Oxidation of Olefins with Benzeneseleninic Anhydride in the Presence of TMSOTf

Article abstract of DOI:10.1021/acs.joc.5b00410

A new oxidizing system for olefins, consisting of benzeneseleninic anhydride and trimethylsilyl triflate, was studied. The highly reactive benzeneseleninyl cation is presumably formed under these conditions. It has been shown that different products are f

Full text of DOI:10.1021/acs.joc.5b00410

Article
pubs.acs.org/joc
Oxidation of Olefins with Benzeneseleninic Anhydride in the
Presence of TMSOTf
Izabella Jastrzębska,*,† Maja Morawiak,‡ Joanna E. Rode,‡,§ Barbara Seroka,† Leszek Siergiejczyk,†
and Jacek W. Morzycki*,†
†Institute of Chemistry, University of Białystok, ul. Ciołkowskiego 1K, 15-245 Białystok, Poland
‡Institute of Organic Chemistry, Polish Academy of Sciences, ul. Kasprzaka 44, 01-224 Warsaw, Poland
§Institute of Nuclear Chemistry and Technology, ul. Dorodna 16, 03-195 Warsaw, Poland
S
* Supporting Information
ABSTRACT: A new oxidizing system for olefins, consisting of
benzeneseleninic anhydride and trimethylsilyl triflate, was
studied. The highly reactive benzeneseleninyl cation is
presumably formed under these conditions. It has been
shown that different products are formed with this species
depending on the specific structure of olefin. The 1,1-
disubstituted olefins afforded mostly α,β-unsaturated carbonyl
compounds. The sterically encumbered tri- or tetrasubstituted
olefins yielded 1,2- or 1,4-dihydroxylated products, presumably
via four-membered cyclic intermediates.
was to test the effect of trimethylsilyl trifluoromethanesulfonate
(TMSOTf) on the oxidation reactions with BSA. Herein, we
report a new and effective one-pot method for the oxidation of
olefins.
INTRODUCTION
■
Benzeneseleninic anhydride (BSA) is a classical organo-
selenium reagent frequently used, among other tetravalent
selenium compounds, for the oxidation of organic compounds.1
At first, BSA was mainly used for alcohol oxidation. The reagent
was introduced in the late 1970s by Barton et al.2 However,
BSA oxidation is also a valuable tool for preparation of various
oxygenated compounds. For instance, previously reported
reactions employing BSA include the oxidation of phenols to
hydroxycyclohexadienones3 or ortho-quinones,4 thiocarbonyl
compounds to their corresponding oxo derivatives,5 hydrazones
and hydroxylamines,6 hydrazones, oximes, and semicarbazones
to ketones,7 and hydrazines to diazenes.8 The dehydrogenation
reactions of ketones9 and lactams,10 as well as the allylic
oxidation of alkenes,11 have also been documented. Further-
more, diphenyl diselenide has been used as a catalytic reagent
in the presence of a stoichiometric oxidizing reagent such as di-
tert-butyl peroxide,12 iodoxybenzene,13 or hydrogen peroxide.14
Apparently, in these processes, BSA (or benzeneseleninic acid)
was formed in situ.15
In the 1990s, certain efforts were directed toward the use of
BSA in the presence of diphenyl diselenide/triflic anhydride.16
Under these conditions, benzeneselenenyl triflate [PhSeOS-
(O)2CF3] was generated. The reaction of terminal alkenes with
this species produced vicinal areneselenenyl triflates. Also,
olefin oxidation with benzeneseleninyl trifluoroacetate [PhSe-
(O)OC(O)CF3], generated in situ from BSA and trifluoro-
acetic anhydride, was briefly studied.16,17 The oxidation of
disubstituted olefins with this reagent effectively yielded 2-
(phenylseleno)-1,3-alkanediols. The aim of the present study
RESULTS AND DISCUSSION
■
At the beginning, the oxidation of Δ5-steroids with BSA/
TMSOTf was studied. The reactions were carried out with 1
equiv of BSA and 0.5 equiv of TMSOTf relative to the starting
steroidal olefin in dry dichloromethane for 45−60 min [thin-
layer chromatography (TLC) control] at room temperature.
Surprisingly, the trans-dihydroxylation of the double bond
occurred during the reaction and afforded 5α,6β-dihydroxy
derivatives (1−5) of starting Δ5-steroids in good yields (Table
1).
Due to the relatively mild reaction conditions, even sensitive
groups such as tosylate, TBDMS, and the spirostane side chain
remained intact during the reaction. Steroidal 17- and 20-
ketone groups also proved to be compatible with the oxidizing
system.
It is well-known that the C5−C6 double bond of steroids is
resistant to BSA. Since blank experiments have shown that the
starting olefins do not react with TMSOTf, it can be concluded
that a new reactive species is formed in situ (Scheme 1). The
most likely candidate is benzeneseleninyl triflate, which is
produced in addition to trimethylsilyl benzeneseleninate. The
Received: February 21, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Oxidation of Δ5-Steroids with BSA/TMSOTf
One can assume that the naked benzeneseleninyl cation, which
is formed from PhSe(O)OTf in a solvent-stabilizing cation
(e.g., dichloromethane), is represented by two mesomeric
structures shown below (Scheme 2). However, due to a
significant difference in electronegativity between selenium and
oxygen atoms, the contribution from the structure on the right
would be rather minimal.
Nevertheless, the reaction mechanism shown in Scheme 3
with an oxonium ion I as a key intermediate, which resulted
from an attack by the oxygen electrophilic center onto the
Scheme 1. Reaction of BSA with Trimethylsilyl Triflate
former is probably the active reagent, and the latter serves as a
reservoir of oxygen because the transfer of oxygen is possible.
Thus, in fact, a 2-fold excess of oxidizing agent was present in
the reaction mixture.
It has been previously reported17 that a similar Se(IV)
electrophile (benzeneseleninyl trifluoroacetate) reacts readily
with alkenes at room temperature to produce β-trifluoroacetoxy
selenoxides that undergo further transformations under the
reaction conditions. However, in our case, the product
structures may suggest that the double bond is stereoselectively
(from the α-side) attacked by the oxygen atom rather than by
the selenium atom. In the case of tri- or tetrasubstituted olefins,
such an attack would be particularly favorable for steric reasons.
Scheme 2. Mesomeric Structures of a Benzeneseleninyl
Cation
B
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 3. Tentative Reaction Mechanism via a Three-
Membered Cyclic Intermediate
olefin from the α-side, was taken into consideration. The ion I
could be cleaved to the trans-diol during the aqueous workup of
the reaction mixture, or the oxirane ring is first opened with
triflate followed by the hydrolysis of a labile diester.
However, there exists an alternative, more likely, reaction
mechanism involving formation of a four-membered inter-
mediate II (Scheme 4), which may result from either an initial
Scheme 4. Alternative Reaction Mechanism via a Four-
Membered Cyclic Intermediate
Figure 1. X-ray structure measured at 158 K showing the
stereochemistry of compound 14. Thermal ellipsoids are drawn at
the 50% probability level. Disordered n-propanol molecule and the
side chain with an occupancy of ca. 20% have been omitted for clarity.
the benzeneseleninyl cation to the double bond. The highly
hindered olefin 13 reacts with a benzeneseleninyl cation,
affording the selenoxetane intermediate III. The reaction is
highly regio- and stereoselective. In spite of the fact that C10 is
more electron-rich than C9 and access to the reagent is easier
from the β-side, the calculations (see below) showed that the
selenium reagent preferably approaches C9 from the α-side due
to the lower energy transition states. In the next step, the
selenoxetane intermediate III is probably cleaved by a
nucleophile (e.g., triflate) with simultaneous elimination of
PhSeOH. A different reaction pathway compared to that
previously described for analogous intermediate II may result
from a high steric strain in III, which can be released by
elimination of PhSeOH. The newly formed C9−C11 double
bond undergoes further electrophilic attack by the benzenese-
leninyl cation. Further syn-elimination of selenoxide afforded
the C11−C12 olefin. Finally, the SN2′ addition of an oxygen
nucleophile from the less hindered α-side to the olefin at C12,
followed by hydrolysis, furnished the final product 14. The
proposed mechanism of C12 hydroxylation of the C9−C11
olefin differs from the usual allylic hydroxylation with the
selenium species.22 Alternative reaction mechanisms can also be
taken into consideration, including the one with an intra-
molecular transfer of the benzeneseleninyl cation.
Further study on BSA/TMSOTf oxidation was carried out
with exomethylenic olefins (Table 2). The reactions of 6- and
12-exomethylene steroids 15 and 17 proved to be much less
efficient than the reactions of more hindered olefins. The 6-
exomethylene 5α-cholestane derivative 15, when treated with
BSA/TMSOTf, afforded 6-formylcholesteryl acetate 16 in a
40% yield. The analogous reaction of 12-exomethylene
tigogenin 17 yielded a mixture of 12ξ-hydroxy-12-formyl
derivatives 18a and 18b in low yields. The reaction of lupenyl
acetate 19 with benzeneseleninic anhydride in the presence of
TMSOTf afforded 3β-acetoxylup-20(30)-en-29-al23 20 in a
satisfactory yield (75%). The obtained results can be explained
by a plausible mechanism shown in Scheme 7. In contrast to
the dihydroxylation reactions observed for tri- and tetrasub-
stituted olefins, in the case of less substituted olefins, the
unsaturated aldehydes or α-hydroxyaldehydes were formed.
Most likely, an attack of a benzeneseleninyl cation onto the
double bond with a soft electrophilic center at selenium
electrophilic attack from the α-side of the selenium species onto
the olefin followed by cyclization or a concerted [2 + 2]
cycloaddition. Then, the selenoxetane ring is opened by a
nucleophilic attack of triflate on the carbon atom linked to
selenium from the opposite side (SN2). The two alternative
mechanisms will be discussed later within this paper.
The conditions of the oxidation reaction were briefly
optimized. The oxidation process was carried out with different
amounts of BSA and TMSOTf (from 0.5 up to 1.25 equiv) at
various temperatures (from −10 °C to rt). The reaction with an
excess of BSA and higher contents of TMSOTf afforded lower
product yields. The best results were achieved with 1.0 equiv of
BSA in the presence of 0.5 equiv of TMSOTf at room
temperature as in the first experiments. The optimal reaction
time was found to be 45−60 min. When the oxidation of
cholesteryl acetate (1) was extended to 2 days, the yield of
5α,6β-diol 2 dropped to 50% and the more oxidized product,
3β-acetoxy-5α-hydroxycholestan-6-one (43%), appeared. The
oxidation of the secondary alcohols to ketones with
benzeneseleninic anhydride is a known reaction,2 but it is
much slower than the reaction of BSA/TMSOTf with olefins.
In the next experiment, the reaction of the tetrasubstituted
olefin 1321 was studied (Scheme 5). The reaction proceeded
smoothly, affording 10β,12α-diol 14 in high yield. The
structure of compound 14 was confirmed by single-crystal X-
ray examination (Figure 1).
The plausible reaction pathway leading to this product is
outlined in Scheme 6. It seems that the initial step was
analogous to that in the Δ5-steroids, that is, the cycloaddition of
Scheme 5. BSA Oxidation of Tetrasubstituted Olefin 13
C
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 6. Tentative Mechanism of the BSA Oxidation of Compound 13
Table 2. Oxidation of 1,1-Disubstituted Olefins with BSA/TMSOTf
occurred. The carbocation thus-formed may undergo a proton
abstraction or an oxygen nucleophile addition. Further
oxidation leads to the α,β-unsaturated (e.g., 16 or 20) or α-
hydroxy (e.g., 18a or 18b) carbonyl compounds, respectively.
However, a different reaction course was observed for
(+)-camphene 21. The initially formed carbocation underwent
various Wagner−Meerwein-type rearrangements (Scheme 8).24
The expected unsaturated aldehyde could not be formed due to
the steric hindrance. Two reaction products were formed, but
none of them showed the presence of a formyl group. Instead,
the oxoselenide 22 and the dioxo compound 23 were obtained
as products. The former compound underwent a trans-
D
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
the desired intermediate ions I (Scheme 3) and II (Scheme 4).
In the route leading to II, according to Markovnikov’s rule, the
O atom attack on the positively charged carbon atom was
considered only (the NBO analysis of the charge distribution
can be seen in Supporting Information Figure S1C). It is a
spontaneous reaction; however, a competition between
thermodynamic and kinetic reaction control is predicted: the
kinetics favors the αR3 intermediate product (I), while
Scheme 7. Tentative Reaction Pathway of Lupenyl Acetate
19 with BSA
α
thermodynamics favors the R4 (II) one. Moreover, in ion I,
the PhSe(O)+ moiety adopts a nonplanar geometry (followed
by spectacular changes in the charge distribution within both
the PhSe(O)+ and M7 moieties in I, Table S4P), and the Se−O
and Ph−Se distances are longer in comparison to the
appropriate distances in the isolated system by ca. 0.3 and
0.02 Å, respectively (Figure S2). Because the difference in
stability of αR3 and αR4 ions is ca. 6 kcal/mol, and the αTS-R4
is above αTS-R3 by ca. 1.3 kcal/mol, the thermodynamic
product II dominates under the reaction conditions employed.
Moreover, this reaction is a concerted [2 + 2] cycloaddition, yet
the process is highly asynchronous: the longer σ(C−Se) bond
in αR4 is formed earlier in the αTS-R4 than is the shorter σ(C−
O) one (Figure 2).
formation to the latter upon further oxidation and rearrange-
ment. The driving force for this rearrangement is a tendency to
assume a bicyclo[2.2.2]octane structure that is less strained
than the starting bicyclo[2.2.1]heptane ring system. It should
be mentioned that diketone 23 is a known compound,25 and
spectral data of the obtained product are consistent with the
proposed structure. Of course, compound 23 obtained from 21
20
was optically active ([α]D +18.2 (c 0.5, CHCl3), unlike the
racemic diketone described in the literature. Interestingly, an
alternative pathway leading to the symmetrical diketone 24 did
not take place.
Also, the PhSe(O)+ β-side attack on the C5−C6 double
bond of the model M7 steroid was taken into account.
However, as expected, this reaction path is not energetically
favored (Supporting Information Diagram S1).
To provide an insight into the mechanism of the PhSe(O)+
attack on the olefin double bond, two alternative pathways,
leading to four- and three-membered intermediate products,
have been studied by quantum-chemical calculations. The
calculations were performed at the B3LYP/6-31G**/PCM-
(CH2Cl2) level by using the Gaussian 09 package of programs
(see the Supporting Information for computational details).
Such an approach requires, at first, finding the most stable
conformers of the interacting systems for the substrates as well
as for the products. As the reagents approach each other during
the process, they reorganize themselves, weakly bond, and an
entrance channel complex (ECC) is formed. To form the
products, the system has to overcome an energetic barrier.
Estimation of the barrier at a given reaction pathway requires
proper identification of the transition state structure (TS).
The first simulated reaction was the PhSe(O)+ α-side attack
on the C5−C6 double bond of model steroid 7 (M7), and the
corresponding energetic diagram is presented in Figure 2.
Considered reaction paths pass through four- and three-
membered TSs (TS-R4 and TS-R3, respectively) and lead to
The other studied reaction was the PhSe(O)+ attack on the
olefin double bond of the model reactant 13 (M13). In this
case, the attacks from both sides (α and β) leading to formation
of the four- or three-membered intermediate III (R4 or R3,
respectively) and the selenium atom attack on the C9 or C10
atoms were considered. The α-side attack is energetically
favored in comparison to the β-attack (Supporting Information
Diagram S2) and is presented in Figure 3. For each of the
reaction paths, the ECCs and the appropriate TSs via the four-
membered ring geometries (TS-R4) were found. Unfortu-
nately, the TSs via the three-membered ring structures (TS-R3)
were not found because of convergence problems. All of the
considered structures are collected in the Supporting
Information Tables S5−S8.
In the preferred reaction path (Figure 3), PhSe(O)+
approaches M13 from the α-side: the reaction starts from
ECC-αR4(SeC9), passes through the lowest αTS-R4(SeC9)
activation barrier, and ends up with the most stable
intermediate product αR4(SeC9). Thus, αR4(SeC9) is both
Scheme 8. Tentative Reaction Pathway of (+)-Camphene 21 with BSA
E
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. Energetic diagram (in terms of ΔG values, kcal/mol) of PhSe(O)+ α-attack on the double bond of model olefin M7 calculated at the
B3LYP/6-31G**/PCM(CH2Cl2) level. Atoms are color-coded: carbon (gray), hydrogen (white), oxygen (red), and selenium (yellow).
kinetically and thermodynamically the most stable ion. What is
is followed by different consecutive reactions that cause the
method to be rather ineffective. However, the BSA/TMSOTf
reactions with encumbered tri- and tetrasubstituted olefins
proceeded smoothly, affording products in high yields. The
dihydroxylated products are formed presumably via the four-
membered cyclic intermediate that is formed by a concerted [2
+ 2] cycloaddition of a benzeneseleninyl cation to the olefin.
The DFT calculations show that such a mechanism is superior
to alternative mechanisms.
α
surprising, to form R4(SeC9), the positively charged Se atom
approaches the less negative charged C9 carbon atom of M13
(Figure S5C). A deeper insight into the charge distribution
indicates that, during the process, the electron density is
reorganized, and in αTS-R4(SeC9), it is changed (Table S7D) in
α
comparison to the noninteracting M13. In the TS-R4(SeC9)
structure, the C9 atom becomes more negative (and is more
likely to form a bond with the positively charged Se atom) and
the C10 atom becomes more positive (and prepared for bond
formation with the negatively charged O atom). The stability of
the reaction species was additionally checked at two different
calculation levels: one accounting for dispersion (B3LYP-D3)
and the other by using a basis set incorporating the diffuse
functions (6-31+G**, Table S3). The energetic order is not
EXPERIMENTAL SECTION
■
General Methods. Reagent-grade chemicals were purchased and
used as received. The methylene chloride was dried prior to use by
distillation over CaH2. Flash column chromatography and dry flash
chromatography were performed with a silica gel, with a pore size of
40 Å (70−230 mesh), unless otherwise stated. Reactions were
monitored by TLC on silica gel plates 60 F254. All reactions were
α
changed; therefore, stability of R4(SeC9) is not a computa-
tional artifact. Analyzing the geometry and visualizing the
α
imaginary frequency of the TS-R4(SeC9) structure, one can
carried out under an argon atmosphere. H and 13C NMR data for all
1
conclude that the PhSe(O)+ α-side attack on the olefin double
bond of the model M13 is a concerted synchronous [2 + 2]
cycloaddition (Figure 3).
previously uncharacterized compounds were recorded at ambient
temperature using 400 and 200 MHz spectrometers and are referenced
to tetramethylsilane (0.0 ppm) and CDCl3 (77.0 ppm), respectively,
unless otherwise noted. NMR resonance multiplicities are reported
using the following abbreviations: br = broad, s = singlet, d = doublet, t
= triplet, q = quartet, and m = multiplet; coupling constants, J, are
reported in hertz. IR spectral data were obtained using an FT-IR
spectrometer (taken in a CHCl3 solution) and are reported in cm−1.
CONCLUSION
■
A new and effective, one-pot method for the oxidation of
olefins has been reported. The method consists of the
generation of a reactive species, presumably the benzenesele-
ninyl cation, by the treatment of BSA with TMSOTf. In the
case of BSA/TMSOTf reactions with simple mono- or
disubstituted olefins, the initial step is usually an electrophilic
attack of selenium at the less substituted carbon atom. This step
Melting points were determined by a Kofler bench (Boetius type)
̈
apparatus and are uncorrected. HR-MS values were obtained on an
electrospray ionization time of flight (ESI-TOF) or electron ionization
(EI) spectrometers. The specific rotations were acquired on a digital
polarimeter.
F
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 3. Energetic diagram (in terms of ΔG values, kcal/mol) of PhSe(O)+ α-attack on the double bond of model olefin M13 calculated at the
B3LYP/6-31G**/PCM(CH2Cl2) level. Atoms are color-coded: carbon (gray), hydrogen (white), oxygen (red), and selenium (yellow).
General Experimental Procedure. TMSOTf (0.5 equiv) was
added to BSA (1.0 equiv) in dry dichloromethane under argon at
room temperature. After 5−10 min of stirring, a solution of alkene
(0.2−0.5 mmol) in dry dichloromethane was added. The reaction
mixture was stirred under argon. The disappearance of the starting
material spot was monitored by TLC. Water was added to the reaction
mixture, and then it was extracted with CH2Cl2, dried over MgSO4,
and evaporated.
5α,6β-Dihydroxycholesterol Acetate (2). The reaction with
cholesterol acetate (1; 100 mg) was carried out (reaction time = 1 h).
Silica gel column chromatography gave pure compound 2 (70 mg;
67%), eluted with 1:1 ethyl acetate/hexane. Compound 2 was proven
in all respects to be identical to the same compound described in the
literature.18
5α,6β-Dihydroxycholesterol p-Toluenesulfonate (4). The
reaction with cholesterol p-tosylate (3; 108 mg) was carried out
(reaction time = 1 h). Silica gel column chromatography gave pure
compound 4 (92 mg; 80%) eluted with 17% ethyl acetate in hexane:
colorless crystals; mp 149−150 °C (acetone); lit.26 mp 148−150 °C;
1H NMR δ 7.79 (d, J = 8.2 Hz, 2H), 7.32 (d, J = 8.0 Hz, 2H), 4.93 (m,
1H), 3.50 (br s, 1H), 2.44 (s, 3H), 2.35 (t, J = 11.9 Hz, 1H), 1.97 (d, J
= 12.7 Hz, 1H), 1.14 (s, 3H), 0.90 (s, 3H), 0.89 (s, 3H), 0.86 (dd, J =
6.5, 1.3 Hz, 6H); ESI-MS (m/z, %) 441 ([MNa+ − Ts− − H]+, 100).
Anal. Calcd for C34H54O5S: C, 71.04; H, 9.47. Found: C, 70.90; H,
9.49.
(25R)-5α,6β-Dihydroxyspirostan-3β-ol Acetate (6). The re-
action with diosgenin acetate (5; 100 mg) was carried out (reaction
time = 3 h). Silica gel column chromatography gave pure compound 6
(70 mg; 65%) eluted with 1:1 ethyl acetate/hexane: colorless crystals;
mp 256−258 °C (CH2Cl2); IR (KBr) νmax (cm−1) 3620, 3461, 1717,
1
1254; H NMR δ 5.15 (m, 1H), 4.40 (m, 1H), 3.54 (br s, 1H), 3.49
(dd, J = 10.9, 4.6 Hz, 1H), 3.37 (t, J = 10.9 Hz, 1H), 2.18 (t, J = 12.7
Hz, 1H), 2.03 (s, 3H), 1.21 (s, 3H), 0.97 (d, J = 6.7 Hz, 3H), 0.80 (s,
3H), 0.79 (d, J = 6.2 Hz, 6H); 13C NMR δ 171.0 (C), 109.2 (C), 80.7
(CH), 75.9 (CH), 75.5 (C), 71.3 (CH), 66.8 (CH2), 62.1 (CH), 55.7
(CH), 45.4 (CH), 41.6 (CH), 40.6 (C), 39.9 (CH2), 38.4 (C), 36.9
(CH2), 34.8 (CH2), 32.1 (CH2), 31.7 (CH2), 31.3 (CH2), 30.2 (CH),
29.8 (CH), 28.8 (CH2), 26.6 (CH2), 21.4 (CH3), 20.9 (CH2), 17.1
(CH3), 16.6 (CH3), 16.5 (CH3), 14.4 (CH3); ESI-MS (m/z, %) 513
([M + Na]+, 100), 1003 ([2M + Na]+, 50). Anal. Calcd for C29H46O6:
C, 70.99; H, 9.45. Found: C, 71.11; H, 9.28.
G
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
3β-Acetoxy-5α,6β-dihydroxyandrostan-17-one (8). The reac-
tion with androst-5-en-3β-ol-17-one acetate (7; 72 mg) was carried
out (reaction time = 1 h). Silica gel column chromatography gave pure
compound 8 (60 mg; 75%) eluted with ethyl acetate. Compound 8
was proven in all respects to be identical to the same compound
described in the literature.19
3β-tert-Butyldimethylsililoxy-5α,6β-dihydroxypregnan-20-
one (10). The reaction with pregn-5-en-3β-ol-20-one tert-butyldime-
thylsilyl ether (9; 96 mg) was carried out (reaction time = 1 h). Silica
gel column chromatography gave pure compound 10 (58 mg; 60%)
eluted with 20% ethyl acetate in hexane. Compound 10 was proven in
all respects to be identical to the same compound described in the
literature.20
compound 16 as an oil (32 mg; 40%) eluted with 10% ethyl acetate in
1
hexane: colorless oil; IR (KBr) νmax (cm−1) 1728, 1662, 1250; H
NMR δ 10.19 (s, 1H), 4.70 (m, 1H), 3.70 (ddd, J = 13.9, 4.4, 1.8 Hz,
1H), 2.49 (ddd, J = 17.8, 5.0, 2.8 Hz, 1H), 2.07 (s, 3H), 1.16 (s, 3H),
0.88 (d, J = 1.8 Hz, 3H), 0.86 (d, J = 1.8 Hz, 3H), 0.69 (s, 3H); 13C
NMR δ 191.0 (CH), 170.3 (C), 159.1 (C), 133.7 (C), 73.0 (CH),
56.6 (CH), 56.1 (CH), 49.3 (CH), 42.2 (C), 39.5 (CH2), 39.4 (CH2),
38.9 (C), 36.3 (CH2), 36.1 (CH2), 35.8 (CH), 30.6 (CH), 29.7
(CH2), 29.6 (CH2), 28.2 (CH2), 28.0 (CH), 27.1 (CH2), 24.2 (CH2),
23.8 (CH2), 22.8 (CH3), 22.5 (CH3), 21.3 (CH3), 21.1 (CH2), 19.8
(CH3), 18.7 (CH3), 11.8 (CH3); ESI-MS (m/z, %) 479 ([M + Na]+,
100). Anal. Calcd for C30H48O3: C, 78.90; H, 10.59. Found: C, 78.66;
H, 10.69.
3β,5α,6β,21-Tetrahydroxypregnan-20-one 3,21-diacetate
(12). The reaction with pregn-5-ene-3β,21-diol-20-one 3,21-diacetate
(11; 82 mg) was carried out (reaction time = 3 h). Silica gel column
chromatography gave pure compound 12 (78 mg; 87%) eluted with
ethyl acetate: colorless crystals; mp 233−235 °C (CH2Cl2); IR (KBr)
νmax (cm−1) 3521, 1750, 1728, 1232; 1H NMR (CD3OD) δ 4.72 (d, J
= 16.8 Hz, 1H), 4.54 (d, J = 16.8 Hz, 1H), 4.11 (m, 1H), 3.56 (br s,
1H), 2.52 (t, J = 8.8 Hz, 1H), 2.18 (s, 3H), 1.19 (s, 3H), 0.69 (s, 3H);
13C NMR (CD3OD) δ 206.3 (C), 172.0 (C), 76.7 (C), 76.4 (CH),
70.5 (CH2), 68.3 (CH), 60.3 (CH), 57.7 (CH), 46.4 (CH), 46.1 (C),
41.4 (CH2), 39.8 (CH2), 39.4 (C), 33.5 (CH2), 31.7 (CH), 31.7
(CH2), 25.5 (CH2), 23.8 (CH2), 22.3 (CH2), 20.3 (CH2), 17.3 (CH3),
13.7 (CH3); ESI-MS (m/z, %) 431 ([M + Na]+, 100). Anal. Calcd for
C25H38O7: C, 66.64; H, 8.50. Found: C, 66.41; H, 8.63.
5β-Methyl-19-norcholest-9(11)-ene-3β,6β,10β,12α-tetraol
3,6-Diacetate (14). The reaction with 5β-methyl-19-norcholest-
9(10)-ene-3β,6β-diol 3,6-diacetate (13; 95 mg) was carried out
(reaction time = 50 min). Silica gel column chromatography gave pure
compound 14 (86 mg, 85%) eluted with 30% ethyl acetate in hexane:
colorless crystals; mp 95−96 °C (n-propanol/CH2Cl2); [α]D20 −1.5 (c
0.5, CHCl3); IR (KBr) νmax (cm−1) 3578, 1732, 1035; 1H NMR δ 5.97
(d, J = 5.6, 1.6 Hz 1H), 5.01 (t, J = 3.1 Hz 1H), 4.78 (bs, 1H), 4.03 (d,
J = 5.6 Hz, 1H), 2.99 (br s, 1H), 2.15 (s, 3H), 2.04 (s, 3H), 0.99 (d, J
= 6.6 Hz, 3H), 0.87 (2d, J = 6.6 Hz, 3H), 0.64 (s, 3H); 13C NMR δ
170.2 (C), 169.5 (C), 141.9 (C), 124.6 (CH), 74.5 (C), 72.0 (CH),
68.8 (CH), 47.2 (CH), 45.7 (CH), 44.8 (C), 41.2 (C), 39.4 (CH2 x
2), 36.2 (CH2), 35.9 (CH2), 35.3 (CH), 31.4 (CH), 31.2 (CH2), 27.9
(CH × 2), 27.7 (CH2 × 2), 24.7 (CH2), 23.8 (CH2), 22.7 (CH3), 22.5
(CH3), 21.4 (CH3), 21.2 (CH2), 18.8 (CH3), 17.5 (CH3), 11.2
(CH3); ESI-MS (m/z, %) 541 ([M + Na]+, 100); HRMS (ESI TOF)
m/z [M + Na]+ calcd for C31H50O6Na 541.3505; found 541.3492.
X-ray Structure Determination. C31H49O6, C3H8O, MW = 577.79,
orthorhombic space group P21212, a = 17.1991(6), b = 32.357(1), c =
5.9359(2) Å, V = 3303.4(2) Å3, F(000) = 1268, dcalc = 1.162 Mg m−3,
Z = 4, m(Cu Kα) = 0.632 cm−1, T = 158 K, crystal size 0.421 × 0.165
× 0.107 mm, 30 264 reflections measured and corrected for numerical
absorption with Tmin = 0.777 and Tmax = 0.935, 5976 reflections were
unique (Rint = 0.0525), final R1 = 0.0576, wR2 = 0.1600, for 5557
observed reflections with [I > 2σ(I)]; GOF = 1.040. Absolute structure
(Flack) parameter 0.1(3).
(25R)-12-Formylo-5α-spirostane-3β,12ξ-diol 3-Acetate (18a
and 18b). The reaction with (25R)-12-methylene-5α-spirostan-3β-ol
acetate (17; 100 mg) was carried out (reaction time = 40 min). Silica
gel column chromatography gave compound 18a (10 mg; 10%)
followed by the more polar 18b (7 mg, 7%) eluted with 10% ethyl
acetate in hexane. 18a (amorphous solid): IR KBr) νmax (cm−1) 3494,
1715, 1256, 1049; 1H NMR δ 10.16 (d, J = 1.0 Hz, 1H), 4.68 (m, 1H),
4.37 (m, 1H), 3.52 (d, J = 1.0 Hz, 1H), 3.45 (ddd, J = 11.0, 4.4, 2.0 Hz,
1H), 3.33 (t, J = 11.0 Hz, 1H), 2.02 (s, 3H), 1.01 (s, 3H), 0.90 (d, J =
6.9 Hz, 3H), 0.88 (s, 3H), 0.78 (d, J = 6.3 Hz, 3H); 13C NMR δ 206.1
(CH), 170.6 (C), 109.5 (C), 80.9 (C), 79.8 (CH), 73.2 (CH), 66.9
(CH2), 57.6 (CH), 52.6 (CH), 51.5 (CH), 48.0 (C), 44.7 (CH), 42.2
(CH), 36.6 (CH2), 35.9 (C), 34.6 (CH), 33.8 (CH2), 32.9 (CH2),
31.7 (CH2), 31.3 (CH2), 31.0 (CH2), 30.2 (CH), 28.7 (CH2), 28.3
(CH2), 27.3 (CH2), 21.4 (CH3), 17.1 (CH3), 13.3 (CH3), 12.5 (CH3),
11.5 (CH3); ESI-MS (m/z, %) 525 ([M + Na]+, 100). Anal. Calcd for
C30H46O6: C, 71.68; H, 9.22. Found: C, 71.45; H, 9.33. 18b
(amorphous solid): IR (KBr) νmax (cm−1) 3505, 1716, 1256, 1051; 1H
NMR δ 9.77 (d, J = 1.0 Hz, 1H), 4.68 (m, 1H), 4.39 (m, 1H), 3.47
(ddd, J = 11.0, 4.2, 2.0 Hz, 1H), 3.34 (t, J = 11.0 Hz, 1H), 3.32 (d, J =
1.0 Hz, 1H), 2.02 (s, 3H), 1.08 (s, 3H), 0.91 (s, 3H), 0.79 (d, J = 7.0
Hz, 3H), 0.78 (d, J = 6.3 Hz, 3H); 13C NMR δ 205.1 (CH), 170.6 (C),
108.9 (C), 80.2 (CH), 79.9 (C), 73.4 (CH), 66.9 (CH2), 52.7 (CH),
49.8 (CH), 48.1 (CH), 47.6 (C), 44.5 (CH), 41.9 (CH), 36.4 (CH2),
35.3 (C), 34.8 (CH), 33.9 (CH2), 31.6 (CH2), 31.5 (CH2), 30.4
(CH2), 30.2 (CH), 28.8 (CH2), 28.4 (CH2), 28.1 (CH2), 27.3 (CH2),
21.4 (CH3), 17.1 (CH3), 14.3 (CH3), 13.8 (CH3), 12.1 (CH3); ESI-
MS (m/z, %) 525 ([M + Na]+, 100). Anal. Calcd for C30H46O6: C,
71.68; H, 9.22. Found: C, 71.49; H, 9.29.
3β-Acetoxylup-20(30)-en-29-al (20). The reaction with lupeol
acetate (19; 94 mg) was carried out (reaction time = 50 min). Silica
gel column chromatography gave pure compound 7 (72 mg; 75%)
eluted with 5% ethyl acetate/hexane. Compound 7A was proven in all
respects to be identical to the same compound described in the
literature.24
2,2-Dimethylbicyclo[2,2,1]-1-phenylselenomethylheptan-5-
one (22) and 3,3-Dimethylbicyclo[2,2,2]octan-2,6-dione (23).
The reaction with (+)-camphene (21; 75 mg) was carried out
(reaction time = 1.5 h). Silica gel column chromatography gave pure
compound 22 as a yellow oil (68 mg; 40%) and compound 23 as an
amorphous solid (27 mg, 30%) eluted with 5 and 7% ethyl acetate in
Intensity data for the single crystals were obtained by collecting
reflections at 158 K using 0.5° ω scans on a Bruker X8 diffractometer
furnished with an APEX II CCD detector using Cu Kα (λ = 1.54178
Å) radiation. Aliphatic group hydrogen atoms were placed at their
idealized positions and allowed to ride on the coordinates of the
parent atom with isotropic thermal parameters (Uiso) fixed at 1.2 Ueq
of the carbon atom to which they are attached. Hydroxyl group
hydrogens were found from the difference electron density maps and
refined with an anisotropic thermal motion model.
Crystallographic data have been deposited at the Cambridge
Crystallographic Data Centre, 12 Union Road, 129 Cambridge CB2
1EZ, UK, and copies can be obtained on request, free of charge, by
quoting the publication citation and the deposition number CCDC
1047802.
1
hexane, respectively. 22: IR (KBr) νmax (cm−1) 1709, 1579, 1478; H
NMR δ 7.52 (m, 2H), 7.23 (m, 3H), 3.84 (s, 1H), 2.93 (t, J = 5.1 Hz,
1H), 2.16 (d, J = 12.6 Hz, 1H), 2.08 (t, J = 5.1 Hz, 1H), 2.01 (m, 2H),
1.85 (2H), 1.23 (s, 3H), 1.01 (s, 3H); 13C NMR δ 208.7 (C), 133.7
(CH), 131.2 (C), 129.0 (CH × 2), 127.1 (CH), 64.6 (CH), 50.5
(CH), 47.8 (CH), 44.2 (C), 34.7 (CH2), 27.7 (CH3), 27.2 (CH2 × 2),
24.7 (CH2), 24.3 (CH3); ESI-MS (m/z, %) 331 ([M + Na]+, 100),
639 ([2M + Na]+, 100), Anal. Calcd for C16H20OSe: C, 62.54; H, 6.56.
20
Found: C, 62.31; H, 6.39. 23: [α]D +18.2 (c 0.5, CHCl3); IR (KBr)
1
νmax (cm−1) 1728, 1717, 1463; H NMR δ 3.13 (q, J = 6.3, 0.6 Hz,
1H), 2.43 (d, J = 13.2 Hz, 1H), 2.21 (d, J = 5.3 Hz, 1H), 2.07 (m, 1H),
1.88 (3H), 1.63 (m, 1H), 1.23 (s, 3H), 1.16 (s, 3H); 13C NMR δ 205.2
(C), 200.0 (C), 51.4 (C), 51.0 (CH), 47.0 (CH), 33.5 (CH2), 25.5
(CH2), 24.6 (CH2), 24.5 (CH3), 21.8 (CH3); ESI-MS (m/z, %) 221
([M + Na + MeOH]+, 100), 189 ([M + Na]+, 100). Anal. Calcd for
C10H14O2: C, 72.26; H, 8.49. Found: C, 72.01; H, 8.35.
6-Formylcholest-5-en-3β-ol acetate (16). The reaction with 6-
methylene-5α-cholestan-3β-ol acetate (15; 80 mg) was carried out
(reaction time = 45 min). Silica gel column chromatography gave pure
H
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Calculations. The hybrid Becke three-parameter Lee−Yang−Parr
density functional theory (DFT) B3LYP functional27,28 and 6-31G**
basis set29 were applied in this study. This computational level
provides an acceptable compromise between computational labor and
basis set error. The first step of the calculations was conformational
analysis performed by the Conflex program,30−32 in which the
MMFF94s force field was implemented. For the reaction modeling,
only the most abundant conformers of the reactants were taken for
further calculations of the reaction paths. All stationary points (minima
and transition structures) were characterized by calculating the
harmonic vibrational frequencies, using analytical second derivatives.
The transition structures were found by using the quadratic
synchronous transit-guided quasi-Newton (QST3) method developed
by Schlegel and co-workers.33 For the lowest reaction barriers of the
model M7 reaction, the intrinsic reaction coordinate calculations34,35
were performed for both forward and reverse directions of the
vibrational mode calculations to confirm that the transition states that
were found connect the appropriate minima.
To account for the dichloromethane solvent effect, the polarizable
continuum model (PCM) was applied.36,37 In this procedure, the
solvent is mimicked by a dielectric continuum with a dielectric
constant (ε = 8.93 for CH2Cl2), surrounding a cavity with a shape and
dimension adjusted on the real geometric structure of the solute
molecule. The latter polarizes the solvent, which, as a response,
induces an electric field (the reaction field) that interacts with the
solute. In the IEF-PCM, the electrostatic part of such an interaction is
represented in terms of an apparent charge density spread on the
cavity surface.
Computational Modelling of University of Warsaw (ICM) is
acknowledged for the computer time. This research was also
supported by PL-Grid Infrastructure. We also thank Mrs.
Jadwiga Maj for her skillful technical assistance.
REFERENCES
■
(1) Wirth, T., Ed. Organoselenium Chemistry; Wiley-VCH Verlag
GmbH & Co.KGaA: Weinheim, Germany, 2012.
(2) Barton, D. H. R.; Brewster, A. G.; Hui, R. A. H. F.; Lester, D. J.;
Ley, S. V. Chem. Commun. 1978, 952−954.
(3) Barton, D. H. R.; Mangust, P. D.; Rosenfeld, M. N. Chem.
Commun. 1975, 301−302.
(4) Barton, D. H. R.; Brewster, A. G.; Ley, S. V.; Rosenfeld, M. N.
Chem. Commun. 1976, 985−986.
(5) Barton, D. H. R.; Cussano, N. J.; Ley, S. V. Chem. Commun. 1978,
393−394.
(6) Barton, D. H. R.; Lester, D. J.; Ley, S. V. Chem. Commun. 1978,
276−277.
(7) Barton, D. H. R.; Lester, D. J.; Ley, S. V. Chem. Commun. 1977,
445−446.
(8) Back, T. G.; Collins, S.; Kerr, R. G. J. Org. Chem. 1981, 46, 1564−
1570.
(9) (a) Barton, D. H. R.; Lester, D. J.; Ley, S. V. Chem. Commun.
1978, 130−131. (b) Barton, D. H. R.; Lester, D. J.; Ley, S. V. J. Chem.
Soc., Perkin Trans. 1 1980, 2209−2212.
(10) Back, T. G. J. Org. Chem. 1981, 46, 1442−446.
(11) Danieli, B.; Lesma, G.; Palmisano, G.; Riva, E. Chem. Commun.
1984, 909−911.
(12) (a) Shimizu, M.; Kuwajima, I. Tetrahedron Lett. 1979, 20, 2801−
2804. (b) Kuwajima, I.; Shimizu, M.; Urale, H. J. Org. Chem. 1982, 47,
837−842.
(13) (a) Barton, D. H. R.; Morzycki, J. W.; Motherwell, W. B.; Ley, S.
V. Chem. Commun. 1981, 1044−1045. (b) Barton, D. H. R.; Godfrey,
C. R. A.; Morzycki, J. W.; Motherwell, W. B.; Ley, S. V. J. Chem. Soc.,
Perkin Trans. 1 1982, 1947−1952. (c) Barton, D. H. R.; Crich, D.
Tetrahedron 1985, 41, 4359−4364.
(14) ten Brink, G.-J.; Vis, J.-M.; Arends, I. W. C. E.; Sheldon, R. A. J.
Org. Chem. 2001, 66, 2429−2433.
(15) Ayrey, G.; Barnard, D.; Woodbridge, D. T. J. Chem. Soc. 1962,
2089−2099.
The natural bond analysis (NBO)38 was performed to estimate the
charge distribution. In this method, natural atomic orbitals that are the
effective orbitals of an atom in the particular molecular environment
are determined, and these are also the maximum occupancy orbitals.
NBOs are localized few-center molecular orbitals that reflect Lewis-like
bonding structures.
All of the calculations were performed using the Gaussian 09
package of programs,39 and the structures were visualized by using the
GaussView program.40
ASSOCIATED CONTENT
* Supporting Information
■
S
Physical and spectroscopic data for compounds 4, 6, 12, 14, 16,
18a, 18b, 22, 23, 26, and 27, and crystallographic information
file (CIF) for compound 14. Computational details (energetics,
NBO charges, Cartesian coordinates, the imaginary frequencies
assignment) for computed structures. The Supporting In-
formation is available free of charge on the ACS Publications
website at DOI: 10.1021/acs.joc.5b00410.
(16) Kutateladze, A. G.; Kice, J. L.; Kutateladze, T. G.; Zefirov, N. S.;
Zyk, N. V. Tetrahedron Lett. 1992, 33, 1949−1952.
(17) Kutateladze, A. G.; Kice, J. L.; Kutateladze, T. G.; Zefirov, N. S.
J. Org. Chem. 1993, 58, 995−996.
́
(18) Carvalho, J. F. S.; Cruz Silva, M. M.; Sa e Melo, M. L.
Tetrahedron 2010, 66, 2455−2462.
(19) Ruddock, P. L.; Williams, D. J.; Reese, P. B. Steroids 2004, 69,
193−199.
AUTHOR INFORMATION
Corresponding Authors
*Tel.: +48 85 7388043. Fax: +48 85 7388103. E-mail: gierdas@
uwb.edu.pl.
■
(20) Duran, F. J.; Ghini, A. A.; Coirini, H.; Burton, G. Tetrahedron
2006, 62, 4762−4768.
(21) Petrow, V. A. J. Chem. Soc. 1939, 998−1003.
́
(22) Lopez, Y.; Jastrzębska, I.; Santillan, R.; Morzycki, J. W. Steroids
2008, 73, 449−457.
*Tel.: +48 85 7388260. Fax: +48 85 7388103. E-mail:
morzycki@uwb.edu.pl.
(23) (a) Ogawa, S.; Hosoi, K.; Makino, M.; Fujimoto, Y.; Iida, T.
Chem. Pharm. Bull. 2007, 55, 247−250. (b) Heisi, K.; Ogawa, S.;
Makino, M.; Mukaiyama, T.; Hori, K.; Iida, T.; Fujimoto, Y. J. Nat.
Med. 2008, 62, 332−335.
Notes
The authors declare no competing financial interest.
(24) Banthrope, D. V.; Whittaker, D. Q. Rev. Chem. Soc. 1966, 20,
373−387.
ACKNOWLEDGMENTS
■
(25) (a) Spreitzer, H. Monatsh. Chem. 1990, 121, 847−851.
(b) Spreitzer, H.; Buchbauer, G.; Reisinger, S. Helv. Chim. Acta
1989, 72, 806−810.
Financial support from the University of Białystok within the
project BST-124 is gratefully acknowledged. The equipment
(i.e., digital polarimeter) for the Center of Synthesis and
Analysis BioNanoTechno of the University of Białystok was
funded by EU, as a part of the Operational Program
Development of Eastern Poland 2007−2013, project
POPW.01.03.00-20-034/09-00. Computational Grant No.
G19-4 from Interdisciplinary Centre for Mathematical and
(26) Witiak, D. T.; Parker, R. A.; Dempsy, M. E.; Ritter, M. C. J. Med.
Chem. 1971, 14, 684−693.
(27) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(28) Burke, K.; Perdew, J. P.; Wang, Y. Electronic Density Functional
Theory: Recent Progress and New Directions; Dobson, J. F., Vignale, G.,
Das, M. P., Eds.; Plenum Press: New York, 1998.
I
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(29) Rassolov, V. A.; Ratner, M. A.; Pople, J. A.; Redfern, P. C.;
Curtiss, L. A. J. Comput. Chem. 2001, 22, 976−984.
(30) CONFLEX 7; Conflex Corp.: Japan, 2012.
(31) Goto, H.; Osawa, E. J. Am. Chem. Soc. 1989, 111, 8950−8951.
(32) Goto, H.; Osawa, E. J. Chem. Soc., Perkin Trans. 2 1993, 187−
198.
(33) Peng, C.; Schlegel, H. B. Isr. J. Chem. 1993, 33, 449−454.
(34) Gonzales, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154−
2161.
(35) Gonzales, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523−
5527.
(36) Mennucci, B. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2,
386−404.
(37) Scalmani, G.; Frisch, M. J. J. Chem. Phys. 2010, 132, 114110.
(38) Weinhold, F.; Landis, C. R. Discovering Chemistry with Natural
Bond Orbitals; John Wiley & Sons: Hoboken, NJ, 2012.
(39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A. et al. Gaussian 09, revision A.1; Gaussian, Inc.:
Wallingford, CT, 2009.
(40) Dennington, R.; Keith, T.; Millam, J. GaussView, version 5;
Semichem Inc.: Shawnee Mission, KS, 2009.
J
DOI: 10.1021/acs.joc.5b00410
J. Org. Chem. XXXX, XXX, XXX−XXX