Fluorinated alcohols as promoters for the metal-free direct substitution reaction of allylic alcohols with nitrogenated, silylated, and carbon nucleophiles

Article abstract of DOI:10.1021/jo301049w

The direct allylic substitution reaction using allylic alcohols in 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) and 2,2,2-trifluoroethanol (TFE) as reaction media is described. The developed procedure is simple, works under mild conditions (rt, 50 and 70 °C), and proves to be very general, since different nitrogenated nucleophiles and carbon nucleophiles can be used achieving high yields, especially when HFIP is employed as solvent and aromatic allylic alcohols are the substrates. Thus, sulfonamides, carbamates, carboxamides, and amines can be successfully employed as nitrogen-based nucleophiles. Likewise, silylated nucleophiles such as trimethylsilylazide, allyltrimethylsilane, trimethylsilane, and trimethylsilylphenylacetylene give the corresponding allylic substitution products in high yields. Good results for the Friedel-Crafts adducts are also achieved with aromatic compounds (phenol, anisole, indole, and anilines) as nucleophiles. Particularly interesting are the results obtained with electron-rich anilines, which can behave as nitrogenated or carbon nucleophiles depending on their electronic properties and the solvent employed. In addition, 1,3-dicarbonyl compounds (acetylacetone and Meldrum's acid) are also successfully employed as soft carbon nucleophiles. Studies for mechanism elucidation are also reported, pointing toward the existence of carbocationic intermediates and two working reaction pathways for the obtention of the allylic substitution product.

Full text of DOI:10.1021/jo301049w

Article
pubs.acs.org/joc
Fluorinated Alcohols As Promoters for the Metal-Free Direct
Substitution Reaction of Allylic Alcohols with Nitrogenated, Silylated,
and Carbon Nucleophiles
Paz Trillo, Alejandro Baeza,* and Carmen Najera*
́
́ ́
Departamento de Química Organica, Facultad de Ciencias, and Instituto de Síntesis Organica (ISO), Universidad de Alicante, Apdo
99, 03080 Alicante, Spain
S
* Supporting Information
ABSTRACT: The direct allylic substitution reaction using allylic alcohols in
1,1,1,3,3,3-hexafluoroisopropanol (HFIP) and 2,2,2-trifluoroethanol (TFE)
as reaction media is described. The developed procedure is simple, works
under mild conditions (rt, 50 and 70 °C), and proves to be very general,
since different nitrogenated nucleophiles and carbon nucleophiles can be
used achieving high yields, especially when HFIP is employed as solvent and
aromatic allylic alcohols are the substrates. Thus, sulfonamides, carbamates,
carboxamides, and amines can be successfully employed as nitrogen-based
nucleophiles. Likewise, silylated nucleophiles such as trimethylsilylazide,
allyltrimethylsilane, trimethylsilane, and trimethylsilylphenylacetylene give
the corresponding allylic substitution products in high yields. Good results
for the Friedel−Crafts adducts are also achieved with aromatic compounds (phenol, anisole, indole, and anilines) as nucleophiles.
Particularly interesting are the results obtained with electron-rich anilines, which can behave as nitrogenated or carbon
nucleophiles depending on their electronic properties and the solvent employed. In addition, 1,3-dicarbonyl compounds
(acetylacetone and Meldrum’s acid) are also successfully employed as soft carbon nucleophiles. Studies for mechanism
elucidation are also reported, pointing toward the existence of carbocationic intermediates and two working reaction pathways
for the obtention of the allylic substitution product.
In the past years, the intermolecular allylic substitution
reaction (and particularly the direct amination reaction) of
allylic alcohols has emerged as a straightforward and environ-
mentally friendly way to get access to allylic amines.6 Our
group, in the search for new and efficient strategies for this
purpose, has recently reported the use of Lewis and Brønsted
acids, such as [(PhO)3P]AuOTf,7a AgOTf,7a FeCl3·6H2O,7b
and TfOH,7bas catalysts able to successfully accomplish this
transformation. In this sense, in the search for a metal-free
strategy to carry out this process, we envisage the possibility of
the use of fluorinated alcohols as effective reaction media able
to activate the hydroxy functionality of allylic alcohols through
hydrogen-bonding, hence promoting the direct allylic nucleo-
philic substitution.8
INTRODUCTION
■
The unique chemical and physical properties of fluoroalkyl
alcohols, which possess a high hydrogen bond donor ability
along with a low nucleophilicity and high polarity and ionizing
power values, have drawn attention about the use of these
particular molecules in organic chemistry transformations
mainly as additives in catalyzed reactions or as solvents.1,2 In
this latter sense and as a result of the mentioned properties,
they have been traditionally used to solubilize those molecules
that are not soluble in the most common organic solvents,
especially in biochemistry when working with peptides and
nucleic acids and in the polymer industry for polyamides and
polyacrylonitriles. In addition, fluoroalkyl alcohols turned out
to be more effective than traditionally employed solvents,
including their nonfluorinated analogues, in some organic
synthesis processes. Thus, results were considerably enhanced
by using these solvents in some oxidation (especially in
epoxidation reactions and oxidation of sulfur compounds) and
in reduction, hydrogenation, and cycloaddition-type reactions.1
Furthermore, acting as solvents, these molecules can also
promote some reactions by themselves. Thus, the group of
RESULTS AND DISCUSSION
■
The direct amination between (E)-1,3-diphenylprop-2-en-1-ol
(1a) and p-toluensulfonamide (2a) was chosen as a model for
the optimization of the reaction conditions (Table 1) using
2,2,2-trifluoroethanol (TFE) and 1,1,1,3,3,3-hexafluoroisopro-
panol (HFIP) as readily available and inexpensive fluorinated
alcohols acting as solvents and reaction promoters (Table 1).9
́ ́
Begue and Bonnet-Delpon have reported a Povarov-type [4 +
2] cycloaddition,3 a ring opening of oxiranes with aromatic
amines,4 and aza-Michael reactions,5 mediated by 2,2,2-
trifluoroethanol (TFE) and by 1,1,1,3,3,3-hexafluoroisopropa-
nol (HFIP) as solvents.
Received: May 29, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
1, entries 8 and 9, respectively). Both solvents were also tested
in a reaction carried out under microwave irradiation for 30
min, but a mixture of 3aa and products resulting from the
competitive reaction of the solvent were observed (Table 1,
entries 10 and 11).
Table 1. Optimization of Reaction Parameters
Once the optimal conditions were established (Table 1,
entries 1 and 3), different nitrogenated nucleophiles were
evaluated for the direct allylic amination reaction onto alcohol
1a mediated by HFIP and TFE (Scheme 2 and Table 2). As
mentioned above p-toluenesulfonamide (2a) gave rise to the
amination product 3aa in almost quantitative yields in both
solvents after purification (Table 2, entries 1 and 2). Under
these optimal conditions benzylcarbamate (2b) led to the
formation of 3ab in excellent yields (Table 2, entries 3 and 4).
However, tert-butylcarbamate (2c) afforded 3ac in modest 45%
and 32% yields, respectively (Table 2, entries 5 and 6).
Benzamide gave only moderate yield when HFIP was used as
solvent (Table 2, entry 7). Surprisingly, a higher 80% yield was
obtained for 3ad by using TFE as solvent (Table 2, entry 8).
Electron-poor aniline such as p-nitroaniline (2e) afforded the
amination product 3ae in low yield in HFIP, but excellent yield
was achieved when the reaction was carried out in TFE (Table
2, entries 9 and 10). Other anilines were also tested, but to our
surprise Friedel−Crafts alkylation products were obtained (see
Table 4). Encouraged by these results, more basic amines,
which have remained elusive in most of the direct allylic
amination reaction reported to date,11 were next tested. When
benzylamine (2f) was used as nucleophile, high yield was
obtained only using HFIP as solvent probably due to the more
acidic nature but lower nucleophilicity and higher hydrogen
bond ability of this fluoroalkyl alcohol compared to TFE (Table
2, entries 11 and 12). The same behavior was observed when
(R)-α-methyl benzylamine (2g) was evaluated, giving product
3ag in a 60% as a 1.2:1 diastereomeric mixture (Table 2, entries
13 and 14). Similar results were also achieved with the more
basic aliphatic n-butylamine (2h) as nucleophile (Table 2,
entries 15 and 16). As mentioned above, TFE did not promote
the amination reaction of any of these more basic amines.
Finally when trimethylsilyl azide (2i), which to the best of our
knowledge has never been successfully employed for the direct
amination of allylic alcohols in a metal-free process,12 was used
as nucleophile, excellent yield was reached for compound 3ai
with HFIP as solvent (Table 2, entry 17). However, the use of
TFE yielded the corresponding product in a lower 51% yield
(Table 2, entry 18). Other nitrogen-based nucleophiles such as
aqueous ammonia, phthalimide, and its corresponding
potassium salt were also tested under the same reaction
conditions, giving low yields at the best (ca. 10%) for the
corresponding amination products. It can be summarized that
HFIP shows in general significantly better performance than
TFE for the direct amination of the allylic alcohol 1a, and in
addition, lower temperatures and equivalents of solvent were
required.
a
Reaction conditions: 1a (0.5 mmol), 2a (1.5 equiv), and 500 μL of
b
the solvent (1 M solution of 1a). Determined from 1H NMR analysis
of the crude mixture. 1 equiv (53 μL) was used. 250 μL of the
solvent was used (2 M solution of 1a). Reaction performed under
MW irradiation (40 W, 36 psi).
c
d
e
In the case of HFIP the reaction was carried out at 50 °C,
giving product 3aa in almost quantitative yields after 24 h
(Table 1, entry 1). At this temperature and using TFE, 84%
yield was obtained for the amination product, obtaining a 16%
of byproduct 4 resulting from the attack of the solvent (Table
1, entry 2). However, raising the temperature to 70 °C
provided quantitative yields for 3aa (Table 1, entry 3). When
the reaction was carried out at this temperature but using
ethanol as solvent7 (Table 1, entry 4), the corresponding ether
5 was obtained as sole product due to the higher nucleophilicity
of this alcohol (NEtOH = 0, NTFE = −2.78, and NHFIP = −4.23).1
For the sake of comparison in terms of Brønsted acidity we
decided to perform the amination reaction in PhOH as reaction
media9 [pKa(PhOH) = 9.95, pKa(TFE) = 12.37, and
pKa(HFIP) = 9.30]1 obtaining the Friedel−Crafts adduct 3ak
as major product (Table 1, entry 5). Finally, when water, which
also possesses a high polarity and hydrogen bond ability, was
used as solvent, dimerization product 6 was the main product
observed (Table 1, entry 6).9,10 Next, reduction of the
fluoroalkyl alcohol quantity was attempted, but unfortunately
the use of a stoichiometric amount of HFIP gave rise to the
formation of dimer 6 (Table 1, entry 7). Using half of the
amount of solvent led to an incomplete reaction in the case of
HFIP, along with the formation of ether 7 as byproduct, and a
complex mixture of products when TFE was employed (Table
Next, other allylic alcohols were submitted to the HFIP- and
TFE-mediated allylic substitution reaction under the standard
conditions and using TsNH2 2a as nucleophile (Table 3). Thus,
alcohol 1b gave rise to the corresponding amination product
3ba in good yields with both solvents (Table 3, entries 1 and
2). Similar results in terms of yields were observed when (Z)-
1b was the used as substrate, obtaining the (E)-isomer as
reaction product (Table 3, entries 3 and 4). However, with the
allylic alcohol 1c, moderate yields in product 3ba were obtained
even when 2 equiv of sulfonamide were employed (Table 3,
B
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 2. Direct Allylic Amination of 1a Mediated by Either HFIP or TFE
a
b
Reaction conditions: 1a (0.5 mmol), 2 (1.5 equiv), and 500 μL of the solvent (1 M solution of 1a). Isolated yields after flash chromatography.
c
d
1
Determined by GC and H NMR. In brackets, yield using 2 equiv of butylamine.
entries 5 and 6). According to these results, alcohol 1b is the
substrate of choice to prepare sulfonamide 3ba. Compound
3da was obtained in 79% yield when cyclohex-2-enol (1d) was
employed as substrate in HFIP as solvent, failing the reaction in
TFE (Table 3, entries 7 and 8). Moderate yields were achieved
in both solvents when the reaction was performed with alcohol
1e (Table 3, entries 9 and 10). In both cases approximately the
same regioisomeric mixture of compounds 3ae and 3ae′ (the
latter arising from the γ-addition)13 were isolated. Next,
alcohols 1f and 1g were evaluated. When the allylic alcohol
bearing a terminal olefin 1f was used, a mixture of compounds
3fa and 3fa′ in 30% and 27% yield was obtained in HFIP and
TFE, respectively. In both cases, compound 3fa′ having the
most substituted olefin was the major regioisomer in a 3:1 ratio
(Table 3, entries 11 and 12). However, when 1g was employed,
3ga was the sole regioisomer observed. Surprisingly, when the
reaction was carried out in HFIP, a low 15% for the amination
product was obtained, whereas 3ga was isolated in a 43% yield
when TFE was the solvent (Table 3, entries 13 and 14). It is
worth noting that the corresponding fluoroalkyl ethers were
obtained as byproduct, the amount of unreacted allylic alcohol
being in all cases negligible (<5%). Other allylic alcohols such
as cinnamyl alcohol, 1-phenyl-2-propen-1-ol, crotyl alcohol,
sorbic alcohol, and penta-1,4-dien-3-ol were tested without
success.
Next, to broaden the scope of this methodology, we turned
our attention onto carbon nucleophiles (Table 4). First,
Friedel−Crafts-type reaction onto allylic alcohol 1a was
examined. Thus, anisole (2j) and phenol (2k), both reacting
only at the para position, gave rise to the corresponding
alkylation adducts 3aj and 3ak, respectively, with higher yields
when using HFIP compared with TFE as solvent (Table 4,
entries 1−4). When indole (2l) was employed as nucleophile,
3al was regioselectively isolated in excellent yields regardless of
the fluoroalkyl alcohol used (Table 4, entries 5 and 6).
Electron-rich anilines, which in principle were expected to
proceed through a direct amination reaction, were next
evaluated as carbon nucleophiles. Thus, p-anisidine afforded
in high yields exclusively the Friedel−Crafts type alkylation
product 3am regardless of the solvent employed (Table 4,
entries 7 and 8). However, p-chloroaniline gave rise to the
formation of alkylation or amination products 3an or 3an′,
respectively, depending on the solvent employed (Table 4,
entries 9 and 10). The formation of the Friedel−Crafts adduct
3an could also be sought as result of a direct amination and
subsequent Hofmann−Martius rearrangement, as described by
other groups.14
Next, trimethylsilylated nucleophiles were evaluated. The use
of allyltrimethylsilane (2o) afforded the corresponding diene
3ao in almost quantitative yields in both solvents (Table 4,
entries 11 and 12). To our delight, the more challenging
C
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 3. Direct Allylic Amination of 1b−g Mediated by
Table 4. Allylic Substitution of 1a with Carbon Nucleophiles
Mediated by Either HFIP or TFE
a
a
Either HFIP or TFE
a
Reaction conditions: 1 (0.5 mmol), 2a (1.5 equiv), and 500 μL of the
b
solvent (1 M solution of 1). Isolated yield after flash chromatography.
c
d
1
2 equiv of nucleophile was used. Product ratio determined by H
NMR analysis of the crude mixture.
phenylacetylene derivative 2p led to the formation of the 1,4-
enyne 3ap in good yields (Table 4, entries 13 and 14).
Encouraged by these results, other silylated carbon nucleophiles
such as trimethylsilyl cyanide and (trimethylsilyl)-
trifluoromethane were also tested, giving unfortunately very
poor results at the best. In contrast, triethylsilane (2q) turned
out be an excellent nucleophile, especially when HFIP was used
as solvent, giving rise to the formation of the deoxygenation
product 3aq in 87% and 62% yield, respectively (Table 4,
entries 15 and 16).
1,3-Dicarbonyl compounds were next evaluated as nucleo-
philes. When the reaction between alcohol 1a and dimethyl
malonate was carried out in both fluorinated alcohols the
expected product was not produced. In contrast, the less basic
acetylacetone (2r) gave the corresponding substitution product
3ar in excellent yields when the less Brønsted acidic solvent
TFE was employed (Table 3, entry 18).15 Meldrum’s acid was
next employed as nucleophile, leading to the desired
substitution product 3as even when more acidic HFIP was
the solvent (Table 4, entry 19). However, the reaction carried
out in TFE gave rise in high yield to the formation of the
malonic acid monoester 3as′ as a result of the cleavage of cyclic
ketal and subsequent transesterification of the product 3as
(Table 4, entry 20). Finally, benzylic alcohol 2t was also tested,
giving mainly ethers 4 and 7, respectively, as main products.
a
Reaction conditions: 1a (0.5 mmol), 2 (1.5 equiv), and 500 μL of the
b
solvent (1 M solution of 1a). Isolated yield after flash
chromatography. Obtained as a 1:1 diastereomeric mixture.
c
Only a 13% yield of compound 3at was obtained in HFIP
(Table 3, entries 21 and 22).
Finally, in a parallel manner, other allylic free alcohols were
also tested in the reaction with different carbon nucleophiles.
Thus, 1b, 1c, and 1d were submitted to the allylic substitution
reaction with anisole (2j), indole (2l), allyltrimethylsilane (2o),
D
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 5. Allylic Substitution of Alcohols 1b−d with Carbon Nucleophiles Mediated by Either HFIP or TFE
a
b
Reaction conditions: 1 (0.5 mmol), 2 (2 equiv), and 500 μL of the solvent (1 M solution of 1a). Isolated yields after flash chromatography.
c
d
Product ratio determined by 1H NMR and/or GC analysis of the crude mixture. Because of the impossible separation of the substitution product
1
from the corresponding fluoroalkyl ether and/or the starting material, the yield was determined by H NMR and/or GC analysis of the crude
mixture. The 31% and 38% missing from the product distribution corresponds to the disubstitution product 2,4-bis(cyclohex-2-enyl)anisole 3dj″.
e
and acetylacetone (2r) as representative carbonucleophiles
under the typical reaction conditions (Table 5). When anisole
(2k) was the nucleophile of choice, regioisomeric alcohols 1b
and 1c produced the same Friedel−Crafts adduct 3bj as a single
isomer, obtaining better yields in both solvents when alcohol 1c
was the substrate employed (Table 5, compare entries 1 and 2
with 9 and 10). Alcohol 1b reacting with indole (2l) as
nucleophile gave rise to the corresponding substitution
products, regioisomers 3bl and 3bl′, in 53% and 50% yield,
respectively, when HFIP and TFE were used as solvents (Table
5, entries 3 and 4). It is worth mentioning that whereas a 1:1
regioisomeric mixture of α- and γ-substitution products 3bl and
3bl′ was obtained in HFIP, the reaction performed in TFE
afforded 3bl as single product. On the other hand,
regioisomeric mixtures of the α- and γ-substitution products
3bl and 3bl′ were obtained in excellent yields in both solvents
when indole (2l) reacted with alcohol 1c (Table 5, entries 11
and 12). Next, allylTMS 2o was tested as nucleophile. Thus, in
the case of alcohol 1b, 71% and 88% yield for the
corresponding diene, which was obtained as a 60/40
regioisomeric mixture 3bo/3bo′, was achieved in HFIP and
TFE, respectively (Table 5, entries 5 and 6). A similar situation
E
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. Reaction kinetics.
Scheme 1. Allylic Amination Reaction onto Ethers 4 and 7
was found when alcohol 1c was allowed to react with the same
nucleophile, giving approximately a 1:1 ratio for 3bo and 3bo′.
In this case, substantially higher yield was obtained when HFIP
was the solvent employed (Table 5, entries 13 and 14). Finally,
diketone 2r was tested in the reaction with regioisomeric
substrates 1b and 1c. In the first case similar yields in both
solvents were achieved for the mixture 3br/3br′ (Table 5,
entries 7 and 8). However, when 1c was the substrate
employed, a drop in the yield, especially in the case of TFE,
for the mixture 3br/3br′ was observed (Table 5, entries 15 and
16).
Contrary to the previous examples, cyclohex-2-enol (1d)
proved to be in general less reactive with the carbon
nucleophiles. Thus, anisole afforded 69% and 58% yield in
HFIP and TFE, respectively, for the Friedel−Crafts adduct,
although a rather complicated inseparable mixture of the para-
and ortho-substituted anisole, 3dj and 3dj′, and dicyclohex-
enylanisole 3dj″ products together with traces of corresponding
fluoroalkyl ether was obtained after flash chromatography
(Table 5, entries 17 and 18). However, indole only produced a
very poor yield at best for the product 3dl when HFIP was the
solvent (Table 5, entries 19 and 20). Next, allyltrimethylsilane
(2o) was evaluated, giving rise to product 3do in 57% yield
only in the HFIP-promoted allylic substitution (Table 5, entries
21 and 22). Unfortunately, when diketone 2r was the
nucleophile, very poor results were achieved (Table 5, entries
23 and 24). In those reactions in which full conversion toward
the substitution product was not achieved, the corresponding
fluoroalkyl ether was obtained. In the cases of alcohols 1b and
1c the same ether (8 or 9, see below in Scheme 3) was
obtained as the consequence of a double bond isomerization.
Mechanistic Considerations. The fact that ethers 4 and 7
were obtained as main byproduct when allylic alcohol 1a was
employed regardless the nucleophile employed (see, for
example, Table 1, entries 2 and 8) drew our attention with
regard to the reaction mechanism. This observation led us to
suppose that an equilibrium between the formation of the
expected substitution product and the ether formation might be
taking place during the process. To corroborate this point a
kinetic study of the reaction of 1a with sulfonamide 2a under
optimal reaction conditions (Table 2, entries 1 and 2) was
carried out in both solvents (Figure 1). From this study
different conclusions can be drawn: first, according to Figure 1,
these ethers (and dimer 6 in the case of HFIP-mediated
reaction) are formed more quickly than the corresponding
amination product, and apparently the allylic amination
mediated by fluoroalkyl alcohols also takes place onto these
ethers, although at a slower rate. Consequently, 3aa can be
considered a result of two operating processes, direct amination
onto alcohol 1a and the indirect allylic substitution onto ethers
4 and 7. Another interesting point is that the reaction carried
out in TFE is faster than in HFIP despite the higher hydrogen
bond ability and acidity of the latter, although this can be
rationalize from the use of higher temperatures and number of
equivalents.9 Finally, as observed in the kinetic study the
reaction was finished in less than 24 h in both cases.
Another experiment that confirms that ethers 4 and 7 can
suffer allylic amination mediated by these fluoroalkyl alcohols
was designed (Scheme 1). Thus, these compounds were
synthesized and submitted to the direct amination reaction
under the standard reaction conditions mentioned above (see
Table 1). The amination product 3aa was cleanly obtained in
high yields when HFIP was used. Similar results were obtained
with TFE although in lower yields (decomposition products
were also observed). Then, the opposite reaction was also
tested. Thus, when allylic sulfonamide 3aa was allowed to react
for 24 h in HFIP or TFE as solvents at the corresponding
temperature, unaltered 3aa together with decomposition
F
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
products (which were more abundant in the case of TFE) were
recovered, suggesting the nonreversibility of the reaction.
Taking advantage of this last experiment we envisioned the
use of fluoroalkyl alcohols as promoters of allylic amination
onto allylic ethers. Thus, ethyl ether 5 reacted with p-
toluensulfonamide 2a, giving rise after 24 h to the amination
product 3aa in high yields when HFIP was employed as
reaction media. The TFE-mediated reaction gave the
corresponding product 3aa in a 70% yield, although several
unidentified byproducts were observed (Scheme 2). The
products was obtained. These results can be rationalized from a
kinetic point of view. In this sense, the reaction between 1b and
acetylacetone (2s) in both solvents was monitored by GC,
demonstrating high conversion toward both regioisomers in the
same ratio as the above-mentioned (Table 5, entries 7 and 8) in
less than 1 h. These results remained practically constant after
24 h and could mean that the reaction is faster than a possible
isomerization reaction toward the most stable olefin, the attack
onto the α- and γ-position of the common π-allyl cationic
intermediate generated from both allylic alcohols (or ethers 8
and 9) being equiprobable. The fact that 3br was the major
regioisomer in front of 3br′ when the reaction was run in TFE
can be explained by the higher temperatures employed with this
solvent, favoring the formation of the thermodynamically
controlled product 3br. As a consequence of this hypothesis,
which in addition would support it, higher yields are obtained
when carbon nucleophiles react with alcohols 1b and 1c in
comparison with the amination reaction with the same
substrates.
Scheme 2. TFE and HFIP Allylic Amination Reaction onto
Ether 5
In addition, we decided to corroborate whether a Hofmann−
Martius-type rearrangement14 was responsible for the allylic
substitution products, 3am and 3an, obtained when electron-
rich anilines were used as nucleophiles. For this purpose, a 1 M
solution of compound 3an′ in HFIP was stirred at 50 °C for 24
h. After this time, fluoroalkyl ether 7 and Friedel−Crafts adduct
3an (in an 80:20 ratio) and p-chloroaniline (2n) were the only
products observed in the mixture, pointing toward a reversible
process between the formation of the amination product 3an′
and ether 7 through an activation of the amine moiety by the
fluorinated alcohol. This equilibrium is displaced by the
irreversible formation of the product 3an (Scheme 4). This
reactivity of the ethers toward allylic substitution reaction can
also explain the low yields obtained when benzylic alcohol 2t
was used as nucleophile (Table 4, entries 21 and 22), since the
presumably formed benzyl ether 3at can suffer another
substitution reaction with the more abundant fluoroalkyl
alcohol forming ethers 4 and 7 (indeed, these ethers were
detected as main products when BnOH was used as
nucleophile).
The fact that the same product 3ba was obtained when
alcohols 1b, (Z)-1b′, and 1c were used (Table 3, entries 1−6)
seems to imply a double bond isomerization and points out a
possible carbocationic intermediate involved in this reaction.
To corroborate these assumptions we decided to submit the
enantioenriched alcohol (S)-1b (65% ee) to the allylic
amination reaction under the standard conditions (Table 3,
entries 1 and 2). After 24 h racemic 3ba was obtained as main
product in both solvents.16 The isomerization of the double
bond was confirmed by submitting allylic alcohol 1c under the
optimized reaction conditions (1 M solution at 50 or 70 °C for
HFIP and TFE, respectively) in the absence of any nucleophile
using both solvents. As expected and according to the
regioselectivity observed for this alcohol (Table 3, entries 5
and 6), the corresponding isomerized ether together with some
alcohol 1b were the main reaction products, suggesting an
isomerization of the double bond toward the thermodinami-
cally most stable olefin prior the amination occurs (Scheme
3).7b,17
Scheme 4. Possible Mechanism for the Formation of
Products 3an and 3an′
hypothesis would also explain the different behavior observed
in both anilines since the more basic allylic amine 3am can be
easily activated by TFE and HFIP through hydrogen bond and/
or protonation, whereas the less basic product 3an is activated
only by the more acidic HFIP with higher hydrogen bond
ability.
However, this isomerization mechanism seems not to be
applicable when alcohols 1b and 1c were allowed to react with
carbon nucleophiles since a mixture of α- and γ-substitution
Finally, it is worth mentioning that when we were trying to
understand the mechanism to explain the excellent perform-
ance of silylated nucleophiles, we realized that the reaction of
allyltrimethylsilane (2o) with allylic alcohol 1a took place in
less than 30 min even at room temperature in both solvents.
The same trend was observed when the reaction was performed
with TMSN3 2i and Et3SiH 2q in HFIP as solvent, with the
reaction, at room temperature, taking less than 1 h to complete.
At this point, we decided to monitor by GC−MS the reaction
between the alcohol 1a and trimethylsilylazide (2i), and at the
early stages of the reaction the formation of the corresponding
trimethylsilylether 10 was observed. This fact would transform
Scheme 3. TFE- and HFIP-Mediated Isomerization of
Alcohol 1c
G
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
128.5, 128.8, 129.0, 129.4, 131.2, 135.2, 136.5, 140.9, 142.1, 156.7
ppm.
the hydroxyl function into a better leaving group, and
consequently the reaction becomes faster (Scheme 5).
(E)-tert-Butyl 1,3-Diphenylallylcarbamate (3ac).7b,34 White
solid (70 mg, 45%); mp 117 °C (lit.31 115−116 °C); 1H NMR
(300 MHz, CDCl3) δ 1.45 (s, 9H), 4.96 (s, 1H), 5.46 (s, 1H), 6.32
(dd, J = 15.9, 6.0 Hz, 1H), 6.54 (dd, J = 15.9, 1.2 Hz, 1H), 7.20−7.38
(m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 28.4, 56.3, 79.7, 126.5,
126.9, 127.5, 127.6, 128.5, 128.7, 129.5, 130.8, 136.5, 141.3, 154.9
ppm.
Scheme 5. Possible Mechanism for Allylic Substitution with
Silylated Nucleophiles
(E)-N-(1,3-Diphenylallyl)benzamide (3ad).7b,21 White solid
(125 mg, 80%); mp 163−164 °C (lit.20 157−159 °C); 1H NMR
(400 MHz, CDCl3) δ 6.03 (t, J = 7.0 Hz, 1H), 6.44 (dd, J = 15.9, 6.0
Hz, 1H), 6.55 (s broad, 1H), 6.62 (d, J = 15.9 Hz, 1H), 7.30−7.43 (m,
13H), 7.83 (d, J = 7.5 Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3) δ
55.2, 126.5, 127.0, 127.1, 127.2, 127.8 128.5, 128.6, 128.7, 128.9, 131.6,
131.7, 134.3, 136.3, 140.7, 166.4 ppm.
(E)-4-Nitro-N-(1,3-diphenylallyl)aniline (3ae).7b,34 Yellow solid
(160 mg, 97%); mp 146−147 °C (lit.31 140−141 °C); 1H NMR (300
MHz, CDCl3) δ 4.89 (d, J = 4.5 Hz, 1H), 5.20 (t, J = 5.4 Hz, 1H), 6.37
(dd, J = 15.9, 6.0 Hz, 1H), 6.56−6.63 (m, 3H), 7.25−7.41 (m, 10H),
8.05 (d, J = 9.3 Hz, 2H) ppm; 13C NMR (75 MHz, CDCl3) δ 60.0,
112.1, 126.2, 126.6, 127.1, 128.1, 128.2, 128.5, 128.7, 129.1, 132.2,
136.0, 138.5, 140.2, 152.1 ppm.
CONCLUSIONS
■
In this work, a new strategy for allylic substitution reaction onto
allylic free alcohols using fluoroalkyl alcohols, 2,2,2-trifluor-
oethanol (TFE) and 1,1,1,3,3,3-hexafluoroisopropanol (HFIP),
as reaction promoters has been developed. This simple
procedure allows the introduction of a wide range of
nucleophiles under mild reaction conditions, giving rise to
the allylic substitution products in high yields, especially when
HFIP was the solvent of choice. This new methodology can be
considered not only as an alternative to metal or Brønsted acid
catalyzed allylic substitution reactions but also as comple-
mentary since some nucleophiles that fail in most of the
existing procedures (i.e., basic aliphatic amines) can be used. In
addition, the mechanistic studies of the direct allylic amination
reaction suggested that the formation of the products can be
sought as a result of two operating pathways: a direct
substitution reaction and an indirect process that implies a
double substitution. In both solvents, a carbocationic
intermediate is postulated due to the racemization observed
when enantioenriched allylic alcohols were employed. Finally,
the regioselectivity of this amination process is seemingly the
result of the isomerization of the double bond, mediated by
these fluorinated alcohols, toward the most energetically stable
olefin prior to the substitution reaction taking place. However,
this isomerization does not apparently occur in the case of
carbon nucleophiles where a mixture of kinetic and
thermodynamic products was obtained.
(E)-N-Benzyl-1,3-diphenylprop-2-en-1-amine (3af).18 Color-
1
less oil (109 mg, 73%); H NMR (400 MHz, CDCl3) δ 1.82 (s, 1H),
3.77 (d, J = 4.36 Hz, 2H), 4.39 (d, J = 7.5 Hz, 1H), 6.31 (dd, J = 15.8,
7.5 Hz, 1H), 6.57 (d, J = 15.8 Hz, 1H), 7.17−7.43 (m, 15H) ppm; 13C
NMR (100 MHz, CDCl3) δ 51.3, 64.5, 126.4, 126.9, 127.3, 127.3,
127.4, 128.1, 128.4, 128.5, 128.6, 130.3, 132.5, 136.8, 140.3, 142.8
ppm.
(E)-1,3-Diphenyl-N-(1-phenylethyl)prop-2-en-1-amine
(3ag).19 Obtained as 1.2:1 inseparable mixture of diatereoisomers.
Yellow oil (94 mg, 60%). Major diastereoisomer: 1H NMR (300 MHz,
CDCl3) δ 1.34 (d, J = 6.7 Hz, 3H), 1.73 (s, 1H), 3.65 (q, J = 6.7 Hz,
1H), 4.18 (d, J = 6.5 Hz, 1H), 6.28 (dd, J = 15.9, 6.6 Hz, 1H), 6.45 (d,
J = 15.9 Hz, 1H), 7.24−7.30 (m, 15H) ppm; 13C NMR (75 MHz,
CDCl3) δ 24.6, 54.8, 62.2, 126.3, 126.7, 127.1, 127.3, 127.5, 128.4,
128.5, 128.6, 128.6, 129.1, 129.5, 133.1, 136.9 ppm. Minor
diastereoisomer: 1H NMR (300 MHz, CDCl3) δ 1.39 (d, J = 6.7
Hz, 3H), 1.73 (s, 1H), 3.95 (q, J = 6.7 Hz, 1H), 4.17 (d, J = 7.9 Hz,
1H), 6.25 (dd, J = 15.8, 7.9 Hz, 1H), 6.42 (d, J = 15.8 Hz, 1H), 7.21−
7.39 (m, 15H) ppm; 13C NMR (75 MHz, CDCl3) δ 24.4, 55.0, 61.9,
126.4, 126.7, 126.9, 127.1, 127.4, 127.9, 128.5, 129.1, 129.4, 131.0,
131.9, 136.9, 143.2, 145.5 ppm.
(E)-N-(1,3-Diphenyl-2-propenyl)butylamine (3ah).20 Colorless
oil (85 mg, 64%); 1H NMR (300 MHz, CDCl3) δ 0.89 (t, J = 7.3 Hz,
3H), 1.32−1.47 (m, 2H), 1.48−1.63 (m, 2H), 2.59 (ddt, J = 33.2, 11.4,
7.1 Hz, 2H), 4.34 (d, J = 7.4 Hz, 1H), 6.30 (dd, J = 15.8, 7.4 Hz, 1H),
7.24−7.38 (m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 14.0, 20.5,
32.3, 47.5, 65.7, 126.3, 127.1, 127.2, 127.3, 128.4, 128.5, 130.0, 132.9,
136.9, 143.1 ppm.
EXPERIMENTAL SECTION
■
General Procedure for Allylic Subsitution Reaction. Onto an
open air tube containing a 1 M solution of the allylic alcohol (0.5
mmol) in HFIP or TFE (500 μL) was added the corresponding
nucleophile (0.75−1 mmol). The reaction mixture was then stirred for
24 h at the indicated temperature for each solvent (see tables). After
the reaction time the volatiles were evaporated, and the crude
compounds were purified by flash chromatography when necessary.
Physical and spectroscopic data are given below. For known
(E)-3-Azido-1,3-diphenyl-1-propene (3ai).12 Colorless oil (113
1
mg, 96%); H NMR (300 MHz, CDCl3) δ 5.20 (d, J = 7.3 Hz, 1H),
6.28 (dd, J = 7.3, 15.6 Hz, 1H), 6.71 (d, J = 15.6 Hz, 1H), 7.23−7.41
(m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 67.2, 126.8, 126.9,
127.1, 128.2, 128.4, 128.6, 128.7, 132.9, 138.5, 139.5 ppm.
(E)-3-(4-Methoxyphenyl)-1,3-diphenylprop-1-ene (3aj).7b
1
compounds, only H and 13C NMR data are listed.
1
Yellow oil (132 mg, 88%); H NMR (300 MHz, CDCl3) δ 3.77 (s,
(E)-N-(1,3-Diphenylallyl)-4-methylbenzenesulfonamide
(3aa).7,33 White solid (174 mg, 96%); mp 140−141 °C (lit.30 136−
137 °C); 1H NMR (300 MHz, CDCl3) δ 2.32 (s, 3H), 4.97 (d, J = 7.1
Hz, 1H), 5.11 (t, J = 6.9 Hz, 1H), 6.10 (dd, J = 15.8, 6.7 Hz, 1H), 6.35
(d, J = 15.9 Hz, 1H,) 7.10−7.30 (m, 12H), 7.65 (d, J = 8.3 Hz, 2H)
ppm; 13C NMR (75 MHz, CDCl3) δ 21.4, 59.7, 126.5, 127.0, 127.3,
127.9, 128.1, 128.4, 128.7, 128.9, 129.4, 132.1, 136.0, 137.7 139.6,
143.3 ppm.
3H), 4.84 (d, J = 5.6 Hz, 1H), 6.32 (d, J = 15.8 Hz, 1H), 6.65 (dd, J =
15.8, 5.6 Hz, 1H), 6.84 (d, J = 8.5 Hz, 2H), 7.13−7.37 (m, 12H); 13C
NMR (75 MHz, CDCl3) δ 53.3, 55.2, 113.8, 126.2, 126.3, 127.2, 128.4,
128.5, 128.6, 129.6, 131.1, 132.9, 135.6, 137.3, 143.8, 158.1 ppm.
(E)-4-(1,3-Diphenylallyl)phenol (3ak).21 Yellow oil (107 mg,
1
75%); H NMR (300 MHz, CDCl3) δ 4.82 (d, J = 7.4 Hz, 1H), 5.21
(s, 1H), 6.31 (d, J = 15.8, 1H), 6.63 (dd, J = 15.8, 7.5 Hz, 1H), 6.76 (d,
J = 8.6 Hz, 2H), 7.08 (d, J = 8.3 Hz, 2H), 7.14−7.42 (m, 10H) ppm;
13C NMR (75 MHz, CDCl3) δ 53.3, 115.3, 126.2, 126.3, 127.2, 128.4,
128.4, 128.5, 129.7, 131.1, 132.8, 135.6, 137.2, 143.7, 154.1 ppm.
(E)-3-(1,3-Diphenylallyl)indole (3al).7b Brownish oil (148 mg,
(E)-Benzyl 1,3-Diphenylallylcarbamate (3ab).7 White solid
(153 mg, 89%); mp 109 °C (lit.7a 110 °C); H NMR (300 MHz,
1
CDCl3) δ 5.15 (m, 3H), 5.54 (br s, 1H), 6.33 (dd, J = 15.9, 6.0 Hz,
1H), 6.56 (d, J = 16 Hz, 1H), 7.10−7.49 (m, 15H) ppm; 13C NMR
(75 MHz, CDCl3) δ 56.7, 66.9, 126.5, 127.0, 127.7, 128.1, 128.2,
1
96%); H NMR (300 MHz, CDCl3) δ 5.12 (d, J = 7.3 Hz, 1H), 6.43
H
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(d, J = 15.8 Hz, 1H), 6.72 (dd, J = 15.8, 7.3 Hz, 1H), 6.90 (s, 1H), 7.02
(t, J = 7.1 Hz, 1H), 7.24−7.35 (m, 13H), 7.99 (br s, 1H) ppm; 13C
NMR (75 MHz, CDCl3) δ 46.1, 111.1, 119.4, 119.9, 122.1, 122.6,
126.3, 126.4, 126.6, 126.7, 127.1, 127.7, 128.4, 128.5, 130.5, 132.5,
136.6, 137.5, 143.3 ppm.
s, 1H) ppm; 13C NMR (100 MHz, CDCl3) δ 49.06, 57.00, 57.07,
60.60−61.35 (m), 124.19, 124.36, 126.31, 126.42, 126.47, 127.47,
127.52, 127.75, 127.80, 128.23, 128.52, 128.75, 128.89, 132.47, 132.58,
136.43, 136.56, 139.23, 139.40, 165.73, 166.05, 171.53, 171.82 ppm.
MS (EI): m/z 334 [M+ − CO2), 18%], 206 (15), 194 (16), 193 (100),
192 (47), 191 (15), 178 (25), 128 (13), 115 (80), 91 (27); HRMS-
CO2 calcd for C19H17F3O2 334.1181, found 334.1183.
(E)-4-Methoxy-2-(1,3-diphenylallyl)aniline (3am). Yellow
sticky oil (147 mg, 93%); Rf 0.41(hexane/ethyl acetate 4:1); IR
(E)-Benzyl 1,3-Diphenylallyl Ether (3at).26 Colorless oil (20 mg,
13%); 1H NMR (400 MHz, CDCl3) δ 4.58 (s, 2H), 5.01 (d, J = 7 Hz,
1H), 6.34 (dd, J = 15.9, 7 Hz, 1H), 6.63 (d, J = 15.9 Hz, 1H), 7.22−
7.44 (m, 15H) ppm; 13C NMR (100 MHz, CDCl3) δ 70.1, 81.6, 126.6,
126.8, 126.9, 127.5, 127.7, 128.1, 128.3, 128.5, 128.7, 130.2, 131.5,
136.5, 138.4, 141.1 ppm.
1
(ATR) ν 3025, 2926, 1599, 1497, 1257, 1041 cm−1; H NMR (300
MHz, CDCl3) δ 3.33 (br s, 2H), 3.74 (s, 3H), 4.93 (d, J = 6.9 Hz, 1H),
6.31 (d, J = 16.0 Hz, 1H), 6.65−6.76 (m, 4H), 7.24−7.41 (m, 10H)
ppm; 13C NMR (75 MHz, CDCl3) δ 49.6, 55.6, 112.2, 115.8, 117.5,
126.4, 126.9, 127.4, 128.5, 128.7, 128.9, 130.1, 131.2, 131.7, 137.1,
137.9, 141.6, 153.0 ppm. MS (EI): m/z 315 [M+, 63%], 313 (22), 311
(43), 224 (100), 223 (80), 209 (20), 193 (17),191 (19), 181 (24), 180
(36), 165 (18), 115 (18), 106 (21), 105 (20), 91 (27); HRMS calcd
for C22H21NO 315.1623, found 315.1646.
(E)-4-Methyl-N-(4-phenylbut-3-en-2-yl)benzenesulfonamide
(3ba).7b,35 White solid (102 mg, 68%); mp 88−89 °C (lit.32 96−98
°C); 1H NMR (400 MHz, CDCl3) δ 1.27 (d, J = 6.6 Hz, 3H), 2.34 (s,
3H), 4.05−4.15 (m, 1H), 5.08 (br s, 1H), 5.83 (dd, J = 16.0, 6.8 Hz,
1H), 6.28 (d, J = 16.0 Hz, 1H), 7.14−7.20 (m, 7H, ArH), 7.75 (d, J =
8.4, 2H) ppm; 13C NMR (100 MHz, CDCl3) δ 21.4, 21.9, 51.6,
126.3,127.2, 127.6, 128.4, 129.5, 130.1, 130.5, 136.1, 138.1, 143.3 ppm.
HPLC: Daicel Chiralcel OD-H column, hexane/2-propanol 95:5, flow
rate 1 mL/min, λ = 254 nm, tR = 27.8 and 31.1 min.
(E)-4-Chloro-2-(1,3-diphenylallyl)aniline (3an). Colorless oil
(144 mg, 90%); Rf 0.44 (hexane/ethyl acetate 4:1); IR (ATR) ν 2917,
1
2849, 1487, 1260, 1091, 1017 cm−1; H NMR (400 MHz, CDCl3) δ
3.61 (br s, 2H), 4.88 (d, J = 6.9 Hz, 1H), 6.33 (d, J = 15.9 Hz, 1H),
6.66 (m, 2H), 7.11 (m, 2H), 7.28−7.43 (m, 10H) ppm; 13C NMR
(100 MHz, CDCl3) δ 49.5, 117.6, 123.6, 126.4, 127.1, 127.4, 127.6,
128.5, 128.7, 128.8, 128.9, 129.8, 130.4, 132.1, 136.9, 140.9, 142.8
ppm. MS (EI): m/z 319 [M+, 29%], 284 (20), 230 (34), 229 (43), 228
(100), 227 (69), 206 (21), 193 (67), 191 (24), 165 (18), 91 (21);
HRMS calcd for C21H18ClN 319.1128, found 319.1121.
N-(Cyclohex-2-enyl)-4-methylbenzenesulfonamide
(3da).7b,36 White solid (99 mg, 79%); mp 102−103 °C (lit.33 97−98
1
°C); H NMR (300 MHz, CDCl3) δ 1.44−1.61 (m, 2H), 1.70−1.84
(m, 2H), 1.87−2.03 (m, 2H), 2.43 (s, 3H), 3.81 (br s, 1H), 4.54 (d, J
= 8.0 Hz, 1H), 5.32−5.37 (m, 1H), 5.74−5.78 (m, 1H), 7.30 (d, J =
8.0 Hz, 2H), 7.77 (d, J = 8.0 Hz, 2H) ppm; 13C NMR (75 MHz,
CDCl3) δ 19.3, 21.5, 24.5, 30.3, 48.9, 126.9, 127.0, 129.6, 131.5, 138.3,
143.2 ppm.
(E)-4-Chloro-N-(1,3-Diphenylallyl)aniline (3an′).7b Yellow oil
1
(147 mg, 92%); H NMR (300 MHz, CDCl3) δ 4.12 (d, J = 6.1 Hz,
1H), 5.03 (d, J = 6.1 Hz, 1H), 6.36 (dd, J = 15.8, 6.1 Hz, 1H), 6.53−
6.58 (m, 3H), 7.07 (d, J = 8.7 Hz, 2H), 7.25−7.39 (m, 10H) ppm; 13C
NMR (75 MHz, CDCl3) δ 60.7, 114.7, 122.3, 126.5, 127.1, 127.7,
127.8, 128.6, 128.9, 129.0, 130.1, 131.3, 136.4, 141.6, 145.7 ppm.
(E)-1,3-Diphenyl-1,5-hexadiene (3ao).7b Yellow oil (115 mg,
N-[1-(Cyclohex-1-en-1-yl)ethyl]-4-methylbenzenesulfona-
mide (3ea).27 Obtained as inseparable mixture with the regioisomer
3ea′. The following data corresponds to 3ea. White solid (80 mg,
1
1
57%); mp 65 °C (3ea/3ea′ 83:17); H NMR (300 MHz, CDCl3) δ
98%); H NMR (300 MHz, CDCl3) δ 2.58 (t, J = 7.5 Hz, 2H), 3.52
1.17 (d, J = 8.0 Hz, 2H), 1.26 (m, 1H), 1.37 (m, 2H), 1.47 (m, 1H),
1.67 (m, 1H), 1.75−1.87 (br m, 3H), 2.43 (s, 3H), 3.84 (p, J = 7.0 Hz,
1H), 4.65 (br s, 1H), 5.46 (br s, 1H), 7,28 (d, J = 8.2 Hz, 2H), 7.74 (d,
J = 8.2 Hz, 2H) ppm; 13C NMR (75 MHz, CDCl3) δ 20.5, 21.5, 22.0,
22.1, 23.5, 24.8, 55.5, 123.9, 127.3, 129.3, 136.9, 138.2, 142.9 ppm.
4-Methyl-N-(3-methylbut-2-en-1-yl)benzenesulfonamide
(3fa′).28 Obtained as diastereomeric mixture 3fa/3fa′ in a 3:1 ratio.
The following data corresponds to the inseparable mixture. Colorless
(m, 1H), 4.97−5.08 (m, 2H), 5.72−5.81 (m, 1H), 6.37 (d, J = 5.4 Hz,
2H), 7.25−7.33 (m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 40.1,
48.9, 116.3, 126.1, 126.3, 127.1, 127.7, 128.4, 128.5, 129.7, 133.4,
136.5, 137.4, 143.8 ppm.
(E)-1,3,5-Triphenylpent-1-en-4-yne (3ap).22 Yellow oil (90 mg,
1
61%); H NMR (300 MHz, CDCl3) δ 4.75 (d, J = 6.5 Hz, 1H), 6.34
(dd, J = 15.7, 6.5 Hz, 1H), 6.78 (dd, J = 15.7, 1.1 Hz, 1H), 7.25−7.39
(m, 13H), 7.48−7.51 (m, 2H) ppm; 13C NMR (75 MHz, CDCl3) δ
41.2, 85.4, 88.8, 123.4, 126.5, 127.1, 127.5, 127.7, 128.0, 128.2, 128.5,
128.7, 129.6, 130.4, 131.7, 136.8, 140.3 ppm.
1
oil (36 mg, 30%); Rf 0.52 (hexane/ethyl acetate 4:1); H NMR (400
MHz, CDCl3) δ 1.30 (s, 6H, 3fa), 1.54 (s, 3H, 3fa′), 1.64 (s, 3H,
3fa′), 2.41 (s, 3H, 3fa), 2.43 (s, 3H, 3fa′), 3.54 (t, J = 6.4 Hz, 2H,
3fa′), 4.25 (br s, 1H, 3fa′) 4.60 (br s, 1H, 3fa), 4.94−5.12 (m, 2H, 3fa
+ 1H, 3fa′), 5.79 (dd, J = 10.4, 17.2 Hz, 1H, 3fa), 7.30 (d, J = 8.4 Hz,
2H 3fa′ + 2H 3fa′), 7.75 (d, J = 8.4 Hz, 2H 3fa′ + 2H 3fa′) ppm; 13C
NMR (100 MHz, CDCl3) δ 17.8, 21.5, 25.5, 27.7, 41.0, 112.8, 118.7,
127.2, 129.3, 129.6, 135.0, 137.6, 143.3 ppm.
(E)-1,3-Diphenylpropene (3aq).23 Yellow oil (94 mg, 97%); H
1
NMR (300 MHz, CDCl3) δ 3.55 (d, J = 6 Hz, 2H), 6.32−6.48 (m,
2H), 7.19−7.37 (m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 39.3,
126.1, 126.2, 127.1, 128.4, 128.6, 129.2, 131.0, 137.4, 140.1 ppm.
(E)-3-(1,3-Diphenylallyl)pentane-2,4-dione (3ar).24 White
solid (139 mg, 95%); mp 85−86 °C (lit.23 85−87 °C); H NMR
1
4-Methyl-N-(4-methylpent-3-en-2-yl)benzenesulfonamide
(3ga). Yellow sticky oil (54 mg, 43%); Rf 0.46 (hexane/ethyl acetate
4:1); IR (ATR) ν 3268, 2953, 1715, 1598, 1453, 1376, 1323, 1157,
1085, 1072, 980 cm−1; 1H NMR (400 MHz, CDCl3) δ 1.14 (d, J = 6.6
Hz, 3H), 1.44 (s, 3H), 1.51 (s, 3H), 2.41 (s, 3H), 4.10 (dq, J = 9.0, 6.6
Hz, 1H), 4.52 (br s, 1H), 4.79 (dd, J = 9.0, 2.6 Hz, 1H), 7.26 (d, J =
8.3 Hz, 2H), 7.71 (d, J = 8.3 Hz, 2H) ppm; 13C NMR (100 MHz,
CDCl3) δ 17.7, 21.5, 22.5, 25.3, 48.1, 126.1, 127.2, 129.3, 134.2, 138.2,
142.9 ppm. MS (EI): m/z 253 [M+, 1%], 239 (14), 238 (100), 154
(44), 98 (39), 91 (77), 83 (28), 82 (31); HRMS calcd for
C13H19NO2S 253.1136, found 253.1117.
(300 MHz, CDCl3) δ 1.92 (s, 3H), 2.25 (s, 3H), 4.32−4.36 (m, 2H),
6.19 (ddd, J = 15.8, 5.1, 2.9 Hz, 1H), 6.43 (d, J = 15.8 Hz, 1H), 7.17−
7.36 (m, 10H) ppm; 13C NMR (75 MHz, CDCl3) δ 29.7, 30.0, 49.1,
74.4, 126.3, 127.2, 127.7, 127.8, 128.5, 129.0, 129.2, 131.6, 136.5,
140.0, 202.6, 202.8 ppm.
(E)-5-(1,3-Diphenylallyl)-2,2-dimethyl-1,3-dioxane-4−6-
dione (3as).25 White powder (101 mg, 60%); mp 142 °C (lit.24 141−
1
142 °C); H NMR (300 MHz, CDCl3) δ 1.47 (s, 3H), 1.71 (s, 3H),
3.96 (d, J = 2.8 Hz, 1H), 4.73 (dd, J = 9.2, 2.8 Hz, 1H), 6.65 (d, J =
15.8 Hz, 1H), 6.92 (dd, J = 15.8 9.2 Hz, 1H), 7.15−7.52 (m, 10H)
ppm; 13C NMR (75 MHz, CDCl3) δ 27.6, 28.2, 47.4, 52.5, 105.3,
126.5, 127.4, 127.7, 127.9, 128.4, 128.5, 128.7, 133.4, 136.6, 139.7,
164.4 ppm.
(E)-4-(4-Phenylbut-3-en-2-yl)anisole (3bj).29 Colorless oil (79
1
mg, 66%); H NMR (300 MHz, CDCl3) δ 1.43 (d, J = 7.0 Hz, 3H),
3.60 (p, J = 6.9 Hz, 1H), 3.78 (s, 3H), 6.35−6.37 (m, 2H), 6.86 (d, J =
8.7 Hz, 2H), 7.16−7.35 (m, 7H) ppm; 13C NMR (75 MHz, CDCl3) δ
21.3, 41.7, 55.2, 113.8, 126.1, 127.0, 128.2, 128.4, 135.5, 137.6, 137.7,
158.0 ppm.
(E)-3,5-Diphenyl-2-[(2,2,2-trifluoroethoxy)carbonyl]pent-4-
enoic acid (3as′). Obtained as 1:1 diastereomeric mixture. White
solid (161 mg, 85%); mp 105−106 °C; Rf 0.28 (hexane/ethyl acetate
4:1); IR (ATR) ν 3026, 2161, 1978, 1761, 1600,1412, 1286, 1258,
1179, 1140 cm−1; 1H NMR (400 MHz, CDCl3) δ 4.08 (d, J = 4.0 Hz,
1H), 4.11 (d, J = 3.6 Hz, 1H), 4.28 (m, 4H), 4.50 (m, 2H), 6.37 (dd, J
= 15.7, 8.6 Hz, 2H), 6.53 (d, J = 15.7 Hz, 1H), 7.29 (m, 10H), 8.72 (br
(E)-3-(4-Phenylbut-3-en-2-yl)-1H-indole (3bl).30 Brown oil (62
1
mg, 50%); H NMR (400 MHz, CDCl3) δ 1.56 (d, J = 7.0 Hz, 3H),
3.93 (p, J = 6.9 Hz, 1H), 6.46−6.48 (m, 2H), 6.98−7.33 (m, 9H), 7.67
I
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(d, J = 7.9 Hz, 1H), 7.90 (s, 1H) ppm; 13C NMR (100 MHz, CDCl3)
δ 20.7, 34.2, 111.1, 119.2, 120.4, 122.0, 126.1, 126.5, 126.8, 128.4,
128.7, 135.4, 136.5, 137.8 ppm.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
General remarks, H and 13C NMR spectra copies of all new
(E)-3-(1-Phenylbut-2-en-1-yl)-1H-indole (3bl′).30 Obtained as
inseparable mixture with the regioisomer 3bl. The following data
1
compounds and H NMR copies of known compounds (GC−
1
MS copy in the case of 3do). This material is available free of
corresponds to 3bl′. Brown oil (122 mg, 99%); H NMR (400 MHz,
charge via the Internet at http://pubs.acs.org.
CDCl3) δ1.71 (d, J = 6.5 Hz, 3H), 4.89 (d, J = 7.5 Hz, 1H), 5.49−5.54
(m, 1H), 5.90−5.94 (m, 1H), 6.85 (d, J = 2.3 Hz, 1H), 6.98−7.33 (m,
8H), 7.67 (d, J = 7.9 Hz, 1H), 7.90 (s, 1H) ppm; 13C NMR (100
MHz, CDCl3) δ 17.9, 46.0, 111.1, 119.6, 120.4, 121.9, 126.1, 126.2,
126.8, 128.1, 128.2, 128.4, 135.4, 137.7 ppm.
AUTHOR INFORMATION
Corresponding Author
*E-mail: alex.baeza@ua.es; cnajera@ua.es.
Notes
■
3-Methyl-1-phenylhexa-1,5-diene (3bo) and 4-Phenylhepta-
1,5-diene (3bo′).31 The mixture of the two regioisomers could not
be separated. Colorless oil (61 mg, 71%); 1H NMR (400 MHz,
CDCl3) δ 1.15 (d, J = 6.7 Hz, 3H, 3bo), 1.71 (dd, J = 6.3, 0.8 Hz, 3H,
3bo′), 2.15−2.27 (m, 2H, 3bo), 2.41−2.52 (m, 1H 3bo+ 2H 3bo′),
3.34 (q, J = 7.5 Hz, 1H 3bo′), 4.98−5.12 (m, 2H 3bo + 2H 3bo′),
5.46−5.91 (m, 1H 3bo + 3H 3bo′), 6.20 (dd, J = 15.9, 7.5 Hz, 1H
3bo), 6.41 (d, J = 15.9 Hz, 1H 3bo), 7.18−7.40 (m, 5H 3bo + 5 H
3bo′) ppm; 13C NMR (100 MHz, CDCl3) δ 17.9, 19.9, 36.9, 40.4,
41.4, 48.8, 115.8, 116.0, 125.1, 126.0, 127.6, 128.3, 128.4, 134.4, 136.0,
136.9, 137.8, 144.8 ppm.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Spanish Ministerio de Ciencia e Innovacion
■
́
(MICINN)
(projects CTQ 2007-62771/BQU, CTQ2010-20387 and
Consolider Ingenio 2010, CSD2007-00006), FEDER, the
Generalitat Valenciana (PROMETEO 2009/039) and the
University of Alicante are gratefully acknowledged for financial
support. A.B. would like to thank MICINN for a Juan de la
Cierva contract (JCI-2009-03710).
(E)-3-(4-Phenylbut-3-en-2-yl)pentane-2,4-dione (3br) and
(E)-3-(1-Phenylbut-2-en-1-yl)pentane-2,4-dione (3br′).32 Ob-
tained as inseparable mixture of regioisomers. Colorless oil (88 mg,
1
76%). 3br: H NMR (400 MHz, CDCl3) δ 1.08 (d, J = 6.7 Hz, 3H),
DEDICATION
■
2.13 (s, 3H), 2.22 (s, 3H), 3.14−3.28 (m, 1H), 3.69 (d, J = 10.4 Hz,
1H), 5.99 (dd, J = 15.9, 8.6 Hz, 1H), 6.44 (d, J = 15.9 Hz, 1H), 7.12−
7.30 (m, 5H) ppm; 13C NMR (100 MHz, CDCl3) δ 18.8, 30.0, 30.1,
37.8, 75.6, 126.9, 127.7, 128.8, 130.7, 130.9, 136.7, 203.0, 203.1, ppm.
3br′: 1H NMR (400 MHz, CDCl3) δ 1.61 (d, J = 6.3 Hz, 3H), 1.87 (s,
3H), 2.23 (s, 3H), 4.10 (dd, J = 11.6, 7.3 Hz, 1H), 4.21 (d, J = 11.7 Hz,
1H), 5.46−5.51 (m, 2H), 7.12−7.30 (m, 5H) ppm; 13C NMR (100
MHz, CDCl3) δ 17.9, 29.5, 29.7, 49.1, 74.7, 126.2, 127.5, 128.5, 130.8,
130.9, 140.7, 203.4, 203.5 ppm.
Dedicated to Prof. Rosa Claramunt on the occasion of her 65th
birthday
REFERENCES
■
(1) For reviews about different uses of fluorinated alcohols in organic
transformations, see: (a) Beg
B. Synlett 2004, 18. (b) Shuklov, I. A.; Dubrovina, N. V.; Borner, A.
́
ue,
́
J. -P.; Bonnet-Delpon, D.; Crousse,
̈
Synthesis 2007, 2925.
3-(4′-Methoxyphenyl)cyclohex-1-ene (3dj′).30 Only a small
amount of this product was isolated. The yield of the mixture (3dj′ +
3dj + 3dj″) was determined by 1H NMR (see Table 5). Colorless oil;
1H NMR (300 MHz, CDCl3) δ 1.46−2.10 (m, 6H), 3.33−3.38 (m,
1H), 3.79 (s, 3H), 5.66−5.71 (m, 1H), 5.83−5.89 (m, 1H), 6.84 (d, J
= 8.5 Hz, 2H), 7.14 (d, J = 8.5 Hz, 2H) ppm; 13C NMR (100 MHz,
CDCl3) δ 21.1, 25.0, 32.7, 40.9, 55.3, 113.6, 128.1, 128.6, 130.5, 138.5,
157.8 ppm.
(2) Berkessel, A.; Adrio, J. A.; Huttenhain, D.; Neudorfl, J. M. J. Am.
̈
̈
Chem. Soc. 2006, 128, 8421.
(3) (a) Spanedda, M. V.; Hoang, V. D.; Crousse, B.; Bonnet-Delpon,
D.; Begue, J. -P. Tetrahedron Lett. 2003, 44, 217. (b) Di Salvo, A.;
Spanedda, M. V.; Ourevitch, M.; Crousse, B.; Bonnet-Delpon, D.
Synthesis 2003, 2231.
(4) (a) Das, U.; Crousse, B.; Kesavan, V.; Bonnet-Delpon, D.; Beg
́
́
́
́
ue,
́
J. -P. J. Org. Chem. 2000, 65, 6749. (b) More recently the
stereoselective and regioselective ring opening of chiral epoxides with
indoles and pyrroles has been reported: Westermaier, M.; Mayr, H.
Chem.Eur. J. 2008, 14, 1638.
3-(2′-Methoxyphenyl)cyclohex-1-ene (3dj).30 This product
could not be separated from the disubstituted product (2,4-
dicyclohex-2-enyl)anisole 3dj″ and small amounts of anisole. The
1
experimental data is given for the mixture and only the peaks for H
(5) De, K.; Legros, J.; Crousse, B.; Bonnet-Delpon, D. J. Org. Chem.
2009, 74, 6260.
and 13C NMR are listed since the integration was not accurate. Yield of
1
the mixture (3dj′ + 3dj + 3dj″) was determined by H NMR (see
(6) For recent reviews about free allylic alcohols in allylic substitution
reactions, see: (a) Muzart, J. Eur. J. Org. Chem. 2007, 3077.
(b) Muzart, J. Tetrahedron 2008, 64, 5815. (c) Emer, E.; Sinisi, R.;
Guiteras-Capdevila, M.; Petruzzielo, D.; De Vicentiis, F.; Cozzi, P. G.
Eur. J. Org. Chem. 2011, 647. (d) Bandini, M. Angew. Chem., Int. Ed.
2011, 50, 994. (e) Biannic, B.; Aponick, A. Eur. J. Org. Chem. 2011,
6605.
Table 5). Colorless oil; 1H NMR (300 MHz, CDCl3) δ 1.46−2.11 (m,
18H, 6H 3dj + 12H 3dj″), 3.32−3.37 (m, 1H, 3dj″), 3.80 (s, 3H,
3dj″), 3.83 (s, 3H, 3dl), 3.84 (m, 2H, 1H 3dl + 1H 3dj″), 5.63−5.72
(m, 3H, 1H 3dl + 2H 3dj″), 5.82−5.93 (m, 3H, 1H 3dl + 2H 3dj″),
6.79 (d, J = 8.9 Hz, 1H, 3dj″), 6.84−6.93 (m, 2H, 3dl), 7.01 (m, 2H,
3dj″), 7.15−7.20 (m, 2H, 3dl) ppm; 13C NMR (100 MHz, CDCl3) δ
21.0, 21.1, 21.3, 25.0, 25.2, 30.1, 30.2, 32.7, 34.2, 34.4, 41.1, 41.2, 55.3,
55.4, 109.9, 110.2, 120.3, 125.6, 126.8, 127.9, 128.2, 128.2, 128.3,
128.4, 130.2, 130.4, 130.6, 130.7, 134.2, 155.2, 156.8 ppm. MS(IE):
3dl m/z 188 [M+, 100%], 173 (31), 160 (25), 159 (48), 145 (27), 115
(26), 91 (32); disubst. m/z 268 [M+, 100%], 187 (54), 159 (20), 121
(17), 81 (17).
(7) (a) Giner, X.; Trillo, P.; Naj
696, 357. (b) Trillo, P.; Baeza, A.; Naj
2929.
́
era, C. J. Organomet. Chem. 2010,
́
era, C. Eur. J. Org. Chem. 2012,
(8) During the preparation of this manuscript an asymmetric α-
alkylation of aldehydes with benzylic alcohols in fluorinated alcohols
was reported: Xiao, J.; Zhao, K.; Loh, T.-P. Chem. Commun. 2012, 48,
3548.
(9) In all cases a 1 M solution of 1a in the indicated solvent was
employed, corresponding to 10 equiv of HFIP, 14 equiv of TFE, 17
equiv of EtOH, 11 equiv of PhOH, and 56 equiv of H2O.
(10) For a SN1 reaction of benzylic alcohols mediated by water, see:
Cozzi, P. G.; Zoli, L. Angew. Chem., Int. Ed. 2008, 47, 4162.
(11) The use of more basic amines seems to be restricted to
processes involving π-allylmetal complexes. Thus, palladium and
3-Allylcyclohex-1-ene (3do).31 Colorless oil. Due to the
impossible separation of the substitution product from the
corresponding fluoroalkyl ether and the starting material, the yield
was determined by GC analysis of the crude mixture (see Table 5); 1H
NMR (300 MHz, CDCl3) δ 1.20 (m, 2H), 1.49 (m, 2H), 1.73 (m,
2H), 1.93−2.03 (m, 4H), 2.14 (m, 1H), 5.00 (m, 2H), 5.59 (m, 1H),
5.67 (m, 1H), 5.83 (m, 1H) ppm; 13C NMR (100 MHz, CDCl3) δ
21.3, 25.3, 28.9, 35.0, 40.4, 115.9, 127.2, 131.2, 137.5 ppm.
J
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
platinum complexes have been used for this purpose. For examples of
palladium-catalyzed allylic substitution reaction with basic amines, see
ref 6a. For a recent example of the use of platinum-based catalyst, see:
Ohshima, T.; Miyamoto, Y.; Ipposhi, J.; Nakahara, Y.; Utsunomiya,
M.; Mashima, K. J. Am. Chem. Soc. 2009, 131, 14317.
(12) Recently an azidation of allylic alcohols catalyzed by AgOTf has
been reported; see: Rueping, M.; Vila, C.; Uria, U. Org. Lett. 2012, 14,
768.
(13) The configuration of the exocyclic doble bond in isomer 3ea′
turned out to be E and was determined by NOESY experiments.
(14) (a) March, J. Advanced Organic Chemistry: Reactions, Mechanism
and Structure, 4th ed.; John Wiley & Sons: New York, 1992; p 560. For
selected recent examples, see: (b) Anderson, L. L.; Arnold, J.;
Bergman, R. G. J. Am. Chem. Soc. 2005, 127, 14542. (c) Magnus, P.;
Turnbull, R. Org. Lett. 2006, 8, 3497. (d) Motokura, K.; Nakagiri, N.;
Mizugaki, T.; Ebitani, K.; Kaneda, K. J. Org. Chem. 2007, 72, 6006 and
references therein.
(15) The same behavior was observed by us and others when a
Brønsted acid was employed as catalyst for the allylic subsitutiton
reaction using dialkyl malonates onto free alcohols. See ref 7b.
(16) The enantioenriched alcohol (S)-1b was obtained according to a
described procedure; for details, see: Fernan
́ ́
dez-Mateos, E.; Macia, B.;
Ramon, D. J.; Yus, M. Eur. J. Org. Chem. 2011, 6851 . The amination
́
product 3ba was analyzed by HPLC. The corresponding fluoroalkyl
ethers obtained as byproducts were also isolated as racemic mixtures in
both cases.
(17) This isomerization of alcohol 1c has been observed by us (see
ref 7b) and by others; see: McCubbin, J. A.; Voth, S.; Krokhin, O. V. J.
Org. Chem. 2011, 76, 8537.
(18) Baudoux, J.; Perrigand, K.; Madec, P. J.; Gaumont, A. C.; Dez, I.
Green Chem. 2007, 9, 1346.
(19) Cimarelli, C.; Palmieri, G. Tetrahedron: Asymmetry 2000, 11,
2555.
(20) Ohta, T.; Sasayama, H.; Nakajima, O.; Kurahashi, N.; Fujii, T.;
Fukukama, I. Tetrahedron: Asymmetry 2003, 14, 537.
(21) Yang, H.; Fang, L.; Zhang, M. Eur. J. Org. Chem. 2009, 666.
(22) Kabalka, G. W.; Yao, M. L.; Borella, S. Org. Lett. 2006, 8, 879.
́
(23) Alacid, E.; Najera, C. Org. Lett. 2008, 10, 5011.
(24) Liu, P. N.; Zhon, Z. Y.; Lau, C. P. Chem.Eur. J. 2007, 13,
8610.
(25) (a) Ross, J.; Chen, W.; Xu, L.; Xiao, J. Organometallics 2001, 20,
138. (b) Haslego, M. L.; Smith, F. X. Synth. Commun. 1980, 10, 421.
(26) Haight, A. R.; Stoner, E. J.; Peterson, M. J.; Grover, V. K. J. Org.
Chem. 2003, 68, 8092.
(27) Pyne, S. G.; Dong, Z. J. Org. Chem. 1996, 61, 5517.
(28) Yamamoto, H.; Ho, E.; Sasaki, I.; Mitsutake, M.; Takagi, Y.;
Imagawa, H.; Nishizawa, M. Eur. J. Org. Chem. 2011, 2417.
(29) Liao, L.; Sigman, M. S. J. Am. Chem. Soc. 2010, 132, 10209.
(30) Malkov, A. V.; Davis, S. L.; Baxendale, I. R.; Mitchell, W. L.;
Kocovsky, P. J. Org. Chem. 1999, 64, 2751.
(31) Meyer, V. J.; Niggemann, M. Eur. J. Org. Chem. 2011, 3671.
́
(32) Sanz, R.; Martínez, A.; Miguel, D.; Alvarez-Gutierrez, J. M. Adv.
Synth. Catal. 2006, 348, 1841.
(33) Zhu, A.; Li, L.; Wang, J.; Zhuo, K. Green. Chem. 2011, 13, 1244.
(34) Ohshima, T.; Nakahara, Y.; Ipposhi, J.; Miyamoto, Y.; Mashima,
K. Chem. Commun. 2011, 8322.
(35) Sreedhar, B.; Ravi, V.; Yada, D. Tetrahedron Lett. 2011, 52, 1208.
́
(36) Giner, X.; Najera, C. Org. Lett. 2008, 10, 2119.
K
dx.doi.org/10.1021/jo301049w | J. Org. Chem. XXXX, XXX, XXX−XXX