The thionophosphate-thiolophosphate photoisomerization proceeds predominantly through a non-chain radical pathway. Synthetically viable benzylation of tetrahydrofuran, propan-2-ol and olefins

Article abstract of DOI:10.1039/b003501g

The photoirradiation of thionophosphates, ROP(S)(OEt)₂, derived from benzyl and vinylogously benzyl alcohols in CH₃CN, with a Hanovia medium-pressure mercury lamp in a quartz vessel leads to the formation of the corresponding thiolophosphates, RSP(O)(OEt)₂, through a non-chain radical pathway. This behavior of thionophosphates is unlike that of the related phosphates, which react through ionic dissociation-recombination processes. When the irradiation is conducted in solvents such as PrⁱOH, THF and toluene, benzylation of these solvents takes place in synthetically respectable yields. Irradiation of thionophosphates in CH₃CN leads to a convenient allylic benzylation of olefins.

Full text of DOI:10.1039/b003501g

View Article Online / Journal Homepage / Table of Contents for this issue
The thionophosphate–thiolophosphate† photoisomerization proceeds
predominantly through a non-chain radical pathway. Synthetically
viable benzylation of tetrahydrofuran, propan-2-ol and olefins‡
1
Veejendra K. Yadav,*a Rengarajan Balamurugan,a Masood Parvezb and Raghav Yamdagnib
a Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India
b Department of Chemistry, University of Calgary, Calgary, Alberta, Canada T2N 1N4
Received (in Cambridge, UK) 2nd May 2000, Accepted 7th December 2000
First published as an Advance Article on the web 11th January 2001
The photoirradiation of thionophosphates, ROP(S)(OEt)2, derived from benzyl and vinylogously benzyl
alcohols in CH3CN, with a Hanovia medium-pressure mercury lamp in a quartz vessel leads to the formation
of the corresponding thiolophosphates, RSP(O)(OEt)2, through a non-chain radical pathway. This behavior of
thionophosphates is unlike that of the related phosphates, which react through ionic dissociation–recombination
processes. When the irradiation is conducted in solvents such as PriOH, THF and toluene, benzylation of these
solvents takes place in synthetically respectable yields. Irradiation of thionophosphates in CH3CN leads to a
convenient allylic benzylation of olefins.
reacted. The thiolophosphates were separated readily from the
Introduction
unchanged thionophosphates by chromatographic techniques
including radial chromatography over silica gel. Except for the
cinnamyl thionophosphate, all other thionophosphates were
stable to thermal conditions2 and silica gel chromatography.
Cinnamyl thionophosphate isomerized3 on chromatography
over silica gel to furnish the same product mixture as that
produced from irradiation (vide infra). However, the ratio of
the isomeric products was different from that produced in the
photochemical reaction.
The species 3 and 4 were also formed in small amounts
from the reaction of methyl mandelate-derived thionophos-
phate (mandelic acid is phenylglycolic acid). The corresponding
thiolophosphate, however, was still the major product. While
the structure of 3 was confirmed from spectroscopic character-
istics, exact mass measurement analysis and comparison with
literature data,4 the structure of 4 was secured from a single-
crystal X-ray structure determination (Fig. 2).5 Cinnamyl
thionophosphate underwent transformation both without and
with allylic shift to furnish 5 and 6, respectively. The efficacies
of the rearrangements of 1-phenylethanol and 1-phenyl-
citronellol derivatives were somewhat low in comparison with
those of methyl mandelate and cinnamyl alcohol derivatives,
and longer irradiation (30 min) was required for у65% con-
version. The results are collected in Table 1.
The photochemistry of phosphates, ROP(O)(OEt)2, has been
studied extensively by Givens and shown to involve an ionic
dissociation–recombination pathway.1 In contrast, the photo-
chemistry of thionophosphates, ROP(S)(OEt)2, has received
little attention. In the present manuscript, we disclose our
results from such a study and show that (a) thionophosphates
are photo-active only when the carbinol (methanol) carbon is
substituted by an aryl or a vinylogous aryl function, (b) thiono-
phosphates are conveniently isomerized to the corresponding
thiolophosphates, RSP(O)(OEt)2, in CH3CN as solvent, (c)
irradiation of thionophosphates in solvents such as PriOH,
tetrahydrofuran and toluene readily leads to their benzylation,
(d) irradiation of thionophosphates in CH3CN brings about
allylic benzylation of olefins, and (e) non-chain radical path-
ways control all the above transformations.
Results and discussion
1. Isomerization to thiolophosphates
The irradiation in CH3CN resulted in a smooth transformation
of the thionophosphates, 1, into the related thiolophosphates, 2
(Scheme 1). Some 65–70% of the initial thionophosphate
1-Phenylhex-5-en-1-ol underwent both rearrangement to the
corresponding thiolophosphate and elimination to generate a
mixture of (Z)- and (E)-1-phenylhexa-1,5-dienes6 in almost
equal amounts. 1,2-Diphenylethyl thionophosphate reacted
similarly to give the corresponding thiolophosphate and a
mixture of (Z) and (E) stilbenes in almost equal amounts.
2. Irradiation in PriOH
In view of the formation of 3 and 4 above, a radical pathway
was considered to be more appropriate than an ionic
dissociation–recombination alternative. This led us to study the
reactions of selected thionophosphates in propan-2-ol, which is
a good hydrogen-atom donor under radical conditions. It was
expected to quench the phosphate radical derived from the
thionophosphate to generate (RO)2P(O)SH, or, a derivative
thereof, and itself combine with the benzyl radical to result in a
C–C bond formation.
Scheme 1 A general representation of the thionophosphate–thiolo-
phosphate photoisomerization.
† The IUPAC name for thionophosphates is thiophosphate O-esters
and that for thiolophosphates is thiophosphate S-esters.
‡ 1H and 13C spectra of the adduct of TEMPO and the radical 14
are available as supplementary data. For direct electronic access see
http://www.rsc.org/suppdata/p1/b0/b003501g
DOI: 10.1039/b003501g
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332
This journal is © The Royal Society of Chemistry 2001
323

View Article Online
Table 1 Results of selected thionophosphate–thiolophosphate photo-
isomerizations
Table 2 Results of the reactions of selected thionophosphates with
PriOH, THF, and toluene
Irradiation time
(t/min)
Conversion
(%)
Thionophosphate
from
Irradiation
Solvent time (t/min) (%)
Conversion Product
Thionophosphate from
yield (%)
Methyl mandelate
Cinnamyl alcohol
1-Phenylethanol
1,2-Diphenylethanol
1-Phenylcitronellol
1-Phenylhex-5-en-1-ol
Benzyl alcohol
4-(tert-Butyl)cyclohexanol
1-Phenylpropan-2-ol
trans-But-2-en-1-ol
15
15
30
30
30
30
30
120
120
120
70
68
65
67
65
65
Methyl mandelate PriOH
Methyl mandelate THF
Methyl mandelate Toluene
Cinnamyl alcohol PriOH
Cinnamyl alcohol THF
30
30
30
120
120
70
75
50
89
91
90
90
85
25
20
65
a
a
a
a = No reaction. In all other cases, the reactions were very clean and
the isolated yield of the thiolophosphate based on the consumed
thionophosphate was 90% and above.
Fig. 2 ORTEP plot of the X-ray structure of 4.
The dimers 10 and 11 (5%, combined) and the isomeric
thiolophosphates 5 and 6 (40%, combined) were the other
products that had formed along with the expected product
12 (25%). The product 13 that is isomeric with 12 was not
formed. The reaction, therefore, had proceeded with very high
regioselectivity. The formation of 12b along with 12a must
be a consequence of photo-induced olefin isomerization.
The 1,2-diphenylethyl thionophosphate underwent smooth
elimination to furnish a mixture of (Z)- and (E)-stilbenes;
PriOH-incorporated product was not formed. 1-Phenylethanol
furnished only the corresponding thiolophosphate.
3. Irradiations in tetrahydrofuran and toluene
The generation of radical species, e.g. 14, from methyl man-
delate-derived thionophosphate, was confirmed further by
conducting the irradiation in non-alcoholic solvents such as
tetrahydrofuran (THF) and toluene. The methyl mandelate
derivative furnished 15 and 167 from the reactions in THF and
toluene, respectively. The reaction in THF was more efficient
than the reaction in toluene. Cinnamyl thionophosphate
furnished an inseparable mixture of products 17–19 (25%).
The E- and Z-olefin geometries in 17 and 18 were determined
from the 1H coupling constants of the vinylic hydrogens,
which were 15.6 and 11.7 Hz, respectively. The species 19 was
a diastereomeric mixture. The dimeric species 10 and 11 (5%)
and the thiolophosphates 5 and 6 (45%) were also formed.
The results of the reactions of selected thionophosphates are
collected in Table 2.
The thionophosphates of 1,2-diphenylethanol and 1-phenyl-
ethanol did not react in the desired manner. The 1,2-diphenyl-
ethanol derivative underwent elimination and furnished a
mixture of (Z)- and (E)-stilbenes. A small amount of the
corresponding thiolophosphate was also formed. 1-Phenyl-
ethanol furnished only the corresponding thiolophosphate.
The reactions in THF may be considered to proceed as
shown in Scheme 2. It is the sulfur radical 20, and not the
carbon radical 14, which abstracts a hydrogen atom from THF.
In support of this, we have never isolated any methyl α-phenyl-
acetate that would be formed if the carbon radical 14 were to
Fig. 1 Substrates 3–19 obtained from some reactions.
In the event, irradiation of a solution of methyl mandelate-
derived thionophosphate in PriOH for 30 min furnished the
β-hydroxy ester 7 as the sole product; none of the cation-
quenched product 8 was formed. This experiment demon-
strated clearly that the benzyl cation 9 was not formed. This is
to be contrasted with the results from the irradiation of
ArCH2OP(O)(OEt)2 in solvents such as alcohols (ROH), moist
CH3CN, and benzene, when ArCH2OR, ArCH2NHCOMe,
and ArCH2C6H5, respectively, constituted the predominant
ϩ
products that arose from the reaction of ArCH2 with the
solvent molecules.1 It is significant to note that we have never
isolated any amide-like products from the reactions in CH3CN.
Likewise, C6H5-incorporated products were also not isolated
from the reactions in benzene.
The cinnamyl thionophosphate also reacted with PriOH.
324
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332

View Article Online
solution (30 mL) of methyl mandelate-derived thionophos-
phate was bubbled with dioxygen for 30 min and then irradiated
for 15 min. 1H NMR analysis indicated 8:1 ratio of thiono- and
thiolophosphate as against a ratio of 3.5:1 when the dioxygen
was replaced by argon under otherwise identical conditions.
The low thiono–thiolo conversion in the presence of dioxygen
indicates involvement of radical(s).
In yet another experiment, a 30 mL CH3CN solution of
methyl mandelate-derived thionophosphate and 2,2,6,6-tetra-
methylpiperidin-1-oxyl (TEMPO), 10 mM in each, was degassed
by bubbling with argon for 30 min and then irradiated for
15 min. Chromatographic purification allowed us to isolate
the product of quench of the benzyl radical 14 with TEMPO.
This experiment confirmed the participation of radicals beyond
doubt.
Scheme 2 A tentative mechanism for the benzylation of tetrahydro-
furan.
5. A tentative mechanistic rationale for the formation of the
species 3 and 4
abstract a hydrogen atom from THF. The products from the
self-couplings of tetrahydrofuran-2-yl radical, 22, and the
carbon radical 14 were not formed either. The cross coupling,
therefore, is greatly favored over self-coupling. This has some
bearing on the reaction mechanism. The THF molecule that
has diffused to form the exciplex 23 is probably readily avail-
able for hydrogen-atom abstraction by the sulfur radical. This is
then followed by a fast capture of the so-generated tetrahydo-
furan-2-yl radical by a benzyl radical as shown in Scheme 3.8
With the participation of radical species confirmed, we can now
speculate on the events in the formation of 3 and 4 as in Scheme
4. The benzylic hydrogen in methyl mandelate-derived thiono-
phosphate is highly reactive. The carbon radical 26, formed
from its abstraction in an intermolecular process via diradical
25 as shown, is stable. The combination of radicals 14 and 26
would give yet another thionophosphate 27, which would react
through a six-membered ring transition state involving a
hydrogen-atom transfer to deliver 4. The self-coupling of the
radical 14 generates 3. The product of self-coupling of 26
was not formed. This may be due to the large steric crowding
that would develop around the carbon–carbon bond that would
form.
6. Reactions of thionophosphates formed from non-benzylic
alcohols
The chemistry of the thionophosphates prepared from trans-4-
tert-butylcyclohexanol, 1-phenylpropan-2-ol and trans-but-2-
en-1-ol were studied to understand the role of the aromatic
ring. There was no reaction even after irradiation for 2 h. It
appears, therefore, that the light is absorbed first by the aro-
matic ring and then transmitted, in some way, to the P᎐S
᎐
function. The failure of the 1-phenylpropan-2-ol derivative to
undergo the thiono–thiolo isomerization suggests further that
the aromatic chromophore must necessarily be present at the
carbinol carbon for a meaningful reaction to occur. The study
of thionophosphates prepared from these alcohols on reaction
with diphenyl chlorothiophosphate turns out to be a distinct
possiblity for future investigations.
Scheme 3 A tentative rationale as to how only the cross-coupled
products are formed.
4. Further evidence for radical participation
Dioxygen is an effective quench for radicals. A 10 mM CH3CN
Scheme 4 A tentative mechanism leading to the formation of species 3 and 4.
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332
325

View Article Online
Table 3 Results of allylic benzylation of selected alkenes with methyl mandelate thionophosphatea
Benzylation
(%)b
Thiolophosphate
Recovered
SMb (%)
Olefin
3 (%)b
4 (%)b
(%)b
Cyclopentene
Cyclohexene
Cyclooctene
Cycloocta-1,5-diene
1-Methylcyclohexene
Dec-1-ene
38.2
32.0
34.4
33.0
38.4
29.7
7.6
6.6
5.5
7.5
8.4
9.7
6.8
6.6
6.9
9.1
7.8
7.8
17.8
23.0
30.8
20.2
18.4
23.0
24.8
29.2
32.1
28.0
25.0
11.7
a The yields are based on the methyl mandelate-derived thionophosphate consumed. b SM = starting material.
Scheme 5 The different possibilities in the allylic benzylation of but-1-
ene.
7. Allylic benzylation of alkenes
The demonstrated ability of sulfur radicals to abstract a hydro-
gen atom from PriOH, THF and toluene raised the possibility
of using olefins as possible hydrogen-atom donors. The abstrac-
tion of a hydrogen atom from an allylic carbon will generate
a radical that could, in principle, combine with the earlier
formed benzyl radical in a subsequent step to generate a
product of net allylic benzylation. This concept is illustrated
in Scheme 5 with but-1-ene as a representative example. The
first formed allylic radical 29, which is in resonance with 30,
will be expected to combine with the radical 14 to generate a
Scheme 6 Allylic benzylation of 1-methylcyclohexene.
mixture of 31 and 32.
The above premise was put to experimental test and we
investigated the irradiation of selected thionophosphates and
olefins such as cyclopentene, cyclohexene, 1-methylcyclohexene,
cyclooctene, cycloocta-1,5-diene and dec-1-ene in CH3CN. It
was gratifying indeed to discover that the reactions proceeded
just as expected. From all these reactions, the dimeric materials
3 and 4 and the corresponding thiolophosphates were also
formed in small to significant amounts. The results of the
reactions of methyl mandelate-derived thionophosphate with
various olefins are collected in Table 3.
1-Methylcyclohexene could, in principle, furnish a total of
five regioisomers, 36–40 (Scheme 6). The regioisomers 36 and
37 represent the products derived from the radical 33. The
regioisomer 38 will be derived from the radical 34. Finally, the
regioisomers 39 and 40 will be formed from the radical 35. Of
these five possible regioisomers, only the first three were formed.
The isomer 37 was separated from the mixture. The isomers 36
and 38 were inseparable from each other even by chromato-
graphy over AgNO3-impregnated silica gel. The ratio (36 ϩ
38):37 was 4:1. Clearly, the radical coupling at the secondary
carbons was 4 times faster than the coupling at the tertiary
carbon. Hydrogen abstraction did not occur at the methyl
group and, thus, the regioisomers 39 and 40 were not formed.
Dec-1-ene reacted with methyl mandelate-derived thiono-
phosphate and furnished an inseparable 3:2 mixture of the
dec-2- and -1-ene, 41 and 42 (Fig. 3), which were benzylated at
C-1 and C-3, respectively. The first formed allylic radical was
resonance stabilized and it reacted through both ends.
The cinnamyl thionophosphate furnished a mixture of the
(E)- and (Z)-3-(cyclohex-2-enyl)-1-phenylpropene, 43a and
43b, respectively, and a diastereomeric mixture of 3-(cyclohex-
2-enyl)-3-phenylpropene, 44 on reaction with cyclohexene. The
1
ratio 43a:43b was estimated at ≈1:1 from the H integrals of
the olefinic hydrogens at C-1. The 1:1 diastereomeric mixture
of 44 was separated into its components by radial chromato-
graphy. The product from 1-phenylethanyl thionophosphate
and cyclohexene was a 1:1 mixture of diastereomeric 1-(cyclo-
hex-2-enyl)-1-phenylethane, 45.
The reaction of cycloocta-1,5-diene was likely to give two
diastereomeric products, 46a and 46b. The structure assignment
as to whether the product was 46a or 46b or a mixture of both
was difficult due to the diastereomeric nature of the products.
To resolve this problem indirectly, we chose to eliminate the
diastereomeric component to simplify product composition
and, hence, irradiated cycloocta-1,5-diene and benzyl thiono-
326
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332

View Article Online
derived thionophosphate and benzyl bromide (5 mol%) in
toluene. The in situ-generated benzyl radical from the reaction
of tributyltin radical and benzyl bromide was expected to initi-
ate the process by adding to the sulfur and, thus, bringing about
the desired overall change. Product analysis indicated that while
the thionophosphate was still present, all the benzyl bromide
had been consumed. Clearly, the benzyl radical formed from
benzyl bromide on reaction with tributyltin radical did not add
to P᎐S. This result is interesting because the dimerization of
᎐
benzyl radical took place even when its concentration was so
very small. It points to the presence of some very strong force
that acts against the addition of a benzyl radical to the P᎐S
᎐
function. To reconfirm this finding, the following experiment
was designed.
A solution of Bu3SnH (1.2 equiv.) in benzene was added
slowly with the help of a syringe pump to a mixture of methyl
mandelate-derived thionophosphate and benzyl bromide (1.0
equiv.) at reflux. The benzyl bromide had apparently reduced
to toluene as it was altogether absent in the mixture and the
thionophosphate was recovered quantitatively. The benzyl
radical, therefore, did not add to P᎐S. The tributyltin radical
᎐
did not add to P᎐S either. This was confirmed further from
᎐
a separate experiment in which a benzene solution of methyl
mandelate-derived thionophosphate, Bu3SnH (1.2 equiv.) and a
catalytic amount of azoisobutyronitrile (AIBN) was refluxed
for 2 h. The thionophosphate was recovered quantitatively.
This last experiment was performed on 1-phenylcitronellyl§
thionophosphate as well, when, again, no reaction took place
and the thionophosphate was recovered quantitatively.
Fig. 3 Examples of allylic benzylations (substrates 41–47).
phosphate under identical conditions and isolated a mixture of
two compounds along with a small amount of 1,2-diphenyl-
ethane. Fortunately, these two compounds were separated easily
by column chromatography over AgNO3-impregnated silica
gel and they were characterized as 47a and 47b from their
individual spectroscopic data including 2D 1H NMR. The
product from allylic shift was, therefore, formed. In compliance
with this, we conclude that a mixture of 46a and 46b was
formed from the reaction of cycloocta-1,5-diene with methyl
mandelate-derived thionophosphate.
The above observations established clearly that neither a
carbon radical nor a tin radical adds to the P᎐S function
᎐
present in thionophosphates. This being so, the transformation
of a thionophosphate into the corresponding thiolophosphate
could only be a non-chain radical process.
To confirm further the above non-chain protocol, we
irradiated a 10 mM solution of methyl mandelate-derived
thionophosphate in CH3CN for 3 min and then waited for 12
min before the solvent was removed. The thiono–thiolo ratio
was determined to be 67:1, which indicated very little con-
version. The above ratio is far from the 3.5:1 ratio observed
on irradiation for 15 min. The low conversion on shorter
irradiation can be explained only by the participation of a
non-chain protocol. With a chain protocol, both the reactions
will be expected to exhibit similar conversions.
8. Evidence in support of the non-chain radical protocol
The thionophosphate–thiolophosphate transformation pro-
ceeds in a non-chain radical fashion. This was demonstrated
using methyl mandelate-derived thionophosphate as the sub-
strate. If the reaction were to proceed in a chain fashion, one
will expect the carbon radical 14 to add to the sulfur in the P᎐S
bond and bring about further cleavage as shown in Scheme 7
᎐
9. Attempts at the cleavage of cyclopropane rings
The successful experiments with cinnamyl thionophosphate led
us to explore the chemistry of cyclopropylmethanol derivatives
as well. This was due largely to the understanding that a cyclo-
propane ring is similar to an olefin. In the event, we irradiated
48 (Fig. 4) in CH3CN and noted absolutely no reaction. This
shows the significance of a true π-bond over a cyclopropane
ring system; the ring does not allow the transfer of energy
absorbed by the aromatic unit to the thionophosphate function.
This above failure is not too difficult to understand. The
transmittance of energy from the aromatic unit to the thiono-
phosphate function across an olefinic linkage is made simple
due to the all-parallel nature of the p orbitals in the aromatic
ring system, the two p orbitals of the olefinic tether and the
C–O bond axis in the thionophosphate function. In the cyclo-
propane derivative, if the p orbitals of the aromatic ring system
are parallel to the adjacent cyclopropane carbon–carbon bond
axis, they are near orthogonal to the cyclopropane carbon–
carbon bond axis at the other carbinol-bearing carbon. This is
likely to break the flow of energy from the aromatic core to the
thionophosphate function. The transfer of energy, therefore,
takes place through the bonds and not through the space. This
Scheme 7 The requirement(s) of the radical-chain process for the
thionophosphate–thiolophosphate isomerization.
to give the desired thiolophosphate with regeneration of the
radical 14. The crucial step in the design of such a radical chain
is the ability of a benzylic radical to add to the sulfur in P᎐S.
᎐
Though such radical additions to thione functions, C᎐S, are
᎐
implicated in the literature,9 we discover below that this is not
so in as much as the addition to the sulfur in thionophosphates
is concerned.
A solution of hexabutylditin (2.5 mol%) in toluene was
added slowly to a refluxing solution of methyl mandelate-
§ Citronellyl = 3,7-dimethyloct-6-enyl.
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332
327

View Article Online
Conclusions
Thionophosphates were conveniently isomerized into the
corresponding thiolophosphates under photochemical condi-
tions. The isomerizations followed exclusively non-chain radical
pathways. The reactions in PriOH and THF constitute conveni-
ent benzylation protocols for these substrates. The irradiation
of a thionophosphate and an olefin in acetonitrile readily
leads to an allylic benzylation of the latter. All these reactions
are, however, limited to alcohols that are either benzylic or
vinologously benzylic. The light absorbed by the aromatic core
is transmitted to the thionophosphate function through the
bonds that trigger the observed transformations.
Experimental
Starting materials
The thionophosphates studied were derived from methyl
mandelate, 1-phenylhex-5-enol, cinnamyl alcohol, 1-phenyl-
ethanol, 1,2-diphenylethanol,12 1-phenylcitronellol,13 trans-4-
tert-butylcyclohexanol, (E)-but-2-en-1-ol, 1-phenylpropan-2-ol
and 1-(2-phenylcyclopropyl)ethanol on reaction of their
sodium salts with diethyl chlorothiophosphate at 0–25 ЊC. It is
important to note that the earlier employed method1 that uses
pyridine as base for the preparation of phosphates was com-
pletely ineffective. Cyclopentene, cyclohexene and dec-1-ene
were distilled before use. Cyclooctene, cycloocta-1,5-diene and
1-methylcyclohexene were used as received.
Fig. 4 Attempts at the cleavage of cyclopropane rings.
hypothesis explains the failure of 1-phenylpropan-2-yl thiono-
phosphate to undergo isomerization to the corresponding
thiolophosphate (vide supra).
General
1H, 13C and 31P data were recorded on Bruker DPX-200,
Bruker DRX-300 and JEOL JNM-LA400 series of instruments
for samples in CDCl3. Signal positions are reported in ppm
(δ scale) relative to SiMe4 for 1H, CDCl3 for 13C and H3PO4 for
31P spectra. The mixtures were purified for their components by
either gravity column chromatography over silica gel (100–200
mesh) or radial chromatography using plates coated with silica
gel 60 PF254 (E. Merck). The components were eluted with mix-
tures of hexanes and EtOAc. IR spectra were measured on a
Perkin-Elmer 1320 spectrometer.
Argon was bubbled through all the solutions for 30 min
before irradiation. Irradiation was carried out using a Hanovia
450W medium-pressure mercury lamp (product no. 679A0360,
arc length 4.3 inches). The center of the magnetically stirred
reaction solution (15–30 mL in a quartz tube of id 2.0 cm) was
kept at a distance of 11–12 cm from the center of the light
source. Water at 25 ЊC was circulated through the outer jacket
of the lamp-housing (all-quartz). The reaction solutions were
only slightly warm to the touch after the irradiation.
We considered we could overcome the above difficulty by
substituting a phenyl group on the carbinol carbon itself and
studying 49. However, the preparation of 49 from the alcohol
50 was difficult. We isolated a host of products, 51–55, but not
the expected thionophosphate from the employment of our
standard conditions. The formation of the products 51–55 can
be explained from a combination of events such as (a) ioniz-
ation leading to the formation of a cyclopropylmethyl cation,
(b) cleavage of the cyclopropane ring in this cation that leads to
the formation of yet another cation, and (c) the quenching
of the rearranged cation by the phosphate ion through both
oxygen and sulfur. The species 49, therefore, is too unstable
under the thionophosphate-forming reaction conditions to
allow its isolation. The all-E 51–53 and 55 show that the cyclo-
propane ring cleaved with very high stereoselectivity.
The alcohols corresponding to the thionophosphates 48 and
49 were prepared from NaBH4 reduction of the corresponding
ketones which, in turn, were prepared from the corresponding
α,β-unsaturated ketones following Corey’s cyclopropanation
method.10
In attempts to suppress or even stop the above rapid ioniz-
ation, we also considered studying the substrate 56 that lacks
the phenyl substituent on the cyclopropane ring. However, the
alcohol 57 could not be prepared from the attempted cyclo-
propanation of 58 under the Simmons–Smith conditions
(CH2I2, Zn/Cu, Et2O, reflux, 2 h). Other methods of cyclo-
propanation were not attempted.
For the thiono–thiolo isomerization, a 10 mM solution of a
thionophosphate in CH3CN was irradiated for 15–30 min. For
reactions with solvents such as PriOH, THF and toluene
and for the allylic benzylation of olefins in CH3CN, 20 mM
solutions of the thionophosphates were used. The molar ratio
of thionophosphate and olefin was 1:2 and the irradiation time
2 h.
General procedure for the formation of thionophosphates
Finally, in our attempts to generate a cyclopropylmethyl
radical to witness the cleavage of the cyclopropane ring and
its further quench by a carbon radical, we irradiated a mixture
of methyl mandelate thionophosphate and 5911 (cis/trans
mixture) in CH3CN to observe only thiono–thiolo isomeriz-
ation. The substrate 59 was recovered and no ring-cleaved
product was formed. This result is surprising particularly
because methyl mandelate-derived thionophosphate has been
found by us (vide supra) to react with toluene. It appears as if
the phosphate radical can quench the carbon radical faster than
it can abstract a hydrogen atom from 59.
To an ice-cold stirred suspension of NaH (1.2 mmol) in dry
THF (2 mL) was added a solution of an alcohol (1.0 mmol) in
THF (2 mL). The resultant mixture was stirred till hydrogen
evolution ceased (30–60 min). A solution of diethyl chlorothio-
phosphate (1.1 mmol) in THF (2 mL) was added, the reaction
mixture allowed to warm to room temperature, and stirring was
continued until the reaction was complete (3–10 h) by TLC. The
reaction mixture was diluted with Et2O (10 mL), mixed with
saturated aq. NH4Cl (5 mL) and stirred for 10 min. The layers
were separated and the aqueous layer was extracted with Et2O
(2 × 5 mL). The combined Et2O solution was dried, filtered and
328
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332

View Article Online
concentrated. The residue was chromatographed over silica gel
and the desired product isolated in near quantitative yield.
Cinnamyl thionophosphate (1, R ؍ PhCH᎐CHCH ). Liquid,
᎐
2
δH 7.41–7.22 (5H, m), 6.68 (1H, d, J 15.8 Hz), 6.29 (1H, td,
J 15.8, 6.3 Hz), 4.72 (2H, dd, J 10.1, 6.3 Hz), 4.22–4.07 (4H, dq,
J 7.1, 9.6 Hz), 1.33 (6H, t, J 7.1 Hz); δC 136.1, 133.9, 128.6,
128.1, 126.6, 123.6 (d, J 7.6 Hz), 68.4 (d, J 4.7 Hz), 64.3 (d, J 5.5
Hz), 15.9 (d, J 7.5 Hz); δP 68.0.
Reaction of methyl mandelate-derived thionophosphate with
benzyl bromide and hexabutylditin
A solution of hexabutylditin (0.025 mmol) in toluene (3.5 mL)
was slowly added to a refluxing solution of the thionophos-
phate (1.0 mmol) and benzyl bromide (0.05 mmol) in toluene
(9 mL) over a period of 2 h using a syringe pump. Reflux was
continued for an additional 1 h. The solvent was removed and
the residue was chromatographed to isolate the starting thiono-
phosphate quantitatively.
Cinnamyl thiolophosphate (2, R ؍ PhCH᎐CHCH ). Liquid,
᎐
2
δH 7.44–7.20 (5H, m), 6.60 (1H, d, J 15.7 Hz), 6.33–6.18 (1H, td,
J 15.7, 7.4 Hz), 4.30–4.03 (4H, m), 3.65 (2H, dd, J 14.9, 7.4 Hz),
1.34 (6H, t, J 7.1 Hz); δC 136.2, 133.4, 128.5, 127.8, 126.3,
124.7 (d, J 4.7 Hz), 63.5 (d, J 5.8 Hz), 33.35 (d, J 4.2 Hz), 15.9
(d, J 7.2 Hz); δP 27.3.
Reaction of methyl mandelate-derived thionophosphate with
benzyl bromide, tributyltin hydride and AIBN
1-Phenylallyl thiolophosphate 6. Liquid, δH 7.36–7.25 (5H,
m), 6.26–6.08 (1H, m), 5.30–5.16 (2H, m), 4.99 (1H, dd, J 11.1,
7.8 Hz), 4.22–3.84 (4H, m), 1.25 (6H, t, J 7.0 Hz); δC 140.4
(d, J 5.3 Hz), 137.8 (d, J 6.2 Hz), 128.6, 127.8, 127.6, 126.3,
116.7, 63.4 (d, J 4.0 Hz), 52.8, 15.8 (d, J 7.5 Hz); δP 25.5.
A solution of Bu3SnH (1.2 mmol) and AIBN (0.06 mmol)
in benzene (3.5 mL) was added to a refluxing solution of
the thionophosphate (1.0 mmol) and PhCH2Br (1.0 mmol) in
benzene (9 mL) over a period of 2 h using a syringe pump. The
reaction mixture was refluxed for an additional 1 h before the
solvent was removed.
1-Phenylcitronellol. Diastereomeric mixture, liquid, δH 7.35–
7.23 (5H, m), 5.11–5.02 (1H, 2t, J 7.1 Hz), 4.76–4.70 (1H, 2t,
J 4.5, 6.4 Hz), 2.05–1.85 (3H, m), 1.68–1.57 (6H, 4s), 1.52–1.30
(2H, m), 1.26–1.10 (1H, m), 0.96–0.93 (3H, 2d, J 6.6 Hz);
δC 145.45, 144.95, 131.17, 131.14, 128.40, 128.39, 127.47,
127.34, 125.95, 125.72, 124.71, 124.68, 72.8, 72.3, 46.7, 46.3,
37.5, 36.8, 29.3, 29.1, 25.6, 25.4, 25.2, 20.1, 19.2, 17.6 (Calc. for
C16H24O: C, 82.69; H, 10.42. Found: C, 82.50; H, 10.30%).
Reaction of methyl mandelate-derived thionophosphate with
tributyltin hydride and AIBN
A solution of Bu3SnH (1.2 mmol) and AIBN (0.06 mmol) in
benzene (3.5 mL) was added to a refluxing solution of the
thionophosphate (1.0 mmol) over a period of 2 h using a
syringe pump. Reflux was continued for an additional 1 h
before the solvent was removed.
1-Phenylcitronellyl thionophosphate. Diastereomeric mixture,
liquid, δH 7.38–7.26 (5H, m), 5.54–5.43 (1H, m), 5.10–5.02 (1H,
m), 4.15–3.93 (2H, m), 3.85–3.65 (2H, m), 2.09–1.73 (2H, m),
1.70–1.56 (6H, 4s), 1.26 (6H, dt, J 7.1, 1.0 Hz), 1.50–1.00 (5H,
m), 0.98–0.93 (3H, 2d, J 6.6 Hz); δC 141.21, 140.60, 131.29,
131.20, 128.37, 128.31, 128.14, 128.03, 126.92, 126.68, 124.57,
80.09, 80.03, 79.75, 79.69, 63.99, 63.93, 63.88, 45.61, 45.53,
44.93, 44.86, 37.19, 36.67, 28.84, 28.67, 25.69, 25.27, 25.16,
19.74, 19.27, 17.62, 15.83, 15.75, 15.58, 15.50; δP 66.6, 66.5.
Methyl mandelate-derived thionophosphate (1, R ؍ PhCH-
CO2Me). Liquid, δH 7.60–7.25 (5H, m), 5.89 (1H, d, J 11.6 Hz),
4.30–4.10 (2H, m), 4.05–3.80 (2H, m), 3.72 (3H, s), 1.35 (3H,
t, J 7.0 Hz), 1.14 (3H, t, J 7.0 Hz); δC 169.7 (d, J 6.4 Hz), 135.4
(d, J 6.0 Hz), 129.7, 129.1, 127.8, 77.9 (d, J 3.2 Hz), 65.1 (d,
J 5.6 Hz), 64.8 (d, J 5.8 Hz), 53.0, 16.1 (t, J 8.3 Hz); δP 67.2.
1-Phenylcitronellyl thiolophosphate. Diastereomeric mixture,
liquid, δH 7.36–7.21 (5H, m), 5.10–4.90 (1H, m), 4.38–4.30 (1H,
m), 4.10–3.98 (1H, m), 3.98–3.82 (2H, m), 3.82–3.72 (1H, m),
2.10–1.70 (2H, m), 1.68–1.54 (6H, 4s), 1.45–1.14 (11H, m), 0.90
(3H, d, J 6.6 Hz); δC 143.26, 142.29, 131.38, 131.27, 128.50,
128.47, 128.44, 127.62, 127.52, 127.42, 127.33, 124.46, 124.41,
63.24, 63.18, 49.12, 49.09, 48.80, 48.77, 45.54, 45.45, 45.01,
44.93, 37.10, 36.20, 30.29, 25.62, 25.15, 19.49, 18.88, 17.61,
15.90, 15.87, 15.82, 15.79; δP 25.8, 25.6.
Methyl mandelate-derived thiolophosphate (2, R ؍ PhCH-
CO2Me). Liquid, δH 7.50–7.30 (5H, m), 5.04 (1H, d, J 11.3 Hz),
4.20–4.10 (1H, m), 4.08–3.90 (3H, m), 3.74 (3H, s), 1.27 (3H,
t, J 7.0 Hz), 1.24 (3H, t, J 7.0 Hz); δC 170.8 (d, J 6.4 Hz), 136.6
(d, J 6.0 Hz), 129.3, 129.0, 128.7, 64.3 (d, J 6.2 Hz), 62.2 (d,
J 6.9 Hz), 53.5, 52.3 (d, J 3.2 Hz), 16.2 (d, J 7.3 Hz); δP 23.4.
Dimethyl 2,3-diphenylsuccinate 3a. More polar, solid, mp
231–234 ЊC; δH 7.18–6.97 (5H, m), 4.25 (1H s), 3.69 (3H, s);
δC 173.7, 135.6, 128.5, 128.4, 128.3, 127.5, 54.7, 52.4 (Calc. for
[M Ϫ MeOH]ϩ: m/z, 266.0943. Observed m/z, 266.0918).
1,2-Diphenylethyl thionophosphate (1, R ؍ PhCH2CHPh).
Liquid, δH 7.30–7.10 (10H, m), 5.62 (1H, dt, J 11.7, 6.4 Hz),
3.94–3.84 (2H, m), 3.82–3.74 (1H, m), 3.72–3.56 (1H, m), 3.27
(1H, dd, J 13.7, 7.7 Hz), 3.08 (1H, ddd, J 13.7, 6.2, 1.9 Hz), 1.13
(3H, dt, J 7.1, 0.7 Hz), 1.00 (3H, dt, J 7.1, 1.0 Hz); δC 139.82 (d,
J 2.5 Hz), 136.5, 129.6, 128.1, 128.0, 126.7, 126.4, 81.7 (d, J 5.8
Hz), 63.74 (d, J 5.8 Hz), 63.64 (d, J 5.0 Hz), 44.3 (d, J 8.2 Hz),
15.57 (d, J 8.2 Hz), 15.43 (d, J 8.2 Hz); δP 66.9.
Dimethyl 2,3-diphenylsuccinate 3b. Less polar, solid, mp 235–
238 ЊC; δH 7.50–7.24 (5H, m), 4.38 (1H, s), 3.39 (3H, s); δC 171.8,
136.3, 128.7, 128.3, 127.9, 54.9, 52.0 (Calc. for [M Ϫ MeOH]ϩ:
m/z, 266.0943. Observed m/z, 266.0923).
1-Phenylethyl thionophosphate (1, R ؍ PhCHCH3). Liquid,
δH 7.40–7.20 (5H, m), 5.65–5.50 (1H, m), 4.20–4.00 (2H, m),
4.00–3.80 (2H, m), 1.62 (3H, d, J 6.6 Hz), 1.30 (3H, t, J 7.2 Hz),
1.14 (3H, t, J 7.2 Hz); δC 141.7, 141.6, 128.4, 128.0, 126.1, 77.4,
64.08, 64.02, 63.96, 23.90, 23.83, 15.83, 15.73, 15.69, 15.58;
δP 66.8.
1,2-Diphenylethyl thiolophosphate (2, R ؍ PhCH2CHPh).
Liquid, δH 7.30–7.08 (10H, m), 4.44 (1H, td, J 11.7, 7.8 Hz),
4.00–3.85 (2H, m), 3.77–3.63 (2H, m), 3.33–3.28 (1H, dd,
J 13.7, 7.6 Hz), 3.25–3.19 (1H, ddd, J 13.7, 7.6, 1.4 Hz), 1.16
(3H, dt, J 7.1, 0.5 Hz), 1.13 (3H, dt, J 7.1, 0.5 Hz); δC 142.1,
138.1, 129.4, 128.4, 128.2, 127.7, 127.5, 126.6, 63.2 (d, J 5.7 Hz),
52.6 (d, J 3.3 Hz), 44.7 (d, J 8.2 Hz), 15.84 (d, J 7.4 Hz), 15.75
(d, J 9.9 Hz); δP 25.8.
1-Phenylethyl thiolophosphate (2, R ؍ PhCHCH3). Liquid,
δH 7.40–7.25 (5H, m), 4.48 (1H, dq, J 10.8, 7.1 Hz), 4.17–3.88
(4H, m), 1.75 (3H, dd, J 7.1, 1.0 Hz), 1.24 (3H, dt, J 7.1, 0.7
Hz), 1.23 (3H, dt, J 7.1, 0.7 Hz); δC 143.5 (d, J 4.9 Hz), 128.6,
127.5, 127.0, 63.8 (d, J 5.7 Hz), 45.8 (d, J 4.1 Hz), 24.6 (d, J 7.5
Hz), 15.9 (d, J 6.6 Hz); δP 26.4.
1-Phenylhex-5-enyl thionophosphate [1, R ؍ PhCH(CH2)3-
CH᎐CH ]. Liquid, δ 7.35–7.28 (5H, m), 5.80–5.70 (1H, m),
᎐
2
H
5.44–5.38 (1H, m), 5.01–4.92 (2H, m), 4.16–4.06 (1H, m),
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332 329

View Article Online
4.06–3.98 (1H, m), 3.89–3.79 (1H, m), 3.79–3.70 (1H, m), 2.10–
2.04 (2H, q, J 7.1 Hz), 2.05–1.95 (1H, m), 1.85–1.77 (1H, m),
1.53–1.43 (1H, m), 1.43–1.33 (1H, m), 1.27 (3H, t, J 7.1 Hz),
1.06 (3H, t, J 7.1 Hz); δC 140.53 (d, J 3.3 Hz), 138.19, 128.28,
128.03, 126.59, 114.84, 81.18 (d, J 5.8 Hz), 63.98, 64.00 (d, J 4.2
Hz), 63.95 (d, J 4.2 Hz), 37.10 (d, J 7.4 Hz), 33.21, 24.41, 15.77
(d, J 8.3 Hz), 15.54 (d, J 7.4 Hz); δP 67.2.
(10H, m), 3.88–3.82 (1H, dd, J 8.7, 6.9 Hz), 3.6 (3H, s), 3.45–
3.38 (1H, dd, J 13.5, 8.7 Hz), 3.06–2.99 (1H, dd, J 13.5, 6.9 Hz);
δC 173.8, 139.0, 138.6, 128.9, 128.6, 128.3, 127.9, 127.4, 126.4,
53.6, 52.0, 39.8 (Calc. for C16H16O2: C, 79.96; H, 6.72. Found:
C, 79.80; H, 6.84%).
2-[(E)-3-Phenylprop-2-enyl]tetrahydrofuran 17. Liquid, δH
7.36–7.17 (5H, m), 6.46 (1H, d, J 15.6 Hz), 6.28–6.20 (1H, td,
J 15.6, 7.1 Hz), 3.98–3.88 (2H, m), 3.78–3.72 (1H, m), 2.53–2.38
(2H, m), 2.10–1.75 (4H, m) (Calc. for C13H16O: C, 82.93; H,
8.57. Found: C, 82.77; H, 8.45%).
1-Phenylhex-5-enyl thiolophosphate [2, R ؍ PhCH(CH2)3-
CH᎐CH ]. Liquid, δ 7.33–7.24 (5H, m), 5.78–5.68 (1H, m),
᎐
2
H
5.00–4.92 (2H, m), 4.27–4.20 (1H, m), 4.10–4.00 (1H, m), 3.99–
3.87 and 3.86–3.76 (3H, m), 2.08–1.90 (4H, m), 1.52–1.41 (1H,
m), 1.37–1.26 (1H, m), 1.21–1.17 (6H, m); δC 142.5, 138.0,
128.5, 127.5, 115.0, 63.3, 50.8, 37.6, 37.5, 33.1, 26.8, 15.88,
15.84, 15.80, 15.76; δP 26.3.
2-[(Z)-3-Phenylprop-2-enyl]tetrahydrofuran 18. Liquid, δH
7.35–7.20 (5H, m), 6.53 (1H, d, J 11.7 Hz), 5.76–5.70 (1H, td,
J 11.7, 7.1 Hz), 3.98–3.84 (2H, m), 3.77–3.70 (1H, m), 2.68–2.60
(1H, m), 2.56–2.49 (1H, m), 2.03–1.95 (1H, m), 1.92–1.83 (2H,
m), 1.55–1.46 (1H, m) (Calc. for C13H16O: C, 82.93; H, 8.57.
Found: C, 82.80; H, 8.42%).
Benzyl thionophosphate (1, R ؍ PhCH2). Liquid, δH 7.40–
7.25 (5H, m), 5.08 (2H, d, J 10.0 Hz), 4.15–4.02 (4H, m), 1.28
(6H, dt, J 7.2, 0.8 Hz); δC 135.8 (d, J 7.4 Hz), 128.3, 128.2,
127.8, 69.3 (d, J 4.9 Hz), 64.1 (d, J 5.8 Hz), 15.7 (d, J 7.4 Hz);
δP 68.0.
Methyl ꢀ-(cyclopent-2-enyl)-ꢀ-phenylacetate. Diastereomeric
mixture, liquid, δH 7.43–7.18 (5H, m), 5.87–5.68 and 5.32–5.26
(2H, m), 3.65 (3H, s), 3.55–3.24 (2H, m), 2.45–2.10 (2H, m),
1.83–1.17 (2H, m); δC 174.4, 174.3, 139.5, 139.4, 133.9, 133.1,
132.8, 129.4, 129.3, 128.2, 123.0, 58.3, 57.9, 52.1, 52.0, 50.4,
50.1, 32.5, 32.1, 29.6, 28.3; νmax (film) 2920, 1718, 1440, 1420,
1160 cmϪ1 (Calc. for C14H16O2: M, 216.1149. Observed Mϩ,
216.1157. Calc. for C14H16O2: C, 77.74; H, 7.46. Found: C,
77.62; H, 7.35%).
Benzyl thiolophosphate (2, R ؍ PhCH2). Liquid, δH 7.38–
7.25 (5H, m), 4.03 (2H, d, J 14.2 Hz), 4.17–4.07 (2H, m), 4.07–
3.97 (2H, m), 1.28 (6H, dt, J 7.1, 0.8 Hz); δC 137.5, 128.9, 128.6,
127.6, 63.5 (d, J 5.8 Hz), 35.0 (d, J 3.3 Hz), 15.9 (d, J 7.4 Hz);
δP 26.8.
trans-4-(tert-Butyl)cyclohexyl thionophosphate. Liquid, δH
4.33–4.22 (1H, m), 4.11–4.02 (4H, m), 2.07 (2H, br d, J 9.0 Hz),
1.76 (2H, br d, J 12.0 Hz), 1.45–1.30 (2H, m), 1.29 (6H, t, J 7.0
Hz), 1.10–0.85 (3H, m), 0.86 (9H, s); δC 78.96, 78.90, 63.92,
63.86, 46.7, 33.61, 33.57, 32.2, 27.5, 25.4, 15.87, 15.80; δP 66.4.
Methyl ꢀ-(cyclohex-2-enyl)-ꢀ-phenylacetate. Diastereomeric
mixture, liquid, δH 7.37–7.18 (5H, m), 5.85–5.75, 5.70–5.60 and
5.20–5.10 (2H, m), 3.66 and 3.65 (3H, s), 3.32 (1H, d, J 11.1),
2.95–2.75 (1H, m), 2.03–1.25 and 1.10–0.95 (6H, m); δC 174.0,
173.9, 137.7, 137.4, 129.2, 129.0, 128.8, 128.5, 128.4, 127.9,
127.3, 57.5, 51.8, 38.5, 38.4, 27.9, 26.4, 25.3, 25.2, 21.1, 20.7;
νmax (film) 2920, 1720, 1442, 1425, 1150 cmϪ1 (Calc. for
C15H18O2: M, 230.1306. Observed Mϩ, 230.1300. Calc. for
C15H18O2: C, 78.22; H, 7.88. Found: C, 78.11; H, 7.73%).
1-Methyl-2-phenylethyl thionophosphate. Liquid, δH 7.30–
7.20 (5H, m), 4.85–4.77 (1H, m), 4.12–3.82 (4H, m), 3.05–3.00
(1H, dd, J 13.7, 6.6 Hz), 2.84–2.78 (1H, dd, J 13.7, 6.6 Hz),
1.30 (3H, d, J 6.1 Hz), 1.29 (3H, dt, J 7.1, 0.8 Hz), 1.22 (3H, t,
J 7.1 Hz); δC 137.3, 129.6, 128.3, 126.5, 77.1 (d, J 5.8 Hz),
63.9 (t, J 5.0 Hz), 43.6 (d, J 6.6 Hz), 20.9 (d, J 3.3 Hz), 15.86
(d, J 4.1 Hz), 15.78 (d, J 5.0 Hz); δP 66.5.
Methyl ꢀ-(cyclooct-2-enyl)-ꢀ-phenylacetate. Diastereomeric
mixture, low-melting solid, δH 7.40–7.18 (5H, m), 5.83–5.71,
5.57–5.46, 5.36–5.27 and 5.02–4.93 (2H, m), 3.65 and 3.62 (3H,
s), 3.49–3.16 (2H, 2H), 2.42–1.93 (2H, m), 1.77–1.17 and 1.03–
0.85 (8H, m); δC 174.3, 174.0, 138.0, 137.7, 131.6, 131.3, 130.4,
128.6, 128.4, 127.3, 127.2, 57.9, 57.7, 51.9, 51.8, 39.5, 38.8,
34.7, 33.0, 29.5, 26.9, 26.8, 25.6, 25.3 (Calc. for C17H22O2:
M, 258.1618. Observed Mϩ, 258.1610. Calc. for C17H22O2: C,
79.02; H, 8.59. Found: C, 78.88; H, 8.45%).
(E)-But-2-enyl thionophosphate. Liquid, δH 5.85–5.77 (1H,
m), 5.65–5.58 (1H, m), 4.51–4.46 (2H, m), 4.18–4.06 (4H, m),
1.74–1.72 (3H, dd, J 6.4, 1.2 Hz), 1.40–1.26 (6H, m); δC 131.4,
125.6, 77.4, 68.6, 64.2, 17.7, 15.8; δP 67.8.
Methyl 3-hydroxy-3-methyl-2-phenylbutanoate 7. Liquid,
δH 7.38–7.26 (5H, m), 3.68 (3H, s), 3.60 (1H, s), 1.34 (3H, s),
1.07 (3H, s); δC 174.6, 135.3, 129.5, 128.3, 127.6, 71.7, 60.4,
52.0, 29.5, 26.6 (Calc. for C12H16O3: C, 69.19; H, 7.75. Found:
C, 68.98; H, 7.81%).
Methyl ꢀ-(1-methylcyclohex-2-enyl)-ꢀ-phenylacetate 37.
Liquid, δH 7.42–7.26 (5H, m), 5.70–5.56 (2H, m), 3.64 (3H, s),
3.54 (1H, s), 1.94–1.25 (6H, m), 1.09 (3H, s); δC 173.2, 133.4,
130.2, 130.1, 127.8, 127.7, 127.1, 127.0, 60.9, 59.9, 51.45, 51.41,
38.3, 37.9, 33.4, 33.1, 25.08, 24.87, 24.82, 24.52, 18.88, 18.66
(Calc. for C16H20O2: C, 78.64; H, 8.26. Found: C, 78.45; H,
8.16%).
(E)-2-Methyl-5-phenylpent-4-en-2-ol 12a and (Z)-2-methyl-5-
phenylpent-4-en-2-ol 12b. ≈1:1 Mixture. Characteristic signals
for the E-derivative: δH 6.46 (1H, d, J 15.9 Hz), 6.32–6.25 (1H,
td, J 15.9, 7.3 Hz), 2.52 (2H, dd, J 7.5, 1.7 Hz). Characteristic
signals for the Z-derivative: δH 6.60 (1H, d, J 11.7 Hz), 5.84–
5.78 (1H, td, J 11.7, 7.5 Hz), 2.38 (2H, d, J 7.5 Hz). The methyl
groups appear at δH 1.27 in one isomer and at δH 1.25 in the
other.
Methyl ꢀ-(3-methylcyclohex-2-enyl)-ꢀ-phenylacetate 36 and
methyl ꢀ-(2-methylcyclohex-2-enyl)-ꢀ-phenylacetate 38. Dia-
stereomeric mixture, liquid, δH 7.39–7.25 (m), 5.49 (br s), 5.40
(br s), 5.33 (br s), 4.87 (br s), 3.78 (d, J 10.0 Hz), 3.68–3.65 (4s),
3.58 (d, J 10.0 Hz), 3.30–3.26 (2d, J 11.0, 11.5 Hz), 2.83 (br s),
2.05–1.70 (m), 1.69, 1.67, 1.58, 1.53 (4s), 1.50–1.20 (m);
δC 174.1, 137.77, 137.54, 136.38, 136.05, 128.78, 128.72, 128.49,
128.44, 128.42, 128.35, 128.32, 127.2, 123.0, 121.8, 57.68, 57.63,
51.98, 51.83, 51.76, 38.79, 38.67, 30.0, 27.7, 25.9, 23.98, 23.89,
21.5, 21.0, 18.3.
Methyl ꢀ-phenyl-ꢀ-(tetrahydrofuran-2-yl)acetate 15. Liquid,
δH 7.33–7.26 (5H, m), 4.58–4.46 (1H, m), 3.98–3.77 (2H, m),
3.70 (3H, s), 3.51 (1H, d, J 9.9 Hz), 1.92–1.74 (2H, m), 1.74–
1.61 (1H, m), 1.51–1.34 (1H, m); δC 173.0, 135.9, 128.7, 128.4,
127.7, 80.6, 68.5, 57.5, 52.1, 29.5, 25.4 (Calc. for C13H16O3: C,
70.87; H, 7.33. Found: C, 70.65; H, 7.18%).
Methyl 2-phenyldodec-4-enoate 41 and methyl 3-heptyl-2-
phenylpent-4-enoate 42. Diastereomeric mixture for 42; liquid,
Methyl 2,3-diphenylpropanoate 16. Liquid, δH 7.27–7.10
330
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332

View Article Online
δH 7.37–7.23 (m), 5.67–5.58 (m), 5.49–5.39 (m), 5.37–5.21 (m),
5.14–5.07 (m), 4.85–4.74 (m), 3.66–3.59 (4s), 3.46–3.41 (2d,
J 11.0 Hz), 2.84–2.68 (m), 2.50–2.36 (m), 1.95–1.90 (q, J 6.8
Hz), 1.40–1.0 (m), 0.89–0.82 (2t, J 7.0 Hz).
1-Phenylhexa-1,5-diene. Mixture of E and Z isomers, liquid,
δH 7.34–7.18 (m), 6.44–6.37 (unsym t), 6.22 (t, J 6.4 Hz), 6.20 (t,
J 6.8 Hz), 5.92–5.78 (m), 5.67 (t, J 7.1 Hz), 5.64 (t, J 7.1 Hz),
5.08–4.96 (m), 2.47–2.40 (dq, J 7.3, 1.7 Hz), 2.33–2.25 (m),
2.65–2.17 (m); δC 138.05, 137.99, 137.71, 137.59, 132.02, 130.16,
130.05, 129.15, 128.72, 128.44, 128.10, 126.85, 126.49, 125.93,
114.95, 114.89, 33.92, 33.52, 32.39, 27.89 (Calc. for C12H14: C,
91.08; H, 8.92. Found: C, 90.87; H, 8.75%).
3-(Cyclohex-2-enyl)-1-phenylpropene 43a and 43b. Liquid,
δH 7.36–7.19 (m), 6.49–6.46 (d, J 12.0 Hz), 6.42–6.38 (d, J 15.8
Hz), 6.26–6.19 (td, J 16.0, 6.8 Hz), 5.74–5.67 (m), 5.64–5.59 (dt,
J 8.3, 1.7 Hz), 2.43–2.28 (m), 2.28–2.18 (m), 2.02–1.98 (m),
1.84–1.75 (m), 1.75–1.65 (m), 1.60–1.45 (m), 1.38–1.22 (m);
δC 137.77, 137.73, 131.34, 131.27, 131.09, 129.74, 129.25,
128.77, 128.46, 128.08, 127.52, 127.42, 126.83, 126.43, 125.94,
39.82, 35.91, 35.49, 34.95, 28.91, 28.88, 25.27, 25.25, 21.42,
21.39 (Calc. for C15H18: C, 90.84; H, 9.16. Found: C, 90.60; H,
9.02%).
Phenyl-(2-phenylcyclopropyl)methanol 50. More polar isomer,
solid, mp 48–50 ЊC; δH 7.53–7.09 (10H, m), 4.36 (1H, d, J 7.5
Hz), 3.25 (1H, br s), 2.10–2.03 (1H, m), 1.63–1.57 (1H, m),
1.27–1.22 (1H, m), 1.14–1.09 (1H, m); δC 143.4, 142.0, 128.2,
128.1, 127.3, 125.90, 125.85, 125.43, 76.3, 29.8, 20.8, 13.6 (Calc.
for C16H16O: C, 85.67; H, 7.19. Found: C, 85.49; H, 7.08%).
Phenyl-(2-phenylcyclopropyl)methanol 50. Less polar isomer,
solid, mp 42–44 ЊC; δH 7.72–7.24 (10H, m), 4.24 (1H, d, J 7.8
Hz), 3.55 (1H, br s), 2.17–2.14 (1H, m), 1.67–1.57 (1H, m),
1.15–1.03 (2H, 2m); δC 143.4, 142.3, 128.2, 128.1, 127.2, 125.9,
125.7, 125.3, 76.7, 30.3, 21.6, 13.3 (Found: C, 85.53; H, 7.10%).
3-(Cyclohex-2-enyl)-3-phenylpropene 44. More polar, liquid,
δH 7.32–7.18 (5H, m), 6.03–5.94 (1H, m), 5.65–5.60 (1H, m),
5.33–5.30 (1H, dd, J 10.2, 2.2 Hz), 5.05–5.00 (2H, m), 3.05 (1H,
t, J 9.3 Hz), 2.51–2.44 (1H, m), 1.99–1.94 (2H, m), 1.87–1.78
(1H, m), 1.76–1.71 (1H, m), 1.56–1.47 (1H, m), 1.37–1.25 (1H,
m); δC 143.4, 140.9, 129.7, 128.4, 128.1, 127.9, 126.1, 115.1,
56.4, 39.7, 27.5, 25.3, 21.5 (Calc. for C15H18: C, 90.84; H,
9.16. Found: C, 90.68; H, 9.05%).
trans-1,4-Diphenylbut-3-enyl thionophosphate 51. Liquid,
δH 7.40–7.19 (10H, m), 6.43 (1H, d, J 15.8 Hz), 6.16–6.08 (1H,
td, J 15.8, 7.3 Hz), 5.57–5.51 (1H, m), 4.13–3.72 (1H each, 4m),
2.92–2.84 (1H, m), 2.78–2.71 (1H, m), 1.21 (3H, t, J 7.1 Hz),
1.06 (3H, t, J 7.1 Hz); δC 140.0, 137.2, 133.2, 128.43, 128.35,
128.19, 127.2, 126.6, 126.1, 124.8, 80.6 (d, J 5.8 Hz), 64.1 (d,
J 3.3 Hz), 64.0 (d, J 3.3 Hz), 41.4 (d, J 7.4 Hz), 15.7 (d, J 8.2
Hz), 15.6 (d, J 7.4); δP 67.0.
3-(Cyclohex-2-enyl)-3-phenylpropene 44. Less polar, liquid,
δH 7.32–7.17 (5H, m), 6.06–5.97 (1H, m), 5.80–5.72 (2H, m),
5.10–5.04 (2H, m), 3.04 (1H, t, J 9.3 Hz), 2.47–2.41 (1H, m),
1.99–1.95 (2H, m), 1.71–1.65 (1H, m), 1.51–1.39 (2H, m), 1.17–
1.07 (1H, m); δC 143.7, 140.6, 129.5, 128.4, 128.1, 127.8, 126.1,
115.7, 56.2, 39.6, 27.7, 25.4, 21.6 (Calc. for C15H18: C, 90.84;
H, 9.16. Found: C, 90.65; H, 9.02%).
Bis-[(E)-1,4-diphenylbut-3-enyl] ether 53. Liquid, δH 7.39–
7.20 (20H, m), 6.47 (2H, d, J 15.9 Hz), 6.10–6.03 (2H, td, J 15.9,
7.1 Hz), 5.87–5.83 (2H, dd, J 7.8, 6.3 Hz), 2.92–2.84 (2H, m),
2.78–2.70 (2H, m); δC 137.4, 136.7, 134.2, 129.0, 128.8, 128.5,
127.6, 126.5, 126.2, 123.1, 85.0, 38.1 (Calc. for C32H30O: C,
89.25; H, 7.03. Found: C, 89.13; H, 6.91%).
1-(Cyclohex-2-enyl)-1-phenylethane 45. Diastereomeric mix-
ture, liquid, δH 7.31–7.16 (10H, m), 5.82–5.79 (1H, m), 5.76–
5.71 (1H, m), 5.65–5.60 (1H, m), 5.39–5.35 (1H, m), 2.62–2.53
(2H, m), 2.35–2.18 (2H, m), 2.00–2.18 (4H, m), 1.85–1.40 (8H,
m), 1.28 (3H, d, J 7.0 Hz), 1.24 (3H, d, J 7.0 Hz); δC 130.73,
129.62, 128.14, 127.76, 127.52, 125.82, 45.03, 44.68, 41.84,
27.60, 26.51, 25.40, 25.27, 21.96, 21.37, 18.81, 18.54 (Calc. for
C14H18: C, 90.25; H, 9.75. Found: C, 90.10; H, 9.58%).
The adduct of TEMPO and the radical 14. Solid, mp 73–
75 ЊC; δH 7.43 (2H, d, J 7.1 Hz), 7.35–7.28 (3H, m), 5.21 (1H, s),
3.65 (3H, s), 1.60–1.20 (6H, m), 1.23 (3H, s), 1.14 (3H, s), 1.07
(3H, s), 0.72 (3H, s); δC 172.4, 138.1, 128.2, 127.8, 126.8, 88.5,
59.8, 51.7, 40.12, 29.99, 33.4, 32.7, 20.1, 17.0.
Methyl ꢀ-(cycloocta-2,6-dienyl)-ꢀ-phenylacetate 46a (dia-
stereomeric mixture) and methyl ꢀ-(cycloocta-2,5-dienyl)-ꢀ-
phenylacetate 46b (diastereomeric mixture). Liquid, δH 7.38–
7.25 (m), 5.67–5.50 (m), 5.47–5.37 (m), 5.07–5.02 (dd, J 11.5,
6.8 Hz), 3.66 (s), 3.65 (s), 3.57–3.45 (m), 3.43–3.36 (2d, J 10.5
Hz), 2.60–2.48 (m), 2.44–2.26 (m), 2.23–2.13 (m), 2.03–1.92
(m); δC 173.96, 173.84, 137.50, 137.36, 130.89, 130.32, 129.58,
129.19, 128.91, 128.86, 128.72, 128.64, 128.47, 128.44, 127.56,
127.49, 127.42, 127.34, 58.38, 58.06, 51.97, 51.94, 42.17, 42.09,
33.07, 31.89, 27.91, 27.78, 27.46 (Calc. for C17H20O2: C, 79.64;
H, 7.87. Found: C, 79.48; H, 7.69%).
Acknowledgements
We thank the Council of Scientific & Industrial Research,
Government of India for financial support, and Professors
H. Ila and H. Junjappa for encouragement.
References
1 R. S. Givens and L. W. Kueper, III, Chem. Rev., 1993, 93, 55 and
references cited therein. Also see: W. W. Epstein and M. Garrossian,
J. Chem. Soc., Chem. Commun., 1987, 532.
2 Heating of a toluene solution in a pressure tube at 140 ЊC (bath
temperature) for 4–5 h gave only the starting material without any
noticeable decomposition.
3 C. D. Poulter and D. S. Mautz, J. Am. Chem. Soc., 1991, 113, 4895;
A. W. Herriott, J. Org. Chem., 1975, 40, 801.
4 Y. Ito, S. Fujiim, T. Konoeke and T. Saegusa, Synth. Commun.,
1976, 6, 429; C. T. Ng, X. Wang and T.-Y. Luh, J. Org. Chem., 1988,
53, 2536.
3-Benzylcycloocta-1,5-diene 47a. Liquid, δH 7.30–7.17 (5H,
m), 5.59–5.51 (3H, m), 5.43–5.38 (1H, dd, J 11.7, 6.3 Hz), 3.05–
2.96 (1H, m), 2.75–2.70 (1H, dd, J 13.4, 6.8 Hz), 2.60–2.55 (1H,
dd, J 13.4, 7.8 Hz), 2.53–2.44 (1H, m), 2.39–2.25 (3H, m), 2.24–
2.13 (2H, m); δC 140.7, 133.6, 129.1, 128.8, 128.2, 128.0, 127.7,
125.9, 43.1, 41.0, 33.7, 28.2, 27.6 (Calc. for C15H18: C, 90.84; H,
9.16. Found: C, 90.67; H, 9.04%).
5 Selected crystal data for dimethyl 2,3-diphenylfumarate 4: CCDC
reference number 137237. See http://www.rsc.org/suppdata/p1/b0/
b003501g/ for crystallographic data in CIF or other electronic
format. M = 296.32, monoclinic, a = 5.845(5), b = 17.134(4), c =
6-Benzylcycloocta-1,4-diene 47b. Liquid, δH 7.28–7.15 (5H,
m), 5.71–5.58 (2H, 2m, 2- and 4-H), 5.43–5.35 (1H, m, 1-H),
5.18–5.12 (1H, m, 5-H), 3.14–3.04 (1H, m, 6-H), 2.86–2.56 (5H,
m, 3-H2, C8-H, PhCH2), 1.97–1.90 (1H, m, 8-H), 1.55–1.47
(1H, m, 7-H), 1.18–1.10 (1H, m, 7-H); δC 141.2, 133.8, 129.8,
129.0, 128.6, 128.4, 128.1, 125.7, 42.3, 37.1, 29.9, 29.8, 24.5
(Calc. for C15H18: C, 90.84; H, 9.16. Found: C, 90.69; H,
9.06%).
7.935(7) Å, β = 107.54(6)Њ, Z = 2, Dc = 1.299 g cmϪ3, µ = 0.91 cmϪ1
.
1533 Data collected at room temperature on a Rigaku AFC6S
diffractometer. 1395 Reflections were used to solve and refine the
structure to R = 0.076 and Rw = 0.051. C7–C7* = 1.402(14) Å, C8–
O1 = 1.190(7) Å, C8–C7–C7* = 111.0(9)Њ, O1–C8–C7 = 125.9(7)Њ,
O1–C8–C7–C7* = Ϫ35.9(14)Њ, C6–C1–C7–C7* = 46.7(11)Њ, O2–
C8–C7–C7* = 149.8(10)Њ, O1–C8–O2–C9 = 0.8(11)Њ, C7–C8–O2–
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332
331

View Article Online
C9 = 175.2(6)Њ. For the preparation of the Z-isomer of 4 (i.e. the
diphenylmaleate) and the data on it, see: H. Oda, M. Morishita, K.
Fugami, H. Sano and M. Kosugi, Chem. Lett., 1996, 811.
6 R. C. Larock and K. Takagi, J. Org. Chem., 1984, 49, 2701;
Y. Yamamoto, K. Maruyama and K. Matsumoto, J. Chem. Soc.,
Chem. Commun., 1984, 548.
7 S. Ghosh, S. N. Pardo and R. G. Salomon, J. Org. Chem., 1982, 47,
4692; M. N. Fox and C.-C. Chen, J. Chem. Soc., Chem. Commun.,
1985, 23.
A. Merz and G. Mark, Angew. Chem., Int. Ed. Engl., 1973, 12, 845;
L. A. Yanovskaya, V. A. Dombrovsky, O. S. Chizhov, B. M.
Zolotarev, O. A. Subbotin and V. F. Kucherov, Tetrahedron, 1972,
28, 1565; D. N. Boykin, A. B. Turner and R. E. Lutz, Tetrahedron
Lett., 1967, 817; Y. D. Vankar, G. Kumaravel and C. T. Rao, Synth.
Commun., 1989, 19, 2181.
11 The substrate 59 was prepared from Simmons–Smith cyclo-
propanation of 1,3-diphenylpropene which, in turn, was made
available from the Wittig reaction of benzaldehyde and the
triphenylphosphorane prepared from 2-phenylethyl bromide.
However, see: R. Hollis, L. Hughes, V. W. Bowry and K. U. Ingold,
J. Org. Chem., 1992, 57, 4284.
8 For an asymmetric C–H insertion of
a
carbenoid into
tetrahydrofuran, see: H. M. L. Davies and T. Hansen, J. Am. Chem.
Soc., 1997, 119, 9075.
9 D. H. R. Barton, D. Crich and W. B. Motherwell, Tetrahedron, 1985,
41, 3901; D. Crich, Aldrichim. Acta, 1987, 20, 35; D. H. R. Barton,
D. Briden, I. F. Picot and S. Z. Zard, Tetrahedron, 1987, 43, 2733.
For an alternative mechanism, see: V. K. Yadav, A. Yadav, P. Pande
and K. K. Kapoor, Indian J. Chem., Sect. B, 1994, 33, 1129.
10 E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 1965, 87, 1353;
12 1,2-Diphenylethanol was prepared from the reaction of PhCHO
with PhCH2MgBr in Et2O following standard reaction conditions in
>90% yield.
13 1-Phenylcitronellol was prepared from the reaction of ( )-citronellal
with phenylmagnesium bromide in Et2O following standard reaction
conditions in >90% yield.
332
J. Chem. Soc., Perkin Trans. 1, 2001, 323–332