Alkyl radical generation using cyclohexa-1,4-diene-3-carboxylates and 2,5-dihydrofuran-2-carboxylates

Article abstract of DOI:10.1039/a606414k

3-Methylcyclohexa-1,4-diene-3-carboxylic acid and 2-methyl-2,5-dihydrofuran-2-carboxylic acid were prepared by Birch reduction and alkylation of benzoic and furoic acid respectively and converted to alkyl esters. Cyclohexadienyl and 2,5-dihydrofuranyl radicals were generated by hydrogen abstraction and characterised by EPR spectroscopy. The esters decomposed thermally in the presence of a radical initiator to generate alkyl radicals which could be trapped with moderate efficiency by halogen donors or alkenes. Loss of methyl to afford an alkyl benzoate was an important side reaction at higher temperatures. From the thermal reaction of hex-5-enyl 3-methylcyclohexa-1,4-diene-3-carboxylate the rate constant for hydrogen abstraction from the ester by hexenyl radicals was determined to be 0.82 × 10⁵ dm³ mol^-1 s^-1 at 140°C.

Full text of DOI:10.1039/a606414k

View Article Online / Journal Homepage / Table of Contents for this issue
Alkyl radical generation using cyclohexa-1,4-diene-3-carboxylates
and 2,5-dihydrofuran-2-carboxylates
Gavin Binmore,a Liberato Cardellinib and John C. Walton*,a
a
University of St. Andrews, School of Chemistry, St. Andrews, Fife, UK KY16 9ST
b Dipartimento di Scienze dei Materiali e della Terra, Università, Via Brecce Bianche,
60131 Ancona, Italy
3-M ethylcyclohexa-1,4-diene-3-carboxylic acid and 2-methyl-2,5-dihydrofuran-2-carboxylic acid were
prepared by Birch reduction and alkylation of benzoic and furoic acid respectively and converted to alkyl
esters. Cyclohexadienyl and 2,5-dihydrofuranyl radicals were generated by hydrogen abstraction and
characterised by EPR spectroscopy. The esters decomposed thermally in the presence of a radical initiator
to generate alkyl radicals which could be trapped with moderate efficiency by halogen donors or alkenes.
Loss of methyl to afford an alkyl benzoate was an important side reaction at higher temperatures. From
the thermal reaction of hex-5-enyl 3-methylcyclohexa-1,4-diene-3-carboxylate the rate constant for
hydrogen abstraction from the ester by hexenyl radicals was determined to be 0.82 × 105 dm3 molϪ1 sϪ1 at
140 ЊC.
CO2H
CO2R
Introduction
The steady increase over the last few years in the use of free
iv
radical methodology in organic synthetic procedures has
emphasised the need for new types of free radical precursors
which will selectively produce a specific radical at low temper-
atures. Organotin and organosilicon reagents continue to be
popular, as do esters of thiohydroxamic acids (1) developed by
Barton and co-workers.1,2 Organocobalt complexes (2) react by
–
CO2H
CO2
ii
5
i
iii
Me
CO2H
Me
CO2R
iv
6
R
N
O
S
a
b
c
C6H13
R
Colll(dmgH)2py
CO2R
CO2R
H2C CH(CH2)4
R
i, ii, iv
O
7
Cyclohexyl
O
CO2H
d Me3C
1
2
O
e
f
3-Cholesteryl
i, iii, iv
PhCH2
O
S
BH2
N
O
Me
H2C CHCH2
g
R
8
O
S
OEt
h Ph2CH
3
4
Scheme 1 i; Li/NH3, ii; NH4Cl, iii; Mel, iv; COCl2/Et2O followed by
ROH/py
homolytic substitution at carbon followed by loss of a cobalox-
ime radical3 and they are effective precursors for constructing
three- and five-membered rings.4 S-Alkoxycarbonyl dithio-
carbonates (3) have recently been exploited, particularly for the
formation of lactones.5 Chiral amine-boranes, e.g. 4, have been
developed as polarity-reversal catalysts for enantioselective
hydrogen abstraction reactions.6 The ideal radical precursor
should smoothly afford a single radical under mild experi-
mental conditions, act as a hydrogen or halogen donor so it can
be easily incorporated into chain sequences, and should not
produce toxic or pungent-smelling by-products. Clearly, none
of the available precursors fulfil all these criteria and hence the
need for improved reagents. We have investigated the induced
decomposition of esters of cyclohexa-1,4-diene-3-carboxylic
acids (5, 6) and esters of 2,5-dihydrofuran-2-carboxylic acids (7,
8), examined the intermediate radicals by EPR spectroscopy
and studied inter- and intra-molecular addition reactions of
carbon-centred radicals generated therefrom. Part of this
research was published as a preliminary communication.7
experimental procedures for the partial reduction of aromatic
carboxylic acids have been reported, employing different alkali
metals, amines and different hydrogen donors.8–10 We found,
however, that best results were obtained by adding just suf-
ficient lithium metal to a solution of benzoic or furoic acid in
liquid ammonia to produce a permanent blue coloration.11–13
Quenching with ammonium chloride gave good yields of 1,4-
dihydrobenzoic acid, or 2,5-dihydrofuroic acid whereas add-
ition of excess iodomethane afforded the 1-methyl-substituted
acids in yields >90%. Conversion to the esters was straight-
forward, except for 6d which was obtained by reaction of the
corresponding acid chloride with lithium tert-butoxide.
Esters 5 to 8 contain allylic or bisallylic hydrogens which will
be readily abstracted by initiator radicals to produce delocalised
cyclohexadienyl radicals 10 or dihydrofuranyl radicals. For-
mation of the aromatic ring in 12 was expected to provide the
driving force for fragmentation of 10 with initial production of
an alkoxycarbonyl radical 11, which would undergo β-scission
ؒ
to give an alkyl radical R together with CO2. The alkyl radicals
could take part in intra- or inter-molecular reactions (e.g. with
Results and discussion
Ester series were synthesised as outlined in Scheme 1. Various
substrate X) before continuing the chain by reacting with
J. Chem. Soc., Perkin Trans. 2, 1997
757

View Article Online
Table 1 EPR parameters for dihydrofuranyl radicalsa
H3
H2
H3
H4
H4
H5
H5′
•
O
•
O
CO2R
OR
H5
O
17
18d
H
H
H
H
H
•
O
•
O
CO2R
CH3
O
H
H
OR
19
18′d
Radical
R
T/K
a(H2)
a(H3)
a(H4)
a(H5)
a(H5Ј)
a(other)
17c
18c
c-C6H11
c-C6H11
But
240
240
220
220
220
390
240
220
220
20.9
—
34.8
—
—
—
34.7
—
—
13.5
2.2
13.5
2.1
2.1
2.3
13.6
13.4
13.4
2.0
10.0
2.1
9.9
9.8
9.8
2.0
1.95
2.0
13.5
32.1
13.5
32.1
31.4
31.3
13.6
13.4
13.4
—
32.1
—
32.1
31.4
31.3
—
17d
18db
18db
18dc
17g
19c
But
But
But
Allyl
c-C6H11
Benzyl
—
—
0.27(3 H)
19h
a
All g-factors = 2.003 ± 0.001, hfs/G (1 mT = 10 G) were checked by computer simulations. b Two conformers. c Average spectrum.
another molecule of ester 9 (Scheme 2). Thus, the products of
•
ButO
•
•
+
the chain would be the adduct, RXH, and aromatic com-
CO2R
CO2R
CO2R
ؒ
O
O
O
pound 12 or, for a suitably unsaturated R , cyclisation could
15
16
PhR1
12
to furan. Spectra assigned to radicals of type 18 showed evi-
dence of conformational plurality. Only for 18d (R = But) were
the signals sufficiently strong and well defined for definitive
analysis. In that case, two conformers with slightly different
hyperfine splittings (hfs) were distinghished at 220 K (Table 1);
line broadening occurred in the range 300–390 K, above which
a single average spectrum was observed. syn- and anti-
Conformers about the ring-carbon to carbonyl–carbon bond
have been detected by NMR spectroscopy for several furan car-
baldehydes16 and thiophene carbaldehydes.17 Similarly, acyl
radicals generated by abstraction of the formyl hydrogens from
these aldehydes show two conformations in their EPR spectra.18
It is probable therefore, that the pair of spectra obtained for 18d
represent syn- and anti-conformers, i.e. 18d and 18Јd, although
it was not possible to assign individual spectra to a particular
conformer. A rough estimate of the barrier to interconversion
of the two conformers, which were nearly equal in concen-
tration, could be made from the coalescence temperature (ca.
350 K) and the low temerature difference in the hfs of H5
[k/sϪ1 = 6.2 × 106(∆a)],19 i.e. k(350 K) ca. 4 × 106 sϪ1 which,
when combined with a normal pre-exponential factor for bond
rotation of 1013 sϪ1, leads to an activation barrier of ca. 40 kJ
molϪ1. This is close to barriers reported for bond rotation in
aromatic carbaldehydes (46, 32 kJ molϪ1)20,21 and significantly
more than the barriers for related acyl radicals (р16 kJ
molϪ1).18 Not surprisingly, therefore, the barrier in radical 18 is
closest to that of the model species (aldehydes) in which the
atoms of the rotor have the same hybridization. It was evident
that hydrogen abstraction took place unselectively at the two
allylic sites.
•
RO2C
CO2
11
RO2C
R1
In•
RO2C
R1
•
10
13 R
–InH
X
9
•
RX
14
R1 = H, Me
RXH
9
Scheme 2
be induced. A similar sequence was envisaged with chain
propagation by dihydrofuranyl radicals which would fragment
to afford 2-methylfuran and an alkyl radical. The aromatic by-
products from these processes (toluene, 2-methylfuran) would
be comparatively benign and could be easily removed because
of their volatility.
In exploratory experiments, reactions of unmethylated esters
5d, 5f, 5g, 7d and 7f with N-bromosuccinimide (NBS) and
bromotrichloromethane were performed in an attempt to trap
the released alkyl radicals as bromides, RBr. Product analysis
showed that benzene or furan was formed, in agreement with
the expected decarboxylation, but yields of ButBr and
PhCH2Br were low (р12%). To probe the reaction in greater
depth, the paramagnetic intermediates were examined by EPR
spectroscopy.
EPR study of intermediates from esters 5–8
Solutions of individual esters and di-tert-butyl peroxide in tert-
butylbenzene were photolysed with UV light in the microwave
cavity of an EPR spectrometer at various temperatures. The
unmethylated, dihydrofuranyl esters 7c,d and g all showed the
presence of the expected dihydrofuranyl radical 15 together
with a second radical identified as 16. The EPR parameters
(Table 1) were similar to those previously reported 14,15 for di-
hydrofuranyl radicals generated by addition of various radicals
EPR spectra obtained on hydrogen abstraction from the
unmethylated ester 5d showed that hydrogen abstraction also
took place unselectively at the two bisallylic sites and both types
of cyclohexadienyl radical were characterised (Table 2). In the
3-methyl series a single cyclohexadienyl radical, with param-
eters similar to those of related radicals reported in the liter-
ature,22 was observed in each case (Table 2). As expected, the hfs
758
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Table 2 EPR parameters of substituted cyclohexadienyl radicalsa
R3 CO2R
R3 CO2R
CO2R
H2
H1
H
H2
H1
H
H2
H1
H
H
ButO•
•
+
•
H
H
H6
H6
H′6
H′6
H6
20
21
22 (R3 = H)
Radical
R
T/K
a(2 H1) a(2 H2) a(H6)
a(HЈ6)
22d
21d
21a
21b
21d
21e
21f
But
But
Hexyl
c-C6H11
But
Cholesteryl
PhCH2
Allyl
210
210
220
215
220
240
220
220
220
8.30
2.70
2.70
2.75
2.70
2.75
2.70
2.65
2.70
2.30
9.20
9.20
9.30
9.25
9.25
9.20
9.20
9.25
45.10
13.70
13.20
13.40
13.40
13.30
13.25
13.35
13.40
46.00
21g
21h
Ph2CH
a
All g-factors 2.003 ± 0.001, hfs/G were checked by computer simulations.
Table 3 Product yieldsa from the reaction of cyclohexa-1,4-diene-3-
Me CO2R
Me CO2R
carboxylates and 2,5-dihydrofuran-2-carboxylates with NBSb
XBr
RBr
–CO2
PhMe + •CO2R
PhCO2R + Me•
NBS–CCI4
(PhCO2)2
Conversion
PhMe or
•
Ester
R
of 6 or 8 (%) 2-MF c
RBr
PhCO2R
6
6a d
6c
6d
6f
C6H13
C6H11
But
PhCH2
PhCH2
<50
7
20
47
nd
nd
3d
4
15
5
Scheme 3
~75
~60
~80
~90
20e
40
70 (63)f
63
10
esters indicated that loss of a methyl radical from the inter-
mediate cyclohexadienyl radical 10 competed to a minor extent
with the main fragmentation via an alkoxycarbonyl radical.
Experiments with cyclohexadienyl carboxylic acids containing
branched 3-substituents (9, R = H, R1 = Pri or But) demon-
strated that alkyl loss became the predominant process with
these substituent types.23
8f
nd g
a
mol% by GLC; nd = not determined. b Volatile hydrocarbon (RH)
yields nd because of losses during reflux. c 2-Methylfuran. d ca. 10%
dibromides observed. e ca. 7% dibromides observed. f Isolated yield.
g PhCO2H and PhCHO also formed.
The low product yields from primary and secondary esters
were due partly to low reactant conversion and partly to
polybromination. Under some reaction conditions alkoxycar-
bonyl radicals 11 can undergo a competing α-scission with for-
mation of an alkoxyl radical and carbon monoxide,24,25 eqn. (1).
of the ring hydrogens showed little variation as the ester group
changed. Spectra weakened significantly when the temperature
of the cavity was raised, but were otherwise unchanged. For
most of the esters, spectra were too weak for detection above
ca. 350 K, but in no case was the spectrum of an alkyl radical
ؒ
R , from decarboxylative fragmentation of 21, observed. Even
ؒ
ؒ
ؒ
R ϩ CO2 ← RO2C → RO ϩ CO
(1)
strongly stabilised radicals (R = allyl, PhCH2, Ph2CH) were not
detected, although the spectra of the parent cyclohexadienyl
radicals appeared to weaken more quickly as the temperature
was raised in these cases.
The EPR observations established that abstraction of hydro-
gen alpha to the carboxylate function (H2 in 7 and H3 in 5) took
place readily to give intermediate radicals which were incapable
of fragmentation with loss of CO2. This would drastically
reduce the yield of the desired products and hence, in sub-
sequent research, esters 6 (and 8) with C(3) [and C(2)] blocked
by a methyl group, were employed.
11
Careful search was made for alcohols derived from the alk-
oxyl radicals but none were detected. A number of unidentified
products were present in the chromatograms obtained from
reaction of the primary ester, 6a, and therefore α-scission can-
not be excluded in this case, but it is probably minor, under the
present reaction conditions, for alkoxycarbonyl radicals derived
from the other esters. To see if decarboxylation could be
improved at higher temperatures, reaction of the cyclohexyl
ester, 6c, was carried out at 100 ЊC with bromoform as the halo-
gen donor. Only low conversion was achieved, yields of toluene
and bromocyclohexane were low and cyclohexyl benzoate was
more important than in the NBS reaction (Table 4). Reactions
were also carried out with tert-butyl hypochlorite and N-
bromobis(trimethylsilyl)amine26 (Table 4). With both reagents
good conversion could be achieved, but yields of the alkyl
halides were not significantly greater than in the NBS reactions,
cyclohexyl benzoate was a major by-product and problems with
polyhalogenation were also encountered.
Reactions of 6 and 8 with halogen sources and with alkenes
Individual esters were refluxed in tetrachloromethane with NBS
and a catalytic amount of lauroyl peroxide as initiator. It
proved difficult to obtain complete consumption of the reactant
under these conditions, particularly for esters of primary and
secondary alcohols but, by adding a second equivalent of NBS,
together with additional peroxide after ca. 10 h, and continuing
the reflux, satisfactory conversions were achieved for secondary,
tertiary and benzylic esters. Product analyses showed toluene
(or 2-methylfuran) and alkyl bromide (RBr) but these were
accompanied by small amounts of the corresponding aromatic
ester (PhCO2R) (Scheme 3). Product yields, which are given in
Table 3, showed that the ester chain decompositions, as out-
lined in Scheme 2, were efficient for benzylic and tertiary esters,
but poor for primary esters. The formation of the aromatic
The usefulness of esters 6 as radical sources for inter-
molecular additions was explored by means of thermally initi-
ated reactions with acrylonitrile. The ester and alkene with di-
tert-butyl peroxide in tert-butylbenzene solvent were degassed
and sealed into a Pyrex tube which was heated at 140 ЊC for ca.
20 h. Product analyses (Table 5) revealed toluene, the expected
J. Chem. Soc., Perkin Trans. 2, 1997
759

View Article Online
Table
4
Reactions of cyclohexyl 3-methylcyclohexa-1,4-diene-3-
In conclusion, hydrogen is readily abstracted from esters of
carboxylate (6c) with various halogen donorsa
type 6 (or 8) to give cyclohexadienyl (10) (or dihydrofuranyl)
radicals. For benzyl, tertiary alkyl and secondary alkyl sub-
stituents, radicals 10 mainly fragment to give alkyl radicals,
although loss of methyl becomes increasingly important as the
temperature is raised above ca. 80 ЊC, and accounts for ca. 50%
of the decomposition by 140 ЊC. The released alkyl radicals can
be trapped by halogen donors to give alkyl halides in moderate
yields. The esters also take part in chain addition reactions with
alkenes to afford moderate yields of adducts. Hydrogen transfer
from 6 to carbon-centred radicals is much slower than from
tributyltin hydride and therefore, in olefin alkylations, and
cyclisations where direct reduction is a problem with tin
reagents, the ester method will be a useful alternative.
Halogen
donor
Reaction
conditions
Conversion
(%)
PhMe RX b PhCO2R b
CHBr3
100 ЊC, ButPh, ca. 50
AIBN, 60 h
80 ЊC, CCl4,
hν, 5 h
20
10
25
ButOCl
ca. 50
ca. 75
20
12c 30
20e 25
(Me3Si)2NBr 80 ЊC, CCl4,
10d
AIBN, 2 h
a
Yields in mol% by GLC. b R = cyclohexyl. c Polychlorocyclohexanes
also formed. d Bromotoluene isomers formed. e Polybromocyclo-
hexanes also formed.
Table 5 Product yieldsa from reaction of esters 6 with acrylonitrile (X)
at 140 ЊC in ButPh b
Experimental
1H NMR spectra were recorded at 200 or 300 MHz and 13C
NMR spectra at 75 MHz, in CDCl3 solution with tetramethylsi-
lane (δH = δC = 0) as reference. Coupling constants are expressed
in Hz. Ether refers to diethyl ether. Light petroleum refers to the
fraction boiling in the range 40–60 ЊC. Mass spectra were
obtained with 70 eV electron impact ionisation on a Kratos
M25RF spectrometer. GLC–MS analyses were run on a Finni-
gan Incos 50 quadrupole instrument coupled to a Hewlett
Packard HP 5890 chromatograph fitted with a 25 m HP 17
capillary column (50% phenyl methyl silicone). For the calcul-
ation of yields from GLC data, the detector response was cali-
brated with known amounts of authentic materials (or close
analogues) and dodecane, or heptane was added as a standard.
Ester
R
PhMe RXH RX2H RX3H RH
PhCO2R
6a
6c
6d
6e
6f
C6H13
c-C6H11
Me3C
Ch c
PhCH2
Allyl
48
46
49
42
57
24
18
22
40
—
26
3
11
13
17
—
—
3
7
4
4
—
—
—
30
28
31
—
d
48
53
41
—
51
20
6g
e
a
Yields in mol% by GLC. b Conversion of 6 was ca. 50% in each case.
c Ch = cholesteryl; no Ch containing products detected. d RH = PhMe in
this case. e Too volatile for accurate analysis; 5% of hexa-1,5-diene
observed.
adducts (RXH) together with oligomers, the direct reduction
products (RH) and aromatic esters (PhCO2R). The sizeable
amounts of RH formed showed that the alkyl radicals R
EPR spectra were obtained with a Bruker ER 200D
spectrometer operating at 9.1 GHz with 100 kHz modulation.
Samples of the substrate (ca. 40 mg) and di-tert-butyl peroxide
(30 µl) in tert-butylbenzene (0.5 cm3) (occasionally cyclo-
propane) were degassed by bubbling nitrogen for 20 min and
photolysed in the resonant cavity by light from a 500 W super
pressure mercury arc lamp. EPR spectra were simulated with a
program due originally to Heinzer.32
ؒ
abstracted hydrogen from ester 6 in competition with addition
to acrylonitrile. The yields of adducts increased for branched
alkyl radicals but, as might be expected, were lower for the
delocalised benzyl and allyl radicals. The aromatic esters
PhCO2R were formed in very significant yields showing that
methyl radical loss from 10 competes more effectively with
alkoxylcarbonyl loss at the higher temperature of the acrylo-
nitrile experiments.
Materials
Cyclohexa-1,4-diene-3-carboxylic acid 9 and 2,5-dihydrofuran-
2-carboxylic acid 12 were made according to the literature pro-
cedures and converted to esters 5 and 7 by standard procedures
via the acid chlorides. tert-Butylcyclohexa-1,4-diene-3-carboxy-
late 5d; bp 200 ЊC/1 Torr (Kugelrohr); δH 1.47 (9 H, s), 2.69 (2
H, m), 3.62 (1 H, m), 5.7–6.0 (4 H, m); δC 25.9, 28.0, 42.7, 80.8,
122.7, 126.0, 171.9. Allyl cyclohexa-1,4-diene-3-carboxylate 5g;
δH 2.71 (2 H, m), 3.78 (1 H, m), 4.62 (2 H, br d, J 7), 5.26 (1 H, d,
J 12), 5.36 (1 H, d, J 17), 5.7–6.0 (4 H, m). Cyclohexyl 2,5-
dihydrofuran-2-carboxylate 7c bp 130 ЊC/1 Torr, δH 1.1–2.0 (11
H, m), 4.7–5.0 (2 H, m), 5.2 (1 H, m), 5.8–6.1 (2 H, m); δC 23.9,
25.7, 27.3, 31.8, 31.9, 73.7, 77.0, 84.9, 125.3, 129.4, 171.0; m/z
The ready formation of oligomers, and the fact that com-
paratively large quantities of initiators were needed to achieve
complete conversion of the esters, seemed to imply that hydro-
gen abstraction from 6 was a relatively slow process. To assess
this aspect of the mechanism, the hexenyl ester 6b (0.91 mol
dmϪ3) was decomposed at 140 ЊC in tert-butylbenzene. The
observed products, i.e. methylcyclopentane, hex-1-ene, cyclo-
hexane, toluene and hex-5-enyl benzoate, were consistent with
the mechanism of Scheme 2. The system was therefore used as a
free radical clock to determine the rate constant of hydrogen
abstraction from 6b (kH ) by primary hexenyl radicals from the
ratio of methylcyclopentane to hex-1-ene and the known con-
centration of 6b {kH = kc[hex-1-ene]/[methylcyclopentane][6b]}.
From an experiment in which the conversion of 6b was limited
to ca. 10% the measured ratio of hex-1-ene to methylcyclo-
pentane was 0.010 hence, kH /kc(140 ЊC) = 1.3 × 10Ϫ2 dm3 molϪ1.
Using the known rate constant for hex-5-enyl radical cyclisation
(6.2 × 106 sϪ1 at 140 ЊC),27–29 we obtain kH = 0.82 × 105 dm3
molϪ1 sϪ1 at 140 ЊC. Thus, kH for 6b lies between values reported
for hydrogen abstraction from cyclohexa-1,4-diene by ethyl30
and hex-5-enyl31 radicals, viz. 0.58 × 105 (at 27 ЊC) and
2.3 × 105 (at 50 ЊC), respectively. In view of the differences in
the temperatures at which the rates were measured, hydrogen
abstraction from 6b is a little slower than from cyclohexa-1,4-
diene. However, most of the difference is probably statistical
because 6b has only half as many available hydrogens as
cyclohexa-1,4-diene. Significantly, esters 6 transfer hydrogen to
alkyl radicals ca. 150-fold slower than does tributyltin hydride
at 140 ЊC.
ϩ
ؒ
196 (M
, 2), 55 (100). tert-Butyl 2,5-dihydrofuran-2-
carboxylate 7d δH 1.48 (9 H, s), 4.7–4.9 (2 H, m), 5.4 (1 H, m),
5.8–6.1 (2 H, m); δC 28.1, 77.0, 84.9, 125.2, 128.9, 170.5 (quat. C
not obs.). Allyl 2,5-dihydrofuran-2-carboxylate 7g δH 4.66 (2 H,
m), 4.7–5.0 (2 H, m), 5.2–5.4 (3 H, m), 5.91 (2 H, m), 6.10 (1 H,
m); δC 65.6, 76.7, 84.4, 118.6, 124.7, 129.4, 131.7, 170.9.
3-Methylcyclohexa-1,4-diene-3-carboxylic acid
Benzoic acid (15 g, 123 mmol) was dissolved in liquid ammonia
(900 cm3) and lithium wire (2.2 g, 314 mmol) was added in small
portions until a permanent deep-blue colour appeared. An
excess of methyl iodide (25 cm3, 400 mmol) was then added
dropwise, leading to discharge of the colour and formation of a
white solid. After evaporation of the ammonia, ice–water (100
cm3) was added and the suspension was acidified with 50%
H2SO4. The mixture was extracted with ether (4 × 100 cm3), the
combined ether layers were washed with saturated sodium thio-
sulfate solution (100 cm3), water (100 cm3) and dried (MgSO4).
760
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Distillation at 85 ЊC/0.4 Torr (Kugelrohr) gave the acid (17.1 g,
95%) as an oily white solid (mp 36 ЊC); δH 1.4 (3 H, s), 2.6 (2 H,
m), 5.8 (4 H, m), 12.3 (1 H, br s); δC 26.6 (CH3), 27.7 (CH2),
44.1 (C), 125.4 (CH), 128.5 (CH), 182.6 (CO2); m/z (%) 138
11.86 (CH3), 18.72 (CH3), 19.37 (CH3), 21.04 (CH2), 22.57
(CH3), 22.83 (CH3), 23.84 (CH2), 24.29 (CH2), 25.94 (CH2),
27.47 (CH3), 27.61 (CH2), 28.02 (CH), 28.24 (CH2), 31.86
(CH2), 31.92 (CH2), 35.80 (CH), 36.18 (CH2), 36.60 (C), 36.97
(CH2), 37.94 (CH2), 39.52 (CH2), 39.73 (CH2), 42.31 (C), 43.80
(C), 50.00 (CH), 56.13 (CH), 56.69 (CH), 74.19 (CH), 122.57
(CH2), 124.22 (CH), 128.90 (CH), 139.69 (CH), 174.61 (CO).
tert-Butyl 3-methylcyclohexa-1,4-diene-3-carboxylate 6d. To
tert-butyl alcohol (1.41 g, 19 mmol) in dry THF (40 cm3), under
nitrogen, was slowly added butyllithium (15 cm3 of a 1.6 
solution in hexane). After 30 min a solution of 3-methyl-
cyclohexa-1,4-diene-3-carbonyl chloride (3.0 g, 19 mol) in THF
(40 cm3) was added dropwise. The mixture was refluxed for 1 h,
cooled and water (100 cm3) was added. The aqueous phase was
extracted with ether (4 ×100 cm3), the combined organic layers
were dried (MgSO4) and distilled to give 6d as a colourless oil
(79%); bp 82 ЊC/0.8 Torr (Kugelrohr); δH 1.25 (3 H, s), 1.4 (9 H,
s), 2.6 (2 H, br s), 5.75 (4 H, m); δC 21.6 (CH3), 23.1 (CH2), 23.6
(CH3), 40.1 (C), 76.1 (C), 119.7 (CH), 124.9 (CH), 170.0 (CO);
m/z (%) 138 (1), 93 (63), 91 (28), 77 (26), 57 (100), 41 (33), 39
(15).
(M ϩ, 2), 123 (1), 93 (100), 91 (80), 77 (90), 65 (26), 51 (31), 45
ؒ
(16), 39 (49), 27 (23).
Typical procedure for preparation of esters
Benzyl 3-methylcyclohexa-1,4-diene-3-carboxylate 6f. To a
stirred solution of 3-methylcyclohexa-1,4-diene-3-carboxylic
acid (4 g, 29 mmol) in dry ether (30 cm3) was added dropwise a
solution of oxalyl chloride (4.42 g, 35 mmol) in dry ether (25
cm3). The solution was stirred at room temperature for 18 h and
then refluxed for 2 h. The product was distilled at 62 ЊC/0.2 Torr
(Kugelrohr) to yield the acid chloride (3.9 g, 86%). The acid
chloride (1.2 g, 7 mmol) in dry ether (25 cm3) was added drop-
wise to equimolar amounts of pyridine (0.56 g, 7 mmol) and
benzyl alcohol (0.79 g, 7 mmol) in dry ether (25 cm3). The
solution was stirred at room temp. for 2 h, filtered, washed with
HCl (50 cm3 of 0.25 ), water (50 cm3), dried (MgSO4) and
distilled to yield 6f as a colourless liquid (1.5 g, 90%); bp 108 ЊC/
0.1 Torr (Found: C, 78.97; H, 7.37. C15H16O2 requires C, 78.92;
H, 7.06%); δH 1.4 (3 H, s), 2.6 (2 H, br s), 5.1 (2 H, s), 5.8 (4 H,
m), 7.3 (5 H, m); δC 26.4 (CH2), 27.9 (CH3), 44.5 (C), 66.9
(OCH2), 125.1 (CH), 128.2 (CH), 128.5 (CH), 129.0 (CH),
129.1 (CH), 136.7 (C), 175.4 (CO); m/z (%) 93 (89), 91 (100), 77
(78), 65 (50), 51 (29), 41 (20), 39 (49); CI MS, found: MHϩ
229.1233, C15H17O2 requires 229.1229.
2-Methyl-2,5-dihydrofuran-2-carboxylic acid.9 The acid was
prepared from furoic acid by a procedure similar to that above
for 3-methylcyclohexa-1,4-diene-3-carboxylic acid. The prod-
uct decomposed on distillation and so was purified by chrom-
atography on SiO2 eluting with cyclohexane–ethyl acetate
(90:10); δH 1.6 (3 H, s), 4.8 (2 H, m), 5.8–6.1 (2 H, m), 8.9 (1 H,
brs).
Hexyl 3-methylcyclohexa-1,4-diene-3-carboxylate 6a. A pro-
cedure similar to the above gave 6a (76%); bp 76 ЊC/0.1 Torr
(Kugelrohr): δH 0.85 (3 H, t, J 7), 1.3 (9 H, m), 1.6 (2 H, m), 2.6
(2 H, br s), 4.05 (2 H, t), 5.8 (4 H, m); δC 14.4 (CH3), 22.9 (CH2),
25.9 (CH2), 26.3 (CH3), 27.7 (CH2), 28.9 (CH2), 31.7 (CH2),
44.2 (C), 65.3 (CH2), 124.5 (CH), 129.2 (CH), 175.5 (CO); m/z
(%) 220 (2), 137 (16), 136 (40), 119 (50), 118 (79), 91 (70), 65
(44), 43 (100), 41 (65), 39 (40); CI MS, found: MHϩ, 223.1701,
C14H23O2 requires 223.1698.
Hex-5-enyl 3-methylcyclohexa-1,4-diene-3-carboxylate 6b. A
similar procedure afforded 6b (60%); bp 82 ЊC/0.1 Torr (Kugel-
rohr); δH 1.33 (3 H, s), 1.45 (2 H, m), 1.65 (2 H, m), 2.07 (2 H,
m), 2.65 (2 H, br s), 4.08, (2 H, t, J 7), 4.95 (1 H, d, J 10), 5.00 (1
H, ddt, Jtrans 17, Jgem 1.6, Jallylic 1.7), 5.78 (4 H, m), 5.80 (1 H, m);
δC 25.2 (CH2), 25.9 (CH2), 27.4 (CH3), 28.0, (CH2), 33.24 (CH2),
43.9 (C), 64.7 (OCH2), 114.8 (CH2), 124.3 (CH), 128.8 (CH),
138.312 (CH), 175.1 (CO); CI MS, found: MHϩ, 221.1533,
C14H21O2 requires 221.1542.
Benzyl 2-methyl-2,5-dihydrofuran-2-carboxylate 8f. 2-Methyl-
2,5-dihydrofuran-2-carboxylic acid (1.7 g, 13 mmol) in dry
ether (15 cm3) was added dropwise to a solution of thionyl
chloride (4.5 cm3) in dry ether (10 cm3). The solution was stirred
for 30 min, the excess thionyl chloride and the ether were
removed on a rotary evaporator and then the crude acid chlor-
ide, in ether (15 cm3), was added to benzyl alcohol (0.75 g, 7
mmol) and pyridine (0.45 g, 6 mmol) in ether (10 cm3). The
solution was stirred for 10 min, acidified with dil. HCl, and the
aqueous layer was extracted with ether (2 × 25 cm3). The com-
bined ether layers were dried (MgSO4) and distilled to give 8f
(bp 165 ЊC/1 Torr, Kugelrohr) which was further purified by
chromatography on SiO2 eluting with cyclohexane–ethyl acetate
(95:5); yield 39%; δH 1.58, (3 H, s), 4.78 (2 H, m), 5.17 (2 H, s),
5.8–6.1 (2 H, m), 7.35 (5 H, m); δC 24.7, 67.1, 76.4, 90.7, 128.4,
128.7, 128.8, 128.9, 129.0, 129.2, 130.4, 136.4, 173.7; m/z (%)
218 (M ϩ, 1), 108 9), 91 (87), 84 (62), 83 (100), 65 (30), 55 (68),
ؒ
43 (29), 39 (32).
Cyclohexyl 3-methylcyclohexa-1,4-diene-3-carboxylate 6c. A
similar procedure yielded 6c (61%); bp 98 ЊC/0.8 Torr (Kugel-
rohr); δH 1.32 (3 H, s), 1.42 (6 H, m), 1.74 (4 H, m), 2.63 (2 H, s),
4.76 (1 H, m), 5.78 (4 H, m); δC 23.4 (CH2), 25.5 (CH2), 26.0
(CH2), 27.4 (CH3), 31.1 (CH2), 43.9 (C), 72.4 (OCH), 124.2
(CH), 129.0 (CH), 174.5 (CO); m/z (%) 138 (3), 93 (100), 92
(32), 91 (41), 83 (60), 77 (33), 55 (78), 41 (42), 39 (21); CI MS,
found: MHϩ 221.1548, C14H21O2 requires 221.1542.
Cyclohexyl 2-methyl-2,5-dihydrofuran-2-carboxylate 8c. A
similar procedure to that for 8f yielded 8c (31%); bp 120 ЊC/1
Torr; δH 1.1–2.0 (11 H, m), 1.52 (3 H, s), 4.75 (2 H, m), 5.8–6.0
(2 H, m); δC 23.8, 24.0, 24.4, 25.7, 31.6, 31.8, 73.2, 76.1, 90.4,
128.5, 130.5, 172.8; m/z (%) 210 (M ϩ, 2), 115 (13), 99 (29), 98
ؒ
(21), 84 (26), 83 (100), 82 (24), 55 (79), 43 (33), 41 (53).
Reaction of 6f with NBS
Diphenylmethyl 3-methylcyclohexa-1,4-diene-3-carboxylate
6h. A similar procedure yielded 6h as white needles (27%); mp
99–100 ЊC (from light petroleum); δH 1.4 (3 H, s), 2.7 (2 H, br s),
5.85 (4 H, m), 6.8 (1 H, s), 7.3 (10 H, m); δC 21.6 (CH3), 22.7
(CH2), 39.9 (CH), 120.3 (CH), 122.5 (CH), 123.4 (CH), 124.1
(CH), 136.2 (C), 169.0 (CO) (Cphenyl obscured by overlap); m/z
(%) 167 (65), 165 (38), 152 (23), 139 (10), 115 (13), 89 (16), 83
(100), 82 (70), 76 (20), 70 (26), 63 (51), 51 (47), 50 (31), 39 (53).
Cholesteryl 3-methylcyclohexa-1,4-diene-3-carboxylate 6e.
The procedure was similar except that THF was used as solvent
and only 0.8 equiv. of cholesten-3β-ol was employed. Ester 6e
was purified by chromatography on neutral alumina the column
being developed with cyclohexane–ethyl acetate (93:7); yield
55%; mp 147–151 ЊC (from light petroleum); δH 0.7 (3 H, s),
0.85 (6 H, s), 0.9 (3 H, s), 1.05 (3 H, s), 1.3 (3 H, s), 1.0–2.1 (28
H, br m), 2.65 (2 H, s), 4.6 (1 H, m), 5.35 (1 H, d), 5.8 (4 H, s); δC
Benzyl 3-methylcyclohexa-1,4-diene-3-carboxylate (1.3 g, 5.6
mmol) was added dropwise, over 2 h, to a refluxing solution of
NBS (1.0 g, 5.6 mmol) and lauroyl peroxide (0.02 g) in CCl4.
The suspension was refluxed for 4 h and then NBS (1.0 g) and
lauroyl peroxide (0.02 g) were added. After refluxing for 5 h the
mixture was filtered and distilled to afford benzyl bromide (0.61
1
g, 63%); bp 81 ЊC/15 Torr; the H and 13C NMR spectra were
identical to those of the authentic material. GC–MS analysis
prior to distillation disclosed the presence of toluene and
benzyl benzoate (Table 3).
Reaction of 6a,c,d and 8f with NBS
For each ester the method was similar to that described for 6f,
except that initial reflux period was 10 h and the final mixtures
were analysed by GLC and GC–MS; products and yields are in
Table 3.
J. Chem. Soc., Perkin Trans. 2, 1997
761

View Article Online
4 M. R. Ashcroft, A. Bury, C. J. Cooksey, A. G. Davies,
Reaction of 6c with bromoform
B. D. Gupta, M. D. Johnson and H. Morris, J. Organomet. Chem.,
1980, 195, 89; M. D. Johnson, G. M. Lampman, R. W. Koops and
B. D. Gupta, J. Organomet. Chem., 1987, 326, 281.
5 J. E. Forbes and S. Z. Zard, J. Am. Chem. Soc., 1990, 112, 2034;
F. Mestre, C. Tailhan and S. Z. Zard, Heterocycles, 1989, 28, 171;
J. E. Forbes and S. Z. Zard, Tetrahedron Lett., 1989, 30, 4367.
6 P. L. H. Mok and B. P. Roberts, J. Chem. Soc., Chem. Commun.,
1991, 150; P. L. H. Mok, B. P. Roberts and P. T. McKetty, J. Chem.
Soc., Perkin Trans 2, 1993, 665; H.-S. Dang, V. Diart and
B. P. Roberts, J. Chem. Soc., Perkin Trans. 1, 1994, 2511.
7 G. Binmore, J. C. Walton and L. Cardellini, J. Chem. Soc.,
Chem. Commun., 1995, 27.
8 J. Birch and H. Smith, Quart. Rev., 1958, 12, 17.
9 M. E. Kuehne and B. F. Lambert, Org. Synth., 1963, 43, 22.
10 E. M. Kaiser, Synthesis, 1972, 391.
11 H. Van Bekkum, C. B. Van Den Bosch, G. Van-Minnen-
Pathius, J. C. Mos and A. M. Van Wijk, Recl. Trav. Chim. Pays-Bas,
1971, 90, 137.
12 A. J. Birch and J. Slobbe, Tetrahedron Lett., 1975, 9, 627.
13 K. Kinoshita, K. Miyano and T. Miwa, Bull. Chem. Soc. Jpn.,
1975, 48, 1865.
14 L. Lunazzi, G. Placucci and L. Grossi, J. Chem. Soc., Perkin
Trans. 2, 1982, 875.
15 R. H. Schuler, G. P. Laroff and R. W. Fessenden, J. Chem. Phys.,
1973, 77, 456.
16 B. P. Roques, S. Combrisson and F. Wehrli, Tetrahedron Lett.,
1975, 1047; D. J. Chadwick, J. Chem. Soc., Perkin Trans. 2, 1976,
451.
17 L. Lunazzi, G. Placucci, C. Chatgilialoglu and D. Macciantelli,
J. Chem. Soc., Perkin Trans. 2, 1983, 819.
18 C. Chatgilialoglu, L. Lunazzi, D. Macciantelli and G. Placucci,
J. Am. Chem. Soc., 1984, 106, 5252.
19 G. A. Russell, G. R. Underwood and D. C. Lini, J. Am. Chem.
Soc., 1967, 89, 6636.
20 K. I. Dahlqvist and S. Forsèn, J. Phys. Chem., 1965, 69, 4062.
21 L. Lunazzi, D. Macciantelli and A. C. Boicelli, Tetrahedron
Lett., 1975, 1205.
22 A. Berndt in Landolt-Bornstein; Magnetic Properties of Free
Radicals, eds. H. Fischer and K.-H. Hellwege, Springer, Berlin, 1977,
vol. 9b, p. 452; 1987, vol. 17c, p. 88.
23 P. A. Baguley, G. Binmore, A. Milne and J. C. Walton, J. Chem.
Soc., Chem. Commun., 1996, 2199.
24 J. Pfenninger, C. Heuberger and W. Graf, Helv. Chim. Acta,
1980, 63, 2328.
25 A. L. J. Beckwith and V. W. Bowry, J. Am. Chem. Soc., 1994,
116, 2710.
Ester 6c (0.2 g, 0.91 mmol), bromoform (0.23 g, 0.91 mmol) and
2,2Ј-azo(2-methylpropionitrile) (AIBN) (0.01 g) were dissolved
in tert-butylbenzene (1 cm3), sealed in a glass tube and heated at
100 ЊC for 60 h. Analysis by GC–MS showed toluene, bromo-
cyclohexane, unreacted ester and cyclohexyl benzoate; yields in
Table 4.
Reaction of 6c with tert-butyl hypochlorite
To a refluxing solution of 6c (0.8 g, 3.6 mmol) in CCl4 (40 cm3)
was added dropwise, over 10 min, freshly prepared tert-butyl
hypochlorite (0.55 g, 5 mmol) in CCl4 (20 cm3). The solution
was refluxed for 5 h under illumination from a 100 W tungsten
lamp. The solvent was evaporated and the products were char-
acterised by GC–MS; toluene, chlorocyclohexane, cyclohexyl
benzoate, polychlorocyclohexanes and an unidentified by-
product were observed (see Table 4).
Reaction of 6c with N-bromobis(trimethylsilyl)amine
To ester 6c (0.6 g, 2.7 mmol), norbornylene (0.05 g) and AIBN
(0.05 g) in CCl4 (40 cm3) was added N-bromobis(trimethyl-
silyl)amine26 (0.5 g, 3 mmol) in CCl4 (20 cm3). The solution was
refluxed for 2 h and analysed by GC–MS which showed: tolu-
ene, bromocyclohexane, cyclohexyl benzoate, bromotoluene
isomers and several polybromocyclohexanes (see Table 4).
Reaction of esters 6 with acrylonitrile: typical procedure for 6f
The benzyl ester (0.1 g, 0.44 mmol), acrylonitrile (0.05 cm3),
di-tert-butyl peroxide (0.05 cm3) and heptane (0.01 cm3) were
dissolved in tert-butylbenzene (0.4 cm3) in a Pyrex tube. The
mixture was degassed on a vacuum line by a series of freeze–
pump–thaw cycles, the tube was flame sealed and heated at
140 ЊC in a GLC oven for 20 h. Products were analysed by GLC
and GC–MS and yields are given in Table 5.
Kinetics of the reaction of hexenyl ester 6b
The ester 6b (0.1 g, 0.45 mmol), di-tert-butyl peroxide (0.015
cm3), heptane (0.01 cm3) were dissolved in tert-butylbenzene
(0.4 cm3) and degassed in a Pyrex tube. The tube was flame
sealed and heated at 140 ЊC for 25 h. GLC analysis revealed
hex-1-ene (0.10%), methylcyclopentane (9.95%) and cyclo-
hexane (0.29%) together with unreacted 6b.
26 M. D. Cook, B. P. Roberts and K. Singh, J. Chem. Soc., Perkin
Trans. 2, 1983, 635; B. P. Roberts and C. Wilson, J. Chem. Soc.,
Chem. Commun., 1978, 752; J. C. Brand, B. P. Roberts and
J. N. Winter, J. Chem. Soc., Perkin Trans 2, 1983, 201.
27 A. L. J. Beckwith and G. Moad, J. Chem. Soc., Perkin Trans. 2,
1980, 1083; J. Chem. Soc., Chem. Commun., 1974, 472.
28 C. Chatgilialoglu, K. U. Ingold and J. C. Scaiano, J. Am. Chem.
Soc., 1981, 103, 7739.
Acknowledgements
We thank Dr K. U. Ingold for valuable discussion, the trustees
of the Lumsden Bequest, the EPSRC, the Italian CNR and the
Academia dei Lincei for financial support of this work.
29 A. L. J. Beckwith and C. H. Schiesser, Tetrahedron, 1985, 41,
392.
30 J. A. Hawari, P. S. Engel and D. Griller, Int. J. Chem. Kinet.,
1985, 17, 1215.
31 M. Newcomb and S. U. Park, J. Am. Chem. Soc., 1986, 108,
4132.
32 J. Heinzer, Quantum Chemistry Program Exchange, University
of Indiana, Bloomington, IN, 1972, No. 209.
References
1 D. H. R. Barton, D. Crich and W. B. Motherwell, J. Chem. Soc.,
Chem. Commun., 1983, 939; D. H. R. Barton, D. Crich and
W. B. Motherwell, Tetrahedron, 1985, 41, 3901; D. Crich, Aldrichim.
Acta, 1987, 20, 35.
2 D. H. R. Barton, D. Bridon, I. Fernandez-Picot and S. Z. Zard,
Tetrahedron, 1987, 43, 2733: D. H. R. Barton, D. Crich and
G. Kretzschmar, J. Chem. Soc., Perkin Trans. 1, 1986, 39.
3 J. H. Espenson and T. D. Sellers, Jr., J. Am. Chem. Soc., 1974,
96, 94.
Paper 6/06414K
Received 17th September 1996
Accepted 20th November 1996
762
J. Chem. Soc., Perkin Trans. 2, 1997