De Luca et al.
possible, compounds were identified by comparison with
authentic samples. All runs were conducted at least in
duplicate.
SCHEME 2
P r ep a r a tion of F or m ic Acid P h en eth yl Ester (Ta ble
1, Ru n 2).15 This procedure is representative for formylation
of alcohols. 2,4,6-Trichloro-1,3,5-triazine (1.0 g, 5.0 mmol) was
added to DMF (2 mL), maintained at 25 °C. After the formation
of a white solid, the reaction was monitored (TLC) until
complete disappearance of TCT, then CH2Cl2 (25 mL) was
added, followed by LiF (0.52 g, 20.0 mmol). After the mixture
was stirred overnight at room temperature, the alcohol (0.68
g, 5.0 mmol) was added, and the mixture was monitored (TLC)
until completion (15-30 min). Water was added, and then the
organic phase was washed three times with 3 N HCl, followed
by a saturated solution of Na2CO3 and brine. The organic layer
was dried (Na2SO4) and the solvent evaporated in vacuo to
yield formic acid phenethyl ester that was isolated without
other purifications (0.75 g, 99%): 1H NMR δ 8.02 (1H, s), 7.32-
24): only the primary function was protected, leaving the
other hydroxyl groups unchanged.
Another characteristic feature of the present formyla-
tion reagent is the possibility to quantitatively convert
O-tert-butyldimethylsilylated alcohols in one step to their
corresponding formates.10 Thus, the removal of silicon
protective groups can occur under extremely mild and
highly selective conditions using fluoride ions that are
compatible with most functional groups. Such an ex-
change of alcohol protecting groups without any inter-
mediate deprotection is of great importance in multistep
synthesis.
On the basis of our previous considerations on the
mechanism of this kind of reactions, it is possible to
suppose that even in this case a Vilsmeier-Haack-type
complex should be formed, containing the triazine moi-
ety.11 The addition of LiF should prevent the formation
of the alkyl chloride7h,12 and allow the attack of the
hydroxyl group of the alcohol to form an imminium
intermediate salt. Subsequent hydrolysis should form the
formate ester (Scheme 2). However, this simplified mech-
anism is not able to explain how the types of hydroxyl
groups can play a different role so determining the high
selectivity of the reaction.
7.21 (5H, m), 4.38 (2H, t, J ) 6 Hz), 2.97 (2H, t, J ) 6 Hz); 13
NMR δ 160.9, 137.5, 128.8, 128.5, 64.3, 34.8.
C
F or m ic a cid 3-p h en ylp r op yl ester 16 (Ta ble 1, r u n 3):
15 min, 99%; 1H NMR δ 8.08 (1H, s), 7.30-7.17 (5H, m), 4.18
(2H, t, J ) 6 Hz), 2.71 (2H, t, J ) 6 Hz), 2.30-1.97 (2H, m).
13C NMR δ 161.0, 138.9, 128.8, 128.3, 128.0, 68.3, 32.0, 31.6.
F or m ic a cid 2-p h en ylsu lfa n yleth yl ester (Ta ble 1, r u n
1
6): 15 min, 80%; H NMR δ 8.08 (1H, s), 7.39-7.19 (5H, m),
4.30 (2H, t, J ) 6 Hz), 3.59 (2H, t, J ) 6 Hz); 13C NMR δ 160.7,
134.2, 130.4, 129.2, 127.0, 72.2, 36.0. Anal. Calcd for C9H10O2S
(182.24): C, 59.32; H, 5.53; S, 17.59. Found: C, 59.32; H, 5.58,
S, 17.57.
F or m ic a cid 3-ter t-bu toxyp r op yl ester 17 (Ta ble 1, r u n
7): 15 min, 85%; 1H NMR δ 8.09 (1H, s), 5.08 (2H, m), 3.42-
3.39 (2H, m), 1.30-1.27 (2H, m), 1.18 (3H, s); 13C NMR δ 160.9,
72.9, 66.1, 61.2, 31.2, 27.5.
F or m ic a cid 2-m eth oxyeth yl ester (Ta ble 1, r u n 8): 15
min, 91%; 1H NMR δ 8.10 (1H, s), 4.33 (2H, t, J ) 6 Hz), 3.63
(2H, t, J ) 6 Hz), 3.41 (3H, s); 13C NMR δ 160.8, 72.4, 69.9,
58.8. Anal. Calcd for C4H8O3 (104.10): C, 46.15; H, 7.75.
Found: C, 46.11; H, 7.78.
F or m ic a cid 3,7-d im eth yloct-6-en yl ester 18 (Ta ble 1,
r u n 9): 15 min, 92%; 1H NMR δ 8.04 (1H, s), 5.07 (1H, t, J )
6 Hz), 4.16 (2H, t, J ) 3 Hz), 2.10-1.94 (2H, m), 1.71 (3H, s),
1.67 (3H, s), 1.45-1.05 (5H, m), 0.11 (3H, d, J ) 6 Hz); 13C
NMR δ 161.1, 131.4, 124.3, 62.3, 36.8, 35.2, 29.2, 25.6, 25.2,
19.2, 17.5.
F or m ic a cid (6,6-d im eth ylbiciclo[3.1.1]h ep t-2-yl)m eth -
yl ester (Ta ble 1, r u n 11): 15 min, 97%; 1H NMR δ 8.02
(1H, s), 4.09 (2H, d, J ) 9 Hz), 2.40-2.29 (2H, m), 1.95-1.82
(6H, m), 1.50-1.17 (1H, m), 1.15 (3H, s), 0.96 (3H, s); 13C NMR
δ 161.2, 68.2, 42.8, 41.1, 40.0, 38.4, 32.7, 27.7, 25.7, 23.1, 18.5.
Anal. Calcd for C11H18O2 (182.26): C, 72.49; H, 9.95. Found:
C, 72.51; H, 9.99.
In conclusion, the procedure reported here is opera-
tionally simple and allow a rapid, high-yielding, and
selective formylation of primary alcohols under very mild
conditions using inexpensive and readily available start-
ing materials. Moreover, it seems to provide a convenient
method for the conversion of O-TBDMS alcohols to their
formate esters in one step.13
Exp er im en ta l Section
F or m ic a cid 2-ben zyloxyca r bon yla m in oeth yl ester
(Ta ble 1, r u n 13): 30 min, 83%; 1H NMR δ 8.08 (1H, s), 7.37
(5H, s), 5.12 (2H, s), 4.26 (2H, t, J ) 6 Hz), 3.70-3.25 (3H,
m); 13C NMR δ 160.7, 156.4, 141.4, 132.5, 128.4, 127.9, 72.2,
66.8, 43.4. Anal. Calcd for C11H13NO4 (223.23): C, 59.19; H,
5.87; N, 6.27. Found: C, 59.21; H, 5.89; N, 6.25.
The N-protected amino acids were prepared according
standard methods, and their purities were established before
utilization by melting point and optical rotation. The N-
protected â-amino alcohols were prepared according to litera-
ture.14 Cyanuric chloride was purchased from Aldrich.
All solvents and reagents were used as obtained from
commercial sources. Standard 1H NMR and 13C NMR were
recorded at 300 and 75.4 MHz, from CDCl3 solutions. When
2-F or m yloxym eth ylp yr r olid in e-1-ca r boxylic a cid ben -
1
zyl ester (Ta ble 1, r u n 14): 30 min, 93%; H NMR δ 8.08
(1H, s), 7.38 (6H, bs), 5.15 (2H, s), 4.27 (1H, bs), 4.16 (2H, bs),
3.44 (2H, bs), 1.97-1.88 (4H, m); 13C NMR δ 162.6, 160.8,
137.4, 128.7, 128.5, 128.0, 72.2, 67.24, 47.3, 46.2, 28.5, 21.8.
Anal. Calcd for C14H17NO4 (263.29): C, 63.87; H, 6.51; N, 5.32.
Found: C, 63.88; H, 6.52; N, 5.34.
(10) The reaction requires the addition of tetrabutylammonium
fluoride (TBAF) together with LiF.
(11) In fact, TCT disappears (TLC) within a few minutes when
dissolved in DMF.
(12) Similar results are obtained using TBAF or tetrabutylammo-
nium bromide; however, in these cases the reaction rate is reduced
and small amounts of alkyl chlorides are formed. On these bases, one
can venture the hypothesis that precipitation of the chloride ion is
responsible for the change of the course of the reaction.
(13) Iranpoor, N.; Zeynizadeh, B. Synth. Commun. 1999, 29, 2123.
Koeller, S.; Lellouche, J . P. Tetrahedron Lett. 1999, 40, 7043.
(14) Falorni, M.; Porcheddu, A.; Taddei, M. Tetrahedron Lett. 1999,
40, 4395.
(15) Buckels, R. E.; Maurer, E. J . J . Org. Chem. 1953, 18, 1585.
(16) Fabre, J . L.; J ulia, M.; Mansour, B.; Saussine, J . J . Organomet.
Chem. 1987, 19, 161.
(17) Maillard, B.; Manigand, C.; Pavlovna Tarassova, N.; Villenave,
J . J .; Flilliatre, C. Bull. Soc. Chim. Fr. 1981, 255.
(18) Eschinasi, E. H. US 3244752, 1961; Chem. Abstr. 1966, 64,
17650f. Sonnet, P. E. Synthesis 1980, 828.
5154 J . Org. Chem., Vol. 67, No. 15, 2002