 cm-1; H NMR
d 1.45 (2H, m, 2¢-H), 1.82 (1H, m, 1¢-H), 2.06 (4H, m, 6ax-H, 3¢-
CH2, 1¢-H), 2.64 (1H, dt, J = 13.5, and 5.0, 6eq-H), 2.76 (1H, dt,
J = 18.5, and 5.0, 5ax-H), 2.95 (1H, m, 5eq-H), 3.72 (3H, s, ester
CH3), 5.01 (2H, m, 5¢-CH2), 5.81 (1H, ddt, J = 17.0, 10.0, and 6.7,
4¢-H), 6.46 (1H, br s, 3-H), 7.42 (3H, m- and p-PhH), and 7.56 (2H,
m, o-PhH); 13C NMR d 23.9 (2¢-C), 25.4 (5-C), 29.9 (6-C), 33.3
(1¢-C), 33,9 (3¢-C), 52.3 (ester CH3), 56.1 (1-C), 114.8 (5¢-C), 124.3
(3-C), 126.0 (o-PhC), 128.7 (m-PhC), 130.1 (p-PhC), 138.0 (4¢-C),
138.1 (ipso-PhC), 158.1 (4-C), 172.0 (ester C O), and 196.1 (2-C).
Found: C, 76.38; H, 7.50%. C19H22O3 requires C, 76.48; H, 7.43.
Methyl 7-oxo-1-phenyltricyclo[4.2.2.03,8]decane-6-carboxylate
17. Methyl 2-oxo-4-phenyl-1-(3¢-butenyl)cyclohex-3-ene-1-carb-
oxylate 8, n = 2 (1.0 g) in acetonitrile (450 mL) was irradiated for
14 h, the reaction being monitored by gc-ms. The solvent was
removed in vacuo to give a dark viscous oil, which was purified by
flash chromatography (hexane/ether, 9 : 1). The resulting oil was
triturated with the same solvent mixture to give some solid, and
the filtrate was cooled to -50 ◦C when more product precipitated.
The solids were combined to give◦the tricyclic ketoester 17
1
(250 mg, 40%) as needles, mp 82–83 C; H NMR d (400 MHz)
1.72 (1H, ddd, J = 13.8, 4.8 and 3.3, 9endo-H), 1.86 (1H, m, 9exo-H),
1.92 (1H, m, 4syn-H), 2.12 (1H, m, 4anti-H), 2.33 (1H, dd, J = 12.6,
and 3.0, 2syn-H), 2.43 (1H, ddd, J = 14.6, 6.0, and 3.0, 5syn-H), 2.59
(2H, m, 10endo-H and 5anti-H), 2.81 (1H, ddd, J = 14.6, 6.0, and 3.0,
10exo-H), 2.96 (2H, m, 3-H and 2anti-H), 3.30 (1H. br d, J = 9.5,
8-H), 3.83 (3H, s, ester CH3), 7.29 (1H, m, p-PhH), 7.40 (4-H, o-
and m-PhH); 13C NMR d (80 MHz) 24.7 (4-C), 29.3 (3-C), 31.4
(10-C), 35.8 (5-C), 37.2 (9-C), 37.8 (2-C), 49.3(1-C), 51.2 (8-C),
52.3 (ester CH3), 57.1 (6-C), 124.7 (o-PhC), 125.9 (p-PhC), 128.3
(m-PhC), 150.0 (ipso-PhC), 173.7 (ester C O), and 213.9 (7-C).
C18H20O3 requires C, 76.03; H, 7.09. Found: C, 75.95; H, 7.00%.
MS (EI) m/z 284. Crystals for X-ray analysis were grown from a
hexane/ether solution (9 : 1) at 5◦ C, collected, washed with the
same solvent at -50 ◦C, and dried in vacuo in a desiccator.
Photochemical reactions
2-Phenyl-bicyclo[3.3.1]non-2-en-9-one
14. 3-Phenyl-6-(2¢-
propenyl)cyclohex-2-en-1-one 13 (1.74 g) in acetonitrile (450 mL)
was irradiated for 2 h. The solvent was removed in vacuo to
give a viscous oil, which was purified by flash chromatography
(hexane/ether, 9 : 1) and trituration with the same solvent mixture
to give white needles. The filtrate was cooled to -50 ◦C, when
more precipitate formed. The solids were combined to afford the
bicyclic ketone 1428 (total yield, 185 mg, 11%), mp 40–41 ◦C; 1H
NMR d (400 MHz) 1.57 (1H, br s, 7exo-H), 1.95 (5H, m, 6exo-H,
6endo-H, 7endo-H, 8exo-H, and 8endo-H), 2.60 (1H, dd, J = 19.0 and
4.0, 4exo-H), 2.62 (1H, m, 5-H), 2.89 (1H, m, 4endo-H), 3.41 (1H, s,
1-H), 6.31 (1H, t, J = 3.5, 3-H), 7.30 (1H, m, p-PhH), and 7.40
(4H, m, o- and m-PhH); 13C NMR d (80 MHz), 17.1 (7-C), 32.7
(6-C), 35.7 (4-C), 36.8 (8-C), 44.4 (5-C), 49.9 (1-C), 125.6 (2-C),
125.7 (3-C), 127.5 (o-PhC), 128.4 (m-PhC), 137.1 (p-PhC), 139.2
(ipso-PhC) and 215.8 (9-C). MS (EI) m/z 212.
Methyl
11-oxo-3-phenyltricyclo[4.3.2.03,10]undecane-6-carbo-
xylate (18). Methyl 2-oxo-1-(4¢-pentenyl)-4-phenylcyclohex-
2-ene-1-carboxylate 8, n = 3 (1.0 g) 1n acetonitrile (450 mL)
was irradiated for 17 h, the reaction being monitored by gc-ms.
The solvent was removed in vacuo to give a dark viscous oil,
which was purified by flash chromatography (hexane/ether,
9 : 1). The resulting oil was triturated with hexane/ether (9 : 1) to
give some solid. The filtrate was cooled to -50 ◦C, when more
solid precipitated. The combined solids were dried in vacuo in a
desiccator to give◦the tricyclic ketoester 18 as needles (271 mg,
Methyl 6-oxo-2-phenyltricyclo[3.3.1.02,7]nonane-5-carboxylate
16. The cyclohexenone ester 8, n = 1 (2.0 g) in acetonitrile
(450 mL) was irradiated for 1.5 h. The solvent was removed
in vacuo to give a viscous oil, which was purified by flash
chromatography, and then dissolved in hexane/ether (9 : 1) and
cooled to -50 ◦C. The precipitate formed was collected, and
the process repeated several times. The combined precipitates
(88 mg, 4.4%) were washed with the same solvent mixture at
1
27%), mp 76–77 C; H NMR d (400 MHz) 1.43 (1H, tt, J =
12.5, and 2.5, 9exo-H), 1.80 (5H, m, 5exo-H, 4endo-H, 9endo-H, 8exo-H,
and 8endo-H), 2.05 (1H, dt, J = 14.0 and 3.3, 5endo-H), 2.13 (2H,
m, 4exo-H and 7exo-H), 2.55 (1H, m, 2endo-H), 2.59 (1H, dd, J =
13.0 and 5.5, 7endo-H), 2.69 (1H, dd, J = 13.5 and 8.0, 2exo-H),
2.86 (1H, m, 1-H), 3.55 (1H, d, J = 11.5, 10-H), 3.74 (3H, s,
ester-CH3), 7.22 (4H, o- and m-PhH), and 7.33 (p-PhH); 13C
NMR d (80 MHz) 19.5 (4-C), 24.4 (1-C), 27.4 (7-C), 28.9 (2-C),
29.0 (9-C), 33.4 (5-C), 36.2 (8-C), 43.6 (3-C), 52.2 (ester-CH3),
55.6 (10-C), 58.6 (6-C), 125.2 (o-PhC), 125.9 (m-PhC), 128.4
(p-PhC) 149.8 (ipso-PhC), 174.6 (ester C O), and 211.3 (11-C).
Found: C, 76.40; H,7.39%. C19H22O3 requires C, 76.48; H, 7.43.
MS (EI) m/z 298. Crystals for X-ray analysis were obtained from
a hexane/ether (9 : 1) solution.
◦
-50 C, to give methyl 6-oxo-2-phenyltricyclo[3.3.1.02,7]-nonane-
◦
5-carboxylate 16, mp 136–137 C; H NMR29 d (400 MHz) 1.69
(1H, d, J = 10.0, 8exo-H), 1.75 (1H, m, 3exo-H), 1.88 (1H, ddd, J =
15.0, 11.0, and 6.6, 3endo-H), 2.25 (2H, m, 4exo-H, 4exo-H), 2.36 (1H,
ddd, J = 12.5, 6.5, and 3.0, 9syn-H), 2.55 (1H, m, 8endo-H), 2.69 (1H,
dd, J = 12.5 and 2.0, 9anti-H), 3.05 (1H, q, J = 6.5, 1-H), 3.35 (1H,
dd, J = 6.5 and 5.5, 7-H), 3.83 (3H, s, ester CH3), 7.29 (1H, m,
p-PhH), and 7.40 (4H, m, o- and m-PhH); 13C NMR d 29.9 (3-C),
31.0 (4-C), 33.9 (8-C), 35.2(9-C), 38.9 (1-C), 52.2 (ester CH3), 55.1
(7-C), 55.9 (2-C), 126.2 (o-PhC), 126.6 (m-PhC), 128.6 (p-PhC),
143.9 (ipso-PhC), 172.0 (ester C O), and 210.8 (6-C). Found: C,
1
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2959–2968 | 2967
©

View Article Online
15 M. Audley and N. W. A. Geraghty, Tetrahedron Lett., 1996, 37, 1641–
1644.
Notes and references
1 1 Part 13. T. B. H. McMurry, A. G. Murphy, D. N. Work, A. G. Avent
and J. P. James, J. Chem. Res. (S), 2002, 317; J. Chem. Res. (M), 2002,
763–773.
2 G. Gowda and T. B. H. McMurry, J. Chem. Soc., Perkin Trans. 1, 1980,
1516–1522.
3 M. T. M. Clements, R. C. Cathcart, I. D. Cunningham, T. B. H.
McMurry and S. N. Rao, J. Chem. Res. (S), 1984, 223; J. Chem. Res.
(M), 2002, 2020.
4 M. T. M. Clements and T. B. H. McMurry, J. Chem. Res. (S), 1993,
344–345.
16 A. G. Griesbeck and H. Heckroth, J. Am. Chem. Soc., 2002, 124, 396–
403.
17 A. Zand, B.-S. Park and P. J. Wagner, J. Org. Chem., 1997, 62, 2326–
2327.
18 P. J. Wagner, A. Zand and B.-S. Park, J. Am. Chem. Soc., 1996, 118,
12856–12857.
19 L. A. Paquette, P. D. Pansegrau, P. E. Wiedeman and J. P. Springer,
J. Org. Chem., 1988, 53, 1461–1466.
20 W. Adam, V. R. Stegmann and S. Weinkoetz, J. Am. Chem. Soc., 2001,
123, 2452–2453.
5 I. D. Cunningham, T. B. H. McMurry, M. P. Napier and S. N. Rao,
J. Chem. Soc., Perkin Trans. 1, 1986, 1235–1242.
6 M. T. M. Clements, K. J. Crowley, P. V. Kavanagh, M. A. Lennon,
T. B. H. McMurry and M. P. Napier, J. Chem. Res. (S), 1991, 318;
J. Chem. Res. (M), 1991, 2920.
7 T. B. H. McMurry, A. Work and B. McKenna, J. Chem. Soc., Perkin
Trans. 1, 1991, 811–816.
21 D. J. Maradyn and A. C. Weedon, J. Am. Chem. Soc., 1995, 117, 5359–
5360; D. Andrew and A. C. Weedon, J. Am. Chem. Soc., 1995, 117,
5647–5663.
22 L. Carlacci, C. Doubleday Jr., T. R. Furlani, H. F. King and J. W.
McIver, J. Am. Chem. Soc., 1987, 109, 5323–5329.
23 C. Doubleday, N. J. Turro and J.-F. Wang, Acc. Chem. Res., 1989, 22,
199–205.
8 R. Srinivasan and K. H. Carlough, J. Am. Chem. Soc., 1967, 89, 4932–
4936.
9 R. S. H. Liu and G. S. Hammond, J. Am. Chem. Soc., 1967, 89, 4936–
4944.
10 R. Gleiter and W. Sander, Angew. Chem., Int. Ed. Engl., 1985, 24,
566–568.
24 S. Wolff and W. C. Agosta, J. Am. Chem. Soc., 1983, 105, 1292–
12991299–1304; A. R. Matlin, C. F. George, S. Wolff and W. C. Agosta,
J. Am. Chem. Soc., 1986, 108, 3385–3394; C. Schroder, S. Wolff and
W. C. Agosta, J. Am. Chem. Soc., 1987, 109, 5491–5497.
25 A. J. Barker, M. J. Begley, M. Mellor, D. A. Otieno and G. Pattenden,
J. Chem. Soc., Perkin Trans. 1, 1983, 1893–1900.
11 P. Margaretha and W. Frostl, Helv. Chim. Acta, 1976, 59, 2244–2248.
12 N. W. A. Geraghty, M. J. Monaghan and N. Hanley, Tetrahedron Lett.,
1987, 28, 4729–4732.
13 D. I. Schuster, in CRC Handbook of Organic Photochemistry and
Photobiology, ed. W. Horspool and F. Lenci, CRC Press, Boca Raton,
2nd edn, 2004, ch. 72, pp. 9–16.
26 Y. Tamura, H. Ishibashi, Y. Kita and M. Takeda, J. Chem. Soc., Chem.
Commun., 1973, 101–102; M. Ikeda, M. Takahashi, T. Uchino, K.
Ohno, Y. Tamura and M. Kido, J. Org. Chem., 1983, 48, 4241–4247.
27 E. Fischer and R. Gleiter, Angew. Chem., Int. Ed. Engl., 1989, 28,
925–927.
28 A. Cope and E. C. Hermann, J. Am. Chem. Soc., 1950, 72, 3405–3410.
29 The assignment of exo, endo, syn and anti in 16 are counter-intuitive
but in accordance with IUPAC and CAS practice.
14 Q.-H. Song, H.-B. Wang, X.-B. Li, X.-M. Hei, Q.-X. Guo and S.-Q.
Yu, J. Photochem. Photobiol., A, 2006, 183, 198–204.
2968 | Org. Biomol. Chem., 2011, 9, 2959–2968
This journal is
The Royal Society of Chemistry 2011
©