Stereochemistry of Reactions Involving Rotationally Restricted, Sterically Hindered Cations, Radicals, and Anions: 9-Fluorenyl Systems

Article abstract of DOI:10.1021/jo0351015

A study of the stereochemical pathways of reactions involving rotationally restricted, sterically hindered cations, radicals, and anions has been undertaken utilizing chiral 9-(o-tert-butylphenyl)-fluorenes. Previous reports of studies with these or related achiral compounds contained erroneous or equivocal conclusions. This study shows that (+)-sp-9-(o-tert-butylphenyl)-9-methoxy-2-methylfluorene, treated with Tf ₂O-CHCl₃ to form 100% of the 9-cation, then with NaOMe-MeOH, provided 29% of re-formed substrate (configurational retention) and 71% of the (-)-sp enantiomer (inversion). The same substrate treated with HI-CHCl₃ was converted into the 9-radical, which was rapidly reduced, affording 100% isolation of (-)-sp-9-(o-tert-butylphenyl)-2-methylfluorene (inversion). Treatment of the latter with n-BuLi-THF provided the 9-anion which, on acidification, yielded 100% of the enantiomeric (+)-sp-9-(o-tert-butylphenyl)-2-methylfluorene (inversion). The substrates in these reactions were the thermodynamically favored sp rotamers. Inversion directly produced the higher energy nonenantiomeric ap rotamers, which rapidly rotated into the sp products that were enantiomeric with the substrates. These results are explained by the rotational restriction and partial steric hindrance by the tert-butyl group to the original face of the sp³ antiaromatic 9-cation (4n π electrons), and the rotational restriction and extensive blockage to the original face of the sp² nonaromatic 9-radical (4n + 1 π electrons) and aromatic (4n + 2 π electrons) 9-anion. The barrier to rotation in some of the ortho-substituted 9-arylfluorenes is great enough to allow their sp and ap rotamers to be detected coexisting in solution, although their crystals were composed exclusively of one. Rotational restriction and steric hindrance at the 9-position have a large influence on the pK_a values of these fluorenes and can offset the classic electronic effects of the substituents.

Full text of DOI:10.1021/jo0351015

Ster eoch em istr y of Rea ction s In volvin g Rota tion a lly Restr icted ,
Ster ica lly Hin d er ed Ca tion s, Ra d ica ls, a n d An ion s: 9-F lu or en yl
System s†
Yuqing Hou‡ and Cal Y. Meyers*
Meyers Institute for Interdisciplinary Research in Organic and Medicinal Chemistry, and Department of
Chemistry and Biochemistry, Southern Illinois University, Carbondale, Illinois 62901-4409
cal@chem.siu.edu
Received J uly 28, 2003
A study of the stereochemical pathways of reactions involving rotationally restricted, sterically
hindered cations, radicals, and anions has been undertaken utilizing chiral 9-(o-tert-butylphenyl)-
fluorenes. Previous reports of studies with these or related achiral compounds contained erroneous
or equivocal conclusions. This study shows that (+)-sp-9-(o-tert-butylphenyl)-9-methoxy-2-meth-
ylfluorene, treated with Tf2O-CHCl3 to form 100% of the 9-cation, then with NaOMe-MeOH,
provided 29% of re-formed substrate (configurational retention) and 71% of the (-)-sp enantiomer
(inversion). The same substrate treated with HI-CHCl3 was converted into the 9-radical, which
was rapidly reduced, affording 100% isolation of (-)-sp-9-(o-tert-butylphenyl)-2-methylfluorene
(inversion). Treatment of the latter with n-BuLi-THF provided the 9-anion which, on acidification,
yielded 100% of the enantiomeric (+)-sp-9-(o-tert-butylphenyl)-2-methylfluorene (inversion). The
substrates in these reactions were the thermodynamically favored sp rotamers. Inversion directly
produced the higher energy nonenantiomeric ap rotamers, which rapidly rotated into the sp products
that were enantiomeric with the substrates. These results are explained by the rotational restriction
and partial steric hindrance by the tert-butyl group to the original face of the sp3 antiaromatic
9-cation (4n π electrons), and the rotational restriction and extensive blockage to the original face
of the sp2 nonaromatic 9-radical (4n + 1 π electrons) and aromatic (4n + 2 π electrons) 9-anion.
The barrier to rotation in some of the ortho-substituted 9-arylfluorenes is great enough to allow
their sp and ap rotamers to be detected coexisting in solution, although their crystals were composed
exclusively of one. Rotational restriction and steric hindrance at the 9-position have a large influence
on the pKa values of these fluorenes and can offset the classic electronic effects of the substituents.
In tr od u ction
consequences of reactions involving the C-9 cation, radi-
cal, and anion of such fluorenes was not addressed. Cram
et al.4 found that chiral 9-substituted fluorenes that were
neither sterically hindered nor rotationally restricted
underwent varying degrees of retention and racemization
during base-catalyzed 9-D/H exchange, which were as-
sociated with the type of base, solvent, and concentration
used. Ford et al.5 provided strong evidence of hindered
rotation around the aryl-C-9 bond of a variety of 9-(2,6-
disubstituted-phenyl)fluorenes and 9-(2-substituted-1-
naphthyl)fluorenes, but they did not explore the com-
bined effects of steric hindrance and restricted rotation
on the stereochemistry of reactions at C-9. Oki and co-
In a series of earlier reports we described our studies
of the properties, crystal structures, and reactions of sp
and ap1a,b 9-acylfluorenes2 and 9-(alkylphenyl)fluorenes,3
some of whose C-9 substituents were rotationally re-
stricting and sterically hindering to attack at that
position. However, the question of the stereochemical
* Author to whom correspondence should be addressed. Fax: 618-
453-6408.
† This paper is dedicated to the late Professor Frederick G. Bordwell,
whose contributions in the field of physical organic chemistry inspired
so many others.
‡ Ph.D. Dissertation, Southern Illinois University, Carbondale, IL,
1997.
(1) (a) The designations sp (syn periplanar) and ap (anti periplanar)
for the conformations discussed here are in accord with Rule E-6.6,
IUPAC Tentative Rules, Section E, Fundamental Stereochemistry (J .
Org. Chem. 1970, 35, 2861). (b) The ap and sp stereochemistry of a
number of crystalline 9-acylfluorenes and 9-(o-alkylphenyl)fluorenes
has been unequivocally determined by X-ray diffraction, and in solution
they are effectively characterized by 1H NMR spectroscopy (refs 2 and
3). In the latter, the most striking differences emanate from the tert-
butyl protons in 9-(o-tert-butylphenyl)fluorenes. In the ap rotamers
these protons are shielded by the fluorene-ring π electrons and resonate
upfield in the 0.7-ppm region, while in the sp rotomers they are
correspondingly deshielded and resonate downfield in the 1.8-ppm
region. Differences are also observed in the resonances of the o-H of
ap vs sp rotamers, and of H-9 of these rotamers of 9-(o-tert-butylphe-
nyl)fluorene itself.
(2) (a) Meyers, C. Y.; Chan-Yu-King, R.; Wahner, A. P.; Manohar,
S. K.; Carr, S. E.; Robinson, P. D. Acta Crystallogr. 1991, C47, 1236-
1239. (b) Meyers, C. Y.; Tunnell, J . L.; Robinson, P. D.; Hua, D. H.;
Saha, S. Acta Crystallogr. 1992, C48, 1815-1818. (c) Robinson, P. D.;
Lutfi, H. G.; Lim, L. W.; Meyers, C. Y. Acta Crystallogr. 1994, C50,
1728-1732. (d) Meyers, C. Y.; Lutfi, H. G. A.; Robinson, P. D. Acta
Crystallogr. 2000, C56, IUC0000220, e418-e419. (e) Robinson, P. D.;
Lutfi, H. G.; Hou, Y.; Meyers, C. Y. Acta Crystallogr. 2000, C56, 1380-
1382. (f) Robinson, P. D.; Lutfi, H. G.; Hou, Y.; Meyers, C. Y. Acta
Crystallogr. 2001, C57, 428-430. (g) Meyers, C. Y.; Lutfi, H. G.; Hou,
Y.; Robinson, P. D. Acta Crystallogr. 2001, C57, 580-581. (h) Robinson,
P. D.; Lutfi, H. G.; Hou, Y.; Meyers, C. Y. Acta Crystallogr. 2001, C57,
582-584.
10.1021/jo0351015 CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/15/2004
1186
J . Org. Chem. 2004, 69, 1186-1195

9-Fluorenyl Systems
SCHEME 1. Con ver sion of 1a to 1 (Ster eoch em ica l
P a th w a y Not Noted )6
SCHEME 2. Dep r oton a tion -Rep r oton a tion of
sp-1: An ion Rota tion -P r oton a tion -Rota tion
P r op osed 6
workers6 subsequently studied reactions at C-9 of achiral
fluorenes whose C-9 substituent concurrently sterically
hindered C-9 and was rotationally restricted. In the
absence of chirality, conversion of sp-9-fluorenol 1a to sp-
fluorene 1 by treatment with HI-HOAc provided no
information regarding the stereochemical pathways
retention or inversion-rotationsas illustrated in Scheme
1. In a related study, they deprotonated sp-1 with
n-BuLi-THF; the lithiated anion solution was then
cooled to -50 °C and acidified. Dynamic 1H NMR at this
temperature showed that the ap rotamer, 1d , had been
formed exclusively, which, when the solution was slowly
warmed above ca. -10 °C, underwent rotation to sp-1.
The Oki group suggested that the initially formed anion
sp-1b was tetrahedral and retained the sp conformation,
but it rotated to the ap-1c anion, which offered less
hindrance to solvation, and it was direcly protonated into
fluorene ap-1, which rotated to the thermodynamically
favored sp-1 (Scheme 2). Again, in the absence of studies
with chiral substrates, the Oki group’s explanation of this
stereochemical pathway as well as that with rotationally
restricted 9-R-naphthylfluorenes (e.g. 2) appeared rea-
sonable.
(3) (a) Meyers, C. Y.; Hou, Y.; Scott, D.; Robinson, P. D. Acta
Crystallogr. 1997, C53, 1149-1151.(b) Robinson, P. D.; Hou, Y.; Lutfi,
H. G.; Meyers, C. Y. Acta Crystallogr. 1998, C54, 73-77. (c) Hou, Y.;
Meyers, C. Y.; Robinson, P. D. Acta Crystallogr. 1998, C54, 1013-
1016. (d) Robinson, P. D.; Hou, Y.; Meyers, C. Y. Acta Crystallogr. 1998,
C54, 1173-1175. (e) Robinson, P. D.; Hou, Y.; Meyers, C. Y. Acta
Crystallogr. 1998, C54, CIF Access paper, DVN IUC9800051. (f)
Meyers, C. Y.; Hou, Y.; Robinson, P. D. Acta Crystallogr. 1999, C55,
626-628. (g) Hou, Y.; Meyers, C. Y. Can. J . Chem, 1999, 77, 960-
966. (h) Meyers, C. Y.; Hou, Y.; Lutfi, H. G.; Saft, H. L. J . Org. Chem.
1999, 64, 9444-9449. (i) Hou, Y.; Robinson, P. D.; Meyers, C. Y. Acta
Crystallogr. 1999, C55, CIF Access, Data Validation No. IUC9900144,
November issue. (j) Robinson, P. D.; Hou, Y.; Meyers, C. Y. Acta
Crystallogr. 1999, C55, CIF Access, Data Validation No. IUC9900147,
November issue. (k) McLean, A. W.; Meyers, C. Y.; Robinson, P. D.
Acta Crystallogr. 2003, E59, o891-o893. (l) Meyers, C. Y.; McLean,
A. W.; Robinson, P. D. Acta Crystallogr. 2003, E59, o978-o980. (m)
Robinson, P. D.; McLean, A. W.; Meyers, C. Y. Acta Crystallogr. 2003,
C59, o539-o540. (n) McLean, A. W.; Meyers, C. Y.; Robinson, P. D.
Acta Crystallogr. 2003, E59, o1228-o1230. (o) Robinson, P. D.; McLean,
A. W.; Meyers, C. Y, Acta Crystallogr. 2003, E59, o1915-o1917. (p)
Meyers, C. Y.; Robinson, P. D.; McLean, A. W. Acta Crystallogr. 2003,
C59, o712-o714. (q) McLean, A. W.; Meyers, C. Y.; Robinson, P. D.
Acta Crystallogr. 2003, E59, accepted for publication. (r) Meyers, C.
Y.; Robinson, P. D.; McLean, A. W. Acta Crystallogr. Submitted for
publication.
(4) (a) Cram, D. J . Fundamentals of Carbanion Chemistry. In
Organic Chemistry-A Series of Monographs; Blomquist, A. L., Ed;
Academic Press: New York, 1965; Vol. 4, Chapter 3 and references
therein. (b) Ford, W. T.; Cram, D. J . J . Am. Chem. Soc. 1968, 90, 2612-
2622 and references therein.
(5) Ford, W. T.; Thompson, T. B.; Snoble, K. A. J .; Timko, J . M. J .
Am. Chem. Soc. 1975, 97, 95-101.
(6) (a) Oki, M. The Chemistry of Rotational Isomers. In Reactivity
and Structure Concepts in Organic Chemistry; Hafner, K., Lehn, J .-
M., Rees, C. W., Schleyer, P. v. R., Trost, B. M., Zahradn´ık, R., Eds;
Springer-Verlag: New York, 1993; Vol. 30, Chapter 4. (b) Moriyama,
K.; Nakamura, M.; Nakamura, N.; Oki, M. Bull. Chem. Soc. J pn. 1989,
62, 485-489. (c) Nakamura, M.; Suzuki, Y.; Oki, M. Bull. Chem. Soc.
J pn. 1985, 58, 2370-2375. (d) Nakamura, M.; Nakamura, N.; Oki, M.
Bull. Chem. Soc. J pn. 1977, 50, 2986-2990.
In related studies, their results with achiral fluorenes
ap-3 and sp-4 suggested that the stereochemistry result-
ing from deprotonation-reprotonation could be controlled
by the intramolecular association of the metalated car-
banion with a polar substituent. However, in the absence
of chirality, whether the intramolecular association
directed this conversion via a rotation or inversion
mechanism could not be ascertained (Scheme 3). A recent
report by Baker et al.7 provided a partial answer by
studies with related rotationally restricted chiral sub-
strates. They found that chiral ap-5, via reprotonation
of its anion, provided chiral sp-6, the inverted product,
thereby ruling out a rotation pathway. The inversion
mechanism in this case was explained by the intramo-
lecular assistance derived from intramolecular ion pair-
ing (5a ) (Scheme 4). However, this explanation and what
stereochemical pathway this reaction would have taken
in the absence of such ion pairing required further study.
In reviewing the earlier studies, it seemed reasonable
to us that the stereochemistry of reactions at C-9 of
fluorenes even without intramolecular assistance, but
(7) Baker, R. W.; Hambley, T. W.; Turner, P. J . Chem. Soc., Chem.
Commun. 1995, 2509-2510.
J . Org. Chem, Vol. 69, No. 4, 2004 1187

Hou and Meyers
SCHEME 3. P r obe of In tr a m olecu la r ly Assisted
Dep r oton a tion -Rep r oton a tion of Ach ir a l a p-3 to
sp-4:6 Rota tion or In ver sion ?
simple chiral 9-arylfluorene without these constraints
would be converted into its achiral 9-anion, -cation, and
-radical and subsequent attack on either face would be
equally favored, leading to racemization. This reasoning
was partly supported by a related example in which
Cram et al. reported 100% racemization of a chiral
9-deuteriofluorene via 9-D/H exchange by treatment with
t-BuOK-t-BuOH.4 In contrast, a chiral 9-arylfluorene
rotationally restricted and sterically hindered at C-9
would be converted into its 9-anion, -cation, and -radical,
respectively, which would also be chiral, rotationally
restricted, and sterically hindered. In the case of sp
rotamers, subsequent attack at the unhindered face is
favored, leading directly to the ap rotamer via inversion
of configuration. The difference between these two types
of fluorenes is apparent from a comparison of their chiral
stereochemistry, illustrated here.
SCHEME 4. Ster eoch em ica l P r obe of
In tr a m olecu la r ly Assisted Rep r oton a tion of a
Ch ir a l 9-F lu or en yl An ion : In ver sion of a p-5 to sp-6
Resu lts a n d Discu ssion
We have now determined the mechanistic stereochem-
istry of reactions of sterically hindered, rotationally
restricted fluorenyl 9-cations, -radicals, and -anions
utilizing chiral fluorenes unencumbered by the possibility
of intramolecular assistance. These results could not have
been predicted on the basis of the previously published
reports and, in some cases, contradict the proposed
stereochemical pathways. Most of our studies described
here with chiral fluorenes either directly utilized (R)-(+)-
sp-9-(o-tert-butylphenyl)-9-methoxy-2-methylfluorene (7)
or chiral compounds prepared from it. The first step in
our planned synthesis of 7 required the preparation of
racemic sp-9-(o-tert-butylphenyl)-9-hydroxyfluorenecar-
boxylic acid (7a ), which was accomplished via the reaction
of o-tert-butylphenylmagnesium bromide with the potas-
sium salt of commercially available fluorenone-2-car-
boxylic acid. Although 7a was then successfully converted
into its two diastereomeric L-cinchonidine salts, they were
1
essentially indistinguishable by H NMR. However, via
with rotationally restricted, sterically hindering C-9
substituents, would differ from that of counterpart fluo-
renes without such substituents. It was reasoned that a
the racemic methyl esters (7b), the epimeric R and S sp-
9-(o-tert-butylphenyl)-9-hydroxyfluorene-2-(S)-N-R-meth-
ylbenzylcarboamides were prepared, from which the R
1188 J . Org. Chem., Vol. 69, No. 4, 2004

9-Fluorenyl Systems
SCHEME 5. P r ep a r a tion of 7
epimer (7c) was successfully isolated and purified, and
was unequivocally chracterized by X-ray diffraction.3c On
treatment of 7c with KOH-MeI, methylation of 9-OH
as well as NH occurred, providing 7d . N-Methylation was
required; otherwise formation of the amide anion in the
subsequent reaction with LiBHEt3 would have prevented
the desired reductive cleavage of the N-methyl-N-R-
methylbenzylamido group. Thus, reductive cleavage of
7d into chiral (R)-sp-9-(o-tert-butylphenyl)-2-hydroxy-
methyl-9-methoxylfluorene (7e) was successful. Treat-
ment with TsCl-LiBr-KOH converted 7e into the cor-
responding 2-bromomethyl compound (7f) which, in turn,
was reductively debrominated with LiBHEt3 to the de-
sired chiral 7. This series of transformations is illustrated
in Scheme 5. The 2-methyl substituent was chosen be-
cause it imparted chirality without imposing intramo-
lecular association or additional steric or rotational hind-
rance in the reactions of the 9-cation, -radical, or -anion.
Thus, the stereochemistry of the reactions unequivocally
determined with chiral 7 or its derivatives would serve
as an excellent model to ascertain the stereochemistry
of these reactions followed by achiral 1 and related
achiral fluorenes via their 9-cation, -radical, and -anion.
Rea ction s via th e 9-F lu or en yl Ca tion . Via a cat-
ionic mechanism, treatment of (R)-(+)-sp-7 at -60 to -40
°C with Tf2O-CHCl3 followed by MeONa-MeOH, then
warming the mixture above -20 °C, produced 71% of the
inverted enantiomer, (S)-(-)-sp-10, along with 29% of re-
formed (R)-(+)-sp-7. Oki et al.6 reported that ap-9-(tert-
butylphenyl)fluorene (1d ) rapidly rotates to the sp rot-
amer (1) above -10 °C, and our studies showed that 1
as well as the corresponding sp-9-fluorenol (1a ) exist
exclusively in their sp rotameric conformations in solution
(NMR) at room temperature as well as their crystalline
states.3b,g,h,8 The fact that the reaction via the 9-fluorenyl
cation 8 occurred with 29% retention-71% inversion of
configuration while the reactions via the corresponding
9-fluorenyl carbanion and radical (described below) pro-
ceeded with exclusive inversion reasonably may be ex-
plained on the basis of the 9-fluorenyl cation’s antiaro-
maticity (4n π electrons).
The antiaromaticity of the 9-fluorenyl cation, which
has been experimentally supported,9a is derived from its
cyclopentadienyl cation moiety, whose antiaromatic char-
acter has been extensively investigated.9b The antiaro-
matic nature of boracyclopentadienes, counterparts of
cyclopentadienyl cations, has likewise been investigated,9c
and that of 9-borafluorenes, electronically related to the
9-fluorenyl cation, was recently reported.9d The antiaro-
matic nature of cation 8 promotes a distorted trigonal
(8) It should be noted that, based on NMR and X-ray analysis of
the large number of C-9-substituted fluorenes, the thermodynamically
favored conformation can be either the sp or ap rotamer, depending
on the C-9 substituents (see refs 2 and 3).
(9) (a) J iao, H.; Schleyer, P. v. R.; Mo, Y.; McAllister, M. A.; Tidwell,
T. T. J . Am. Chem. Soc. 1997, 119, 7075-7083. Amyes, T. L.; Richard,
J . P.; Novak, M. J . Am. Chem. Soc. 1992, 114, 8032-8041. Olah, G.
A.; Prakash, G. K. S.; Liang, G.; Westerman, P. W.; Kunde, K.;
Chandrasekar, J .; Schleyer, P. v. R. J . Am. Chem. Soc. 1980, 102,
4485-4492. (b) Lambert, J . B.; Lin, L.; Rassolov, V. Angew. Chem.,
Int. Ed. 2002, 41, 1429-1431. Otto, M.; Scheschkewitz, D.; Kato, T.;
Midland, M.; Lambert, J . B.; Bertrand, G. Angew. Chem., Int. Ed. 2002,
41, 2275-2276. Mu¨ller, T. Angew. Chem., Int. Ed. 2002, 41, 2276-
2278. Allen, A. D.; Sumonja, M.; Tidwell, T. T. J . Am. Chem. Soc. 1997,
119, 2371-2375. (c) Eisch, J . J .; Galle, J . E.; Kozima, S. J . Am. Chem.
Soc. 1986, 108, 379-385. (d) Wehmschulte, R. J .; Diaz, A. A.; Khan,
M. A. Organometallics 2003, 22, 83-92.
J . Org. Chem, Vol. 69, No. 4, 2004 1189

Hou and Meyers
a
SCHEME 6. Rea ction of 7 via Ch ir a l 9-F lu or en yl
SCHEME 7. 100% Con ver sion of
Ca tion 8 (71% In ver sion a n d 29% Reten tion )
sp-9-MeO-F lu or en e 11 to sp-9-EtO-F lu or en e 13
via 9-F lu or en yl Ca tion s 1b a n d 1b
forms (R)-(+)-sp-7 on reaction with MeONa, and the less
hindered, major ion pair provides the inverted product,
ap-9 (Scheme 6, b). In either case, the initially formed
inversion product, 9, rotates to the thermodynamically
favored, isolated conformer, (S)-(-)-sp-10, the enantiomer
of 7. The percentages of products 7 and 10 were calcu-
lated from the specific rotation of the isolated mixture,
[R]20 -19.2 (c 0.106, acetone); ee of 10 ) 42.2%. That
D
isolated 7 did not contain any unchanged starting mate-
rial 7 was determined from the quantitative exchange of
9-methoxy by ethoxy on converting the related sp-9-o-
tert-butylphenyl-9-methoxyfluorene (11) to sp-9-(o-tert-
butylphenyl)-9-ethoxyfluorene (13) by similar treatment
with triflic anhydride in CHCl3 followed by NaOEt-
EtOH (Scheme 7). Thus, 7, with only one asymmetric
center, offers an example wherein inversion at that center
does not directly provide the enantiomeric structure. A
similar result was reported by Baker et al.7 (shown in
Scheme 4). Both of these observations are rare examples
of this phenomenon.
Rea ction s via th e 9-F lu or en yl Ra d ica l. As shown
in Scheme 1, Oki et al.6 reported that achiral sp-9-
(o-tert-butylphenyl)-9-fluorenol (1a ), treated with HI-
HOAc, was converted into sp-9-(o-tert-butylphenyl)fluo-
rene (1), but neither the mechanistic nor stereochemical
aspects of the reaction were noted.
We recently reported on our effort to examine these
aspects via a dynamic NMR study also involving achiral
1a .3h In that radical-reduction study, carried out with HI
in CDCl3 at -50 °C in an NMR tube, sp-1a was initially
converted into the 9-cation, which was reduced via single-
electron transfer from I- into the 9-radical, which, in
turn, underwent hydrogen atom transfer. Complete
replacement of the 9-OH by H was signaled by the
exclusive formation of ap-9-(o-tert-butylphenyl)fluorene
(1d ). As the temperature was warmed from -50 to -20
°C, rotation of the o-tert-butylphenyl group of ap-1d
slowly and irreversibly converted it into its thermody-
namically favored sp-1 rotamer, the isolated product. We
have already shown by 1H NMR (CDCl3) that the sp
rotamers 1a and 1 show no tendency to undergo rotation
a
See reaction with EtONa-EtOH under identical conditions,
Scheme 7.
character to C-9+ (see structure 8),9 affording a geometry
in which attack leading to inversion is favored over the
o-tert-butyl-hindered frontal attack resulting in retention
(Scheme 6, a). A related explanation for the relative
amounts of inversion vs retention products involves the
participation of two cationic species ion-paired with
triflate anion (8): the more hindered, minor ion pair re-
1190 J . Org. Chem., Vol. 69, No. 4, 2004

9-Fluorenyl Systems
SCHEME 8. Illu str a tion of th e Con ver sion of
Ach ir a l sp 1a to sp 1 via a p 1d w ith HI-CHCl3 by a
F r ee-Ra d ica l In ver sion P a th w a y
SCHEME 9. 100% In ver sion via Ch ir a l Ra d ica l 14:
Con ver sion of 7 in to 16 w ith HI-CHCl3
i.e., 100% kinetically controlled inversion, discussed
above. The complete blockage to attack on the original
face strongly suggests an sp2 geometry on C-9 of the
radical 14, although it is nonaromatic, having 4n + 1 π
electrons. Warming the solution >-20 °C effected rota-
tion of ap-15 to the thermodynamically favored (-)-sp
to their ap rotamers in solution, their spectra respectively
exhibiting a sharp downfield singlet of nine hydrogens
for their deshielded sp t-Bu group but no sign of any
corresponding upfield resonance for the shielded t-Bu
hydrogens of their ap rotamers; moreover, they crystallize
exclusively as their sp rotamers.3b Under identical NMR
conditions, equilibration of the sp and ap rotamers of
the less rotationally restricted 9-(o-isopropylphenyl)-
fluorene does occur, both rotamers being readily ob-
served.3a Thus, the initial formation of ap-1d in this
reaction proceeding through the rotationally restricted,
sterically hindered 9-fluorenyl radical, together with the
inversion mechanism substantiated in the reactions
involving the corresponding 9-fluorenyl cations (and
anions; see below), strongly supported an inversion
mechanism in the reactions with HI involving these
9-fluorenyl radicals (Scheme 8).
product, 16, [R]20 -33.5 (c 0.2, acetone), which is con-
D
figurationally enantiomeric with substrate 7. These
transformations from (+)-sp-7 to (-)-sp-16 are illustrated
in Scheme 9.
We intended to carry out a parallel study of the
stereochemistry of these C-9 radical reactions by X-ray
crystallographic examination of the absolute structures
of the epimeric products formed from HI reductions of
diastereomeric substrates related to the chiral 2-meth-
ylfluorenes discussed above. In our first attempt we
utilized chiral diastereomer 7c, whose absolute stereo-
chemical structure we had determined by X-ray analysis.3c
As expected, treatment of 7c with HI in CHCl3 at -60
°C provided a reddish-brown solution, as in the related
reductions, indicative of the generation of I2 from I- by
transfer of an electron to the fluorenyl cation, forming
the fluorenyl radical.3i,j,11 Workup afforded an off-white
solid (ca. 100% yield) shown by NMR to be an almost
pure, single epimer. Unfortunately, this product resisted
all attempts at crystallization, thereby precluding X-ray
analysis.
The mechanistic stereochemistry involving the 9-radi-
cal was further investigated by repeating the reaction
in HI in CHCl3 with a corresponding chiral substrate,
R-(+)-(sp)-9-(o-tert-butylphenyl)-9-methoxy-2-methylfluo-
rene (7), [R]20 +45.5 (c 0.2, acetone). At -60 to -40 °C,
D
chiral sp 7 was converted exclusively into the ap reduction
product, (R)-ap-9-(o-tert-butylphenyl)-2-methylfluorene
(15), via chiral cation 8 and chiral radical 14, the latter
subsequently undergoing hydrogen atom transfer from
HI.3h,10 This free-radical reduction mechanism is sup-
ported by our related studies.11 H-atom transfer from HI
occurred exclusively via kinetic control from the less
hindered face, the large iodine atom enhancing the effect
of steric hindrance, to provide only the ap product (15),
However, based on the indication that it was a single
epimer and that the related radical reductions appeared
(11) The purple-brown coloration of I2 was indicative of free-radical
formation. The related conversion of 9-hydroxy-9-(o-isopropylphenyl)-
fluorene into 9-(o-isopropylphenyl)fluorene by treatment with HI-
CHCl3 likewise produced this coloration. Moreover, formation of the
free-radical intermediate in the latter transformation was verified by
the isolation of its air-oxidation product, ap-9-(o-isopropylphenyl)-9-
fluorenyl peroxide, which was unequivocally characterized by X-ray
analysis.3i,j Under the same conditions, the corresponding peroxide from
radical 14 was not observed, most likely for steric reasons.
(10) Aside from our report3h and that of Oki et al.,6 there are only
a few reports of HI as a reducing agent for organic compounds: (a)
Gordon, P. E.; Fry, A. J . Tetrahedron Lett. 2001, 42, 831-833 and
references therein. (b) Penso, M.; Mottadelli, S.; Albanese, D. Synth.
Commun. 1993, 23, 1385. (c) Gemal, A. L.; Luche, J . L. Tetrahedron
Lett. 1980, 21, 3195.
J . Org. Chem, Vol. 69, No. 4, 2004 1191

Hou and Meyers
SCHEME 10. Red u ction of Dia ster eom er 7c w ith
Ep im er iza tion via C-9 Ra d ica l: Con ver sion of 7g
to proceed via inversion, it is reasonable to presume that
this reaction likewise proceeded via C-9 inversion, pro-
viding the single epimeric product 7g. This reaction
sequence is illustrated in Scheme 10.
In our second attempt to utilize X-ray analysis of the
product to confirm the stereochemistry of the radical
reduction, we used (R)-sp-9-(o-tert-butylphenyl)-9-meth-
oxy-2-(S)-(3-methyl-1-pentyl)fluorene (17) as the sub-
strate. Diastereomer (R)-sp-(S) 17 was prepared from the
reaction of (R)-sp-7f with (S)-2-methyl-1-butylMgBr,
which also provided the dimer (R)-sp,(R)-sp 17a . Chro-
matographic separation afforded (R)-sp-(S) 17, an oil,
identified and shown to be pure by 1H and 13C NMR.
Bubbling HI gas into a solution of 17 in CHCl3 at -60
°C for 30 min, followed by the usual workup and
chromatographic purification, afforded a colorless oil
identified by NMR to be the reduction product, a single
epimer, telling us that the reaction was stereospecific.
Crystallization was unsuccessful from a variety of sol-
vents which, again, precluded X-ray identification of its
absolute stereochemistry. On the basis of the observa-
tions discussed above, it is strongly indicated that this
reduction via the 9-fluorenyl radical, which provided a
single epimer, 18, proceeded via inversion (Scheme 11).
Rea ction s via th e 9-F lu or en yl An ion . After (R)-sp-
(-)-9-(o-tert-butylphenyl)-2-methylfluorene (16) was
treated with n-BuLi in THF for 6 h at 25 °C, a small
sample was removed, treated with excess D2O, and
1
worked up. H NMR showed that the 9-position of the
compound isolated contained 9.6% H and 90.4% D,
indicating that almost all of 16 was converted into its
anion. Acidification of the remainder of the solution pro-
SCHEME 11. Red u ctive Ep im er iza tion by
In ver sion via C-9 Ra d ica l: Con ver sion of 17 to 18
w ith HI-CHCl3
duced colorless crystals whose optical rotation, [R]20
D
+28.0 (c 0.093, acetone), indicated a composition of 8.2%
16 and 91.8% of its (S)-sp-(+)-enantiomer, 20. This result
supports the D/H exchange result, confirming that 16 was
almost all converted into its aromatic 9-fluorenyl anion
(16a ) (sp2, 4n + 2 π electrons). Reprotonation exclusively
from the less hindered face produced the (S)-ap rotamer,
19, i.e., 100% inversion associated with kinetic control,
which underwent rapid rotation to the thermodynami-
SCHEME 12. 100% In ver sion via Ch ir a l An ion
16a : Qu a n tita tive Con ver sion of 16 to 20 by
Dep r oton a tion -Rep r oton a tion
1192 J . Org. Chem., Vol. 69, No. 4, 2004

9-Fluorenyl Systems
a
SCHEME 13.
Effect of Rota tion a l Restr iction a n d Ster ic Hin d r a n ce on th e p Ka of 9-Ar ylflu or en es
a
pKa values (DMSO, 25 °C): fluorene, 22.6; 21, 17.92; 22, 18.32; 23, 18.78; 24, 18.55;12a,b and 1, 20.20.13
cally favored rotamer, sp-(+)-20, the enantiomer of 16
(Scheme 12).
found in this study, viz. steric hindrance to deprotonation,
and reprotonation with inversion, leading directly to the
geometrically different ap rotamer, 1d . The effects of
these structural features are illustrated in Scheme 13.
In view of these observations, the question arises: Can
constants be assigned to these rotational restriction and
steric hindrance effects, like resonance and inductive
effects, to correlate with their contribution to pKa values?
We are currently addressing this question.
In their study with related but achiral 1, Oki et al.6
found that reprotonation of its lithiated anion at -50 °C
indeed produced the ap rotamer, which rotated to the sp
substrate on warming. However, in the absence of chiral
information, they erroneously ascribed their results to
rotation of the o-tert-butylphenyl group of an sp3-hybrid-
ized anion/Li+ pair prior to reprotonation which, in turn,
would also erroneously indicate retention of configuration
(see Scheme 2). In our studies of the stereochemistry of
various reactions of thermodynamically favored sp fluo-
rene rotamers, inversions directly produced the thermo-
dynamically less favored ap rotamers which underwent
rapid rotation to the sp rotamer product, which was
enantiomeric with the substrate. In comparison, depro-
tonation of the ap rotamer 5 was followed by intramo-
lecularly assisted reprotonation with inversion, directly
providing sp rotamer 6. Because of the high barrier to
rotation, conversion of 6 into its ap rotamer, the enan-
tiomer of 5, did not occur (see Scheme 4).7
Exp er im en ta l Section
Gen er a l Meth od s. Unless stated otherwise, melting points
are corrected and 1H and 13C NMR spectra were taken in
CDCl3 at 300 and 75 MHz, respectively.
R esolu t ion of 7a via t h e Dia st er eom er ic (S)-N-r-
Meth ylben zylca r boa m id es. F or m a tion of (R)-sp-9-(o-ter t-
Bu tylp h en yl)-2-h yd r oxym eth yl-9-m eth oxyflu or en e (7e).
(a ) sp-Meth yl 9-(o-ter t-Bu tylp h en yl)-9-flu or en ol-2-ca r -
boxyla te (7b). A mixture of 7a (0.62 g, 1.73 mmol) and 10
mL of DMF was stirred, forming a light yellow solution.
Iodomethane (0.5 mL, 8.0 mmol) and anhyd NaHCO3 (3 g, 36
mmol) were added, the flask was flushed with argon, and the
reaction mixture was stirred at 25 °C for12 h, after which time
TLC indicated the absence of starting material. The mixture
was extracted with ether and the extracts were washed with
water, dried, and evaporated in vacuo, leaving a light blue
Bordwell et al.12a,b determined the pKa values of a very
large number of fluorenes and correlated them mainly
with the resonance, inductive, and polarizability effects
of the substituents. However, some of the changes in pKa
values could not be explained on this basis. For example,
in going from 9-phenylfluorene (21) to 9-p-tolylfluorene
(22) the acidity is reduced, but it is substantially further
reduced in 9-o-tolylfluorene (23); yet 9-mesitylfluorene
(24) is more acidic than 23. They related these unex-
pected observations to the restricted rotation of the 9-(o-
methylphenyl) and 9-(o,o′-dimethylphenyl), respectively,
in 23 and 24 and their 9-anions, and their relative anion
vs ground-state stabilities.12a We recently verified the
rotational restriction in 23 in solution (25 °C), NMR
showing 60% sp and 40% ap, although the crystalline
state is composed exclusively of the former.3r Our results
suggest that the reduction in the acidity of 1 compared
to 23 by 1.42 pKa units13 simply by replacing the o-methyl
with an o-tert-butyl (1, sp), emanates from the enhanced
rotational restriction in sp-1 and its anion14 and the
additional effects engendered by the o-tert-butyl group
1
solid, 0.70 g, shown by H NMR to be almost pure 7b (100%
yield). Recrystallization (hexanes) provided a colorless solid,
mp 172-173 °C. 1H NMR δ 1.82 (s, 9 H), 2.48 (br s, 1 H), 3.83
(12) (a) Bordwell, F. G.; Drucker, G. E.; McCollum, G. J . J . Org.
Chem. 1982, 2504-2510. (b) Bordwell, F. G. Acc. Chem. Res. 1988,
21, 456-463. (c) Structural and solvent effects evaluated from acidities
in DMSO and in the gas phase are described by: Taft, R. W.; Bordwell,
F. G. Acc. Chem. Res. 1988, 21, 463-469.
(13) The determination of the pKa of 1 by the standard method, using
4-chloro-2-nitroaniline as indicator, required a full day to get a stable
reading: Meyers, C. Y.; Hou, Y.; Lutfi, H. G.; Robinson, P. D. Abstracts
of Papers; 210th National Meeting of the American Chemical Society,
Chicago, IL, Aug 20-24, 1995; American Chemical Society: Wash-
ington, DC, 1995; ORGN 297.
(14) sp-1 is the exclusive rotamer in solution as well as the
crystalline state,3b while 9-(o-isopropylphenyl)fluorene exists in an sp:
ap rotamer ratio of 70:30 in CDCl3, and 58:42 in CD3OD, but 100% as
the sp rotamer in the crystalline state,3a and 9-(m-tert-butylphenyl)-
fluorene rotates rapidly in solution showing no conformational prefer-
ence, but crystallizes as its ap rotamer exclusively.3m
J . Org. Chem, Vol. 69, No. 4, 2004 1193

Hou and Meyers
(s, 3 H), 6.37 (dd, J ) 8.1, 1.5 Hz, 1 H), 6.74 (ddd, J ) 7.5, 1.2
Hz, 1 H), 7.06 (ddd, J ) 7.5, 1.5 Hz, 1 H), 7.33 (m, 3 H), 7.59
(dd, J ) 8.1, 1.5 Hz, 1 H), 7.71 (dd, J ) 7.8, 0.9 Hz, 2 H), 7.92
(s, 1 H), 8.03 (d, J ) 7.8 Hz, 1 H); 13C NMR δ 34.0, 37.1, 52.1,
87.2, 119.9, 120.9, 125.1, 125.2, 126.1, 126.4, 127.7, 128.9,
130.0, 130.2, 130.6, 131.9, 137.4, 138.5, 143.3, 149.6, 154.6,
155.3, 166.6. Anal. Calcd for C25H24O3: C, 80.62; H, 6.49.
Found: C, 80.58; H, 6.41.
(b ) (R)-sp -9-(o-ter t-Bu t ylp h en yl)-9-flu or en ol-2-(S)-N-
r-m eth ylben zylca r boa m id e (7c). A stirred solution of (S)-
(-)-R-methylbenzylamine (1 mL, 7.7 mmol) in THF (5 mL) was
cooled in an ice-water bath, n-BuLi-hexane (3.0 mL, 1.6 M,
4.8 mmol) was added, stirring was continued for 10 min, and
7b (0.569 g, 1.42 mmol) was added. After the mixture was
stirred under argon at 25 °C for 17 h, TLC indicated the
absence of 7b, but exhibited two very close spots after the TLC
plate was developed in hexanes-ether (2:1) three times. This
mixture was extracted with ether, and the extracts were
washed successively with dil HCl, NaHCO3, and water, dried,
and evaporated in vacuo, leaving a tan product (0.717 g).
Purification by dry column chromatography afforded a light
yellow solid, 0.658 g (100% yield). 1H NMR showed the product
to be a mixture of equal amounts of diastereomeric (R)- and
(S)-sp-9-(o-tert-butylphenyl)-9-fluorenol-2-(S)-N-R-methylben-
zylcarboamide.
7.55 (dd, J ) 8.4, 1.2 Hz, 1 H), 7.70 (d, J ) 7.8 Hz, 1 H), 7.73
(d, J ) 7.8 Hz, 1 H); 13C NMR δ 17.0 (br), 27.8 (br), 31.8 (br),
33.9, 37.6, 49.7, 50.8 (br), 56.5 (br), 92.9, 120.0, 126.6 (br),
124.9, 125.6, 126.1, 126.4, 127.3, 127.8, 128.5, 128.7, 129.0,
131.7, 136.3, 139.6, 139.6, 139.8 (br), 143.6 (br), 149.1, 150.4
(br). MS (HR, FAB, m/z) calcd for C34H36NO2 (M + H)+
490.2746, obsd 490.2742.
(d ) (R)-sp-9-(o-ter t-Bu tylp h en yl)-2-h yd r oxym eth yl-9-
m eth oxyflu or en e (7e). A solution of LiBHEt3 in THF (10 mL,
1.0 M, 10 mmol) was injected into a septum-sealed, argon-
flushed flask containing 7d (1.70 g, 3.48 mmol) and maintained
in an ice bath. The mixture, which became light blue, was
stirred for 30 min at 0 °C, then for 30 min at 25 °C, and the
reaction was quenched by the addition of dil aq NaOH-ethanol
and 30% H2O2. This mixture was stirred at 25 °C for 10 h and
extracted with ether; the extracts were washed with water,
dried (anhyd MgSO4), and evaporated in vacuo, leaving a
colorless thick oil (1.22 g, 97.9% yield) that crystallized from
hexanes solution; mp 118-119 °C. The product was character-
1
ized as 7e. H NMR δ 1.64 (br s, 1 H), 1.76 (s, 9 H), 2.76 (s, 3
H), 4.63 (s, 2 H), 6.42 (dd, J ) 8.1, 1.5 Hz, 1 H), 6.71 (ddd, J
) 8.1, 1.5 Hz, 1 H), 7.01 (ddd, J ) 8.4, 1.5 Hz, 1 H), 7.17-7.26
(m, 3 H), 7.31-7.38 (m, 2 H), 7.56 (dd, J ) 8.4, 1.5 Hz, 1 H),
7.66 (d, J ) 7.8 Hz, 2 H); 13C NMR δ 34.0, 37.7, 49.7, 65.4,
93.0, 119.6, 119.7, 124.1, 124.9, 125.6, 126.0, 127.5, 127.8,
128.4, 128.6, 131.9, 140.1, 140.2, 140.3, 141.2, 149.2, 150.4,
150.7. Anal. Calcd for C25H26O2: C, 83.76; H, 7.31. Found: C,
83.21; H, 7.32.
Over a 12-h period at 25 °C, an acetone solution of the two
diastereomers gave rise to a mass of colorless crystals formed
from an acetone solution containing the two diastereomers.
The liquid was decanted off and the crystalline residue (0.23
g) was washed with hexanes. TLC (silica gel, hexanes:ether
3:2) indicated that only the more polar diastereomer was
Con ver sion of 7e to (R)-sp-9-(o-ter t-Bu tylp h en yl)-2-
m eth ylflu or en e (7). (a ) (R)-sp-2-Br om om eth yl-9-(o-ter t-
bu tylp h en yl)-9-m eth oxyflu or en e (7f). After a mixture of
powdered KOH (1.0 g, 17.8 mmol), freshly recrystallized
p-toluenesulfonyl chloride (1.07 g, 5.6 mmol), and 7e (1.0 g,
2.79 mmol) in THF (10 mL) was stirred at 25 °C for 5 min,
powdered anhydrous lithium bromide (4.0 g, 46.2 mmol) was
added. The mixture became hot and stirring was continued
without external heating for 6 h, after which the reaction was
quenched with water, the mixture was extracted with hexanes,
and the extracts were washed with water, dried, and evapo-
rated in vacuo, yielding a colorless oil, shown by NMR to
contain 7e along with product 7f and some sp-(R)-2-chlorom-
ethyl-9-(o-tert-butylphenyl)-9-methoxyfluorene. Purification by
dry column chromatography (silica gel, hexanes) provided a
colorless solid. It was dissolved in acetone (30 mL) and lithium
bromide (10 g) was added while the solution was stirred at 25
°C for 10 h, then concentrated in vacuo. The solid residue was
triturated with ether and the ether washings were evaporated,
leaving colorless crystals, 0.98 g (87.1% yield), mp 113-114
1
present. H NMR likewise exhibited the presence of a single
diastereomer, which was unequivocally identified by X-ray
crystallography3c as the (R)-sp-(S) isomer, 7c; mp 194-195 °C
1
(crystals became brown). H NMR δ 1.55 (d, J ) 7.2 Hz, 3 H),
1.80 (s, 9 H), 2.61 (br s, 1 H), 5.24 (dq, J ) 6.6 Hz, 1 H), 6.26
(d, J ) 6.9 Hz, 1 H), 6.37 (dd, J ) 8.1, 1.5 Hz, 1 H), 6.75 (ddd,
J ) 7.5, 1.2 Hz, 1 H), 7.06 (ddd, J ) 7.8, 1.5 Hz, 1 H), 7.30 (m,
8 H), 7.58 (dd, J ) 7.8, 1.2 Hz, 1 H), 7.68 (m, 3 H), 7.75 (dd, J
) 8.1, 1.2 Hz, 1 H); 13C NMR δ 21.7, 33.9, 37.1, 49.3, 87.3,
120.1, 120.8, 123.6, 125.2, 126.3, 126.5, 127.4, 127.9, 128.1,
128.7, 128.9, 129.8, 132.1, 134.8, 137.6, 142.1, 143.1, 149.7,
154.9, 155.3, 166.2. On standing, the mother liquor gave rise
to more crystals, somewhat enriched in one diastereomer.
After the second crystallization, the mother liquor was
highly enriched in the other diastereomer, (S)-sp-(S), which
separated as crystals upon evaporation of the solvent and was
purified by column chromatography (silica gel; hexanes:ether
3:2) as white crystals (0.12 g). After being washed with
hexanes their mp was 179.5-181 °C (became brown). 1H NMR
δ 1.57 (d, J ) 6.9 Hz, 3 H), 1.81 (s, 9 H), 2.43 (br s, 1 H), 5.28
(dq, J ) 6.9 Hz, 1 H), 6.23 (d, J ) 7.2 Hz, 1 H), 6.38 (dd, J )
8.1, 1.5 Hz, 1 H), 6.75 (ddd, J ) 7.5 Hz, 1 H), 7.06 (ddd, J )
7.8, 1.5 Hz, 1 H), 7.31 (m, 8 H), 7.58 (dd, J ) 8.1, 1.2 Hz, 1 H),
7.68 (m, 3 H), 7.75 (dd, J ) 8.1, 1.2 Hz, 1 H); 13C NMR δ 21.8,
34.0, 37.1, 49.3, 87.3, 105.6, 120.1, 120.7, 123.6, 125.1, 126.2,
126.4, 127.4, 127.8, 127.8, 128.6, 128.9, 129.7, 132.0, 134.8,
142.0, 142.9, 149.5, 154.8, 155.1, 166.0.
(c) (R)-sp -9-(o-ter t-Bu t ylp h en yl)-9-m et h oxy-2-(S)-N-
m eth yl-r-m eth yben zylca r boa m id e (7d ). Freshly powdered
KOH (2.0 g, 35.7 mmol) and iodomethane (2 mL, 32 mmol)
were added to a solution of 7c (1.655 g, 3.59 mmol) and DMSO
(10 mL), the suspension was stirred at 25 °C for 5 h, and water
was added. The mixture was extracted with ether, the extracts
were washed with water, dried (anhyd MgSO4), and evapo-
rated in vacuo, leaving a light yellow sticky oil (1.77 g, 100%
yield). Crystals formed over a period of several weeks, mp 87-
88.5 °C, which were identified as 7d by NMR. 1H NMR δ 1.33
(br s, 2 H), 1.56 (br s, 1 H), 1.73 (s, 9 H), 2.54 (br s, 0.6 H),
2.73 (br s, 5.4 H), 4.82 (br s, 0.62 H), 6.1 (br s, 0.37 H), 6.38
(d, J ) 8.1 Hz, 1 H), 6.68 (dd, J ) 7.8 Hz, 1 H), 7.00 (ddd, J )
7.5, 1.5 Hz, 1 H), 7.3 (m, H), 7.49 (dd, J ) 7.5, 1.5 Hz, 1 H),
1
°C, identified as 7f by NMR. H NMR δ 1.77 (s, 9 H), 2.77 (s,
3 H), 4.44 (s, 2 H), 6.41 (dd, J ) 8.1, 1.5 Hz, 1 H), 6.72 (ddd,
J ) 1.2, 6.9, 8.1 Hz, 1 H), 7.03 (ddd, J ) 6.9, 1.5 Hz, 1 H), 7.22
(m, J ) 7.2 Hz, 3 H), 7.34 (ddd, J ) 7.5, 2.1 Hz, 1 H), 7.40 (dd,
J ) 7.8, 1.8 Hz, 1 H), 7.57 (dd, J ) 8.1, 1.5 Hz, 1 H), 7.64 (d,
J ) 7.8 Hz, 1 H), 7.66 (dd, J ) 7.5, 0.9 Hz, 1 H); 13C NMR δ
34.0, 37.7, 49.8, 92.9, 119.8, 120.0, 124.9, 125.6, 125.9, 126.1,
127.8, 128.7, 128.7, 129.8, 131.8, 138.0, 139.8, 139.9, 140.9,
149.2, 150.5, 150.9.
(b) (R)-sp-(+)-9-(o-ter t-Bu tylph en yl)-9-m eth oxy-2-m eth -
ylflu or en e (7). A solution of 7f (0.457 g, 1.12 mmol) in freshly
distilled dry THF (5 mL) was added to a solution of LiBHEt3
in THF (1.0 M, 10 mL, 10 mmol) in a flask sealed with a rubber
septum and flushed with argon, and immersed in an ice-water
bath. After the mixture was stirred at 25 °C under argon for
2 h, TLC indicated the absence of 7f and the presence of a
single spot chracteristic of a compound less polar than 7f. After
residual LiBHEt3 was decomposed by the addition of water,
the mixture was extracted with hexanes and worked up as
usual, leaving a colorless oil. Column chromatography (silica
gel, hexanes) provided a colorless sticky oil (0.36 g, 97.7% yield)
that crystallized overnight in the freezer. Recrystallization
(ethanol) yielded colorless crystals of 7, mp 106-107 °C. 1H
NMR δ 1.77 (s, 9 H), 2.30 (s, 3 H), 2.76 (s, 3 H), 6.44 (dd, J )
1194 J . Org. Chem., Vol. 69, No. 4, 2004

9-Fluorenyl Systems
8.1, 1.5 Hz, 1 H), 6.73 (ddd, J ) 8.1, 1.5 Hz, 1 H), 7.02 (ddd,
J ) 8.4, 1.5, Hz, 1 H), 7.04 (d, J ) 0.9 Hz, 1 H), 7.13-7.24 (m,
3 H), 7.32 (ddd, J ) 6.9, 1.5 Hz, 1 H), 7.55 (d, J ) 7.8 Hz, 1
H), 7.56 (dd, J ) 8.1, 1.5 Hz, 1 H), 7.63 (dd, J ) 7.5, 1.2 Hz,
1 H); 13C NMR δ 21.8, 34.0, 37.6, 49.6, 93.0, 119.3, 119.3, 124.9,
125.5, 125.9, 126.1, 127.6, 127.9, 128.4, 129.3, 132.1, 138.0,
138.38, 140.6, 149.1, 150.2, 150.5. Anal. Calcd for C25H26O: C,
87.68; H, 7.65. Found: C, 87.77; H, 7.71. The optical rotation
was measured from a 25-mL solution of 50.0 mg of 7 in
17 (0.175 g, mmol) and CHCl3 (5 mL) in a flask sealed with a
rubber septum was cooled in a liquid nitrogen-acetone bath
(-60 °C) and then stirred while HI gas (prepared by refluxing
a mixture of I2 and tetralin) was bubbled in for 30 min. The
reaction was stopped by removing the solvent and HI in vacuo.
The black oil residue was purified by dry column chromatog-
raphy (silica gel, hexanes) to a brownish oil, which was
dissolved in hexanes, and the solution washed with aq
Na2S2O3. The organic layer was dried and evaporated in vacuo,
leaving a colorless oil shown by NMR to be a single epimer of
(R?)-sp-9-(o-tert-butylphenyl)-2-(S)-(3-methylpentyl)fluorene (18).
Crystallization was unsuccessful from a variety of solvents.
Attempts to obtain crystals by forming a complex with picric
acetone: [R]20 +45.5 (c 0.2, acetone).
D
Ma jor In ver sion of 7 via Ch ir a l Ca tion 8. Tr ea tm en t
w it h Tr iflic An h yd r id e follow ed b y Na OCH3-CH 3OH .
F or m a tion of (S)-sp-(-)-9-(o-ter t-Bu tylp h en yl)-9-m eth -
oxy-2-m eth ylflu or en e (9). Triflic anhydride (0.5 mL, 6 mmol)
was added to a stirred solution of 7 (55 mg) in chloroform (2
mL) in a flask sealed with a rubber septum, flushed with
argon, and cooled in a liquid nitrogen-acetone bath (-60 °C).
The blood-red mixture, maintained at -60 to -40 oC, was
stirred for 30 min. A suspension of NaOMe-MeOH [43.5 mmol
(1.0 g of sodium metal in 25 mL of anhyd MeOH)] was added,
which immediately discharged the red color, and the mixture
was extracted with hexanes. The extracts, worked up as usual,
provided a pale yellow sticky oil, 49 mg (89.1% of substrate
1
acid in 95% ethanol or benzene also failed. H NMR δ 0.82 (t,
J ) 7.5 Hz, 3 H), 0.87 (d, J ) 6.3 Hz, 3 H), 1.15 (m, 1 H), 1.34
(m, 3 H), 1.58 (m, 1 H), 1.72 (s, 9 H), 2.57 (m, 2 H), 5.82 (s, 1
H), 6.25 (dd, J ) 7.5, 1.5 Hz, 1 H), 6.88 (ddd, J ) 7.5, 1.2 Hz,
1 H), 7.03 (s, 1 H), 7.11 (ddd, J ) 7.5, 1.5 Hz, 1 H), 7.19 (m, 3
H), 7.34 (m, 1 H), 7.49 (dd, J ) 8.1, 1.5 Hz, 1 H), 7.72 (d, J )
7.8 Hz, 1 H), 7.77(d, J ) 7.5 Hz, 1 H); 13C NMR δ 11.4, 19.2,
29.4, 32.7, 33.8, 34.2, 35.7, 38.7, 51.0, 119.4, 119.5, 125.0, 125.6,
126.1, 126.3, 126.8, 126.9, 127.1, 131.0, 138.6, 140.14, 140.08,
142.8, 148.0, 149.8, 150.1.
1
weight), whose H NMR spectrum matched that of substrate
Exclu sive In ver sion of (R)-sp-(-)-16 via Ch ir a l An ion
16a . Qu a n tita tive Dep r oton a tion -Rep r oton a tion of 16
Lea d in g t o (S)-sp -(+)-9-(o-ter t-Bu t ylp h en yl)-2-m et h yl-
flu or en e (20). A solution of n-BuLi in hexanes (2.2 M, 2 mL,
4.4 mmol) in THF was added to 16 (60 mg, 0.19 mmol) in a
septum-sealed flask. The mixture became red immediately and
was stirred at 25 °C under argon for 6 h, during which time
substantial precipitation occurred. A 0.5-mL sample of the
suspension was removed and diluted with D2O, while the
major part of the suspension was diluted with CHCl3 contain-
ing CF3CO2H, the red color of the anion being quickly
discharged in both cases, giving way to a light tan color. The
sample was diluted with D2O and extracted with hexanes and
the extracts were washed with water and worked up as usual,
7. However, recrystallization (ethanol) afforded colorless crys-
tals, mp 114-115.5 °C, with an optical rotation (determined
from a 25-mL solution of 26.5 mg in acetone) of [R]20 -19.2
D
(c 0.106, acetone), corresponding to a mixture of 71.1% 10
(inversion) and 28.9% 7 (retention); ee 42.2%. It was convinc-
ingly shown that isolated 7 was not unreacted substrate but
was formed in the reaction (see below).
Qu a n tita tive Con ver sion of sp-9-(o-ter t-Bu tylp h en yl)-
9-m eth oxyflu or en e (11) via Its Ca tion 1b to sp-9-(o-ter t-
Bu tylp h en yl)-9-eth oxyflu or en e (13) by Tr ea tm en t w ith
Tr iflic An h yd r id e F ollow ed by Na OEt-EtOH. The pro-
cedure described in the preceding reaction was repeated with
the same amounts of triflic anhydride and chloroform, but with
11 (50 mg) and NaOEt in ethanol (43.5 mmol; prepared by
dissolving 1.0 g of sodium metal into 25 mL of absolute
ethanol). The product was a slightly yellow sticky oil, 48 mg
(ca. 100% yield), characterized by 1H NMR to be exclusively
sp-9-(o-tert-butylphenyl)-9-ethoxyfluorene (sp-13).3g This prod-
uct was not purified further.
1
leaving light tan crystals, 13 mg, shown by H NMR to be sp-
9-(o-tert-butylphenyl)-2-methylfluorene containing 90.4% D
and 9.6% H at C-9. The portion of the suspension treated with
CF3CO2H-CHCl3 was worked up similarly, except that the
extracts were washed with aq NaHCO3 before being worked
1
up to provide light tan crystals, 44 mg, shown by H NMR to
Exclu sive In ver sion of 7 via Ch ir a l Ra d ica l 14. Red u c-
tion w ith HI (g) in CHCl3. F or m a tion of (R)-sp-(-)-9-(o-
ter t-Bu tylp h en yl)-2-m eth ylflu or en e (16). Hydrogen iodide
gas was bubbled into a colorless solution of 7 (145 mg, 0.29
mmol) in chloroform (2 mL) in a flask sealed with a rubber
septum, flushed with argon, and immersed in liquid nitrogen-
acetone bath (-60 °C). The mixture, which turned dark brown
very rapidly, was stirred at -60 to -40 °C for 60 min. TLC
exhibited a spot very close to that of the starting material.
The reaction was quenched by the addition of aq NaHCO3, the
mixture was extracted with ether, and the extracts were
washed with aq Na2S2O3, then water, and worked up as usual,
providing a yellow solid, 0.129 g (yield 97.2%). Recrystalliza-
tion (hexanes) afforded colorless crystals, mp 194-195 °C,
characterized as 16. 1H NMR δ 1.71 (s, 9 H), 2.32 (s, 3 H),
5.82 (s, 1 H), 6.34 (dd, J ) 7.8, 1.5 Hz, 1 H), 6.88 (ddd, J )
7.8, 1.5 Hz, 1 H), 7.03 (s, 1 H), 7.11 (ddd, J ) 8.1, 1.5 Hz, 1 H),
7.18 (m, 3 H), 7.35 (m, 1 H), 7.49 (dd, J ) 8.1, 1.2 Hz, 1 H),
7.71 (d, J ) 7.8 Hz, 1 H), 7.77 (d, J ) 7.8 Hz, 1 H); 13C NMR
δ 21.8, 32.7, 35.7, 50.9, 119.4, 119.5, 120.2, 125.6, 125.7, 126.1,
126.3, 126.8, 127.9, 130.9, 137.2, 138.4, 140.1, 141.0, 148.0,
be 100% sp-9-(o-tert-butylphenyl)-2-methylfluorene. Recrys-
tallization (ethanol) provided colorless crystals, mp 157-170
°C; optical rotation (measured on a 25-mL solution of 23.2 mg
in acetone), [R]20D +28.0 (c 0.093, acetone), corresponding to a
mixture of 8.2% (R)-sp-(-) substrate (16) and 91.8% of its (S)-
sp-(+) enantiomer, product 20; ee 83.6%. The 9.6% residual
hydrogen on C-9 of the sample diluted with D2O corresponded
well with the determination of 8.2% unreacted substrate 16,
showing an inversion of virtually 100%.
1
Su p p or tin g In for m a tion Ava ila ble: Synthesis, H and
13C NMR spectra, and elemental analysis of sp-9-(o-tert-
butylphenyl)-9-fluorenol-2-carboxylic acid (7a ), attempted reso-
lution of sp-9-(o-tert-butylphenyl)-9-fluorenol-2-carboxylic acid
(7a ) through its diastereomeric L-cinchonidine salts, and
attempted enantiomeric identification via X-ray analysis,
reduction of (R)-sp-9-(o-tert-butylphenyl)-9-fluorenol-2-(S)-N-
R-methylbenzylcarboamide (7c) via its 9-fluorenyl radical;
formation of a single epimer N-(S)-R-methylbenzyl (R?)-sp-9-
(o-tert-butylphenyl)fluorene-2-carboamide (7g), conversion of
(R)-sp-2-bromomethyl-9-(o-tert-butylphenyl)-9-methoxyfluo-
rene (7f) to (R)-sp-9-(o-tert-butylphenyl)-9-methoxy-2-(S)-(3-
methylpentyl)fluorene (17) and (R)-sp,(R)-sp-1,2-bis[9-(o-tert-
butylphenyl)-9-methoxy-2-fluorenyl]ethane (17a). This material
is available free of charge via the Internet at http://pubs.acs.org.
149.8, 150.1. Anal. Calcd for C24H24
: C, 92.26; H, 7.74.
Found: C, 92.48; H, 7.65. Optical rotation was determined on
a 25-mL solution of 50.0 mg of product in acetone; [R]20D -33.5
(c 0.2, acetone).
Red u ction of 17 via Its 9-F lu or en yl Ra d ica l. F or m a -
tion of a Sin gle Ep im er : (R?)-sp-9-(o-ter t-bu tylp h en yl)-
2-(S)-(3-m eth ylp en tyl)flu or en e (18). A colorless solution of
J O0351015
J . Org. Chem, Vol. 69, No. 4, 2004 1195