Oxyfunctionalization of non-natural targets by dioxiranes. 2. Selective bridgehead dihydroxylation of fenestrindane

Full text of DOI:10.1021/jo961316l

J . Org. Chem. 1996, 61, 8681-8684
8681
atives10a and several steroids.10b,c To carry out these
Oxyfu n ction a liza tion of Non -Na tu r a l
Ta r gets by Dioxir a n es. 2. Selective
1
1
oxidations, methyl(trifluoromethyl)dioxirane (1b) is
often the reagent of choice, since it provides higher
substrate conversions and shorter reaction times with
Br id geh ea d Dih yd r oxyla tion of
F en estr in d a n e1
1
2
respect to dimethyldioxirane (1a ) with no loss in
selectivity.6
a
Caterina Fusco, Michele Fiorentino, Anna Dinoi, and
Ruggero Curci*
Centro CNR “MISO”, Dipartimento Chimica, Universit a` di
Bari, Via Amendola 173, I-70126 Bari, Italy
Ralph A. Krause and Dietmar Kuck*,†
Fakult a¨ t f u¨ r Chemie, Universit a¨ t Bielefeld,
Universit a¨ tsstrasse 25, D-33615 Bielefeld, Germany, and
Universit a¨ t-Gesamthochschule Paderborn, Warburger
Strasse 100, D-33098 Paderborn, Germany
We have shown recently that representative cen-
tropolyindanes undergo selective oxygen atom insertion
into their bridgehead C-H bonds by dioxiranes. In
Received J uly 12, 1996X
1
particular, by varying the excess of the methyl(trifluo-
romethyl)dioxirane (1b ) reagent, tetrabenzo[5.5.5.5]-
fenestrane (fenestrindane) (2)13 could be converted either
into the corresponding monoalcohol or into the corre-
sponding all-bridgehead tetraalcohol, i.e., 4b,8b,12b,16b-
tetrahydroxyfenestrindane (3);1 the oxidation products
were isolated and characterized as the corresponding
mono- and tetraacetate, respectively.1
As an expansion on this breakthrough, it appeared
convenient to explore the access to fenestrindanes being
only partially functionalized at the bridgeheads, since
these are foreseen to be valuable intermediates on the
In tr od u ction
The regioselective functionalization of complex molec-
ular framework is an important goal in organic synthesis;
in particular, the bridgehead functionalization of poly-
cyclic compounds can provide access to derivatives bear-
ing quaternary carbon centers or strained bridgehead
double bonds. In this context, the controlled function-
,14
1
-3
alization of centropolyindanes
at their benzylic and/
or benzhydrylic bridgehead positions is of particular
interest since it allows the synthesis of complex three-
dimensionally fused polyquinane carbon skeletons, in-
cluding highly strained cycloolefins such as those of the
acepentalene family.4
way to the highly unsaturated fenestrindanes containing
,5
15-17
strained isoindene moieties.
In this context, the
6
elucidation of the stereo- and regiocontrol in the sequence
of multiple oxyfunctionalization of fenestrindane ap-
peared of varied interest. We report herein on the
selective 2-fold bridgehead oxyfunctionalization of fenes-
trindane 2 using the powerful dioxirane 1b, along with
kinetic data that provide useful hints concerning the
mechanism of the oxygen insertion reaction.
Dioxiranes (1) have been proven to be useful reagents
for the selective oxidation of a variety of polycyclic
hydrocarbons under mild conditions.
6
-9
In addition to
7
,8a
representative alicyclic systems (e.g., the decalins,
adamantane,8b and 2,4-didehydroadamantane ) they have
9
been applied to the selective oxyfunctionalization of
natural target compounds, such as vitamin D deriv-
3
Resu lts a n d Discu ssion
†
University of Paderborn.
Abstract published in Advance ACS Abstracts, November 1, 1996.
1) Part 1: Kuck, D.; Schuster, A.; Fusco, C.; Fiorentino, M.; Curci,
X
F en estr in d a n e Dih yd r oxyla tion Usin g Meth yl-
tr iflu or om eth yl)d ioxir a n e. Treatment of 2 with a ca.
(
(
R. J . Am. Chem. Soc. 1994, 116, 2375.
2) Schuster, A.; Kuck, D. Angew. Chem., Int. Ed. Engl. 1991, 30,
699.
3) For accounts on centropolyindane chemistry, see: (a) Kuck, D.
(
1
(10) (a) Curci, R.; Detomaso, A.; Prencipe, T.; Carpenter, G. B. J .
(
Am. Chem. Soc. 1994, 116, 8112. (b) Bovicelli, P.; Lupattelli, P.;
Mincione, E.; Prencipe, T.; Curci, R. J . Org. Chem. 1992, 57, 2182. (c)
Bovicelli, P.; Lupattelli, P.; Mincione, E.; Prencipe, T.; Curci, R. J . Org.
Chem. 1992, 57, 5052.
(11) Mello, R.; Fiorentino, M.; Sciacovelli, O.; Curci, R. J . Org. Chem.
1988, 53, 3890.
(12) (a) Murray, R. W.; J eyaraman, R. J . Org. Chem. 1985, 50, 2847.
(b) Cassidei, L.; Fiorentino, M.; Mello, R.; Sciacovelli, O.; Curci, R. J .
Org. Chem. 1987, 52, 699.
(13) (a) Kuck, D. Chem. Ber. 1994, 127, 409. (b) Kuck, D.; B o¨ gge,
H. J . Am. Chem. Soc. 1986, 108, 8107.
Synlett 1996, in press. (b) Kuck, D. In Quasicrystals, Networks, and
Molecules of Fivefold Symmetry; Hargittai, I., Ed.; VCH: New York,
1
990; Chapter 19.
4) (a) Kuck, D.; Schuster, A.; Paisdor, B.; Gestmann, D. J . Chem.
(
Soc., Perkin Trans. 1 1995, 721. (b) Kuck, D.; Schuster, A.; Gestmann,
D.; Posteher, F.; Pritzkow, H. Chem. Eur. J . 1996, 2, 58. (c) Kuck, D.;
Schuster, A.; Gestmann, D. J . Chem. Soc., Chem. Commun. 1994, 609.
(
5) (a) Haag, R.; Ohlhorst, B.; Noltemeyer, M.; Fleischer, R.; Stahlke,
D.; Schuster, A.; Kuck, D.; de Meijere, A. J . Am. Chem. Soc. 1995,
17, 10474. (b) Kuck, D.; Schuster, A.; Ohlhorst, B.; Sinnwell, V.; de
1
Meijere, A. Angew. Chem., Int. Ed. Engl. 1989, 28, 595. (c) Haag, R.;
Ohlhorst, B.; Noltemeyer, M.; Schuster, A.; Kuck, D.; de Meijere, A.
J . Chem. Soc., Chem. Commun. 1993, 1727. (d) Haag, R.; Kuck, D.;
Fu, X.-Y.; Cook, J . M.; de Meijere, A. Synlett 1994, 340.
(14) Kuck, D.; Schuster, A.; Krause, R. A. J . Org. Chem. 1991, 56,
3472.
(15) [5.5.5.5]Fenestranes containing more than four double bonds
in the fenestrane core have been discussed in the literature (ref 16) in
the context of the planar-tetracoordinate-carbon problem (ref 17).
(16) (a) Gupta, A. K.; Fu, X.; Snyder, J . P.; Cook, J . M. Tetrahedron
1991, 47, 3665. (b) Luef, W.; Keese, R. In Advances in Strain in Organic
Chemistry; Halton, B., Ed.; J AI Press: Greenwich, CT, 1993; Vol. 3,
pp 229-267. (c) Chadrasekhar, J .; W u¨ rthwein, E. U.; Schleyer, P. v.
R. Tetrahedron 1981, 37, 921. (d) W u¨ rthwein, E.-U.; Chadrasekhar,
J .; J emmis, E. D.; Schleyer, P. v. R. Tetrahedron Lett. 1981, 22, 843.
(e) B o¨ hm, M. C.; Gleiter, R.; Schang, P. Tetrahedron Lett. 1979, 20,
2575.
(6) For recent reviews on dioxirane reactivity, see: (a) Curci, R.;
Dinoi, A.; Rubino, M. F. Pure Appl. Chem. 1995, 67 , 811. See references
to the pertinent articles quoted therein. (b) Adam, W.; Hadjiarapoglou,
L. P.; Curci, R.; Mello, R. In Organic Peroxides; Ando, W., Ed.; Wiley:
New York, 1992; Chapter 4, pp 195-219. See references therein.
(
7) Murray, R. W.; J eyaraman, R.; Mohan L. J . Am. Chem. Soc.
986, 108, 2470.
8) (a) Mello, R.; Fiorentino, M.; Fusco, C.; Curci, R. J . Am. Chem.
1
(
Soc. 1989, 111, 6749. (b) Mello, R.; Cassidei, L.; Fiorentino, M.; Fusco,
C.; Curci, R. Tetrahedron Lett. 1990, 31, 3067-3070.
(17) (a) Hoffmann, R.; Alder, R.; Wilcox, C. F. J . Am. Chem. Soc.
1970, 92, 4992. (b) Hoffmann, R. Pure Appl. Chem. 1971, 28, 181.
(9) Teager, D. S.; Murray, R. K., J r. J . Org. Chem. 1993, 58, 5548.
S0022-3263(96)01316-3 CCC: $12.00 © 1996 American Chemical Society

8
682 J . Org. Chem., Vol. 61, No. 24, 1996
Notes
Sch em e 1. Con for m a tion a l Equ ilibr iu m betw een
th e Tw o Rota m er s of F en estr a n ed iol 5a
group; this feature is presented twice (once each face) by
the 4b,8b-diol 5a . Thus, it is unlikely that the isolated
diol has instead the structure of the 4b,12b-diol 5b, since
the said stereospecific water elimination processes would
(
S
4
Sym m etr y Nota tion Refer s to th e Ca r bon
Sk eleton On ly)
2
0
not be expected for its gaseous molecular radical cation.
The NMR spectra of the diol 5a corroborate the
structure assignment; in fact, they also allow one to rule
out the alternative structure 5b, provided one makes the
assumption that the two possible conformers of the
[5.5.5.5]fenestrane skeleton undergo fast equilibration on
the NMR time scale (Scheme 1). This assumption should
hold safely for both diols 5a and 5b; in fact, even tetraol
3
swhich bears two sets of 1,3-syn diol groupingsshas
1
4,21
been shown to be conformationally relatively flexible.
Given this conformational flexibility, the apparent
molecular symmetry of 4b,8b-diol 5a and of 4b,12b-diol
1
5
b would be C
2
and C2v, respectively. Then, both the H
0
.82 M solution of methyl(trifluoromethyl)dioxirane (1b)
1
3
and the C NMR spectra of the dialcohol product indicate
(
1.5 equiv) at the conditions given in eq 1 resulted in a
that it has the 4b,8b-diol structure 5a , while they rule
ca. 60% substrate conversion within 20 min, yielding the
monoalcohol 4 as the only product (isolated yield 77%,
based on converted 2).
1
3
1
out the corresponding regioisomer 5b. Indeed, the
C
NMR spectrum displays just three (of the four possible)
resonance lines for the carbon atoms of the intraindan
bonds and all of the eight lines expected for the remaining
benzo nuclei resonances in 5a . By way of contrast, 4b,-
1
2b-diol 5b would exhibit only a set of two and four
resonances for the quaternary and tertiary carbon atoms
of the benzo nuclei, respectively. Careful inspection of
1
1
1
the H NMR and the COSY- H- H spectra of the diol
isolated reveals the presence of two distinct AA′BB′ spin
systems along with two identical ABCD spin systems;
2
this also points to apparent C molecular symmetry, as
indeed presented by 5a . Thus, the further dioxirane
hydroxylation of monoalcohol 4 occurs with surprising
high regio and facial selectivity at the back side of the
[
5
5.5.5.5]fenestrane core to yield exclusively the 4b,8b-diol
a .
Kin etics a n d Mech a n ism . Rate data were gathered
in order to gain insight into the nature of the O-insertion
reactions at hand. In the kinetic experiments performed
with comparable initial concentrations of substrate and
oxidant under “normal” conditions (i.e., air blanket,
protection from light), the rate of dioxirane consumption
with time was monitored (iodometry); the formation of
products was verified by GC and/or GC/MS. In all of the
cases examined, the kinetics were found to obey a clean
overall second-order rate-law (order one each in dioxirane
and fenestrindane 2 or 4); integrated second-order rate
law plots were found to be linear to over 80% reaction,
Then, isolated 4 was allowed to react with an ad-
ditional 1.7 equiv of 1b under similar conditions (eq 2).
Within 15 min, a 60% conversion of the starting material
was attained (GC/MS monitoring); from the reaction
mixture a dialcohol could be isolated in 88% yield (based
on converted 3). This novel derivative was identified as
4
b,8b-dihydroxyfenestrindane 5a , i.e., the regioisomer
bearing both hydroxyl groups at one and the same
indane unit. Its regioisomer 5b bearing the hydroxy
groups at the 4b,12b-bridgehead positions was not de-
tected in the reaction mixture.
yielding reproducible k
2
values. In Table 1 rate data are
collected for the transformation of fenestrindane 2 into
its monoalcohol 4 and in turn of this into diol 5a , along
The identity of diol 5a could be unambigously estab-
1
13
lished on the basis of its H and C NMR spectra, as
well as by EI and FAB mass spectrometry. In fact, while
the FAB mass spectrum (NBA/NaI) does show the (quasi)
(20) Direct support for this assumption is given by the recent finding
that 8b,16b-centropentaindane diol (6), a derivative of 5b bearing two
strictly 1,3-syn oriented hydroxy groups, exhibits only moderate water
+
molecular ion ([M + Na] , m/z 423), the EI mass
•
+
•+
•+
2
loss upon EI-MS, i.e., ([M] /[M - H O] ) ca. 0.5 (ref 4b).
spectrum exhibits no significant molecular ion peak (M ,
•
+
m/z 400), but it displays an intense [M - H
2
O] signal
(
m/z 382) as the base peak. As recorded for a number of
similar cases in centropolyindane chemistry and indeed
1
8
recorded for the precursor alcohol 4 itself, an extremely
•+
19
fast water loss from M ions points to a stereochemical
disposition having a weakly bound hydrogen atom in a
favorable 1,3-syn orientation with respect to a hydroxyl
(
21) It is interesting that the tetrabromo analog of tetrahydroxy-
fenestrindane (3), i.e., 4b,8b,12b,16b-tetrabromofenestrindane, instead
behaves as in an apparently static equilibrium of two conformers with
(
18) Kuck, D.; Neumann, E.; Schuster, A. Chem. Ber. 1994, 127, 151.
4
S molecular symmetry on the NMR time scale. Clearly, the presence
(19) For a general discussion of stereospecific water elimination
of bulky bromine atoms at the bridgehead positions effectively restrains
conformational equilibration of the [5.5.5.5]fenestrane skeleton. (See,
e.g.: Kuck, D.; Schuster, A. Angew. Chem., Int. Ed. Engl. 1988, 27,
1192.)
processes in gaseous radical cations of alcohols, see: Mandelbaum, A.
In Stereochemistry; Kagan, H. B., Ed.; Thieme: Stuttgart, Germany,
977; Vol. 1, pp 153 ff.
1

Notes
J . Org. Chem., Vol. 61, No. 24, 1996 8683
Ta ble 1. Kin etic Da ta a n d Rela tive Ra tes for
O-In ser tion by Dioxir a n e 1b in to Ben zh yd r ylic C-H
Bon d s of Selected Su bstr a tes in Aceton e a t 0.0 ( 0.1 °C
Ch a r t 1
10 k2 (M-1 s-1
)
2
a
rel rate per Hb
0.57c
entry
substrate
1
2
3
4
fenestrindane (2)
4b-fenestrindanol (4)
PhCH(CH3)2 (7)
Ph2CH(CH3) (8)
9.37 ( 0.03
13.9 ( 0.04
4.10 ( 0.02
2.50 ( 0.05
d
1.7
1.0
0.61
a
Kinetic constants calculated from integrated second order rate-
law plots that were linear to over 80% reaction; average values
shown are from two or more independent runs agreeing within
b
(
5%. From kinetic constants divided by the appropriate statis-
c
tical factor. Statistical factor of 4 (four equivalent benzhydrylic
d
C-H). A statistical factor of two applies here since only insertion
into the two benzhydrylic C-H anti to the OH group of 4 can lead
to the anti diol 5a (the sole reaction product).
with the oxidation rates measured for cumene (7) and
1
,1-diphenylethane (8).
Similar to fenestrindane (2), the oxyfunctionalization
of these reference compounds 7 and 8 occurs selectively
7
,8b,22
at the benzhydrylic C-H bonds.
examined could the presence of esters ROAc23 or ROC(:
O)CF , nor (in the case of substrates 7) of acetophenone
PhC(:O)CH
In none of the cases
3
1
3
, be revealed (GC and H NMR) in the
reaction mixture. Should the reaction involve free radi-
cals, formation of these side products is to be expected.23
Furthermore, the oxidation of 1,1-diphenylethane (8) is
slower than that of cumene (7) (Table 1, entries 3 and
migration terminus while the O-O bond is being bro-
28
ken.
According to this preferential direction of ap-
proach, as shown by ts structures 9 and 10 (Chart 1),
fenestrindane (2)sdespite its complex architecturesshould
present no additional steric hindrance to stereoalignment
with respect to 1,1-diphenylethane (8). Indeed, after the
statistical factor is taken into account, the hydroxylation
rate for the two substrates is practically identical (Table
2
3
4
); this is also at odds with the expectation of free-
radical involvement in view of the higher stability of Ph
2
-
•
• 24
(CH
3 3 2
)C with respect of Ph(CH ) C .
These facts and the simple kinetics observed rule out
2
3
the recent claim by Minisci et al. that under the
conditions adopted the O-insertion into C-H bonds
proceeds via a free-radical chain mechanism. Instead,
1
, entries 1 and 4).
Of particular interest is the case of fenestrindanol (4)
8
a,25
findings hereinsalongside with previous
and recent
hydroxylation, leading selectively to the corresponding
b,8b-diol 5a only. In fact, stereoselectivities in dioxirane
evidence,26 including the application a radical-clock probe
by the Ingold group sprovide further support to our
earlier proposal of a rather concerted O-insertion into
4
2
7
epoxidations of allylic and homoallylic alcohols have been
rationalized in terms of opposing steric effects and
8
C-H bonds in the rate-determining step, possibly leading
to radical pairs capable of fast in-cage collapse (“oxygen
cooperative H-bonding exercised by the OH functionality
6a
in the ts, sometimes with conflicting outcome.
For
rebound”).6 In line with this, we have reported that
the transformation of optically active (S)-(-)-2-phenylbu-
tane Ph(Et)(Me)C*H into the corresponding alcohol Ph-
a
25
instance, Murray, Singh, et al. have reported a rather
clear-cut case of OH participation in the dioxirane
2
epoxidation of homoallylic -CH OH-substituted cyclo-
(Et)(Me)C*OH by dioxirane 1b occurs in high yield with
29
hexene, leading to distinct syn-facial selectivity. More
recently, Adam and Smerz have discussed the regio- and
diastereoselective epoxidations of acyclic allylic alcohols
by dimethyldioxirane (1a ) in terms of intramolecular
complete retention of configuration and no loss of optical
purity.
Once a free-radical chain mechanism is discounted, a
rationalization of the kinetic and stereochemical data can
be attempted on the grounds of a previously described
transition-state (ts) model for the rate-determining O-
insertion.1 On the basis of careful ab initio calculations
and FMO analysis,28 this envisages electrophilic attack
by the dioxirane to be directed along the O-O bond axis
at the carbon atom of the C-H bond; the latter should
be favorably oriented to shift the H atom to the ap-
proaching oxygen, which employs an electron pair as
30
H-bonding with the incoming dioxirane. On the other
hand, dioxirane epoxidations of allylic alcohols possessing
sufficient conformational rigidity and of 4â-hydroxycho-
31
32
lesterol (as well as other steroids ) were found to
proceed largely with anti stereochemistry. Since diox-
iranes are highly polar species (cf., µ ) 2.5 D for the
6
2 2
parent dioxirane H CO ), it is also recognized that
specific solvation and electrostatic effects (such as dipole-
dipole interactions) in the ts could become a major factor
6
a
in determining stereoselectivity of dioxirane epoxidations.
(
(
22) Kuck, D.; Schuster, A. Z. Naturforsch. 1991, 46B, 1223.
23) (a) Bravo, A.; Fontana, F.; Fronza, G.; Mele, A.; Minisci, F. J .
In addition to the OH functionality, fenestrindanol 4
contains only benzhydrylic C-H bonds at the bridgehead
positions, each pointing to one or the other of the two
concave molecular surfaces in this centropolyindane. If
Chem. Soc., Chem. Commun. 1995, 1573. (b) Minisci, F.; Zhao, L.;
Fontana, F.; Bravo, A. Tetrahedron Lett. 1995, 36, 1697.
(
24) See, e.g.: Robaugh, D. A.; Stein, E. S. J . Am. Chem. Soc. 1986,
08, 3224 and references therein.
25) Adam, W.; Asensio, G.; Curci, R.; Gonzal e` s-Nu n˜ ez, M. E.; Mello,
R. J . Org. Chem. 1992, 57, 953.
26) Curci, R.; Dinoi, A.; Fusco, C.; Lillo, M. A. Tetrahedron Lett.
996, 37, 249.
27) Vanni, R.; Garden, S. J .; Banks, J . T.; Ingold, K. U. Tetrahedron
Lett. 1995, 36, 7999.
28) Bach, R. D.; Andr e´ s, J . L.; Owensby, A. L.; McDouall, J . J . W.
J . Am. Chem. Soc. 1993, 115, 5768.
1
(
(29) Murray, R. W.; Singh, M.; Williams, B. L.; Moncrieff, H. M. J .
Org. Chem. 1996, 61, 1830.
(30) Adam, W.; Smerz, A. K. J . Org. Chem. 1996, 61, 3506.
(31) Cicala, G.; Curci, R.; Fiorentino, M.; Laricchiuta, O. J . Org.
Chem. 1982, 47, 2670.
(32) Marples, B. A.; Muxworthy, J . P.; Baggaley, K. H. Tetrahedron
Lett. 1991, 32, 533.
(
1
(
(

8
684 J . Org. Chem., Vol. 61, No. 24, 1996
Notes
the hydroxyl group were to excercise cooperative H-
bonding to the approaching dioxirane, face-selective
oxyfunctionalization at the C-H in 1,3-syn orientation
with respect the OH group is expected. By way of
contrast, the opposite stereochemical course is verified
since the anti diol 5a is produced exclusively. Inspection
of the likely ts arrangements (Chart 1) provides a clue.
In fact, for syn hydroxylation to occur with the given
stereoalignment, the hydroxy group is in no position to
assist the approaching dioxirane by cooperative H-
bonding (cf.,11b). Then, the hydroxylation preferentially
occurs with anti stereochemistry (11a ), perhaps with
such an orientation of the existing OH moiety that
O-insertion from the opposite face becomes favored by
electrostatic interactions. This would explain the notice-
able increase in rate on going from 2 to 4 (Table 1).
4b-Hyd r oxy-4br,8bâ,12br,16bâ-tetr a h yd r od iben zo[a ,f]-
d iben zo[2,3:4,5]p en t a len o[1,6-cd ]p en t a len e (4): colorless
1
solid; mp 258-260 °C; H NMR (CDCl
3
, 400 MHz): δ 7.72 (m,
2
4
H, 4/5-H), 7.53 (m, 4 H), 7.49 (m, 2 H), 7.34 (m, 4 H), 7.25 (m,
H), 5.23 (s, 1 H, 12b-H), 4.71 (s, 2 H, 8b/16b-H), 2.20 (s, 1 H,
13
OH); C NMR (100.5 MHz, CDCl
3
): δ 145.24 (s), 144.72 (s),
1
44.00 (s), 143.84 (s), 129.28 (d), 128.17 (d), 127.73 (d), 127.45
(d), 124.67 (d), 124.56 (d), 124.48 (d), 124.28 (d), 92.31 (s, C-4b),
75.34 (C-16d), 60.93 (d, C-8b/16b), 54.65 (d, C-12b); MS (EI, 70
•
+
2
eV) m/ z (r.i.) 366 (100, [M - H O] ), 364 (12), 363 (21), 350 (9),
2
+
2
(
89 (7), 183 (10, [M - H
7). Anal. Calcd. for C29
H, 5.31.
2
O] ), 181 (11), 180 (7), 175 (15), 168
H20O: C, 90.6; H, 5.24. Found: C, 89.5;
Next, a cool aliquot of 0.50 M dioxirane 1b in TFP (0.8 mL,
.400 mmol) was added to a stirred solution of the above-
described monoalcohol 4 (90 mg, 0.230 mmol) in CH Cl (7 mL)
2 2
at 0 °C; GC/MS monitoring of the reaction mixture indicated
60% starting material conversion after 15 min and formation of
diol 5a as essentially the sole product. Removal of solvent and
0
column chromatography (silica gel, CHCl
the monoalcohol 4 (38 mg, 0.100 mmol) and isolation of diol 5a
3
) allowed recovery of
Con clu sion s
(
46 mg, 87% yield based on converted 4).
b,8b-Dih yd r oxy-4br,8bâ,12br,16bâ-tetr a h yd r od iben zo-
a ,f]d iben zo[2,3:4,5]p en ta len o[1,6-cd ]p en ta len e (5a ): color-
Our results support the notion thatsprovided one
avoids conditions that trigger free-radical decomposition
of the peroxide2 sdioxirane oxyfunctionalizations can
4
[
3,26
1
1
1
less solid; mp 275-278 °C; H NMR (CDCl , 400 MHz), H- H
COSY (300 MHz): δ 7.72 (m, 2 H, 5/8-H), 7.68 (m, 2 H, 4/9-H),
7.54 (m, 2 H, 1/12-H), 7.50 (m, 2 H, 13/16-H), 7.41 (m, 2 H, 6/7-
3
6
a
lead to remarkable regio- and anti-facial selectivity.
Data presented herein corroborate the view that, along
with steric effects, electrostatic effects (such as dipole-
dipole interactions) in the ts might become a relevant
factor in discriminating between anti and syn attack of
functionalized substates by the dioxirane, favoring an
anti stereochemistry in the case of 4b-hydroxyfenestrin-
dane at hand.
Apart from the mechanistic clues, the feat of partial
oxyfunctionalization of fenestrindane 2 at two bridgehead
positions belonging to the same indan unit by using
methyl(trifluoromethyl)dioxirane paves the road to ex-
H), 7.33 (m
c
, 4 H, 3/10/2/11-H), 7.24 (m, 2 H, 14/15-H), 5.21 (s,
): δ 145.3 (s),
44.91 (s), 144.51 (s, 2 C), 129.91 (d), 129.55 (d), 128.05 (d),
27.63 (d), 124.76 (d), 124.43 (d), 124.38 (d), 124.15 (d), 91.31
2 H, 12b/16b-H); 13C NMR (100.5 MHz, CDCl
3
1
1
(
s, C-4b/8b), 78.56 (C-16d), 53.75 (d, C-12b/16b); MS (EI, 70 eV)
•
+
m/ z (r.i.) 382 (100, [M - H
3
181 (10); HRMS (FAB , 3-nitrobenzyl alcohol/NaI) 423.1365 of
[M + Na] , calcd for C29
2
O] ), 366 (10), 365 (10), 354 (19),
2+
2
52 (16), 351 (10), 350 (11), 276 (9), 190.5 (15, [M - H O] ),
+
+
20 2
H O + Na 423.1361.
Kin etic Mea su r em en ts. Runs were performed by following
the decay of dioxirane concentration (by iodometry) with time,
according to described analytical techniques.8
,11
Absolute rates
plore the conversion into other 4b,8b-disubstituted fenes-
trindanes, hence into the corresponding isoindenes,33 and
were determined under second-order conditions, with the diox-
irane and hydrocarbon initial concentrations kept in the range
novel [5.5.5.5]fenestranes bearing two unsaturated bridge-
-2
(
4-6) × 10 M, and differing by 8-20%. At zero time an aliquot
head positions.2
,5,15
The attainment of these targets
(0.5-1.0 mL) of a thermostated dioxirane 1b solution was added
remains a challenging goal in the chemistry of cen-
tropolyindanes.
to 10-20 mL of a solution (also thermostated) of the given
hydrocarbon substrate; aliquots (10-20 µL) of the reaction
solution were sampled periodically and quenched with excess
KI/EtOH, and the liberated I
Linear ln[(a - x)/ (b - x)] vs time plots were obtained to over
2
was determined by iodometry.
Exp er im en ta l Section
-
1
-1
8
0% reaction; from these k
(Table 1). In each case, at least two independent runs were
performed and the k values averaged (estimated error e(8%).
2
(M
s ) values could be estimated
Methylene chloride, acetone, and 1,1,1-trifluoro-2-propanone
TFP) (bp 22 °C) were purified by standard methods, stored over
1
(
5
2
Å molecular sieves at 2-5 °C, and routinely redistilled prior
to use. Curox triple salt 2KHSO ‚KHSO ‚K SO (a gift by
5
4
2
4
Peroxid-Chemie GmbH, Munich, Germany) was our source of
potassium peroxymonosulfate to be employed in the synthesis
of dioxiranes. Solutions of 0.8-1.0 M methyl(trifluoromethyl)-
Ack n ow led gm en t. We thank the Ministry of Uni-
versity, Scientific and Technological Research of Italy
MURST 40) and the CNR-Progetto Strategico “Tec-
(
8
,11
dioxirane (1b)
in TFP were obtained by adopting procedures,
nologie Chimiche Innovative” (Rome, Italy) for partial
support. Work in Bielefeld was supported by the
Deutsche Forschungsgemeinschaft (DFG, Ku 663) and
the Fonds der Chemischen Industrie (FCI). Thanks are
also due to Dr. R. Mello (University of Valencia, Spain)
for performing some of the HRMS analyses. One of us
equipment, and precautions already described in detail. High
purity commercial (Aldrich) cumene and 1,1-diphenylethane
were further purified by distillation. The synthesis and spectral
characteristics of 4bR,8bâ,12bR,16bâ-tetrahydrodibenzo[a,f]-
dibenzo[2,3;4,5]pentaleno[1,6-cd]pentalene (fenestrindane) (2)
have been given. Melting points were not corrected. Equip-
ment and analytical methods have been previously reported.1
Step w ise Dih yd r oxyla tion of F en estr in d a n e (2) Usin g
Meth yl(tr iflu or om eth yl)d ioxir a n e (1b). To a stirred solution
1
3
(R.C.) is grateful for hospitality during sabbatical leave
spent (November 1995-October 1996) at the Depart-
ment of Chemistry of Brown University, a most gracious
host institution.
of 2 (150 mg, 0.407 mmol) in CH
2 2
Cl (15 mL) kept at 0 °C was
gradually added a standardized solution of 0.82 M dioxirane 1b
(0.6 mL, 0.490 mmol) in TFP. As monitored by GC or GC/MS
[
°
column SE 30, 30 m × 0.25 mm i.d., temperature program 240
1
Su p p or tin g In for m a tion Ava ila ble: H NMR (CDCl
3
,
C (5 min), 240-280 °C (10 °C/min)], during 20 min over 50%
1
1
13
4
00 MHz), 1H- H COSY (300 MHz), { H} C NMR (100.5
MHz, CDCl ), and HRMS of diol 5a (4 pages). This material
is contained in libraries on microfiche, immediately follows
this article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
substrate conversion was achieved. Removal of the solvent in
vacuo followed by column chromatography (silica gel, CHCl
afforded recovery of the starting material (65 mg, 0.176 mmol)
and the monoalcohol 4 (68 mg, 0.177 mmol, 77% yield based on
converted 2).
3
3
)
(33) Kuck, D.; Krause, R. A.; Barth, D. To be published.
J O961316L