Studies on the hydrogenolysis of benzyl ethers

Article abstract of DOI:10.1016/j.tetlet.2006.05.109

Selective hydrogenolysis of benzyl ethers can be achieved by the appropriate choice of the catalyst, solvent, and concentration, which facilitates the design of O-benzyl based protecting group strategies for the synthesis of natural products.

Full text of DOI:10.1016/j.tetlet.2006.05.109

Tetrahedron Letters 47 (2006) 5815–5818
Studies on the hydrogenolysis of benzyl ethers
*
*
´
`
`
Enric Llacer, Pedro Romea and Felix Urpı
`
´ ´
Departament de Quı´mica Organica, Universitat de Barcelona, Martı i Franques 1–11, 08028 Barcelona, Catalonia, Spain
Received 6 April 2006; accepted 17 May 2006
Available online 16 June 2006
Abstract—Selective hydrogenolysis of benzyl ethers can be achieved by the appropriate choice of the catalyst, solvent, and concen-
tration, which facilitates the design of O-benzyl based protecting group strategies for the synthesis of natural products.
Ó 2006 Elsevier Ltd. All rights reserved.
Benzyl ether is one of the most widely used protecting
groups of alcohols since it is easily installed, stable to
a broad scope of reaction conditions and readily re-
moved in the presence of many common functionalities
through hydrogenation, dissolving metal reduction, oxi-
dative treatments, or Lewis acid mediated cleavage.1
Among these methodologies, catalytic hydrogenolysis
offers the mildest option for removing benzyl protecting
groups because of its technical simplicity and the quanti-
tative yields usually achieved without the requirement of
further purification of the reaction mixture. In spite of
these advantages, studies on the selective hydrogenolysis
of benzyl ethers are scarce and, more importantly, there
is a lack of information about the experimental condi-
tions that permit the removal of true benzyl protecting
groups of alcohols when other oxygenated benzylic posi-
tions present in the same structure have to remain
intact.2–4 These limitations have traditionally been
addressed through the development of more elaborate
benzyl-like protecting groups. However, the usefulness
of some of these is compromised by a loss of robustness
that often introduces additional synthetic difficulties.5–7
Therefore, it would be worth gaining insight into the
experimental conditions required for the selective cleav-
age of benzyl protecting group of alcohols in the pres-
ence of other ethers placed at a benzylic position.
We came across the abovementioned limitations during
the preparation of the C9–C21 fragment of debromo-
aplysiatoxin and oscillatoxin D.8 As represented in
Scheme 1, we envisaged that protection of C11 and
C20 alcohols as benzyl ethers would fulfil the stability
requirements expected along the synthesis in such a
way that they could be cleanly and simultaneously
removed by catalytic hydrogenolysis in a late step.9
Surprisingly, this approach earlier proved troublesome.
Indeed, preliminary trials of hydrogenolysis on the
C9–C21 advanced intermediate 1 shown in Scheme 2
in the presence of 10% Pd/C in methanol (1 atm, rt,
O
O
O
OBn
OMe
O
O
OMe
O
OMe
OH
O
9
9
15
21
X
O
O
9
15
21
O
O
15
21
O
O
OBn
Benzyl–protected C9–C21 fragment
OH
OH
OH
Debromoaplysiatoxin
Oscillatoxin D
Scheme 1.
Keywords: Protecting groups; Benzyl ether; Selective hydrogenolysis.
*
Corresponding authors. Tel.: +34 93 402 12 47/403 91 06; fax: +34 93 339 78 78 (F.U.); e-mail addresses: pedro.romea@ub.edu; felix.urpi@ub.edu
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.109

`
E. Llacer et al. / Tetrahedron Letters 47 (2006) 5815–5818
5816
However, it was difficult to anticipate the behavior of an
O
OBn
OMe
aliphatic benzyl ether and it was advisable to look for
more selective and reliable conditions. Therefore, other
catalysts were next surveyed. The results are summa-
rized in Table 2.12
N
15
MeO
1
OBn
H2, 10% Pd/C
Pearlman’s catalyst, Pd(OH)2/C, turned out to be more
active than 10% Pd/C and led to the overreduced diol 5
both in methanol and toluene more easily (compare
entries 1 and 2 in Table 2 to entries 3 and 5 in Table 1,
respectively). Interestingly, this side reaction was com-
pletely suppressed using rhodium or Raney nickel cata-
lysts (see entries 3–5 and 6–8 in Table 2, respectively).
Indeed, both catalysts were able to promote the depro-
tection of C20 phenol without affecting methyl ether at
the C15 benzylic position in such a way that hydrogeno-
lyses of 3 in their presence afforded the desired alcohol 4
in quantitative yields irrespective of the reaction times
(see entries 3–4 and 6–8 in Table 2). It is worth mention-
ing that Raney nickel removed the aromatic benzyl pro-
tecting group of 3 without hydrogen atmosphere (see
entries 6 and 7 in Table 2), which undoubtedly increased
its appeal for such reactions.13 Having found the exper-
imental conditions that allowed for deprotection of C20
benzyl ether without affecting the integrity of C15
methyl ether, we focused back our attention on the
Weinreb amide 1. Unexpectedly, application of the
O
OH
N
MeO
15
2
OH
Scheme 2.
3 h) produced quantitatively the corresponding diol 2
lacking methyl ether at C15.10
It was clear that the overreduction of the C15 benzylic
position jeopardized the whole synthetic approach and
more selective hydrogenolysis conditions were required.
Thus, we first carefully examined the hydrogenolysis of
alcohol 3, a precursor of Weinreb amide 1, catalyzed
by 10% Pd/C in a wide range of solvents. Keeping in
mind that the experimental conditions evolved from
such a model should be applied to tiny amounts of a
structurally more complex substrate, we restricted our
analysis to hydrogenolysis (1 atm, rt) of highly diluted
solutions (0.035 M) in the presence of moderately
important amounts of catalyst (25 mol % of Pd). The
results are summarized in Table 1.
Table 2. Hydrogenolysis of 3 employing other catalysts than 10%
Pd/C
Entry Catalyst
Solvent
Time (h) 3a
4a
5a
(%) (%) (%)
1
2
3
4
Pd(OH)2/C MeOH
Pd(OH)2/C Toluene
1
1
4
18
5
1
—
40
—
—
>95
10
90 10
Analysis of the former data clearly proved that the
kinetics of the deprotection of 3 was highly solvent
dependent (EtOAc ꢀ hexane > methanol > toluene).11
Additionally, concentration played a crucial role (com-
pare entries 3–4 and 6–7) and must be carefully consid-
ered in order to minimize the formation of diol 5. Thus,
undesired overreduction of C15 methyl ether could only
be avoided when hydrogenolysis in the presence of 10%
Pd/C was carried out in toluene at low concentrations.
55
100
100
<5
5
—
—
—
—
—
—
Rh/Al2O3
Rh/Al2O3
Rh/Al2O3
Ra Ni
MeOH
EtOH
Toluene
MeOH
MeOH
MeOH
5
6b
7b
8
90
Ra Ni
2.5
24
—
—
100
100
Ra Ni
a Determined by analysis of the 1H NMR of the reaction mixture.
b Without H2 atmosphere.
Table 1. Hydrogenolysis of 3 in the presence of 10% Pd/C
OMe
OMe
H2, 10% Pd/C
+
HO
HO
HO
15
15
OBn
OH
OH
4a (%)
3
4
5
Entry
Solvent
Time (h)
3a (%)
5a (%)
1
2
3
4b
5
EtOAc
Hexane
MeOH
MeOH
Toluene
Toluene
Toluene
1
1
1
1
1
5
5
—
—
50
—
80
10
60
—
25
50
85
20
70
40
100
75
—
15
—
20
—
6
7c
a Determined by analysis of the 1H NMR of the reaction mixture.
b Concentration 0.1 M.
c Concentration 0.02 M.

`
E. Llacer et al. / Tetrahedron Letters 47 (2006) 5815–5818
5817
O
OH
OMe
O
OBn
OMe
O
O
O
OMe
OMe
H2, 10% Pd/C
toluene
H2, Rh/Al2O3
Me
Me
Me
N
N
N
MeO
MeO
MeO
8
OBn
6
OH
OH
1
OBn
H2, Rh/Al2O3
MeOH
O
OH
OMe
OBn
Ra Ni
Me
N
N
H
95%
MeO
9
OH
7
Scheme 3.
mildest conditions based on rhodium catalyst to 1 re-
moved the C20 benzyl group but produced the reduction
of the aromatic ring of the protecting benzyl group of
C11 secondary alcohol, leading to ether 6. Similarly,
Raney nickel easily promoted the cleavage of C20 aro-
matic benzyl ether without affecting sensitive methyl
ether at C15 but removal of benzyl protecting group
of secondary alcohol proved to be too slow. In this case,
long reaction times were required and competitive
reduction of N–OMe bond took place instead, leading
to N-methyl amide 7 as a major component of the reac-
tion mixture (see Scheme 3). These unsatisfactory results
suggest that both catalysts are unable to cleave benzyl
protecting groups from sterically hindered positions as
C11 secondary alcohol or C15 disubstituted methyl
ether.14,15
BQU2002-01514), and from the Generalitat de Catalunya
(2001SGR00051 and 2005SGR00584).
References and notes
1. (a) Greene, T. W.; Wuts, R. G. M. Protective Groups in
Organic Synthesis; John Wiley & Sons: New York, 1999; (b)
Kocienski, P. J. Protecting Groups; Thieme: Stuttgart, 1994.
2. Surprisingly and to the best of our knowledge, this
problem has been rarely identified. For instance, see: (a)
Bolitt, V.; Mioskowski, C.; Kollah, R. O.; Manna, S.;
Rajapaksa, D.; Falck, J. R. J. Am. Chem. Soc. 1991, 113,
6320–6321; (b) Jennings, M. P.; Clemens, R. T. Tetra-
hedron Lett. 2005, 46, 2021–2024.
3. For a recent review on ether dealkylation, see: Weissman,
S. A.; Zewge, D. Tetrahedron 2005, 61, 7833–7863.
4. For a recent evaluation of selectivity on the hydrogeno-
lysis of benzyl amines, see: Kanai, M.; Yasumoto, M.;
Kuriyama, Y.; Inomiya, K.; Katsuhara, Y.; Higashiyama,
I. Org. Lett. 2003, 5, 1007–1010.
5. For a classical study on the selective deprotection of
benzyl-like protecting groups for alcohols, see: Horita, K.;
Yoshioka, T.; Tanaka, T.; Oikawa, Y.; Yonemitsu, O.
Tetrahedron 1986, 42, 3021–3028.
6. For a recent survey on debenzylation conditions, see:
Trost, B. M.; Wrobleski, S. T.; Chisholm, J. D.; Harring-
ton, P. E.; Jung, M. J. Am. Chem. Soc. 2005, 127, 13589–
13597.
Having failed the mildest options, we turned our atten-
tion to the catalyst first evaluated, 10% Pd/C. After
careful optimization, it was established that cleavage
of benzyl ether from secondary aliphatic C11 alcohol
took place faster than of the C20 aromatic one.16 Then,
hydrogenolysis of 1 in toluene at low concentrations
using 10% Pd/C (1 atm, rt, 3 h) selectively deprotected
the C11 aliphatic alcohol to give hydroxy amide 8,
whose benzyl protecting group of C20 phenol was finally
removed in the presence of Rh/Al2O3. This two-step
hydrogenolysis sequence afforded the desired dihydroxy
amide 9 in 95% yield (see Scheme 3).
7. For other O–Bn-like protecting groups recently reported,
see: (a) Falck, J. R.; Barma, D. K.; Baati, R.; Mioskowski,
C. Angew. Chem., Int. Ed. 2001, 40, 1281–1283; (b)
Sharma, G. V. M.; Rakesh Tetrahedron Lett. 2001, 42,
5571–5573.
In summary, the design of benzyl-based protecting
group strategies should consider that other oxygenated
benzylic positions can be affected. Keeping in mind this
potential drawback, careful evaluation of the catalyst,
solvent, and concentration allows for selective deprotec-
tive sequences. In our case, hydrogenolysis of aliphatic
and aromatic O-benzyl protecting groups in the presence
of benzylic methyl ether has been carried out through a
suitable two-step hydrogenolysis sequence using 10%
Pd/C and Rh/Al2O3 as catalysts consecutively.
8. See the following communication. Tetrahedron Lett. 2006,
47, doi:10.1016/j.tetlet.2006.05.187.
9. For previous reports on hydrogenolysis of debromoaply-
siatoxin derivatives and oscillatoxin D, see: (a) Park, P.-u.;
Broka, C. A.; Johnson, B. F.; Kishi, Y. J. Am. Chem. Soc.
1987, 109, 6205–6207; (b) Toshima, H.; Suzuki, T.;
Nishiyama, S.; Yamamura, S. Tetrahedron Lett. 1989,
30, 6725–6728; (c) Toshima, H.; Goto, T.; Ichihara, A.
Tetrahedron Lett. 1995, 36, 3373–3374.
10. Methyl ether at C15 could be considered as methanol
protected with a complex benzyl-like group.
11. Solvent effect (THF > hexanol ꢀ hexane > methanol >
toluene) on the catalytic hydrogenation of (PhCH2)2O
has been previously reported: Hawker, S.; Bhatti, M. A.;
Griffin, K. G. Chim. Oggi 1992, 10, 49–51; See also:
Perosa, A.; Tundo, P.; Zinovyev, S. Green Chem. 2002, 4,
492–494.
Acknowledgments
We thank financial support from the Ministerio de
´
Ciencia y Tecnologıa and Fondos FEDER (Grant

`
E. Llacer et al. / Tetrahedron Letters 47 (2006) 5815–5818
5818
12. Other reductive methodologies (Pd-mediated transfer
hydrogenation or LDDP) were also tested ineffectively.
13. For a successful application of Ra Ni to the cleavage of
C20 benzyl ether in a late step of the synthesis of
oscillatoxin D, see Ref. 9c.
14. Similar results were obtained in the case of analogous
methyl ester. For instance, its hydrogenolysis in the
presence of Rh/Al2O3 afforded quantitatively the corre-
sponding cyclohexylmethyl ether.
15. It has been reported that hydrogenation (1 atm) of
benzyl protecting groups from primary alcohols in the
presence of Pt/C and Rh/Al2O3 can produce partial
aromatic ring reduction: Oikawa, Y.; Tanaka, T.; Horita,
K.; Yonemitsu, O. Tetrahedron Lett. 1984, 25, 5397–
5400.
16. Selective Pd/C-mediated hydrogenolysis of a benzyl ester
in the presence of an aromatic benzyl ether has been
reported: Sajiki, H.; Hattori, K.; Hirota, K. J. Org. Chem.
1998, 63, 7990–7992.
O
OBn
OMe
O
O
OMe
H2
MeO
MeO
Rh/Al2O3
100%
OBn
OH