Mechanistic Aspects of the Phosphine-Catalyzed Isomerization of Allenic Sulfones to 2-Arylsulfonyl 1,3-Dienes

Article abstract of DOI:10.1021/acs.joc.5b02097

When an allenic sulfone is treated with a phosphine nucleophile and a proton shuttle, an isomerization to a 2-arylsulfonyl 1,3-diene occurs. Mechanistic aspects of the process were investigated leading to the formulation of a mechanism for the reaction. S

Full text of DOI:10.1021/acs.joc.5b02097

Article
pubs.acs.org/joc
Mechanistic Aspects of the Phosphine-Catalyzed Isomerization of
Allenic Sulfones to 2‑Arylsulfonyl 1,3-Dienes
Carissa S. Hampton and Michael Harmata*
Department of Chemistry, University of MissouriColumbia, Columbia, Missouri 65211, United States
S
* Supporting Information
ABSTRACT: When an allenic sulfone is treated with a phosphine nucleophile
and a proton shuttle, an isomerization to a 2-arylsulfonyl 1,3-diene occurs.
Mechanistic aspects of the process were investigated leading to the formulation of
a mechanism for the reaction. Some further optimization studies of this process
are reported.
Scheme 2. Proposed Mechanism for the Isomerization of an
Alkynyl Ester Catalyzed by a Phosphine
INTRODUCTION
■
In 1988, both the Trost and Lu groups simultaneously reported
the isomerization of α,β-alkynones to conjugated (E,E)-α,β:γ,δ-
dienones by palladium and ruthenium catalysis, respectively.1
This reaction was performed stereoselectively, resulting in the
formation E,E-dienones. The mechanism proposed for the
process suggested a hydridometal species and the formation of
an allene as an intermediate in the isomerization (Scheme 1).
The isomerization was initially thought to require transition-
metal catalysis to induce the rearrangement.
Scheme 1. Proposed Mechanism for the Isomerization of
Alkynones Catalyzed by a Metal−Hydride Complex
Scheme 3. Isomerization of Acetylenic Phosphorus
Compounds
isomerization could be differentiated into two types: a reaction
that is catalyzed by phosphines alone or a reaction in which a
transition metal is required. Thus, phosphines and transition
metals play a different role in the isomerization of acetylenic
compounds.6
During a study of the functionalization of allenic sulfones, we
discovered two new reactions.7 When an allenic sulfone was
treated under palladium catalysis, a 1-arylsulfonyl 1,3-diene was
formed, but when treated under nucleophilic catalysis, using a
phosphine in the presence of a proton transfer agent, a 2-
arylsulfonyl 1,3-diene was isolated (Scheme 4).
The isomerization of an allenic sulfone to a 1-arylsulfonyl
1,3-diene is similar to the isomerization of allenic and acetylenic
carbonyl compounds1a,8 and acetylenic phosphorus com-
pounds4,9 previously reported. The ability of these carbonyl
and phosphorus derivatives to isomerize solely by phosphine
catalysis is possible because these functional groups do not
behave as leaving groups. This is the key to the regiodivergent
In 1992, Trost and Kazmaier reported that the same
isomerization could be performed by phosphine catalysis
alone.2 The reaction was highly chemoselective, requiring an
alkyne conjugated to the carbonyl group. Observations
supported a mechanistic path involving a series of prototopic
shifts induced by nucleophilic addition of the phosphine to the
triple bond as presented in Scheme 2.3
A similar type of isomerization was demonstrated in the mid-
1990s in which acetylenic phosphorus compounds were
isomerized by palladium catalysis4 and then separately by
phosphine catalysis5 (Scheme 3). Mechanistic proposals were
similar to that shown above for the carbonyl compounds.
Guo and Lu’s reinvestigation of the isomerization of
acetylenic carbonyl compounds found that the catalytic
Received: September 7, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
reaction required a slightly longer time of 1.5 h to complete
(Table 1, entry 2). Lower and higher concentrations proved
detrimental to the yield of the process (Table 1).
Scheme 4. Regiodivergent Synthesis of 1- and 2-Arylsulfonyl
1,3-Dienes
Table 1. Concentration Effects on the Phosphine-Catalyzed
a
Conversion of 17 to 19
nature of our reactions. The sulfone functional group can act as
a leaving group,10 thereby generating a new product when an
allenic sulfone is treated with a nucleophile such as a phosphine
in the presence of a proton shuttle. However, prior to our work,
the isomerization to the 2-arylsulfonyl 1,3-diene had no
precedent in the literature to the best of our knowledge. We
recently reported the discovery of this reaction7 and herein
report our studies of the mechanism of this process and further
examination of its reaction parameters.
entry
conc (M)
time (h)
yield (%)
1
2
3
4
5
6
0.025
0.05
0.075
0.10
0.15
0.25
3
59
82
70
68
57
54
1.5
1.2
1
0.67
0.67
a
RESULTS AND DISCUSSION
The reactions were conducted with 20 mol % of catalyst and
cocatalyst.
■
Examination of Reaction Parameters. As we previously
described,7 the best catalyst system for the conversion of aryl
allenic sulfones to 2-arylsulfonyl 1,3-dienes was a mixture of
triphenylphosphine (20 mol %) as the nucleophile and phenol
(20 mol %) as a proton shuttle agent. The scope of this
reaction was examined with several substrates; the products
formed along with their yields are shown in Figure 1.7
While the isomerization proceeded with as little as 5 mol %
of triphenylphosphine (Table 2, entry 2), the reaction did not
Table 2. Effects of Catalyst Loading on the Phosphine-
Catalyzed Reaction
a
entry
catalyst loading (mol %)
time (h)
yield (%)
1
2
3
4
2.5
5.0
24
25
1.5
1
b
62
83
68
10.0
20
a
b
The reactions were conducted at 0.1 M. 17 was recovered in 73%
yield.
proceed at all at half that amount of catalyst (Table 2, entry 1).
However, the yield peaked at 10 mol % of phosphine and
phenol (Table 2, entry 3) with more catalyst resulting in a
lower yield of the target diene (Table 2, entry 4).
Next, the effect of solvent on the reaction was examined.
THF was the solvent we used initially and we found that it was
the best solvent for the nucleophilic isomerization among the
solvents we examined, giving 19 in 68% yield in 1 h (Table 3,
entry 1). The only other comparable solvent was tert-butyl
methyl ether, which afforded a 63% yield of 19 after reflux for 2
h at 55 °C. All other solvents either required longer reaction
times or afforded decreased yields of the product. It is also
noteworthy that this reaction is able to proceed in water with
no addition of a proton-transfer agent, as the water serves this
purpose. This is similar to the reports by Xue11 and Li12 of
triphenylphosphine-catalyzed reactions in aqueous solutions.
Under these conditions the reaction is complete within 3 h
(Table 3, entry 8) but the yield is low (42%). Using a mixed
solvent system of THF/H2O (1:8) to increase solubility, with
water as the proton-transfer agent in the absence of phenol,
gave a yield of 63% of the diene 19 in only 2 h at reflux
temperature (Table 3, entry 9).
Figure 1. 2-Arylsulfonyl-1,3-dienes prepared via isomerization.
The results shown in Figure 1 were obtained with a catalyst
loading of 20 mol %. We wondered if the process might be
improved and thus pursued limited optimization of the
reaction, using the triphenylphosphine and phenol reactants
that were initially successful using allene 17 as our substrate.
The control conditions employed 20 mol % of triphenylphos-
phine and phenol in THF at a 0.1 M concentration of allene
starting material and were varied from there.
We first examined the effect of concentration and catalyst
loading. Diluting the reaction concentration to 0.05 M (based
on 17) led to a significant increase in yield although the
During the course of these studies, the role of the
phosphorus catalyst was surveyed by using different trivalent
phosphorus compounds. In the phosphine-catalyzed isomer-
B
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
comparable to that of simple triphenylphosphine. Although
tricyclohexylphosphine is formally more nucleophilic than
triphenylphosphine from an electronic perspective, steric effects
appear to dominate in this case. The cone angle of PCy3 is
170°, while that of triphenylphosphine is 145°. Steric effects
thus slow the reaction significantly such that after 26 h 84% of
17 was recovered (Table 4, entry 3). The sterically hindered tri-
o-tolylphosphine with a cone angle of 194° also failed to
produce any diene product. In addition, less nucleophilic
trivalent phosphites such as triisopropyl phosphite failed to
catalyze the reaction, with 71% of 17 being recovered after 29 h
at reflux (Table 4, entry 4). These results correlate well with
Trost and Kazmaier’s work.2
Table 3. Solvent Effects on the Phosphine-Catalyzed
Reaction
a
entry
solvent
THF
temp (°C)
time (h)
yield (%)
1
2
3
4
5
6
7
8
9
65
80
1
68
53
35
52
63
57
39
42
63
toluene
7.5
1
b
DMF
80
c
DMC
90
2.5
2
d
TBME
MeCN
CH2Cl2
55
82
4.75
9
At the suggestion of a reviewer, we have also examined a
selection of proton shuttle agents. We observed that the more
acidic the proton shuttle source the longer the time required for
complete consumption of starting material and the lower the
yields (Table 5, entries 1 and 2). Phenols with an acidity similar
40
e
H2O
100
100
3
e,f
THF:H2O
2
a
The reactions were conducted with 20 mol % of triphenylphosphine
b
c
and phenol at 0.1 M. DMF = N,N-dimethylformamide. DMC =
dimethyl carbonate. TBME = tert-butyl methyl ether. No phenol was
used in this reaction; water is the proton-transfer agent. THF/H2O =
d
e
f
Table 5. Examination of Proton Shuttle Agents
1:8.
izations of acetylenic carbonyls, the proposed mechanisms
show the addition of the phosphine to the acetylene followed
by proton transfer with an acidic compound such as phenol.3
The phosphorus catalyst must be chosen in consideration of
both steric and electronic effects.13 The nature of steric and
electronic effects of trivalent phosphine ligands was detailed in
a review by Tolman in 1977.13 The steric environment of a
trivalent phosphorus ligand could be evaluated by determining
the approximate amount of “space” the ligand consumes about
a metal center, referred to as the cone angle.13,14 The electronic
properties of phosphorus ligands could be ranked on the basis
of the change in the CO stretching frequency of mono-
substituted transition metal carbonyls, which can be quanti-
fied.13,15
entry
proton shuttle
AcOH
time (h)
yield (%)
1
2
3
4
5
6
7
5
15
1
31
43
68
75
66
56
54
4-(NO2)C6H4OH
PhOH
2,6-di-MeC6H3OH
3,5-di-MeC6H3OH
4-(OMe)C6H4OH
1.3
1.3
4.5
4
a
BHT
a
BHT = 2,6-di-tert-butyl-4-methylphenol.
to that of phenol, such as 3,5-dimethylphenol and the slightly
more hindered 2,6-dimethylphenol, produced the diene in
similar yields in relatively the same amount of time (Table 5,
entries 4−5). In addition, the use of a sterically hindered
phenol significantly slowed the reaction and lowered the yield
(Table 5, entry 7).
After these studies with 17, the best conditions included a 10
mol % loading of triphenylphosphine and phenol at 0.1 M in
THF at reflux. This improved the yield of 19 to 83% as
compared to 68% in our original work (Scheme 5).7
The results of using different trivalent phosphorus catalysts
are shown in Table 4. When the more nucleophilic tri-n-
Table 4. Effect of the Phosphine/Phosphite on the
Nucleophilic Reaction
entry
phosphine(ite)
PPh3
time
yield (%)
Scheme 5. Optimum Reaction Conditions for the
Phosphine-Catalyzed Formation of 19
1
2
3
4
5
3
4
1 h
68
40
77
84
a
PBu3
12 min
15 min
20 min
29 h
P(p-(OMe)C6H4)3
P(p-tol)3
P(o-tol)3
PCy3
26 h
b
P(Oi-Pr)3
29 h
c
a
b
c
17 was recovered in 50% yield. 17 was recovered in 84% yield. 17
While these reaction conditions appeared ideal for 17, it is
likely that the other dienes would also experience an increase in
yield when treated under these reaction conditions. Therefore,
a few allenic sulfones were reacted, and the results are shown in
Table 6. It was seen that the results were varied. For dienes 19
and 29 the yield was increased by ∼20%, whereas for 28 the
yield was almost the same as in our earlier study, and for the
heteroaromatic 24, the yield actually decreased slightly relative
to our original results. This outcome is not unexpected because
was recovered in 71% yield.
butylphosphine was used, the reaction was complete within 12
min but with a large sacrifice in yield (Table 4, entry 2). It is
possible that oligomeric products were formed as in the report
by Trost and Kazmaier,2 but these were not pursued. The
reactions employing tris(p-methoxyphenyl)phosphine and tri-
p-tolylphosphine were completed quickly and in yields
C
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
phosphine to the electrophilic unsaturation and, through a
series of proton shifts and elimination, results in an isomer-
ization. Addition of phosphines to unsaturated bonds has also
been used to catalyze other reactions as well.19 A similar type of
addition has been reported in the sulfonylation of an
unsaturated bond.20
Table 6. Comparison of Reaction Conditions
In due consideration of what was known in the literature, we
proposed the mechanism shown in Scheme 7 for the
entry
product
conditions
time
yield (%)
1
2
3
4
5
6
7
8
19
19
24
24
28
28
29
29
A
B
A
B
A
B
A
B
1h
68
83
71
61
68
64
63
83
Scheme 7. Proposed Mechanism of the Phosphine-Catalyzed
Reaction
1.5h
35 min
1.5h
15 min
50 min
45 min
1.5h
these conditions were optimized for a single allene example and
there may be other factors that influence the reaction for other
allenes.
It should also be noted that this reaction can proceed in the
presence of sulfinate anions and a proton shuttle without
phosphine to initiate the reaction. The competition between
phenol and p-toluenesulfinate as nucleophiles and the
insolubility of the sodium sulfinate salt in the THF solvent
led to a drastic reduction in the rate of reaction. This problem
was presumably exacerbated due to the depletion of the proton
shuttle, the phenol, which was consumed via nucleophilic attack
on the allene, resulting in the formation of 30. As a result, while
some of the diene 19 was formed, the vinyl ether 30 and other
unidentifiable products were produced as well. All of these side
products had the same Rf value on TLC and could not be
separated by column chromatography and were thus not
rigorously characterized. This process improved when BHT
was used in place of phenol. Steric factors reduced the tendency
for nucleophilic attack on the allene, thus allowing the BHT to
act as a simple proton shuttle. The reaction could be completed
within 36 h in this case, but at only a 60% yield (Scheme 6).
phosphine-catalyzed generation of 2-arylsulfonyl 1,3-dienes
from allenic sulfones. For the formation of 19, nucleophilic
addition of triphenylphosphine to 17 produces the sulfonyl-
stabilized carbanion 31, which then deprotonates phenol to
produce 32 (Scheme 7). Elimination of a p-tolylsulfinate anion
to afford the phosphonium salt 33 is achieved by deprotonation
of 32 by phenoxide. This is the critical step in generating the
small amounts of p-tolylsulfinate ion required to produce
19.20d,e,21 Thus, nucleophilic addition of the p-tolylsulfinate
anion to 17 produces 34. This is followed by a protonation/
deprotonation sequence (the proton shuttle) that leads to 19
(Scheme 7).
Scheme 6. Sulfinate-Promoted Generation of a 2-
Arylsulfonyl 1,3-Diene
Crossover Experiments. In order to support our proposed
mechanism of the phosphine-catalyzed reaction, a mixture of
two allenes with different aryl groups and different alkyl
substituents, chosen so that we could easily differentiate the
products, was treated under standard reaction conditions with
triphenylphosphine and phenol. An intramolecular sulfone
transposition should lead to only two products, but a mixture of
four products should be obtained if free sulfinate anions are
generated during the course of the transformation as we had
proposed, assuming they were productively involved in
generating the products. In the event, the reaction of a 1:1
mixture of 36 and 37 afforded a mixture of all four possible 2-
arylsufonyl 1,3-dienes (Scheme 8A). The percentages represent
Mechanistic Studies. A sulfone group can act as both a
leaving group and a nucleophile, whereas a carbonyl group
cannot. The phosphine-catalyzed reaction that we discovered is
thus a beautiful example of the “chameleon” nature of the
sulfone group.8,10,16 In contrast to the 1-substituted-1,3-diene
obtained from alkynes or allenes bearing functional groups such
as carbonyls2,3,6,17 or pentavalent phosphorus functionalities,5,18
when an allenic sulfone is treated under phosphine catalysis in
the presence of a proton-transfer agent, a 2-arylsulfonyl 1,3-
diene is generated. The phosphine-catalyzed isomerization of
acetylenic and allenic carbonyls and acetylenic and allenic
phosphorus compounds is initiated by addition of the
1
percent composition based on H NMR analysis of the crude
reaction mixture (compared with spectra of the pure
compounds prepared individually), not absolute yields, as the
dienes were inseparable. This result implies that after conjugate
addition of triphenylphosphine to the allene a sulfinate anion is
liberated. This may attack other allene molecules liberating
more sulfinate anions and generating the 2-arylsulfonyl 1,3-
D
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
30 min in an attempt to isolate a phosphonium phenoxide salt
Scheme 8. Nucleophilic Crossover Reactions
of 32 or a phosphonium sulfinate salt of 33 (Scheme 10).
Scheme 10. Attempted Isolation of a Reaction Intermediate
A crude 31P NMR of the phosphine/phenol-catalyzed
reaction showed three different phosphorus peaks (Figure 2).
a
Percentages represent composition, not isolated yields.
dienes. This process was also examined with another mixture of
allenes, 17 and 38 (Scheme 8B). Again, all four possible diene
products were generated in this crossover reaction.
Since we have shown that the reaction could proceed in the
presence of p-toluenesulfinate hydrate and BHT,7 this would
imply that the sulfinate ions necessary to advance the reaction
must come as a result of expulsion after the addition of the
sulfinate anion to the allene.20d,e,21,22 Therefore, if a sulfinate
that differed from the sulfone group on the allene was
employed, two different diene products should be produced,
one containing a sulfone group from the added sulfinate anion
and the other containing the sulfone group initially on the
allene. To examine this hypothesis, the reaction of 37 with
sodium p-toluenesulfinate hydrate was performed and did
indeed lead to both dienyl products 19 and 24 (Scheme 9).
Figure 2. Crude 31P NMR of the phosphine-catalyzed reaction.
The peak at −4.75 ppm corresponds to the literature value of
triphenylphosphine,23 and the peak at 29.7 ppm corresponds to
the literature value of triphenylphosphine oxide.24 At 22.43
ppm, there is a very small phosphorus peak that we have
assigned to either 32 or the vinyl phosphonium ion 33. This
chemical shift compares very closely to the vinyl phosphonium
salts prepared by Arisawa and Yamaguchi in their report of the
metal-catalyzed addition of triphenylphosphine and methane-
sulfonic acid to alkynes.25 For example, they reported that
treatment of 39 with triphenylphosphine, methanesulfonic acid,
and tetrakis(triphenylphosphine)palladium (0), followed by
anion exchange with lithium hexafluorophosphate, generated
the vinyl phosphonium salt 40 (Scheme 11). This compound
Scheme 9. Sulfinate Crossover Reaction
Scheme 11. Preparation of Vinyl Phosphonium Salts from
Alkynes
After 3 days the reaction was still not complete, so it was
stopped and the components were analyzed. On the basis of
the amount of p-toluenesulfinate used, 19 was formed in 94%
yield, and based on the amount of 37 recovered, 24 was formed
in 68% yield.
Reaction Intermediates. In an attempt to identify
intermediates in the proposed mechanism, we wanted to
pursue the isolation and/or identification of phosphonium salts
(i.e., 32) from the left-hand side of the proposed mechanism
(Scheme 7). In order for this to work, once the phosphine adds
to the allene further reaction must be avoided. To this end, 1
equiv of phosphine was added to a solution of 17 in the
presence of 10 mol % of phenol. The reaction was stopped after
displays a peak at 24.3 ppm in the 31P NMR. Other examples of
similar reactions generate vinyl phosphonium salts with
phosphorus peaks in the range of 13.7−25.2 ppm. Therefore,
our proposal here is reasonable. Even in the shortened reaction
time in Scheme 10, the small amount of phenol in the solution
allows the reaction to continue and diene formation to occur.
All attempts to isolate the phosphonium salts 32 and/or 33
were unfruitful.
E
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
During the reaction shown in Scheme 10, a compound much
more polar than 17 or 19 was visualized by TLC analysis. This
compound was isolated in 12% yield and was determined to be
the disulfone 41, a proposed intermediate in the right-hand side
of the proposed mechanism (Scheme 7). The isolation of 41
was interesting, but we needed to be sure that this intermediate
could lead to dienyl products. To examine this possibility, 41
could be prepared directly by the reaction of 17 with sodium p-
toluenesulfinate and p-toluenesulfonic acid in a yield of 88%.
This disulfone could then be converted to diene 19 by
treatment with in situ-generated sodium phenoxide (from
phenol and sodium hexamethyldisilylamide) in THF in 20 min
in 91% yield (Scheme 12).
Scheme 14. Crossover of Phosphonium Salt 42 with Allene
37
Scheme 12. Preparation of 41 and Its Conversion to 19
anion to be expelled from the phosphonium salt and to add to
other allenes present in solution to produce dienes. The sparse
solubility of potassium carbonate in THF may be the cause of
the long reaction time, and it is possible that the reaction is
taking place on the surface of the potassium carbonate. We
therefore used sodium hexamethyldisilazide, a soluble base, to
deprotonate the phenol to promote the reaction of the
phosphonium salt to dienyl products (Scheme 14). This
proved much more effective in rapidly producing dienyl
products. We therefore conclude that the phosphonium salt
42 is a viable intermediate in the reaction pathway leading to
the diene products that are observed and that it is likely to be
an intermediate in the formation of diene 19 as detailed in
Scheme 7.
In an effort to prepare a vinyl phosphonium salt that was
isolable, we developed a process similar to that performed by
Arisawa and Yamaguchi but that did not require a precious
metal.25 Nucleophilic addition of triphenylphosphine to an
electrophilic allene could lead to a salt if the initial adduct could
be quenched. We reasoned that the use of p-toluenesulfonic
acid would interrupt the reaction because the tosylate conjugate
base would not be basic enough to deprotonate a compound
like 32 (see Scheme 7) at the methyl group. This would
prevent any further reaction so that formation of a
phosphonium sulfonate salt would be possible. We thus treated
17 with 1 equiv of triphenylphosphine in the presence of 1
equiv of p-toluenesulfonic acid. The crude product from this
reaction was then treated with potassium hexafluorophosphate
in ethanol to effect anion exchange upon which a white solid
precipitated (Scheme 13). This was the phosphonium salt 42,
CONCLUSION
■
The evidence we have obtained supports the proposed
nucleophile-catalyzed mechanism for the formation of 2-
arylsulfonyl 1,3-dienes from the corresponding allenic sulfones.
Crossover experiments support the generation of free sulfinate
anions in the process. The fact that both 41 and 42 lead to
diene products supports the idea that they are intermediates in
the process. 31P NMR evidence supports the generation of 32
and/or 33 as intermediates in the catalytic cycle. Although we
have not pursued it, we note that the synthesis of 42 represents
a potentially general way of producing vinylphosphonium salts
in a straightforward fashion without the need for transition-
metal catalysis.
Scheme 13. Formation of the Phosphonium Salt 42
EXPERIMENTAL SECTION
■
All reactions were carried out in oven-dried glassware and cooled in a
desiccator prior to use. Tetrahydrofuran (THF) was purchased and
distilled from sodium benzophenone prior to use. The starting allenes
were prepared by methods akin to those introduced by Toru,
Sharpless, Braverman, and Harmata.26 All reactions were monitored by
thin-layer chromatography (TLC) on glass-backed silica gel plates with
fluorescent UV indicator. Flash chromatography was carried out using
230−400 mesh silica gel with HPLC-grade solvents. Unless stated
otherwise, all commercially available reagents were used as received.
which was identified by NMR and an X-ray crystal structure. At
room temperature, the methylene group of 42 has a hindered
rotation and the peaks are broadened so significantly in the 1H
NMR that they almost blend in with the baseline. If the probe
is cooled to 253 K, the peaks can be resolved, each to a broad
doublet of doublets with different chemical shifts. The 31P
NMR chemical shift of 42 is 23.6 ppm for the phosphonium
ion, which also correlates well with Arisawa and Yamaguchi’s
vinyl phosphonium salts as well as the peak proposed to be
either 32 or 33 in the spectrum presented in Figure 2.
If 42 is an intermediate in the overall process, it should lead
to a dienyl product under the appropriate conditions. Thus, the
phosphonium salt 42 was treated with potassium phenoxide,
generated in situ from phenol and potassium carbonate; then
37 was added to the reaction and the mixture was stirred for 6
days at reflux (Scheme 14). Both 19 and 24 were formed in the
reaction, indicating that it is possible for the p-toluenesulfinate
All compounds were characterized by H and 13C nuclear magnetic
1
resonance (NMR) spectroscopy using either a 300, 500, or 600 MHz
spectrometer. Proton spectra were reported in δ units, parts per
million (ppm), relative to trimethylsilane internal standard (0.00
ppm). Carbon spectra were reported in ppm relative to deuterated
chloroform (77.0 ppm). 31P NMR spectra were obtained on a 250
MHz spectrometer at 101.25 MHz in CDCl3 with H3PO4 (δ 80.0
ppm) as an external reference. Melting points were determined with a
Fisher-Johns melting point apparatus and are uncorrected. Infrared
spectra were recorded on a FT-IR spectrometer. High-resolution mass
spectra were acquired on a FTICR-MS with ESI.
F
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Procedure for Isomerization. To an oven-dried round-bottom
flask with an argon balloon was added the catalyst (phosphorus
reagent or sulfinate salt, 20 mol % unless otherwise stated), co-catalyst
(phenol, 20 mol % unless otherwise stated), and dry solvent (0.1 M,
unless otherwise stated). Then the allenic sulfone was added to the
reaction flask and heated to reflux. The reaction was monitored by
TLC (15% EtOAc/hexanes). Either upon reaction completion or
prolonged reaction time with no change in TLC, the reaction was
cooled to room temperature. Water was added to the reaction flask,
and the mixture was extracted three times with CH2Cl2. The organic
layers were combined and washed with water and brine. The solution
was dried over anhydrous sodium sulfate, filtered, and concentrated by
rotary evaporation. The crude residue was purified by column
chromatography (5% EtOAc/hexanes) to isolate 19.
to room temperature, diluted with CH2Cl2, and washed with brine.
The organic layer was dried over anhydrous sodium sulfate,
concentrated by rotary evaporation, and purified by column
chromatography (10% EtOAc/hexanes) to yield 19 in 91%.
Formation of Phosphonium salt 42. (3-Methyl-1-tosylbut-2-
en-2-yl)triphenylphosphonium Hexafluorophosphonate (42). In an
oven-dried round-bottom flask were dissolved 1-methyl-4-((3-
methylbuta-1,2-dien-1-yl)sulfonyl)benzene 17 (0.500 g, 2.249
mmol), triphenylphosphine (0.590 g, 2.249 mmol), and p-
toluenesulfonic acid hydrate (0.387 g, 2.249 mmol) in dry THF
(22.5 mL), and the mixture was heated to reflux for 9 h until the
triphenylphosphine was no longer visible by TLC analysis. The
reaction was then cooled to room temperature and concentrated by
rotary evaporation. Crude NMR was acquired. The crude reside was
then dissolved in absolute ethanol (22.5 mL), and potassium
hexafluorophosphate (0.621 g, 3.374 mmol) was added. The reaction
was stirred at room temperature for 4 h (a precipitate began forming at
2 h). The reaction was filtered over filter paper and washed with
absolute ethanol. The solid material was dissolved in chloroform and
filtered over filter paper, rinsing with chloroform. The filtrate was
concentrated by rotary evaporation to yield the phosphonium salt 42
Procedure for Nucleophilic Crossover Experiments. In an
oven-dried round-bottom flask with an argon balloon were dissolved
phenol (0.011 g, 0.120 mmol, 20 mol %) and triphenylphosphine
(0.032 g, 0.120 mmol, 20 mol %) in dry THF (6.0 mL). Then 1-
methyl-4-((3-methylbuta-1,2-dien-1-yl)sulfonyl)benzene 17 (0.067 g,
0.300 mmol) and 2-((2-cyclohexylidenevinyl)sulfonyl)thiophene 38
(0.076 g, 0.300 mmol) were added to the reaction flask and heated to
reflux. After 45 min at reflux, the reaction was cooled to room
temperature. Water was added to the reaction flask, and the mixture
was extracted with CH2Cl2 three times. The organic layers were
combined and washed with dilute NaOH, water, and brine. The
solution was dried over anhydrous sodium sulfate, filtered, and
concentrated by rotary evaporation. The crude mixture was analyzed
1
(1.07 g, 75%) as a white solid (mp = 212−213 °C): H NMR (300
MHz, CDCl3, at 253 K) δ 7.96−7.65 (m, 15H), 7.32−7.20 (m, 4H),
4.71 (t, J = 18.6 Hz, 1H, diastereotopic CH2), 3.52 (t, J = 13.5 Hz, 1H,
diastereotopic CH2), 2.55 (d, J = 1.5 Hz, 3H), 2.40 (s, 3H), 1.79 (d, J
= 1.5 Hz, 3H); 13C NMR (125 MHz, CDCl3) δ 174.6 (m), 145.5,
135.6, 134.8, 134.6 (d, J = 10.0 Hz), 130.3 (d, J = 12.5 Hz), 130.1,
127.4, 119.0 (d, J = 87.5 Hz), 101.9 (d, J = 87.5 Hz), 56.4 (d, J = 13.8
Hz), 29.3 (d, J = 8.8 Hz), 26.6 (d, J = 12.5 Hz), 21.6; 31P NMR
(101.25 MHz, CDCl3) δ −143.8 (PF6, septet, J = 712.8 Hz), 23.6; IR
(cm−1) 3060, 3019, 2966, 2921, 1593, 1487, 1434, 1401, 1311, 1213,
1144, 1103, 1037, 997, 829, 751; HRMS m/z calcd for (C30H30O2PS)+
485.1698, found 485.1695.
1
by H NMR.
Procedure for Sulfinate Crossover Experiment. In an oven-
dried round-bottom flask with an argon balloon were dissolved BHT
(2,6-di-tert-butyl-4-methylphenol, 0.021 g, 0.093 mmol, 20 mol %)
and sodium p-toluenesulfinate (0.027 g, 0.140 mmol, 30 mol %) in dry
THF (9.3 mL). Then 2-((3-methylbuta-1,2-dien-1-yl)sulfonyl)-
thiophene 37 (0.100 g, 0.467 mmol) was added to the reaction flask
and the mixture heated to reflux. After 3 days at reflux, the reaction
was cooled to room temperature. Water was added to the reaction
flask, and the mixture was extracted with CH2Cl2 three times. The
organic layers were combined and washed with dilute NaOH, water,
and brine. The solution was dried over anhydrous sodium sulfate,
filtered, and concentrated by rotary evaporation. The crude product
was purified by flash column chromatography (5% EtOAc/hexanes) to
yield a mixture of 1.0:0.9 mixture of 24 (68% brsm)/19 (94% based on
TsNa) and recovered starting material (0.024 g, 24%).
4,4′-(3-Methylbut-2-ene-1,2-diyldisulfonyl)bis(methylbenzene)
(41). In an oven-dried round-bottom flask were dissolved 1-methyl-4-
((3-methylbuta-1,2-dien-1-yl)sulfonyl)benzene 17 (0.100 g, 0.4498
mmol), p-toluenesulfonic acid monohydrate (0.0801 g, 0.4498 mmol),
and sodium p-toluene sulfinate hydrate (0.0736 g, 0.4498 mmol) in
THF (4.5 mL) and the mixture heated to reflux for 5 h until complete
consumption of starting material by TLC (25% EtOAc/hexanes).
Water was added to the reaction flask, and the mixture was extracted
with CH2Cl2 three times. The organic layers were combined and
washed with brine. The solution was dried over anhydrous sodium
sulfate, filtered, and concentrated by rotary evaporation. The crude
product was purified by flash column chromatography (15% EtOAc/
hexanes) to yield 41 in 88%: 1H NMR (500 MHz, CDCl3) δ 7.88 (d, J
= 8.5 Hz, 2H), 7.87 (d, J = 8.0 Hz, 2H), 7.38 (d, J = 8.0 Hz, 2H), 7.30
(d, J = 8.5 Hz, 2H), 4.54 (s, 2H), 2.46 (s, 3H), 2.42 (s, 3H), 2.11 (s,
3H), 2.04 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 158.7, 145.1,
144.0, 139.2, 136.6, 129.9, 129.6, 128.5, 127.7, 127.6, 56.1, 25.9, 23.4,
21.7, 21.6; IR (cm−1) 3023, 2919, 1616, 1592, 1493, 1441, 1398, 1318,
1298, 1219, 1183, 1135, 1084, 817, 754, 714, 659; HRMS m/z calcd
for (C19H22O4S2)Na+ 401.0852, found 401.0849.
Procedure for Crossover of 42 with 37. In a oven-dried round-
bottom flask was dissolved phenol (0.003 g, 0.033 mmol) in dry THF
(2 mL), sodium hexamethyldisilylamide (0.032 mL, 0.032 mmol, 1.0
M in THF) was added, and the mixture was stirred for 20 min. Then a
solution of 42 (0.064 g, 0.102 mmol) in dry THF (0.5 mL) was added
to the reaction flask and the mixture heated to reflux for 20 min. The
solution was cooled slightly, and a solution of 37 (0.034 g, 0.159
mmol) in dry THF (0.5 mL) was added. The reaction was then heated
to reflux again and monitored by TLC (20% EtOAc/hexanes). The
reaction was cooled to room temperature, diluted with CH2Cl2, and
washed with brine. The organic layer was dried over anhydrous
sodium sulfate, concentrated by rotary evaporation, and purified by
column chromatography (7−10% EtOAc/hexanes) to yield an
1
inseparable mixture of 24:19 that was analyzed by H NMR.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02097.
NMR data for compounds 41 and 42 and NMR of
reaction mixtures of crossover experiments (PDF)
X-ray crystal structure data of 42 (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: harmatam@missouri.edu.
■
Notes
Conversion of 41 to 19. In a oven-dried round-bottom flask was
dissolved phenol (0.0373 g, 0.396 mmol) in dry THF (3.5 mL),
sodium hexamethyldisilylamide (0.037 mL, 0.0376 mmol, 1.0 M in
THF) was added, and the mixture was stirred for 20 min. Then a
solution of 41 (0.150 g, 0.396 mmol) in dry THF (0.5 mL) was added
to the reaction flask and the mixturew heated to reflux for 20 min and
monitored by TLC (25% EtOAc/hexanes). The reaction was cooled
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Department of Chemistry at
the University of MissouriColumbia and the National
Science Foundation (CHE-1463724). We thank Dr. Charles
G
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
L. Barnes (University of MissouriColumbia) for acquisition
of X-ray data.
REFERENCES
■
(1) (a) Ma, D.; Lin, Y.; Lu, X.; Yu, Y. Tetrahedron Lett. 1988, 29,
1045−1048. (b) Trost, B. M.; Schmidt, T. J. Am. Chem. Soc. 1988, 110,
2301−2303.
(2) Trost, B. M.; Kazmaier, U. J. Am. Chem. Soc. 1992, 114, 7933−
7935.
(3) Rychnovsky, S. D.; Kim, J. J. Org. Chem. 1994, 59, 2659−2660.
(4) Ma, C.; Lu, X.; Ma, Y. Main Group Met. Chem. 1995, 18, 391−
397.
(5) Ma, C. L.; Lu, X. Y.; Ma, Y. X. Chin. Chem. Lett. 1995, 6, 747−
750.
(6) Guo, C.; Lu, X. J. Chem. Soc., Perkin Trans. 1 1993, 1921−1923.
(7) Hampton, C. S.; Harmata, M. Org. Lett. 2014, 16, 1256−1259.
(8) (a) Inoue, Y.; Dmaizumi, S. J. Mol. Catal. 1988, 49, L19−L21.
(b) Lu, X.; Ma, D. Pure Appl. Chem. 1990, 62, 723−730. (c) Ma, D.;
Yu, Y.; Lu, X. J. Org. Chem. 1989, 54, 1105−1109.
(9) Srebnik, M.; Azab, A.; Quntar, A. A. A.; Antebi, T. Heterocycles
2010, 82, 417−429.
(10) Trost, B. M. Bull. Chem. Soc. Jpn. 1988, 61, 107−124.
(11) Zhou, Q. F.; Yang, G.; Guo, Q. X.; Xue, S. Chin. Chem. Lett.
2007, 18, 1029−1032.
(12) Skouta, R.; Varma, R. S.; Li, C.-J. Green Chem. 2005, 7, 571−
575.
(13) Tolman, C. A. Chem. Rev. 1977, 77, 313−348.
(14) Tolman, C. A. J. Am. Chem. Soc. 1970, 92, 2956−2965.
(15) (a) Strohmeier, W.; Muller, F. J. Chem. Ber. 1967, 100, 2812−
̈
2821. (b) Tolman, C. A. J. Am. Chem. Soc. 1970, 92, 2953−2956.
(c) Kuhl, O. Coord. Chem. Rev. 2005, 249, 693−704.
̈
(16) El-Awa, A.; Noshi, M. N.; du Jourdin, X. M.; Fuchs, P. L. Chem.
Rev. 2009, 109, 2315−2349.
(17) (a) Fu, M. Y.; Guo, J.; Toy, P. H. Synlett 2011, 2011, 989−991.
(b) Kwong, C. K.-W.; Fu, M. Y.; Lam, C. S.-L.; Toy, P. H. Synthesis
2008, 2008, 2307−2317. (c) Lu, X.; Zhang, C.; Xu, Z. Acc. Chem. Res.
2001, 34, 535−544. (d) Zhou, Q. F.; Yang, F.; Guo, Q. X.; Xue, S.
Chin. Chem. Lett. 2007, 18, 1029−1032.
(18) (a) Pudovik, A. N.; Aladzheva, I. M. Zh. Obshch. Khim. 1963, 33,
3096−3100. (b) Pudovik, A. N.; Aladzheva, I. M. Zh. Obshch. Khim.
1963, 33, 708−709.
(19) (a) Bull, J. R.; Desmond-Smith, N. S.; Heggie, S. J.; Hunter, R.;
Rees-Jones, S. C. M. ARKIVOC 2002, 55−61. (b) Lu, C.; Lu, X.
Tetrahedron 2004, 60, 6575−6579. (c) Xie, P.; Lai, W.; Geng, Z.;
Huang, Y.; Chen, R. Chem. - Asian J. 2012, 7, 1533−1537.
(20) (a) Back, T. G.; Lai, E. K. Y.; Muralidharan, K. R. J. Org. Chem.
1990, 55, 4595−4602. (b) Flasz, J. T.; Hale, K. J. Org. Lett. 2012, 14,
3024−3027. (c) Padwa, A.; Gareau, Y.; Harrison, B.; Rodriguez, A. J.
Org. Chem. 1992, 57, 3540−3545. (d) Padwa, A.; Yeske, P. E. J. Am.
Chem. Soc. 1988, 110, 1617−1618. (e) Padwa, A.; Yeske, P. E. J. Org.
Chem. 1991, 56, 6386−6390. (f) Plobeck, N. A.; Baeckvall, J. E. J. Org.
Chem. 1991, 56, 4508−4512.
(21) (a) Padwa, A.; Bullock, W. H.; Dyszlewski, A. D. J. Org. Chem.
1991, 56, 3556−3564. (b) Padwa, A.; Gareau, Y.; Harrison, B.;
Norman, B. H. J. Org. Chem. 1991, 56, 2713−2720.
(22) (a) Altenhofer, E. F.; Harmata, M. Tetrahedron Lett. 2015, 56,
3176−3178. (b) Knight, D. J.; Lin, P.; Whitham, G. H. J. Chem. Soc.,
Perkin Trans. 1 1987, 2707−2713.
(23) Allen, D. W.; Taylor, B. F. J. Chem. Soc., Dalton Trans. 1982,
51−54.
(24) Bemi, L.; Clark, H. C.; Davies, J. A.; Fyfe, C. A.; Wasylishen, R.
E. J. Am. Chem. Soc. 1982, 104, 438−445.
(25) Arisawa, M.; Yamaguchi, M. J. Am. Chem. Soc. 2000, 122, 2387−
2388.
(26) (a) Braverman, S.; Pechenick, T.; Zafrani, Y. Tetrahedron Lett.
2001, 42, 1391−1393. (b) Harmata, M.; Huang, C. Adv. Synth. Catal.
2008, 350, 972−974. (c) Klunder, J. M.; Sharpless, K. B. J. Org. Chem.
1987, 52, 2598−602. (d) Watanabe, Y.; Mase, N.; Tateyama, M.-a.;
Toru, T. Tetrahedron: Asymmetry 1999, 10, 737−745.
H
DOI: 10.1021/acs.joc.5b02097
J. Org. Chem. XXXX, XXX, XXX−XXX