Intramolecular site selectivity in cation radical Diels-Alder cycloadditions of difunctional and trifunctional dienophiles

Article abstract of DOI:10.1002/1099-1395(200009)13:9<518::AID-POC288>3.0.CO;2-9

The cation radical Diels-Alder reactions of several difunctional and trifunctional dienophiles with 1,3-cyclopentadiene, catalyzed by tris(4-bromophenyl)aminium hexachloroantimonate, were determined. When both of the reactive sites are of the 4-methoxysty

Full text of DOI:10.1002/1099-1395(200009)13:9<518::AID-POC288>3.0.CO;2-9

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2000; 13: 518–522
Intramolecular site selectivity in cation radical Diels±Alder
cycloadditions of difunctional and trifunctional dienophiles
Nathan L. Bauld* and Jingkui Yang
Department of Chemistry and Biochemistry, University of Texas at Austin, Austin, Texas 78712, USA
Received 1 February 2000; revised 19 April 2000; accepted 21 April 2000
ABSTRACT: The cation radical Diels–Alder reactions of several difunctional and trifunctional dienophiles with 1,3-
cyclopentadiene, catalyzed by tris(4-bromophenyl)aminium hexachloroantimonate, were determined. When both of
the reactive sites are of the 4-methoxystyrene type, reaction occurs smoothly at both styrene moieties to give the bis-
adducts. However, when one reactive moiety is of the styrene type and one is of the enol ether type, reaction occurs
exclusively at the styrene moiety to give a mono-adduct, even in the presence of a large excess of cyclopentadiene.
When three potentially reactive moieties are included, two of them being of the styrene type and one of the 1,2-
diaryloxyethene type, the reaction is again specific for the styrene moieties, giving only the bis-adduct. The high site
selectivity observed in these reactions is especially noteworthy in view of the observation that monofunctional
molecules containing all three of these moieties are found to be reactive toward Diels–Alder adduct formation under
the same reaction conditions and that these adducts are relatively stable under the reaction conditions. Copyright
2000 John Wiley & Sons, Ltd.
KEYWORDS: Cation radical; radical cation; specificity; Diels–Alder; Aminium ion; conjugated systems
INTRODUCTION
RESULTS AND DISCUSSION
Aminium salt-catalyzed cation radical Diels–Alder
cycloadditions (For a review of the cation radical
Diels–Alder reaction and references to the many
contributors to this general area, see Ref. 1) of such
readily ionized substrates as trans-anethole,2 aryl prope-
nyl ethers,3 and 1,2-diaryloxyethenes4 to 1,5-cyclopenta-
diene have all been observed (Scheme 1). However,
relatively little research has been reported on cation
radical Diels–Alder reactions of di- or polyfunctional
substrates. The goal of the present research was to
investigate the feasibility of multiple cation radical
Diels–Alder reactions of a single substrate and more
especially to establish the level of selectivity of such
reactions when different potential dienophilic reaction
sites are present in the same molecule. A further unique
aspect of two of the substrate molecules chosen for this
study is that the two (or three) functional groups
constitute part of a continuous conjugated system, so
that the preferred site of reactivity is not determined by
the ionization step (presumably involving reaction with
the aminium salt), but as an inherent preference of the
extensively conjugated cation radical.
The ionizable functionalities selected for the present
study (Scheme 2) were precisely the ones mentioned
above, the behavior of which in aminium salt-catalyzed
cation radical Diels–Alder reactions is now well
established. The linking conjugated system, in two cases,
is the 1,4-benzenediyl (1,4-phenylene) moiety, which, in
the present work, is not considered to be a further
potential site of reactivity, since benzene and other
aromatics have been found not to participate in the cation
radical Diels–Alder reaction.
The feasibility of consecutive cation radical Diels–
Alder additions to a substrate having two equivalent
unconjugated reactive sites was demonstrated by the
reaction of substrate 1 (Scheme 2) with a fivefold excess
of 1,3-cyclopentadiene in dichloromethane solvent at 0–
5°C. After a reaction time of only 15 s, the bis-adducts
are generated in 47% yield. The isolated yield in the
corresponding monofunctional reaction of trans-anethole
was 52%.2 The ratio of endo to exo Diels–Alder linkages
was found from the proton NMR spectrum to be 5.2:1,
but the percentage composition of the three diastereoi-
someric Diels–Alder adducts could not be directly
determined from the NMR (the endo moieties in the
endo, endo and endo, exo isomers are NMR equivalent, as
are the exo moieties in the exo, exo and endo, exo
diastereomers) or by GC (because of the instability of the
bis adducts at the high temperatures required for its
elution). The monoadduct of 1 with 1,3-cyclopentadiene
could not be isolated because the bis-adduct was the
*Correspondence to: N. Bould, Department of Chemistry and
Biochemistry, University of Texas at Austin, Austin, Texas 78712,
USA.
E-mail: bauld@mail.utexas.edu
Contract/grant sponsor: Robert A. Welch Foundation; Contract/grant
number: F-149.
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 518–522

SITE SELECTIVITY IN CATION RADICAL
519
Scheme 1
major product even at very short reaction times or when
only one mole of 1,3-cyclopentadiene was employed.
This interesting observation is consistent with the
hypothesis that, at least in part, intramolecular hole
transfer occurs in the initially formed monoadduct cation
radical to generate a cation radical site at the second
trans-anethole-like moiety, which then reacts with
another molecule of the diene to yield the bis-adduct
without proceeding through the neutral mono-adduct.5
Selectivity in the context of non-equivalent, conju-
gated reaction sites was then investigated in the reaction
of substrate 2 with cyclopentadiene. Because enol ether
functionalities have been found to be sensitive to the
strong acids which are generated during aminium salt-
catalyzed reactions, the two-phase dichloromethane–
water solvent mixture which had previously been shown
in this laboratory to work especially well for acid
sensitive groups was employed.3 The reaction is highly
site selective for the trans-anethole-like moiety in
preference to the enol ether moiety, the alternative
mono-adducts not being detectable by either GC or
NMR. Further, bis-adducts were also absent from the
reaction mixture. The predominant endo isomer of the
mono-adduct was isolated in pure form in 32% yield, but
Scheme 2
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 518–522

520
N. L. BAULD AND J. YANG
the minor, exo isomer was difficult to separate from the
neutral triarylamine. Although the higher reactivity of a
conjugated cation radical containing both a trans-anet-
hole moiety and an arylpropenyl ether moiety seems not
unreasonable based upon the lower oxidation potential of
trans-anethole than of simple aryl propenyl ethers and the
consequently higher hole density (cation radical char-
acter) on the former moiety, it is at least somewhat
surprising that, in the presence of such a large excess of
the diene (10-fold), the aryl propenyl ether moiety of the
mono-adduct is essentially unreactive. Evidently, intra-
molecular hole transfer is not operative in this instance,
possibly because hole transfer to the enol ether moiety is
less energetically favorable than to the trans-anethole
moiety involved in the reaction of 1. We suggest that the
short reaction time (3 min) along with retardation of the
subsequent cycloaddition to the enol ether moiety by the
neutral triarylamine generated in the initial stage of the
reaction6 could account for the failure to observe any bis-
adducts. In any case, the selectivity for the trans-anethole
moiety is impressive. It is important to note that, although
the yield in this reaction is not especially high, the
possibility that alternative adducts corresponding to
addition to the enol ether moiety could have been formed
and subsequently decomposed is rendered highly un-
likely by the demonstrable stability of the corresponding
adducts derived from the monofunctional enol ether
under the same reaction conditions.
Substrate 3 provides a novel opportunity to probe,
simultaneously, multiple Diels–Alder reactivity and site
selectivity. The reaction of this trifunctional substrate
with an excess of 1, 3-cyclopentadiene in dichloro-
methane for 3 min yields the bis-Diels–Alder adducts
corresponding to reaction exclusively at the two trans-
anethole sites in 56.7% yield. The diastereoisomeric
adduct mixture consisted predominantly of the endo,
endo diastereomer, but smaller amounts of the endo, exo
isomer were evident in the NMR spectra. Once again, the
diaryloxyethene moiety, although reactive in the simple
monofunctional context, shows no sign of reactivity
toward cation radical cycloaddition, since neither the
appropriate bis-adducts nor a tris-adduct could be
detected.
for reaction at the trans-anethole moities, yielding bis-
adducts, but no tris-adducts.
EXPERIMENTAL
Equipment. Proton NMR spectra were recorded on a
Bruker AC250 or a Varian UNITY INOVA 500 spectro-
meter. Carbon spectra were recorded on the Bruker
AC250 machine. COSY and NOESY spectra were
recorded on the Varian UNITY INOVA 500 spectro-
meter. High-resolution mass spectra were recorded on a
VG ZAB-2E mass spectrometer.
Chemicals. All chemicals used as starting materials were
purchased from Aldrich and used as received. The
catalyst, tris(4-bromophenyl)aminium hexachloroanti-
monate, was also obtained from Aldrich. The dichlor-
omethane solvent was dried by refluxing it over calcium
hydride.
Reaction of 1 with 1,3-cyclopentadiene. To 1,2-bis(4-
trans-propenylphenoxy)ethane (1, 50 mg, 0.17 mmol)7
and 1,3-cyclopentadiene (56 mg, 0.85 mmol) in anhy-
drous dichloromethane (10 ml) at 0–5°C were added, all
at once, 21 mg (0.25 mmol) of tris(4-bromophenyl)ami-
nium hexachloroantimonate dissolved in 5 ml of anhy-
drous dichloromethane. After a reaction time of 15 s, the
mixture was quenched with saturated potassium carbo-
nate–methanol solution and water was added. The
dichloromethane layer was then dried and the solvent
removed by a vacuum aspirator, followed by column
chromatography on silica gel. The adduct mixture was
eluted with 4:1 hexanes–dichloromethane, yielding
34 mg (47%) of the endo, endo isomer admixed with
smaller amounts of the endo, exo and, perhaps, the exo,
exo isomer. The ratio of endo to exo linkages in the
diastereoisomer mixture was found to be 5.2:1. However,
the endo linkages of the predominant endo, endo isomer
were indistinguishable from those of the endo, exo
isomer, and the exo linkages of the exo, exo and endo, exo
1
isomers were also indistinguishable. NMR (250 MHz,
CDCl3, endo linkages), ꢀ 1.19 (d, 6H, J = 6.57 Hz), 1.45–
1.51 (m, 2H, C7-anti proton on the bicycloheptadiene
ring), 1.61–1.70 (m, 4H, C7-syn proton plus C6 proton),
2.49 (s, 2H, C1 proton), 2.67–2.70 (t, 2H, J = 4.02, C4
proton), 2.95 (s, 2H, C5 proton), 4.24–4.28 (m, 4H, ether
methylenes), 5.84–5.88 (m, 2H, C2 proton), 6.28–6.32
(m, 2H, C3 proton), 6.77–6.81 (d, 4H, J = 8.67,
aromatic), 7.05–7.08 (d, 4H, J = 8.49); exo linkages:
0.93–0.96 (d, 6H, J = 6.57), 1.74–2.10 (m, 6H, C7 syn
and anti and C6 protons), 2.73–2.77 (m, 6H, C1, C4, and
C5 protons), 4.24–4.28 (m, 4H, ether methylenes), 6.09–
6.11 (m, 2H, C2 proton), 6.28–6.32 (m, 2H, C3 proton),
aromatic protons same as in the endo diastereoisomer.
CONCLUSIONS
Double cation radical Diels–Alder reactions of substrates
containing two trans-anethole-type moieties are found to
be relatively facile and even difficult to arrest at the stage
in which a single molecule of the diene is incorporated. In
contrast, corresponding reactions of difunctional sub-
strates containing one trans-anethole moiety and one of
the enol ether type are highly selective for the trans-
anethole moiety and no bis-adducts are formed. Simi-
larly, a substrate containing two trans-anethole moieties
and one of the enediol diether type are highly selective
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 518–522

SITE SELECTIVITY IN CATION RADICAL
521
Synthesis of cis-(1-propenyl) 4-(trans-1-propenyl)phe-
nyl ether (2). To a mixture of 4-propionylphenol
(50.35 g, 0.335 mol) and potassium carbonate (55.61 g,
0.40 mol) in acetone (250 ml) was added allyl bromide
(101 g, 0.83 mol). The solution was heated to 75–90°C
and vigorously stirred for 6 h. After allowing the reaction
mixture to cool, the inorganic solids were removed by
filtration and the solvent was removed by rotary
evaporation. Water and dichloromethane were then
added, the dichloromethane layer was separated and
dried with sodium sulfate and the solvent dichloro-
methane was evaporated to obtain 56.8 g (89.4%) of the
product, allyl 4-propionylphenyl ether. 1H NMR
(250 MHz, CDCl3), ꢀ 1.12–1.19 (t, 3H, J = 7.25 Hz),
2.84–2.29 (q, 2H, J = 7.26), 4.52–4.55 (d, 2H, J = 7.26),
5.23–5.30 (dt, 1H, J = 10.59, 1.46), 5.32–5.42 (dt, 1H,
J = 17.37, 1.59), 5.92–6.05 (m, 1H), 6.85–6.09 (m, 2H),
7.85–7.91d (m, 2H); 13C NMR (250 MHz, CDCl3), ꢀ
8.283, 31.248, 68.703, 114.232, 117.969, 129.946,
130.034, 132.410, 162.153, 199.291; HRMS, calculated
for C12H14O2 ‡ H (M ‡ 1), 191.107205; found,
191.1107770.
To allyl 4-propionylphenyl ether (the product of the
previous reaction, 49.94 g, 0.26 mol) dissolved in 50 ml
of ethanol was added 5.28 g (0.14 mol) of sodium
borohydride. The solution was stirred for 4 h and then
quenched with 50 ml of water and the solvents were
removed by a vacuum aspirator. Water and dichloro-
methane were then added and the latter layer was
separated, dried and the solvent removed, providing
36.82 g (0.192 mol, 73.7%) of the oily alcohol product,
allyl 4-(1-hydroxypropyl)phenyl ether. 1H NMR
(250 MHz, CDCl3), ꢀ 0.87 (t, 3H, J = 7.38), 1.65–1.82
(m, 2H, J = 7.21), 4.46–4.52 (m, 3H), 5.24–5.30 (dd, 1H,
J = 10.49, 1.35), 5.34–5.44 (dd, 1H, 17.34, 1.61), 5.96–
6.10 (m, 1H), 6.84–7.24 (m, 4H); 13C NMR (250 MHz,
CDCl3), ꢀ 10.142, 31.688, 68.768, 75.528, 114.521,
117.550, 127.123, 133.261, 136.908, 157.925; HRMS,
calculated for C12H16O2 ‡ H (M ‡ 1), 193.122855;
found, 193.122719.
J = 6.75), 5.23–5.29 (d, 1H, J = 10.53), 5.34–5.43 (d, 1H,
J = 17.21), 5.95–6.11 (m, 2H), 6.28–6.35 (d, 1H,
J = 15.78), 6.80–6.85 (d, 2H, J = 6.87), 7.20–7.24 (d,
2H, J = 6.70); 13C NMR (250 MHz, CDCl3), ꢀ 18.390,
68.817, 114.733, 117.581, 123.535, 126.827, 130.307,
130.957, 133.332, 157.577; HRMS, calculated for
C12H14O ‡ H (M ‡ 1), 175.112290; found, 175.111787.
To allyl 4-(trans-1-propenyl)phenyl ether (14.58 g,
83.7 mmol) dissolved in 150 ml of dry dimethyl sulfoxide
was added potassium tert-butoxide (9.32 g). The reaction
mixture was heated at 120–140°C for 6 h, followed by
cooling and addition of 300 ml of water. The organic
product was extracted into dichloromethane and, after
drying and removal of the solvent, the desired product, 1-
propenyl 4-(trans-1-propenyl)phenyl ether (2), was
obtained in 82% yield (12.0 g) as a mixture of cis/
trans-propenyl isomers (cis:trans = 2.5:1). A small
amount (0.175 mg, 1.2% yield) of the pure cis-1-propenyl
ether was then obtained by column chromatography on
silica gel (hexanes eluent). 1H NMR (250 MHz, CDCl3),
ꢀ 1.68–1.72 (dd, 3H, J = 5.96,1.70), 1.84–1.87 (dd, 3H,
J = 6.56,1.53), 4.82–4.88 (m, 1H), 6.09–6.15 (m, 1H),
6.32–6.38 (m, 2H), 6.89–7.27 (m, 4H); 13C NMR
(250 MHz, CDCl3), ꢀ 9.350,18.404, 107.275, 116.157,
124.272, 126.895, 130.195, 132.432, 140.935, 156.458;
HRMS, calculated for C12H14O ‡ H (M ‡ 1),
174.104465; found, 174.105042.
Reaction of cis-(1-propenyl) 4-(trans-1-propenyl)phenyl
Ether (2) with 1,3-cyclopentadiene. The title compound
(2, 61 mg, 0.35 mmol) and 1, 3-cyclopentadiene (245 mg,
3.5 mmol) were dissolved in a solvent mixture consisting
of 8 ml of dichloromethane and 2 ml of water and the
solution was cooled to 0°C. Tris(4-bromophenylaminium
hexachloroantimonate (60 mg, 0.074 mmol) was then
added and the reaction mixture stirred for 3 min prior to
quenching with saturated methanolic potassium carbo-
nate. After workup with additional water and dichloro-
methane and drying, the product was chromatographed
on silica gel [hexanes–dichloromethane (50:1)] to yield
27 mg (32%) of the endo Diels–Alder adduct correspond-
ing to addition to the anisyl moiety. 1H NMR (250 MHz,
CDCl3), ꢀ 1.16–1.25 (d, 3H, J = 6.89), 1.43–1.49 (m, 1H),
1.50–1.54 (m, 1H), 1.61–1.72 (m, 4H), 2.48 (s, 1H),
2.67–2.71 (t, 1H, J = 4.29), 2.94 (s, 1H), , 4.76–4.87 (m,
1H), 5.83–5.87 (m, 1H), 6.30–6.34 (m, 2H), 6.80–7.10
(m, 4H); HRMS, calculated for C17H20O, 241.159240;
found, 241.160156.
To the product of the previous reaction, allyl 4-(1-
hydroxypropyl)phenyl ether (32.7 g, 0.169 mol), dis-
solved in 50 ml of anhydrous dichloromethane was
added 56.5 ml of a 1.0 M solution of phosphorus
tribromide in dichloromethane. After stirring the reaction
mixture overnight, 50 ml of water were added, the
dichloromethane layer was separated and dried and the
solvent removed. Then, a solution of potassium tert-
butoxide (20 g, 0.18 mol) dissolved in 30 ml of dimethyl
sulfoxide was added and the reaction mixture was heated
under reflux (120–140°C) for 4 h. After cooling, 100 ml
of water were added and the organic product was
extracted into dichloromethane solvent. The product,
allyl 4-(trans-1-propenyl)phenyl ether, was obtained in
56% yield (16.5 g) after silica gel chromatography using
Synthesis of cis-1,2-bis(4-trans-propenylphenoxy)-
ethene (3). A solution of cis-1,2-bis(4-bromophenoxy)-
ethene (3.1 g, 8.37 mmol)8 in 100 ml of anhydrous
diethyl in ether was cooled to 20 to 25°C and treated
with 18.3 ml of 1.6 M butyllithium in hexane). After
stirring the reaction mixture for 2 h, propionaldehyde
(2.61 g, 45 mmol) was added and the resulting mixture
stirred for another 2 h at the same temperature before
1
hexanes as the eluent. H NMR (250 MHz, CDCl3), ꢀ
1.82–1.85 (dd, 3H, J = 6.51, 1.45), 4.48–4.51 (d, 2H,
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 518–522

522
N. L. BAULD AND J. YANG
quenching with water. Aqueous workup then afforded,
after chromatography on alumina (dichloromethane
eluent), the corresponding diol, cis-1,2-bis[4-(1-hydro-
usual way. Silica gel chromatography yielded 10 mg
(56.7%) of the bis-Diels–Alder adduct, corresponding to
reaction at the two propenyl groups, predominantly as the
endo, endo isomer, but with a minor amount of the endo,
exo isomer, which could not be readily separated. The
spectroscopic data are given for the predominant endo,
1
xypropyl)phenyoxy]ethene, in 47% yield (1.28 g). H
NMR (250 MHz, CDCl3), ꢀ 0.84–0.89 (t, 6H, J = 7.41),
1.66–1.82 (m, 4H), 4.51–4.56 (t, 2H, J = 6.63), 6.12 (s,
2H), 7.01–7.29 (m, 8H); 13C NMR (250 MHz, CDCl3), ꢀ
10.075, 31.843, 75.462, 116.059, 127.251, 128.351,
139.041, 156.704; HRMS, calculated for C20H22O4 ‡ 2H
(M ‡ 2), 328.167460; found, 328.167084.
1
endo isomer. H NMR (250 MHz, CDCl3), ꢀ1.19–1.23
(d, 6H, J = 6.85), 1.46–1.53 (m, 2H, anti-C7 proton on the
norbornane ring), 1.65–1.67 (m, 4H, syn-C7 and C6
protons), 2.49 (s, 2H, bridgehead proton at C1), 2.68–
2.71 (m, 2H, bridgehead proton at C4), 2.96 (s, 2H,
benzylic proton at C5), 5.84 (m, 2H, olefinic proton at
C2), 6.06–6.08 (s, 2H, olefinic protons on the central
double bond), 6.29–6.32 (m, 2H, olefinic protons at C3),
6.91–6.95 (d, 4H, aromatic protons, J = 8.69), 7.08–7.11
(d, 4H, aromatic protons, J = 8.52); 13C NMR (250 MHz,
CDCl3), ꢀ 21.053, 41.366, 46.870, 49.198, 49.492,
52.579, 115.526, 128.320, 128.806, 133.475, 138.089,
139.122, 155.603; HRMS, calculated for C30H32O2 ‡ H
(M ‡ 1), 425.248056; found, 425.247317.
This diol (205 mg, 0.625 mmol) was dissolved in 25 ml
of anhydrous dichloromethane and treated with phos-
phorus tribromide (0.4 ml of a 1 M solution in dichlor-
omethane) and allowed to react at room temperature for
1 h. After evaporating the solvent, triethylamine (50 ml)
was added and the solution refluxed for 2 h. After
removal of the triethylamine, water and dichloromethane
were added and the organic layer was separated, dried
and the solvent removed. Following chromatography on
alumina [hexanes–dichloromethane (6:1)] the desired
compound (3) was isolated in 52% yield (95 mg). 1NMR
(250 MHz, CDCl3), ꢀ 1.83–1.86 (d, 6H, J = 6.51), 6.07–
6.16 (m, 2H), 6.10 (s, 2H), 6.30–6.37 (d, 2H, J = 17.1),
6.98–7.02 (d, 4H, J = 6.73), 7.24–7.28 (d, 4H, J = 6.78);
13C NMR (250 MHz, CDCl3), ꢀ 18.405,116.252,
124.601, 126.915, 128.290, 130.114, 132.914, 156.298;
HRMS, calculated for C20H20O2 ‡ H (M ‡ 1),
293.154155; found, 293.153920.
Acknowledgement
The authors thank the Robert A. Welch Foundation (grant
F-149) for support of this research.
REFERENCES
Diels±Alder reaction of cis-1,2-bis(trans-4-propenyl-
phenoxy)ethene (3) with 1,3-cyclopentadiene. To
cyclopentadiene (23 mg, 0.35 mmol) and the substrate 3
(17 mg, 0.058 mmol) dissolved in anhydrous dichloro-
methane (10 ml) were added 9.5 mg (0.0116 mmol) of
tris(4-bromophenyl)aminium hexachloroantimonate in
5 ml of dichloromethane at 0°C. The reaction was
quenched after 3 min with saturated methanolic potas-
sium carbonate and the organic phase worked up in the
1. Bauld NL. Tetrahedron 1989; 45: 5307–5363.
2. Pabon RA, Bellville DJ, Bauld NL. J.Am.Chem.Soc. 1983; 105:
5158–5159.
3. Bauld NL, Yang J. Tetrahedron Lett. 1999; 40: 8519–8522.
4. Bauld NL, Yang J. Org.Lett. 1999; 1: 773–774.
5. Bauld NL, Aplin JT, Yueh W, Sarker H, Bellville D. Macromol-
ecules 1996; 29: 3661–3662.
6. Lorenz KT, Bauld NL. J.Am.Chem.Soc. 1987; 109: 1157–1160.
7. Bauld NL, Gao D, Aplin JT. J.Phys.Org.Chem. 1999; 12: 1–11.
8. Yang J, Bauld NL. J.Org.Chem. 1999; 64: 9251–9253.
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 518–522