Regio- and diastereoselective catalytic epoxidation of chiral allylic alcohols with hexafluoroacetone perhydrate. Hydroxy-group directivity through hydrogen bonding

Article abstract of DOI:10.1021/jo982025a

The threo diastereoselectivity in the catalytic epoxidation of chiral allylic alcohols with 1,3-allylic strain by hexafluoroacetone perhydrate and its regioselectivity in the epoxidation of 1-methylgeraniol establishes a hydroxy-directing effect through hydrogen bonding between the oxidant and substrate. The higher syn selectivity for the cis than trans isomer of 5- tert-butylcyclohexen-3-ol suggests a hydrogen-bonded transition-state structure similar to that of peracids for this catalytic oxygen-transfer process.

Full text of DOI:10.1021/jo982025a

1
274
J . Org. Chem. 1999, 64, 1274-1277
Regio- a n d Dia ster eoselective Ca ta lytic Ep oxid a tion of Ch ir a l
Allylic Alcoh ols w ith Hexa flu or oa ceton e P er h yd r a te.
Hyd r oxy-Gr ou p Dir ectivity th r ou gh Hyd r ogen Bon d in g
,
†
Waldemar Adam,* Hans-Georg Degen, and Chantu R. Saha-M o¨ ller
Institute of Organic Chemistry, University of W u¨ rzburg, Am Hubland, D-97074 W u¨ rzburg, Germany
Received October 7, 1998
The threo diastereoselectivity in the catalytic epoxidation of chiral allylic alcohols with 1,3-allylic
strain by hexafluoroacetone perhydrate and its regioselectivity in the epoxidation of 1-methylgeraniol
establishes a hydroxy-directing effect through hydrogen bonding between the oxidant and substrate.
The higher syn selectivity for the cis than trans isomer of 5-tert-butylcyclohexen-3-ol suggests a
hydrogen-bonded transition-state structure similar to that of peracids for this catalytic oxygen-
transfer process.
In tr od u ction
Epoxidation reactions are of great synthetic interest
lytic epoxidant. For this purpose, the chiral allylic
alcohols 1 (cf. Table 1 for structures) were to be oxidized.
These serve as mechanistic tools to define the transition-
state structure of the oxygen-transfer process by com-
parison of the observed regio- and diastereoselectivities
in organic chemistry, as attested by the large volume of
1
work on this subject. Of special interest are catalytic
procedures, as they allow economic use of the employed
resources.2 A good number of selective methods are
presently available which utilize transition-metal cata-
4
with those of the established oxidants m-CPBA and
7
DMD. The structural similarities (Figure 1) between the
perhydrate and both the peracid and dioxirane are clearly
evident. While the peroxidic functionality in the perhy-
drate is internally hydrogen-bonded as in the peracid,
3
4
lysts, most prominently TBHP/Ti(O-i-Pr)
4
, VO(Acac)
2
,
5
6
methyltrioxorhenium (MTO), and Mn(Salen) complexes.
As purely organic nonmetal oxidants, the most widely
3
its central carbon atom is sp -hybridized as in the dioxi-
4
used are the peracid m-CPBA and the isolated dioxirane
rane. Thus, the perhydrate is a composite of the peracid
and the dioxirane structural features and it should be of
mechanistic interest to assess what geometrical factors
control the hydroxy-group directivity in the perhydrate
epoxidation of the chiral allylic alcohols 1.
DMD,7 which function stoichiometrically. A potential
catalytic case constitutes hexafluoroacetone perhydrate,
8
which has been applied to epoxidations, the oxidation
of heteroatoms,9 arenes, and aldehydes, and the
10
11
9
Baeyer-Villiger rearrangement. This nonmetal oxida-
tion catalyst is generated in situ from hexafluoroacetone
hydrate and hydrogen peroxide as the oxygen donor,
which are both commercially available.
The incentive of this study was to assess the efficiency
and selectivity of hexafluoroacetone perhydrate as cata-
Resu lts a n d Discu ssion
The chiral allylic alcohols 1a -l were prepared accord-
12
ing to literature procedures or were purchased. The
epoxidations were conducted with a catalytic amount (0.1
equiv) of hexafluoroacetone sesquihydrate in the presence
of 2 equiv of 85% hydrogen peroxide and disodium
hydrogenphosphate as buffer. A general procedure is
given in the Experimental Section.
†
Fax: +49(0)931/8884756. E-mail: adam@chemie.uni-wuerzburg.de.
(1) (a) Schwesinger, R.; Willaredt, J .; Bauer, T. In Methods of
Organic Chemistry (Houben Weyl), 4th ed.; Helmchen, G., Hoffmann,
R. W., Mulzer, J ., Schaumann, E., Eds.; G. Thieme: Stuttgart, New
York, 1995; Vol. E 21, Chapter 4.5.1. (b) Oehlschlaeger, A. C. In
Methods of Organic Chemistry (Houben Weyl), 4th ed.; Helmchen, G.,
Hoffmann, R. W., Mulzer, J ., Schaumann, E., Eds.; G. Thieme:
Stuttgart, New York, 1995; Vol. E 21, Chapter 4.5.2.
The diastereoselectivities for the chiral, acyclic allylic
alcohols 1a -h are listed in Table 1, together with the
4
7
literature data for the m-CPBA and DMD epoxidations.
The epoxidations of substrates 1a ,b (entries 1 and 2) with
no allylic strain display a modest (ca. 62:38) threo
selectivity. For the substrates 1c,d (entries 3 and 4) with
1,2-allylic strain, a slight (38:62) preference for the
erythro diastereomer was observed. In contrast, the 1,3-
allylic strain present in the derivatives 1e,f (entries 5
and 6) induces a high (>90:10) threo preference. When
(
2) Sheldon, R. A.; Kochi, J . K. Metal-Catalyzed Oxidations of
Organic Compounds; Academic Press: New York, 1981.
3) Adam, W.; Corma, A.; Reddy, T. I.; Renz, M. J . Org. Chem. 1997,
2, 3631-3637.
4) (a) Sharpless, K. B.; Verhoeven, T. R. Aldrichimica Acta 1979,
2, 63-74. (b) Rossiter, B. E.; Verhoeven, T. R.; Sharpless, K. B.
(
6
1
(
Tetrahedron Lett. 1979, 19, 4733-4736.
5) Adam, W.; Mitchell, C. Angew. Chem., Int. Ed. Engl. 1996, 35,
33-535.
6) (a) J ohnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric
(
5
(
Synthesis; Ojima, I., Ed.; VCH: New York, 1993; Chapter 4.1. (b)
J acobsen, E. N. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.;
VCH: New York, 1993; Chapter 4.2. (c) Katsuki, T. Coord. Chem. Rev.
1
,2- and 1,3-allylic strain are both acting in the same
1
995, 140, 189-214.
(12) (a) Morgan, B.; Oehlschl a¨ ger, A. C.; Stokes, T. M. J . Org. Chem.
(
7) Adam, W.; Smerz, A. K. J . Org. Chem. 1996, 61, 3506-3510.
8) (a) Heggs, R. P.; Ganem, B. J . Am. Chem. Soc. 1979, 101, 2484-
1992, 57, 3231-3236. (b) Renz, M. Ph.D. Thesis, University of
W u¨ rzburg, 1996. (c) Adam, W.; Mitchell, C. M.; Paredes, R.; Smerz, A.
K.; Veloza, L. A. Liebigs Ann./ Recl. 1997, 1365-1369. (d) Fatiadi, A.
J . Synthesis 1976, 65-92. (e) House, H. O.; Wilkins, J . M. J . Org. Chem.
1978, 43, 2443-2454. (f) Chamberlain, P.; Roberts, M. L.; Witham, G.
H. J . Chem. Soc. B 1970, 1374-1381. (g) Ho, N. H.; le Noble, W. L. J .
Org. Chem. 1989, 54, 2018-2021. (h) Schalley, C. A.; Schr o¨ der, D.;
Schwarz, H. J . Am. Chem. Soc. 1994, 116, 11089-11097.
(
2
8
486. (b) Biloski, A. J .; Heggs, R. P.; Ganem, B. Synthesis 1980, 810-
11.
(9) Adam, W.; Ganeshpure P. A. Synthesis 1996, 179-188.
(10) Adam, W.; Ganeshpure P. A. Synthesis 1993, 280-282.
(11) Ganem, B.; Heggs, R. P.; Biloski, A. J .; Schwartz, D. R.
Tetrahedron Lett. 1980, 21, 685-688.
1
0.1021/jo982025a CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/04/1999

Catalytic Epoxidation of Chiral Allylic Alcohols
J . Org. Chem., Vol. 64, No. 4, 1999 1275
function of the dihedral angle between the allylic hydroxy
group and the plane of the π system.14 Table 2 shows the
diastereoselectivities for the epoxidations of these sub-
strates by hexafluoracetone perhydrate and for compari-
son the literature data of DMD14 and m-CPBA.
12f,15
The
2 2
HFAH/H O oxidant is more syn-selective for the sub-
strates 1i,j (94:6) with the larger dihedral angle (140°
versus 110°) than for 1k (70:30).
F igu r e 1. Comparison of the peracid, perhydrate, and diox-
irane structures.
Ta ble 1. Dia ster eoselectivities for th e Ep oxid a tion of
th e Ch ir a l Allylic Alcoh ols 1a -h by HF AH/H2O2, m -CP BA,
a n d DMD
For the diolefinic allylic alcohol 1l, its regioselectivity
allows one to test the ease of epoxidation of a plain alkene
12c
and an allylic alcohol within the same molecule.
Moreover, the chiral allylic alcohol functionality provides
additional information on the diastereoselectivity of the
epoxidation. The regio- and diastereoselectivities for the
2 2
HFAH/H O system are given in Table 3, again together
with the relevant literature data for DMD and m-CPBA
for comparison.1 These results display a good match
between the perhydrate and the peracid regioselectivities.
The low regioselectivities for both express that an ap-
preciable amount of the 3,4 epoxide is formed, despite
the fact that the allylic double bond is electronically
deactivated by the inductive effect of the hydroxy group,
compared to the plain one. The high (95:5) threo diaste-
reoselectivity is again characteristic for chiral allylic
alcohols with 1,3-allylic strain.
2c
The relative rates as a function of the substitution
pattern of unfunctionalized cis-, trans-, and gem-disub-
2 2
stituted alkenes are given for HFAH/H O in Table 4 and
are compared with those for m-CPBA16 and DMD. The
good correspondence between the perhydrate and the
peracid is again clearly evident.
17
The composite set of selectivity and reactivity data in
Tables 1-4 unequivocally proclaims that the observed
hydroxy-group directivity derives from hydrogen bonding
between the perhydrate oxidant and the allylic alcohol
substrate 1, akin to that established for peracids. The
salient supporting experimental facts are the following:
(a) the pronounced threo selectivity for the 1,3-allylically
strained acyclic substrates (Table 1, entries 5-7), (b) the
high syn selectivity for the cyclic allylic alcohols with
dihedral angle R > 110° (Table 2), (c) the appreciable
formation of the hydroxy-epoxide regioisomer (Table 3),
and (d) the nearly identical epoxidation rates for cis-,
trans-, and gem-substituted alkenes (Table 4). In analogy
to the peracid transition-state structure I, we propose
structure II for the perhydrate (Figure 2). This structure
differs from the DMD transition state III in that the R
angle is smaller and the planar rather than the spiro
geometry applies. As for peracids, also for the perhydrate
the â oxygen atom is transferred, while for the dioxirane
it is necessarily the R oxygen atom. Consequently, the
steric demand is more pronounced in the latter case. In
a
Determined by 1H NMR analysis of characteristic signals
directly on the crude product mixture (error ( 5% of the stated
values); pentachlorobenzene was used as internal standard. Ref-
erence 4. Reference 7.
b
c
molecule, as in the substrates 1g,h (entries 7 and 8), the
epoxidation is also highly (>90:10) threo-selective. For
1
3
example, for the stereochemical probe 1g (entry 7), the
observed threo selectivity of 95:5 is within the error limits
the same as that found for the substrates 1e,f (entries 5
and 6), for which solely 1,3-allylic strain is acting. The
correspondence in the sense (threo versus erythro) and
the extent of diastereomeric control between the perhy-
drate and peracid in Table 1 is impressive.
The hydroxy-capped derivatives of mesitylol, namely,
the methyl ether 1f-Me and the acetate 1f-Ac, showed
within the experimental error the same threo selectivity
3
view of the proximate sp -carbon center the spiro geom-
17
etry (structure III) is preferred, as supported by the
1
8
reactivity data in Table 4 and theoretical work. For
peracids, the experimental data does not suffice to
(
92:8) as that of the parent alcohol 1f. This is contrary
to m-CPBA and DMD, since both result in proportionally
more erythro epoxide; in fact, for most cases (cf. entries
(14) Adam, W.; Smerz, A. K. Tetrahedron 1995, 51, 13039-13044.
(15) Chautemps, P.; Pierre, J .-L. Tetrahedron 1976, 32, 549-557.
(16) Rebek, J ., J r.; Marshall, L.; Wolak, R.; McManis, J . J . Am.
6
a,b) this stereoisomer is predominantly formed.
The cyclic allylic alcohols 1i-k were epoxidized in the
Chem. Soc. 1984, 106, 1170-1171.
(
17) (a) Baumstark, A. L.; McCloskey C. J . Tetrahedron Lett. 1987,
same manner to assess the syn:anti selectivity as a
2
1
8, 3311-3314. (b) Vasquez, P. C.; Baumstark, A. L. J . Org. Chem.
988, 53, 3437-3439.
(
13) Adam, W.; Nestler, B. J . Am. Chem. Soc. 1993, 115, 5041-
(18) Bach, R. D.; Owensby, A. L.; Andres, J . L.; Schlegel, H. B. J .
5
049.
Am. Chem. Soc. 1991, 113, 7031-7032.

1
276 J . Org. Chem., Vol. 64, No. 4, 1999
Adam et al.
Ta ble 2. Dia ster eoselectivities for th e Ep oxid a tion of th e Cycloh exa n ols 1i-k by m -CP BA, DMD, a n d HF AH/H2O2
substrate
R2
syn:anti diastereoselectivity
entry
R1
R3
convn (%)
m.b. (%)
HFAH/H2O2a
m-CPBAb
DMDc
dihedral angled (deg)
1
2
3
1i
1j
1k
Me
Bu
H
Me
H
Bu
Me
H
H
67
49
47
>95
86
91
92:8
94:6
70:30
96:4
96:4
84:16
94:6
82:18
58:42
137
140
110
t
t
a
Determined by 1H NMR analysis of characteristic signals directly on the crude product mixture (error ( 5% of the stated values);
pentachlorobenzene was used as internal standard. CH2Cl2 (ref 15). c 1:9 acetone/CCl4 (ref 7). d Calculated for the preferred ground-
b
state conformations (ref 7).
Ta ble 3. Regio- a n d Dia ster eoselectivities for th e
Ep oxid a tion of 1-Meth ylger a n iol (1l) by HF AH/H2O2,
m -CP BA, a n d DMD
selectivitya
regio
diastereob
entry
oxidant
3,4:7,8
threo:erythro
c
1
2
3
HFAH/H2O2
52:48
46:54
68:32
95:5
89:11
94:6
F igu r e 2. Transition-state structures for the epoxidation by
m-CPBA (I), hexafluoroacetone perhydrate (II and V), DMD
d
m-CPBA
d
DMD
3 3
(III), and CF CO H (IV).
a
Determined by 1H NMR analysis of characteristic signals
directly on the crude product mixture (error ( 5% of the stated
arrangement remains to be seen, but at this time, we
b
values); pentachlorobenzene was used as internal standard. The
c
adhere to the planar transition-state structure I (Figure
diastereoselectivity of the 7,8-epoxide is 50:50. 26% conversion,
4
,16,17
d
2) for the epoxidation of allylic alcohols by peracids.
mass balance >95%. CCl4 (ref 12c).
Therefore, in view of the excellent match in the selectivi-
ties between m-CPBA and the perhydrate, the planar
transition-state geometry II is proposed as well for the
perhydrate.
Ta ble 4. Effect of th e Su bstitu tion P a tter n on th e
Ep oxid a tion Ra te by Hexa flu or oa ceton e P er h yd r a te,
m -CP BA, a n d DMD
The major discrepancy between the m-CPBA and
perhydrate diastereoselectivities is exposed by the hy-
droxy-protected ether 1f-Me and the acetate 1f-Ac
derivatives of the allylic alcohols 1f (Table 1, entries
relative reaction rate
6
a,b). This very high threo selectivity requires some
substitution pattern
HFAH/H2O2a
m-CPBAb
DMDc
efficient association between these capped substrates and
the perhydrate. A relevant case has been disclosed by
Ganem,20 in which silyl ethers of 3-cyclohexenol were
epoxidized expectedly anti-selectively by m-CPBA but
trans
cis
gem
1
1.5
1.5
1
1.2
1.4
1
7
3.5
a
Alkenes used were cis,trans-2-heptenes and 2-methylhexene;
3 3
syn-selectively by CF CO H. For the latter, hydrogen
b
cis,trans-2-octenes, cyclohexene, and methylenecyclohexane (ref
c
bonding from the peracid to the substrate oxygen func-
tionality was proposed, as portrayed in the transition-
state structure IV. Such multiple hydrogen bonding to
1
6); cis,trans-2-hexenes and 2-methyl-1-pentene (ref 17).
differentiate between the planar and the spiro geom-
etries, but recent computational results favor the spiro
arrangement in the epoxidation of unfunctionalized alk-
enes;19 however, for allylic alcohols with allylic strain this
point remains open because no theoretical work is
available. Whether the optimal dihedral angle (R) of ca.
(19) (a) Bach, R. D.; Owensby, A. L.; Gonzalez, C.; Schlegel, H. B.
J . Am. Chem. Soc. 1991, 113, 2338-2339. (b) Bach, R. D.; Winter, J .
E.; McDouall, J . J . W. J . Am. Chem. Soc. 1995, 117, 8586-8593. (c)
Singleton, D. A.; Merrigan, S. R.; Liu, J .; Houk, K. N. J . Am. Chem.
Soc. 1997, 119, 3385-3386. (d) Bach, R. D.; Estevez, C. M.; Winter, J .
E.; Glukhotsev, M. N. J . Am. Chem. Soc. 1998, 120, 680-685.
1
20° for hydrogen bonding between the peracid and the
hydroxy group corresponds better with the spiro or planar
(20) McKittrick, B. A.; Ganem, B. Tetrahedron Lett. 1985, 26, 4895-
4898.

Catalytic Epoxidation of Chiral Allylic Alcohols
J . Org. Chem., Vol. 64, No. 4, 1999 1277
three nucleophilic oxygen centers has been shown kineti-
cally to be unlikely for the deprotonation of phenylazore-
sorcinol;21 however, such a structure does account for the
observed diastereoselectivity in the epoxidation by per-
which hydrogen bonding is an essential feature to achieve
stereochemical control through the hydroxy-group direc-
tivity. The additional advantage of this catalytic non-
metal process is that besides allylic alcohols with 1,3-
allylic strain, their ether and ester derivatives are also
epoxidized in high threo selectivity.
2
0
oxytrifluoroacetic acid. Moreover, recent computational
results1 have suggested a similar transition-state struc-
ture for the epoxidation of allylic alcohols with peroxo-
9d
formic acid. The stronger acidity of CF
3
CO
3
H versus
Exp er im en ta l Section
m-CPBA is presumably responsible for the effective
Typ ica l P r oced u r e for th e Ca ta lytic Ep oxid a tion of
th e Allylic Alcoh ols by Hexa flu or oa ceton e P er h yd r a te.
Into a 25-mL, two-necked, round-bottomed flask, equipped
with a reflux condenser (thermostated at ca. -75 °C) and
topped with a nitrogen-filled balloon, were placed 95.6 mg (944
mmol) mesitylol (1f), 0.2 equiv of pentachlorobenzene (as
hydrogen bonding to the oxygen atom of the silyl ether
to achieve this syn stereocontrol.20 Ganem
8,20
has also
suggested that a similar effect accounts for the syn
selectivity (80:20) observed in the epoxidation of 3-cyclo-
hexenyl acetate by hexafluoroacetone perhydrate. Ac-
cordingly, structure V is assigned for the perhydrate
epoxidation of the hydroxy-capped derivatives 1f-Me,-
Ac. The acidity of the perhydrate is apparently sufficient
to hydrogen bond strongly enough with the ether and
acetate oxygen atoms to account for the high threo
stereocontrol. In view of the recent theoretical work,1
the multiply coordinated structure V is also more likely
for the allylic alcohol 1f.
internal standard), 142 mg (1.00 mmol) of NaH
2.00 mmol) of 85% H , 13.0 µL (0.11 mmol) of hexafluoro-
acetone sesquihydrate, and 5 mL of CDCl as solvent. The
2 4
PO , 60.0 µL
(
2 2
O
3
contents were heated at reflux for 5.5 h, and after cooling to
1
room temperature (ca. 20 °C), a sample was taken for H NMR
analysis. The diastereomeric epoxides 2f were obtained in a
ratio of 92:8 (threo:erythro) at 89% conversion and a mass
balance of >95%.
9d
Ack n ow led gm en t. The generous financial support
by the Deutsche Forschungsgemeinschaft (Schwerpunkt-
programm “Peroxidchemie”: Mechanistische und pr a¨ -
parative Aspekte des Sauerstofftransfers) and the Fonds
der Chemischen Industrie is gratefully appreciated.
In conclusion, our diverse selectivity and reactivity
data for the perhydrate (HFAH/H
2 2
O ) epoxidation dis-
close a transistion state analogous to that of peracids, in
(
21) Bethell, D. In Advances in Physical Organic Chemistry; Bethell,
D., Ed.; Academic Press: London, San Diego, 1990; Vol. 26, pp 255-
67.
3
J O982025A