Selective mono-acylation of meso- and C2-symmetric 1,3- and 1,4-diols

Article abstract of DOI:10.1016/S0040-4039(02)00935-8

The direct one-pot mono-acylation of meso and C₂-symmetric 1,3- and 1,4-diols has been achieved using carboxylic acid anhydrides and catalytic amounts of cerium trichloride.

Full text of DOI:10.1016/S0040-4039(02)00935-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4761–4763
Selective mono-acylation of meso- and C2-symmetric 1,3- and
1,4-diols
Paul A. Clarke*
School of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, UK
Received 19 April 2002; accepted 13 May 2002
Abstract—The direct one-pot mono-acylation of meso and C2-symmetric 1,3- and 1,4-diols has been achieved using carboxylic acid
anhydrides and catalytic amounts of cerium trichloride. © 2002 Elsevier Science Ltd. All rights reserved.
The mono-functionalisation of symmetrical 1,n-diols is
a non-trivial process. This is because the hydroxyl
groups are in identical or similar chemical environ-
ments, and as such they react at a similar, if not the
same, rate. The effect of this is that reactions run with
a stoichiometric quantity of reagent tend to generate
statistical mixtures of starting material, mono-function-
alised and bis-functionalised products. Those reactions,
however, which are run using an excess of reagent tend
to favour the formation of bis-functionalised products.
Several attempts have been made to obviate these prob-
lems.1–3 These include continual extraction of the
mono-acylated product from the reaction medium,1
heterogeneous or solid supported methods2 and the
formation/cleavage of cyclic ‘acetal-like’ systems.3 The
problems inherent with these strategies are the labori-
ous and/or the multi-step nature of the processes.
In this letter it is disclosed that these reaction condi-
tions may be modified to enable the one-step mono-
acylation of both meso- and C2-symmetric 1,3- and
1,4-diols in excellent yields (Scheme 2).
The optimal procedure for the mono-acylation of 1,2-
diols involved the use of YbCl3 (10 mol%) as a catalyst
in the presence of acetic anhydride (10 equiv.). When
2-phenylpropane-1,3-diol was subjected to these condi-
tions it was found that the diol was rapidly consumed
1
(ꢀ2 h). However, H NMR (400 MHz) analysis of the
crude reaction mixture indicated that the acylation was
unselective, generating equal quantities of both the
mono- and the bis-acylated product. This is in marked
contrast to the rapid yet mono-selective reaction of
1,2-diols under these conditions. To rule out the possi-
bility of an unusually rapid uncatalysed background
reaction, 2-phenylpropane-1,3-diol was treated with
acetic anhydride (10 equiv.) in the appropriate amount
of THF. Even after 2 days only a trace amount of
We have recently reported a novel and much simpler
procedure for the one-step highly selective mono-
acylation of meso- and C2-symmetric 1,2-diols.4 This
procedure utilised an excess of carboxylic acid anhy-
dride, as an acylating agent, and a catalytic amount of
lanthanide trichloride as an acylation promoter
(Scheme 1).5–7
1
acylated product could be detected by H NMR (400
MHz). In an attempt to slow the catalysed reaction and
increase mono selectivity CeCl3 (10 mol%) was next
used as a catalyst.8 In the case of 1,2-diols, CeCl3 was
as equally effective as YbCl3 (10 mol%), but a longer
reaction time was required. Indeed, the rate of the
reaction was reduced, but after 23 h, 2-phenylpropane-
1
1,3-diol had been consumed. Gratifyingly the H NMR
O
HO
R
OH
R
YbCl3, THF
HO
R
O
R1
O
O
OH OH
( )n
OH OAc
( )n
R
CeCl3, THF
Ac2O
O
R1
R1
Scheme 1.
n= 1 or 2
* Tel.: +44 115 9513566; e-mail: paul.clarke@nottingham.ac.uk
Scheme 2.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00935-8

4762
P. A. Clarke / Tetrahedron Letters 43 (2002) 4761–4763
(400 MHz) of the crude reaction mixture indicated that
85% of the material was the mono-acylated product.
This was isolated via flash column chromatography in a
77% overall yield. These results are summarised in
Table 1.
Having successfully extended the methodology to 1,3-
diols, the next step was to investigate the possibility of
the mono-acylation of 1,4-diols (Table 3). The first
1,4-diol to be investigated was 2R,5R-hexanediol.
Acylation in the presence of CeCl3 (10 mol%) was slow
(only 30% conversion after 23 h) but completely mono-
selective. In an attempt to increase the rate of the
acylation reaction, YbCl3 (10 mol%) was used as a
catalyst. In this case the acylation was rapid and gener-
ated 65% of the mono-acetate within 8 h (the remainder
was the bis-acylated product). Diols 5 and 6 were
formed by the LiAlH4 reduction of the appropriate
cyclic anhydrides. The anhydrides being generated by
the Diels–Alder cyclisations of maleic anhydride and
the appropriate diene. It was found that meso-diols 5
and 6 also underwent mono-acylation smoothly and in
high yields, 80 and 83%, respectively. In the case of diol
5 the reaction was extended to the formation of the
mono-benzoate, although the rate of benzoylation (5
equiv. of Bz2O was used) was such to warrant the use
of YbCl3 as a catalyst.
It was next sought to employ these conditions for the
mono-acylation of other symmetrical 1,3-diols with a
selection of anhydrides. As can be seen from Table 2,
pro-chiral and C2-symmetric 1,3-diols can be treated
with the appropriate anhydride to form the mono-ace-
tate, mono-benzoate and even the mono-tert-butyloxy-
carbonate, all in good to excellent yields. In the case of
the benzoylation of 3, the CeCl3-promoted reaction was
sluggish so YbCl3 was employed.8
Table 1.
OH OH
OH OH OH OAc OAc OAc
Ph
Ph
Ph
Ph
1
1a
1b
It is intriguing that the nature of the lanthanide trichlo-
ride catalyst is very influential on both the rate and the
selectivity of these reactions. While the increased rate of
reaction with lanthanides of greater atomic mass can be
rationalised in terms of the increasing Lewis acidity of
the lanthanide metal (i.e. the well known lanthanide
contraction effect), it is more difficult to rationalise the
mono- versus bis-selectivity differences. One possibility
Catalyst
(10 mol%)
Yield 1a (%)
Time (h)
Ratio 1/1a/1b
100/trace/0
0/50/50
None
YbCl3
48
2
0
48
0/85/15
CeCl3
23
77
Table 3.
Catalyst
(10 mol%)
Table 2.
Yield (%)
Diol
OH
Product
OH
30a
CeCl3
Anhydride (10 eq.)
Yield (%)
Diol
Product
4
OH
OH
4a OAc
OH OH
OH OBz
Bz2O
70
OH
OH
YbCl3
CeCl3
Ph
Ph
65
80
1
1c
OAc
4
4a
OH OH
OH OBoc
H
H
OH
OH
OAc
OH
Boc2O
Ac2O
83
93
Ph
Ph
H
H
5a
1
1d
5
OH OH
OH OAc
H
H
OH
OH
OBz
OH
64b
YbCl3
CeCl3
H
H
2
2a
5b
5
OH OH
OH OAc
Ac2O
Bz2O
75
83
3
3a
OH
OH
OAc
OH
OH OH
OH OBz
57a
6
6a
3
3b
a) Ac2O (10 eq.), 100% mono-acetate at 30% conversion after 23 h.
b) Bz2O (5 eq.), 76% mono-benzoate at 84% conversion after 24 h.
a) Bz2O (5 eq.), YbCl3 (10 mol%), 57% conversion at 22.5 h.

P. A. Clarke / Tetrahedron Letters 43 (2002) 4761–4763
4763
is that the diol chelates to the lanthanide salt and the
References
differences in the chelate ring size/stability and dis-
tances between the co-ordinating hydroxyl groups are
more or less compatible with different sized lan-
thanides. The role of the lanthanide trichloride catalyst
and its interaction with the diol is currently the subject
of investigation and will be reported in due course.
1. Babler, J. H.; Coghlan, M. J. Tetrahedron Lett. 1979, 20,
1971.
2. (a) Nishiguchi, T.; Taya, H. J. Am. Chem. Soc. 1989, 111,
9102; (b) Nishiguchi, T.; Fujisaki, S.; Ishii, Y.; Yano, Y.;
Nishida, A. J. Org. Chem. 1994, 59, 1191; (c) Nishiguchi,
T.; Fujisaki, S.; Kuroda, M.; Kajisaki, K.; Saitoh, M. J.
Org. Chem. 1998, 63, 8183; (d) Ogawa, H.; Amano, M.;
Chihara, T. Chem. Commun. 1998, 495.
3. (a) Oikawa, M.; Wada, A.; Okazaki, F.; Kusumoto, S. J.
Org. Chem. 1996, 61, 4469; (b) Maezaki, N.; Sakamoto,
A.; Nagahashi, N.; Soejima, M.; Li, Y.-X.; Imamura, T.;
Kojima, N.; Ohishi, H.; Sakaguchi, K.-i.; Iwata, C.;
Tanaka, T. J. Org. Chem. 2000, 65, 3284; (c) Zhu, P. C.;
Lin, J.; Pittman, C. U., Jr. J. Org. Chem. 1995, 60, 5729;
(d) Choudary, B. M.; Reddy, P. N. Synlett 1995, 959.
4. Clarke, P. A.; Holton, R. A.; Kayaleh, N. E. Tetrahedron
Lett. 2000, 41, 2687.
In summary, however, it has been shown that it is now
possible to mono-acylate meso- and C2-symmetric 1,3-
and 1,4-diols directly in good to excellent yields. The
methodology outlined will be of use to those in the
synthetic chemistry community who seek to differenti-
ate identical hydroxyl functionalities.
Representative experimental procedure
Acetic anhydride (0.8 ml, 8.6 mmol) was added to a
stirred solution of diol (0.86 mmol) and CeCl3 (21 mg,
0.086 mmol) in THF (2.4 ml) under a nitrogen atmo-
sphere. When judged complete by TLC (EtOAc–hexane
mixtures) the reaction was diluted with Et2O and
washed twice with a saturated NaHCO3 solution and
once with brine. The organics were dried (MgSO4) and
the solvent was evaporated to yield the crude product.
The product was purified via flash column chromatog-
raphy (EtOAc–hexane mixtures) to provide the mono-
acetate in good to excellent yields (Tables 2 and 3).
5. For the selective acylation of primary over secondary
hydroxyl groups using lanthanide salts see: Bianco, A.;
Brufani, M.; Melchioni, C.; Romagnoli, P. Tetrahedron
Lett. 1997, 38, 651.
6. For an example of an esterification using AcOH as the
acyl donor and a Yb(OTf3) catalyst see: Barrett, A. G. M.;
Braddock, D. C. Chem. Commun. 1997, 351.
7. For the selective acylation of 10-DAB using lanthanide
salts, see: (a) Holton, R. A.; Zhang, Z.; Clarke, P. A.;
Nadizadeh, H.; Procter, D. J. Tetrahedron Lett. 1998, 39,
2883; (b) Damen, E. W. P.; Braamer, L.; Scheeren, H. W.
Tetrahedron Lett. 1998, 39, 6081.
8. We have found that the rate of the acylation reaction
increases as the atomic mass of the lanthanide used
increases: YbCl3 provides for the fastest reaction while
CeCl3 for the slowest.
Acknowledgements
I thank Mr. S. J. Bird for preliminary work and the
EPSRC and GlaxoSmithKline for financial support.