Not surprisingly, attempts to remove the TCE group using
Zn/HOAc resulted in complete desulfation (loss of the TCE
group and cleavage of the aryl S-O bond). Removal of the
TCE group was attempted using the hydrogenolysis condi-
tions of Paquet (10 wt % of 10% Pd/C, H2 (balloon) in
aqueous or dry ethanol). However, this also resulted in
complete desulfation. It was hypothesized that HCl was
formed during the reaction that resulted in desulfation of the
deprotected product. An alternative to using H2 for hydro-
genolyses is to use an alkene, formic acid, or ammonium
formate as the hydrogen source (catalytic transfer hydro-
genolysis, CTH).14 We reasoned that CTH using ammonium
formate would be suitable for TCE sulfate deprotection
because the ammonium formate would not only act as a
source of hydrogen but also buffer the solution and prevent
cleavage of the S-O bond of the monoester product. After
some experimentation, it was found that the TCE group could
be readily removed, usually within 5 h, using 6 equiv of
ammonium formate and 10% Pd/C (10 wt %) in methanol.
When the reaction was complete, the Pd catalyst was
removed and the supernatant concentrated. The crude mate-
rial was rapidly passed through a small silica column using
CH2Cl2/MeOH/NH4OH (10:2:0.5) as the eluant (to remove
the ammonium chloride that is formed in the reaction). The
pooled fractions were concentrated, and the residue was
dissolved in water and then lyophilized to give the aryl
sulfates as their ammonium salts. In general, the deprotection
proceeded in good yield (for substrates 16-22 and 26, Table
1) except in cases where certain groups on the aryl ring were
present (substrates 23-25, Table 1). Aryl halides 23 and 24
underwent both deprotection and dehalogenation.15 The
reaction with 25 resulted in a complex mixture of products
due to reduction of the carbonyl group to the alcohol as well
as reductive amination of the carbonyl group to the amine.16
However, the ketone moiety at the 17-position in estrone
derivative 26 remained intact. Surprisingly, the 4-nitro group
on ester 18 was not reduced to the amine. Ram and
Ehrenkaufer have reported that aromatic nitro groups are
reduced to amino groups using CTH with ammonium
formate.17 However, these workers used anhydrous am-
monium formate, 25-40 wt % of 10% Pd/C and dry MeOH.
Conditions that would selectively remove the TCE group
in the presence of other reducible groups were explored. It
has been reported that the TCE group can be removed in
moderate yields from TCE esters of carboxylic acids using
Zn/1 M NH4OAc.18 Therefore, we anticipated that the TCE
groups in 16-26 could be removed using Zn/NH4HCO2.
Subjecting compounds 16-26 to a mixture of 2 equiv of Zn
dust and 6 equiv of NH4HCO2 in MeOH resulted in removal
of the TCE group usually in less than 1 h. When the reactions
were complete, they were filtered through Celite and the
concentrated filtrates purified and lyophilized as described
above. This gave the aryl sulfates as their ammonium salts
Scheme 2
generally in very good yield (Table 1). Using the Zn
procedure, aryl halides 23 and 24 did not undergo dehalo-
genation and the carbonyl group in 25 was not reduced.
However, the nitro group in substrate 18 was partially
reduced to the amine.
Paquet reported that TCE-protected phosphates were stable
to hydrogenolysis when the reaction was performed in
AcOH/TFA 1:1 (H2, 10% Pd/C) and that benzyl esters could
be selectively debenzylated in the presence of TCE-protected
phosphates group using these conditions.8,9 Using compound
39 as a model system, it was found that the benzyl ester
moiety could be selectively deprotected under these condi-
tions to give acid 41 in 82% yield (Scheme 2). Moreover,
the sulfate group could be selectively deprotected using the
Zn procedure to give 42 in 95% yield or both esters could
be deprotected simultaneously using CTH to give 40 in 81%
yield.
Stability studies were conducted using 4-fluoro ester 16
as a model system. This diester is remarkably stable to acids
such as TFA, TFA containing 5% of 30% HBr/AcOH, 30%
HBr/AcOH, and 4 M HCl in dioxane over a period of 24 h.
It was also stable to weak organic bases such as 10% Et3N
in CH2Cl2, 20% N-ethylmorpholine in CH2Cl2 for at least
24 h, 2 equiv of aqueous LiOH in THF at 0 °C for 30 min,
and NaBH4/MeOH. However, it is not stable to good
nucleophiles or stronger organic bases such as 20% piperi-
dine or 20% DBU in DMF or CH2Cl2, MeONa/MeOH, or
t-butoxide in THF.
This new methodology was applied to the synthesis of
compounds 1 and 2 (Scheme 3). Heating a mixture of
estrone, paraformaldehyde, MgCl2, and tributylamine in a
sealed tube for 11 h gave a mixture of 2-formyl and 4-formyl
estrone in a 14:1 ratio. These two isomers could be separated
by chromatography to give the 2-isomer, 43, in a 52% yield.19
(14) Ram, S.; Ehrenkaufer, R. E. Synthesis 1988, 91-95.
(15) Rajagopal, S.; Spatola, A. F. J. Org. Chem. 1995, 60, 1347-1355.
(16) Ram, S.; Spicer, L. D. Synth. Commun. 1992, 22, 2673-2681.
(17) Ram, S.; Ehrenkaufer, R. E. Tetrahedron Lett. 1984, 25, 3415-
3418.
(18) Jou, G.; Gonzalez, F.; Albericio, P. L.-W.; Giralt, E. J. Org. Chem.
1997, 62, 354-366.
(19) This procedure is based on the work of Hofslokken and Skattebol:
Hofslokken, N. U.; Skattebol, L. Acta Chim. Scand. 1999, 53, 258-262. A
more detailed account on the formylation of estrogens using this and other
procedures is forthcoming.
Org. Lett., Vol. 6, No. 2, 2004
211