12072 J. Am. Chem. Soc., Vol. 118, No. 48, 1996
Lew et al.
redistilled over molecular sieves through a 30 cm Vigreux column.
Solvents used in all the HPLC experiments were of spectroscopic grade
(Ominsolv). All other solvents and reagents were of highest purity
available and were used as received from commercial sources.
for the equilibrium constant. Similar curve fitting of the 560
nm data for 3 gave values of k1 ) (8.5 ( 1.2) × 105 s-1, k2 )
4.4 × 107 M-1 s-1, and K ) 1.9 × 10-2 M. Values of k1 are
subject to large errors when k2 is comparable or larger than k1,
thus resulting in a substantial variation in both k1 and K in
comparing the two sets of data. Nevertheless, the average
estimate of 1.1 × 10-2 M for the equilibrium constant for the
acid dissociation of 3 compares favorably with the value of K
) 1 × 10-2 M obtained above from equilibrium concentration
measurements.62
Melting points are uncorrected. UV-visible spectra were measured
on a Varian Cary 3 double-beam spectrophotometer. Proton and
carbon-13 NMR spectra were recorded in CDCl3 on a Bruker AM-200
or an AMX-500 instrument. Chemical shifts are reported in parts per
million relative to tetramethylsilane as an internal standard. HPLC
experiments were run on a HP 1090 liquid chromatograph equipped
with a HP ODS Hypersil column (5 µm, 200 × 2.1 mm) and a UV
detector. LC-MS experiments were carried out on a HP 1090 liquid
chromatograph with a HP ODS-2 column (5 µm, 250 × 4.6 mm) linked
to a HP 5988 mass spectrometer with a thermospray interface.
Conclusions
Photolysis of alcohol 1 in HFIP generates transient zwitterion
2, which has an absorption maximum at 495 nm and reacts
rapidly with alcohols but is insensitive to oxygen. The
assignment of the 495 nm transient to zwitterion 2 is confirmed
by a time-resolved infrared experiment in which a species that
absorbs in the carboxylate region and that matches kinetically
the UV-visible transient is observed. Consistent with this,
steady state photolysis of 1 in methanol leads to the formation
of the expected nucleophilic trapping product, ether 7. This
work provides the first direct detection at room temperature of
the type of zwitterionic intermediate that is thought to be
responsible for the efficient polyesterifications observed for
reactions involving R-lactone intermediates.37,39,63 However, the
experiments do not provide any direct evidence for the interven-
tion of the cyclic R-lactone in the room-temperature photo-
chemistry of alcohol 1. In the presence of TFA the zwitterion
is protonated to give its conjugate acid, cation 3, which has
λmax at 560 nm. Both are formed at intermediate acid
concentrations, and experiments in the presence of variable
concentrations of TFA lead to an estimated pKa (in HFIP) of
∼2 for 3, substantially lower than the pKa of the parent
hydroxycarboxylic acid 1 (∼4 in water). Similar increases in
acidity have been observed for radical cations of p-aminobenzoic
acid and various methoxylated benzoic acids which yield
transient radical zwitterions upon deprotonation.64-66
Both zwitterion 2 and cation 3 react with alcohols more
slowly than the parent 9-fluorenyl cation, indicating that both
R-carboxyl and carboxylate groups lead to kinetic stabilization
of the 9-fluorenyl cation. This is more pronounced in the
additions of 2 and 6 to substituted benzenes, for which the rate
constants vary by 3 orders of magnitude. Reactions of zwit-
terion 2 with anionic nucleophiles are also several orders of
magnitude slower than those for other 9-fluorenyl cations,
although in this case the difference in reactivity is largely due
to electrostatic effects. Similar rate constants are measured for
the addition of both zwitterion 2 and the 9-fluorenyl cation to
several substituted diphenylmethanes, although there are several
possible mechanisms for these reactions.
Steady State Photolysis of 1. (a) NMR. 9-Hydroxy-9-fluorene-
carboxylic acid (1) (50 mg) was dissolved in a mixture of HFIP/
methanol (1 mL, 9:1 v/v), and the solution was photolyzed at 300 nm
in a Rayonet photoreactor for 20 min. The solvent was evaporated
under vacuum. A 200 MHz 1H NMR spectrum of the residual material
in CDCl3 consists of a singlet (2.92 ppm), a very broad peak (6.2 ppm),
and a multiplet in the aromatic region (7.25-7.70 ppm). The broad
peak at 6.2 ppm was assigned to alcoholic/acidic protons by a deuterium
exchange experiment with D2O. Addition of an authentic sample
showed that the signal at 2.92 ppm is due to the methoxy group of
9-methoxy-9-fluorenecarboxylic acid (7). In the particular experiment
described here, the ratio of the MeO signal at 2.95 ppm and aromatic
protons at 7.25-7.70 ppm indicated ∼20% conversion of the starting
material 1 to the ether.
(b) HPLC/LC-MS. A solution prepared as described above was
photolyzed for varying times, and aliquots were periodically withdrawn
and examined by HPLC. As photolysis in the HFIP/methanol (9:1 v/v)
mixture proceeded, the peak due to starting material 1 decreased in
intensity and a new peak with a longer retention time was formed.
The chemical ionization (ammonium acetate) mass spectrum of the new
peak had m/z 258 (M + NH4+, M ) CH3OC13H8CO2H). This peak
was assigned to 7 by co-injection of an authentic sample. Longer
photolysis times (>30 min) resulted in the formation of 9-fluorenone
and unidentified products with longer retention times.
When the photolysis was carried out in HFIP alone, a similar new
peak was detected by HPLC; the chemical ionization (ammonium
acetate) mass spectrum of this peak had m/z 394 (M + NH4+, M )
(CF3)2CHOC13H8CO2H). Based on the above results, the new peak
was assigned to 9-(1,1,1,3,3,3-hexafluoro-2-propanoxy)-9-fluorenecar-
boxylic acid. At longer photolysis times (>30 min), 9-fluorenone and
unidentified products with longer retention times were also detected.
Preparation of 9-Methoxy-9-fluorenecarboxylic acid (7). 7 was
prepared by saponification of methyl 9-methoxy-9-fluorenecarbox-
ylate,67,68 which was synthesized by MeI/Ag2O O-methylation69 of
methyl 9-hydroxy-9-fluorenecarboxylate.70 A solution of methyl
9-methoxy-9-fluorenecarboxylate (0.5 g, 2 mM) in carbon tetrachloride
(10 mL) and an aqueous solution of sodium hydroxide (2 M, 25 mL)
was refluxed for 24 h. After cooling to room temperature, the organic
layer was removed and the aqueous layer was further washed with
carbon tetrachloride (2 × 25 mL). Aqueous hydrochloric acid (1 M)
was added to the aqueous layer until the pH was <3. A white solid
precipitated out and was extracted with chloroform (3 × 20 mL). After
drying over magnesium sulfate, the solvent was removed at reduced
pressure to give 7 as a white solid (0.26 g, 55%). NMR and HPLC
experiments indicated that 7 prepared in this manner was at least 97%
pure, mp 182-5 °C, lit.68 mp 188 °C: 1H NMR (CDCl3, 200 MHz) δ
7.27-7.70 (m, 8H, aromatic), 2.92 (s, 3H, OMe); (lit.68 1H NMR
(CDCl3) δ 2.9 (s, 3H)): FABMS for C15H12O3 m/z 253 [M - H +
2Li+], 247 [M + Li+] (using 3-nitrobenzyl alcohol as the matrix,
positive mode).
Experimental Section
General Information. 9-Hydroxy-9-fluorenecarboxylic acid and
9-fluorenol were commercial samples (Aldrich) and were recrystallized
twice from ethanol twice before use. 1,1,1,3,3,3-Hexafluoro-2-propanol
(HFIP) was distilled under a dry nitrogen atmosphere from a mixture
of anhydrous sodium bicarbonate and 4A molecular sieves and then
(62) It should be noted that both approaches have assumed that the [H+]
is equivalent to the bulk TFA concentration, which may introduce an
additional error in the determination of the equilibrium constant.
(63) Adam, W.; Rucktaschel, R. J. Am. Chem. Soc. 1971, 93, 557-559.
(64) Tripathi, G. N. R.; Su, Y. J. Am. Chem. Soc. 1996, 118, 2235-
2244.
(67) Schlenk, W.; Bergmann, E. Justus Liebig’s Ann. Chem. 1928, 463,
98.
(65) Steenken, S.; O’Neill, P.; Schulte-Frohlinde, D. J. Phys. Chem. 1977,
81, 26-30.
(66) O’Neill, P.; Steenken, S.; Schulte-Frohlinde, D. J. Phys. Chem. 1977,
81, 31-34.
(68) Toussaini, O.; Capdevielie, P.; Maumy, M. Tetrahedron 1984, 40,
3229.
(69) Davidson, D.; Bernhard, S. J. Am. Chem. Soc. 1948, 70, 3426.
(70) Cannon, J. G.; Darko, L. L. J. Org. Chem. 1964, 29, 3419.