Solid-supported acids for debenzylation of aryl benzyl ethers

Article abstract of DOI:10.1021/jo052337v

Solid-supported acids have been investigated for aromatic debenzylation reactions. Stoichiometric amounts of solid-supported acids in refluxing toluene with or without 4 equiv of methanol effectively provided the desired aromatic debenzylation products of various systems in moderate to excellent yields (up to 98%).

Full text of DOI:10.1021/jo052337v

bound materials can be economically recycled. Their use in
organic synthesis simplifies the reaction procedures, as these
reactions generally require simple filtration to remove the
polymer-bound materials without the necessity of aqueous
workup, resulting in less aqueous and organic waste than
otherwise generated from the workup procedure. Consequently,
development and use of polymer-supported reagents in various
reactions are crucial for practicing green chemistry. In addition,
reactions employing solid-supported reagents can be analytically
monitored in a manner similar to those using other conventional
reagents.
Solid-Supported Acids for Debenzylation of Aryl
Benzyl Ethers
Thaninee Petchmanee,† Poonsakdi Ploypradith,*,† and
Somsak Ruchirawat†,‡
Laboratory of Medicinal Chemistry, Chulabhorn Research
Institute, VipaVadee-Rangsit Highway,
Bangkok 10210, Thailand, and Program on Research and
DeVelopment of Synthetic Drugs, Institute of Science and
Technology for Research and DeVelopment, Mahidol UniVersity,
Salaya Campus, Nakhon Pathom, Thailand
A number of protecting groups for phenol have been
developed and used extensively in organic synthesis for many
years.2 The benzyl group and its derivatives have been employed
as hydroxy-protecting groups for their ease of preparation and
removal and their chemical stability toward a number of reaction
conditions. Their versatility makes them ideal for protecting
group manipulation. In general, conditions employed for
aromatic debenzylation fall into the categories of (1) catalytic
hydrogenolysis3 (H2 in the presence of Pd/C or Raney Ni), (2)
reductive cleavage4 (Na or Ca in NH3), (3) iodide-mediated
debenzylation5 (TMSI or NaI in the presence of BF3‚Et2O), (4)
metal-mediated debenzylation6 (TiCl3 in the presence of Mg),
(5) Lewis-acid-mediated debenzylation7 ((a) AlCl3 in the
presence of N,N-dimethylaniline or (b) BF3‚Et2O in the presence
of (1) ethanethiol or (2) 2-bromo-1,3,2-benzodioxaborole), and
(6) acid-mediated cleavage8 ((a) CF3CO2H in the presence of
(1) thioanisole and/or CH3OSO3F or CF3OSO3F and (2)
pentamethylbenzene or (b) concentrated HCl in refluxing EtOH).
Each method offers its own advantages but possesses some
limitations. As a result, there still exists a need for developing
new methods for aromatic debenzylation.
poonsakdi@cri.or.th.
ReceiVed NoVember 10, 2005
Solid-supported acids have been investigated for aromatic
debenzylation reactions. Stoichiometric amounts of solid-
supported acids in refluxing toluene with or without 4 equiv
of methanol effectively provided the desired aromatic de-
benzylation products of various systems in moderate to
excellent yields (up to 98%).
Our interest in developing a new method for aromatic
debenzylation arises in part from our synthetic program for
lamellarins (1) whose structures contain substituted polyoxy-
genated aromatics on their periphery. Two of our synthetic
approaches9 for lamellarins require substituted benzyl-protected
aromatic synthons 2-5 (Scheme 1). In addition, we have
The development and utility of solid-supported reagents in
organic synthesis have been well documented over a number
of years as they have found wide application by serving as
reagents, catalysts, or scavengers.1 In principle, these polymer-
(2) (a) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic
Synthesis, 3rd ed.; John Wiley & Sons: New York, 1999; pp 246-286. (b)
Kocienski, P. J. Protecting Groups, 3rd ed.; Georg Thieme Verlag: Stuttgart,
2003; Chapter 4.
† Chulabhorn Research Institute.
‡ Mahidol University, Salaya Campus.
(1) (a) Ley, S. V.; Baxendale, I. R.; Bream, R. N.; Jackson, P. S.; Leach,
A. G.; Longbottom, D. A.; Nesi, M.; Scott, J. S.; Storer, R. I.; Taylor, S.
J. J. Chem. Soc., Perkin Trans. 1 2000, 3815-4195. (b) Flowers, R. A.;
Xu, X.; Timmons, C.; Li, G. Eur. J. Org. Chem. 2004, 2988-2990. (c)
Pennington, T. E.; Kardiman, C.; Hutton, C. A. Tetrahedron Lett. 2004,
45, 6657-6660. (d) Chiang, G. C. H.; Olsson, T. Org. Lett. 2004, 6, 3079-
3082. (e) Tashino, Y.; Togo, H. Synlett 2004, 2010-2012. (f) Jaunzems,
J.; Kashin, D.; Scho¨nberger, A.; Kirschning, A. Eur. J. Org. Chem. 2004,
3435-3446. (g) Petricci, E.; Mugnaini, C.; Radi, M.; Correlli, F.; Botta,
M. J. Org. Chem. 2004, 69, 7880-7887. (h) Donati, D.; Morelli, C.;
Porcheddu, A.; Taddei, M. J. Org. Chem. 2004, 69, 9316-9318. (i) Li, X.;
Wu, X.; Chen, W.; Hancock, F. E.; King, F.; Xiao, J. Org. Lett. 2004, 6,
3321-3324. (j) Deprele, S.; Montchamp, J. Org. Lett. 2004, 6, 3805-3808.
(k) Weno, M.; Togo, H. Synthesis 2004, 2673-2677. (l) Ballini, R.; Fiorini,
D.; Palmieri, A. Tetrahedron Lett. 2004, 45, 7027-7029. (m) Marinescu,
L. G.; Pedersen, C. M.; Bols, M. Tetrahedron 2005, 61, 123-127. (n)
Dueymes, C.; Scho¨nberger, A.; Adamo, I.; Navarro, A.-E.; Meyer, A.;
Lange, M.; Imbach, J.-L.; Link, F.; Morvan, F.; Vasseur, J.-J. Org. Lett.
2005, 7, 3485-3488. (o) Yadav, J. S.; Reddy, B. V. S.; Vishnumurthy, P.
Tetrahedron Lett. 2005, 46, 1311-1313. (p) Chretien, J.-M.; Zammattio,
F.; Le Grognec, E.; Paris, M.; Cahingt, B.; Montavon, G.; Quintard, J.-P.
J. Org. Chem. 2005, 70, 2870-2873. (q) Maki, T.; Ishihara, K.; Yamamoto,
H. Org. Lett. 2005, 7, 5043-5046.
(3) (a) Bu¨chi, G.; Weinreb, S. M. J. Am. Chem. Soc. 1971, 93, 746-
752. (b) Deng, W.-P.; Zhong, M.; Guo, X.-C.; Kende, A. S. J. Org. Chem.
2003, 68, 7422-7427.
(4) (a) Loev, B.; Dawson, C. R. J. Am. Chem. Soc. 1956, 78, 6095-
6098. (b) Hwu, J. R.; Wein, Y. S.; Leu, Y.-J. J. Org. Chem. 1996, 61,
1493-1499.
(5) (a) Lott, R. S.; Chauhan, V. S.; Stammer, C. H. J. Chem. Soc., Chem.
Commun. 1979, 495-497. (b) Vankar, Y. D.; Rao, C. T. J. Chem. Res.,
Synop. 1985, 232-233.
(6) Kadam, S. M.; Nayak, S. K.; Banerji, A. Tetrahedron Lett. 1992,
33, 5129-5132.
(7) (a) Akiyama, T.; Hirofuji, H.; Ozaki, S. Tetrahedron Lett. 1991, 32,
1321-1324. (b) Fuji, K.; Ichikawa, K.; Node, M.; Fujita, E. J. Org. Chem.
1979, 44, 1661-1664. (c) King, P. F.; Stroud, S. G. Tetrahedron Lett. 1985,
26, 1415-1418.
(8) For CF3SO3H, see: (a) Kiso, Y.; Isawa, H.; Kitagawa, K.; Akita, T.
Chem. Pharm. Bull. 1978, 26, 2562-2564. (b) Kiso, Y.; Ukawa, K.;
Nakamura, S.; Ito, K.; Akita, T. Chem. Pharm. Bull. 1980, 28, 673-676.
(c) Yoshino, H.; Tsuji, M.; Kodama, M.; Komeda, K.; Niikawa, N.; Tanase,
T.; Asakawa, N.; Nose, K.; Yamatsu, K. Chem. Pharm. Bull. 1990, 38,
1735-1737. For concentrated HCl in refluxing EtOH, see: (d) Sae-Lao,
P.; Kittakoop, P.; Rajviroongit, S. Tetrahedron Lett. 2006, 47, 345-348.
10.1021/jo052337v CCC: $33.50 © 2006 American Chemical Society
Published on Web 03/04/2006
2892
J. Org. Chem. 2006, 71, 2892-2895

SCHEME 1. Substituted Benzyl Ethers of Aromatic
Synthons 2-5
SCHEME 2. Aromatic Debenzylationa of Benzyl Ethers 6
and 7
a Reagents and conditions: (a) Amberlyst-15 (3 equiv), toluene, 110 °C,
2 h, 57% (8), 75% (9), 23% (10).
SCHEME 3. Proposed Mechanism of Acid-Mediated
Aromatic Debenzylation
investigated the utility of solid-supported reagents in two
synthetic approaches developed by our group. We have shown
that three polymer-supported reagents could be employed
sequentially for the synthesis of either natural or unnatural
lamellarins.9d During this work, we have encountered a novel
Amberlyst-15-mediated aromatic debenzylation reaction. We
now wish to report the results of our investigation of this
aromatic debenzylation reaction employing solid-supported
acids.
Because the basic units required for the synthesis of lamel-
larins are the substituted benzaldehyde derivatives whose
aromatic hydroxy groups were protected as benzyl, the systems
we first investigated were the benzyl ethers of isovanillin 6 and
vanillin 7. Their preparation employed a standard benzylation
reaction with K2CO3 serving as base in refluxing acetone and
benzyl bromide as the benzylating agent.10 Such reaction
conditions provided the corresponding benzyl ether products
in excellent yields (86-98%). With the starting materials in
hand, we first explored aromatic debenzylation using 3 equiv
of Amberlyst-15, a sulfonic acid on polymer support, in toluene
at 110 °C for 2 h, which were the conditions previously
employed in the synthesis of the lamellarin framework.9d
Isovanillin and vanillin were obtained from 6 and 7 in 57%
and 75% yields, respectively. Aromatic debenzylation of 7 was
thus chosen for a more detailed study because of its significantly
higher yield. Examination of the reaction mixture showed that
C-benzylated vanillin 10 was formed in 23% yield. In addition,
Friedel-Crafts-type benzylation of the solvent toluene gave rise
to an inseparable mixture of presumably ortho- and para-
benzylated toluene 11 (Scheme 2).
and type of solid-supported acids (Amberlyst-15, Lewatit,
Dowex 2030, and Dowex 50X), (2) the temperature, (3) the
effect of solvents, and (4) the effect of different benzyl
scavengers (bromoanisole, thioanisole, and methanol) as addi-
tives. It was found that all solid-supported acids utilized in this
study except Dowex 50X were effective in aromatic debenzy-
lation. A stoichiometric amount of solid-supported acid was
required for complete consumption of starting material. Lower-
ing the temperature may improve the product ratios but result
in a longer reaction time. The best solvent was found to be
toluene. Our proposed reaction mechanism, which involves
initial protonation at the oxygen atom followed by C-O
cleavage to generate the corresponding benzylic carbocation 13
along with the desired phenol and quenching of the benzyl cation
with toluene, reasonably suggested that the formation of 10
could partly arise from the intermolecular benzylation of vanillin
under this reaction condition (Scheme 3) and prompted us to
consider the use of benzyl scavengers as additives that could
scavenge the benzyl cation faster than vanillin to suppress the
formation of 10. Among the benzyl scavengers employed in
this study, we found that a mixed solvent system of 4 equiv of
methanol in toluene gave the best result, providing the desired
product vanillin in excellent 98% yield with only trace amounts
of C-benzylated vanillin 10. Larger amounts of methanol (10:1
by volume) resulted in no reaction presumably because of the
competing initial protonation between the substrate and the
methanol. (For full details of optimization for aromatic deben-
zylation of 7, see the Table and the following discussion in the
Supporting Information.)
The parameters we have investigated and optimized for solid-
supported acid in aromatic debenzylation were (1) the amount
(9) (a) Ruchirawat, S.; Mutarapat, T. Tetrahedron Lett. 2001, 42, 1205-
1208. (b) Ploypradith, P.; Jinagleung, W.; Pavaro, C.; Ruchirawat, S.
Tetrahedron Lett. 2003, 44, 1363-1366. (c) Ploypradith, P.; Mahidol, C.;
Sahakitpichan, P.; Wongbundit, S.; Ruchirawat, S. Angew. Chem., Int. Ed.
Engl. 2004, 43, 866-868. (d) Ploypradith, P.; Kagan, R. K.; Ruchirawat,
S. J. Org. Chem. 2005, 70, 5119-5125.
(10) (a) Kotecha, N. R.; Ley, S. V.; Monte´gani, S. Synlett 1992, 395-
398. (b) Mayer, A. A.; Murphy, B. P. Synth. Commun. 1985, 15, 423-
429.
The electronic nature of the benzyl group toward debenzy-
lation was also investigated. Compounds 14-16 (Figure 1) were
J. Org. Chem, Vol. 71, No. 7, 2006 2893

SCHEME 4. Aromatic Debenzylation of Benzyl Ethers 6
and 19
FIGURE 1. Structure of PMB, PNB, and ONB ethers of vanillin 14-
16.
synthesized from the corresponding substituted benzyl halides
in the presence of K2CO3 as base in 97-99% yields.11,12 The
electron-donating p-methoxy group is expected to favorably
stabilize the positive charge on the benzylic carbon, and the
electron-withdrawing o- or p-nitro group would destabilize such
charge on that position. We then anticipated that the p-methoxy
benzyl group (PMB) would undergo a more facile aromatic
debenzylation reaction, whereas a similar reaction for the o-nitro
benzyl (ONB) or p-nitro benzyl (PNB) group would require
harsher conditions.
After some experimentation, we found that aromatic deben-
zylation of PMB ether required 5 equiv of Amberlyst-15 in
toluene at 110 °C for 6 h and gave cleanly the desired product
in 61% yield without any corresponding C-benzylated byprod-
uct.13 In the presence of 4 equiv of methanol, only 2.2 equiv of
Amberlyst-15 was required and the reaction proceeded to give
the desired product 9 in 86% yield without any detectable
amount of the C-benzylated product. The result suggested that
the PMB cation is more stabilized and more selective toward
nucleophilic attack. Under the reaction conditions, the presence
of a PMB toluene adduct and the absence of the C-benzylated
byproduct reflected the more effective quenching of the PMB
cation by toluene than by vanillin. In addition, the improved
yield with the added methanol further confirmed the results
obtained from aromatic debenzylation reactions of benzyl ether
7 and could be attributed to the lower equivalent of the acid
employed, the presence of methanol, or both effects.
It is, however, perplexing to observe the longer reaction time
for cleaving PMB ether in comparison with that for normal
benzyl ether. It is less conceivable to propose that the rate of
C-O cleavage of PMB ether is slower than that of benzyl ether
because the resulting cation of PMB is more stabilized by the
presence of the p-methoxy group. Thus, we propose that the
longer reaction time for cleaving PMB ether is a consequence
of the slower rate of quenching the resulting less reactive PMB
cation by toluene. Our proposed explanation was confirmed by
competitive debenzylation reactions between vanillin benzyl
ether 7 and PMB ether 14.14 In the presence of both ethers in
1.1 equiv of Amberlyst-15 in toluene at 110 °C for 20 h, only
the PMB ether was selectively cleaved (>95% conversion)
whereas benzyl ether remained intact (virtually 100%); no
C-benzylated byproduct was observed. The result confirmed that
the rate of C-O cleavage of PMB ether is indeed faster than
that of benzyl ether. The overall longer reaction time reflected
the time required for the most thermodynamically favorable
pathway which includes initial protonation and quenching of
the resulting cation with an available nucleophile. Excess acid
(2.2 equiv) with similar length of reaction time resulted in
cleavage of the benzyl ether as well.15 With 4 equiv of methanol,
the reaction proceeded to selectively cleave PMB ether 14 in 6
h, providing 9 in quantitative yield without any C-benzylated
byproduct, and 93% yield of benzyl ether 7 was recovered. It
is noteworthy that various acid-mediated conditions, e.g.,
p-TsOH, TFA, and Amberlyst-15, were not effective in cleaving
either ONB or PNB ethers.16
We then applied the optimized reaction conditions for
aromatic debenzylation reactions of other systems (Scheme 4).
With 4 equiv of methanol, isovanillin benzyl ether 6 gave the
desired product 8 in 83% yield along with ortho- and para-C-
benzylated vanillin 17 and 18 in 10% combined yield.17 The
formation of 18 further confirmed the proposed intermolecular
mode of C-benzylation under these reaction conditions. We
found that, even with the optimized reaction conditions,
salicylaldehyde benzyl ether 19 gave the desired product 20 in
only 10% yield along with C-benzylated byproducts.18
We then turned to the nitrostyrene system which is chemically
prone to metal-mediated reduction/hydrogenation frequently and
is similarly employed in hydrogenolysis of aromatic debenzy-
lation (Scheme 5). Nitrostyrene benzyl ethers 21 and 22 were
prepared from the Henry reaction of 8 and 9 with nitromethane.9c
In the absence of methanol, nitrostyrene benzyl ether 21 gave
cleanly the desired product 2319 in 69% yield together with the
(14) The crude reaction mixture was analyzed by 1H NMR. Piperonal
was used as an internal standard, and the amount of the remaining starting
material was determined by relative integration.
(15) Approximately 83% of benzyl ether was consumed.
(16) After some experimentation, we found that ONB and PNB ethers
were effectively cleaved by potassium tert-butoxide in refluxing tert-butyl
alcohol. Vanillin was obtained in 84% and 80% yields from 15 and 16,
respectively. See Supporting Information for details.
(17) Without methanol, the reaction gave 8 in 57% yield as well as 17
and 18 in 12% combined yield.
(18) Compound 19 and 20 are commercially available (Aldrich). The
low yield of salicylaldehyde may arise from its instability under reaction
conditions. In a separate experiment, only 11% of salicylaldehyde was
recoverable when treated with Amberlyst-15 in toluene at 110 °C for 4 h.
The C-benzylated byproducts were generated in approximately 15%
combined yield presumably as a mixture of 3-benzyl- and 5-benzyl-
salicylaldehydes.
(19) (a) Schmidt, M.; Eger, K. Pharmazie 1996, 51, 11-16. (b) Rao, T.
V.; Ravishankar, L.; Trivedi, G. K. Indian J. Chem., Sect. B 1990, 29, 207-
214. (c) Ramirez, F. A.; Burger, A. J. Am. Chem. Soc. 1950, 72, 2781-
2782. In addition, compound 25 is commercially available (Aldrich).
(11) para-Methoxy benzyl ether 14 was synthesized from vanillin (1.5
equiv), p-methoxy benzyl chloride (1.0 equiv), and K2CO3 (1.5 equiv) in
the presence of n-Bu4NI (0.2 equiv) in refluxing acetone, and the para-
and ortho-nitro benzyl ethers 15 and 16 were prepared from vanillin (1.0
equiv), the corresponding nitro benzyl bromide (1.1 equiv), and K2CO3
(1.5 equiv) in refluxing acetone.
(12) (a) Pe´rez, R. A.; Ferna´ndez-Alvarez, E.; Nieto, O.; Piedrafita, F. J.
J. Med. Chem. 1992, 35, 4584-4588. (b) Plourde, G. L.; Spaetzel, R. R.
Molecules 2002, 7, 697-705.
(13) A number of other reaction conditions employing 1.1-3.0 equiv
of Amberlyst-15 in toluene at 110 °C for 3-6 h gave 9 together with some
detectable amount of 10. When 5 equiv was used and the reaction time
was less than 6 h, the reaction did not proceed to completion and the
formation of 10 was observed. Interestingly, p-TsOH (1.5 equiv) under
similar reaction conditions for 3 h effectively cleaved the PMB ether and
only a trace amount of 10 was observed in the crude mixture.
2894 J. Org. Chem., Vol. 71, No. 7, 2006

SCHEME 5. Aromatic Debenzylation Reactiona of
Nitrostyrene Benzyl Ethers 21 and 22
it requires only filtration to remove the polymer from the crude
mixture, resulting in less organic as well as aqueous waste from
the reaction. Taken together, use of solid-supported acids in
the aromatic debenzylation reaction represents an approach
suitable for practicing green chemistry and thus should find wide
applications in protecting group manipulation.
Experimental Section
Aromatic Debenzylation Reaction Using 7 as an Example
(Supporting Information, Entry 6, Table 1). To a solution of
vanillin benzyl ether 7 (0.021 g, 0.087 mmol) in toluene (3.5 mL)
at room temperature was added Amberlyst-15 (0.021 g, 0.096
mmol). The resulting mixture was heated to 80 °C for 6 h. At that
time, the reaction was cooled to room temperature and filtered and
the polymer was washed with DCM and EtOAc. The solution was
then concentrated under reduced pressure. The crude material was
purified by preparative thin-layer chromatography (PTLC) to give
the desired product vanillin 9 (0.012 g, 0.079 mmol, 89%), the
C-benzylated product 10 (0.002 g, 0.008 mmol, 9%), and benzyl
toluene adduct 11.20
a Reagents and conditions: (a) Amberlyst-15, toluene, 110 °C, 69% (23),
19% (24), 63% (25), 9% (26); (b) Amberlyst-15, toluene, methanol (4 equiv),
110 °C, 80% (23), 19% (24), 75% (25), 13% (26).
corresponding ortho-C-benzylated nitrostyrene phenol 24 in 19%
yield. With 4 equiv of methanol, the reaction improved to give
the desired product in 80% yield. Similarly, nitrostyrene 22 gave
the corresponding products 2519 in 63% yield and C-benzylated
nitrostyrene 26 in 9% yield in the absence of methanol. With 4
equiv of methanol, the desired product 25 was obtained in 75%
yield together with 26 in 13% yield. Therefore, these reaction
conditions are chemically compatible with the nitrostyrene
moiety.
In summary, we have developed and demonstrated a novel
use of solid-supported acids for aromatic debenzylation reac-
tions. The optimized reaction conditions were found to be
Amberlyst-15 in toluene at 110 °C with 4 equiv of methanol as
the benzyl scavenger. When compared with other previously
reported methods of acid-mediated aromatic debenzylation, our
optimized conditions offer several advantages. The conditions
are compatible with some functional groups such as the
nitrostyrene which is susceptible to hydrogenation and/or
reduction conditions otherwise frequently employed for aromatic
debenzylation reactions. In addition, the procedure is simple as
Acknowledgment. Financial support from the Thailand
Research Fund (RTA/07/2544 and Senior Research Scholar for
S.R. and DBG4780006 for P.P. and T.P.) and from the Thailand
Toray Science Foundation (TTSF) for P.P. is gratefully ac-
knowledged.
Supporting Information Available: General methods, experi-
1
mental procedures, spectroscopic data, and copies of H and 13C
NMR of all new compounds (10, 11, 17, 18, 24, and 26). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO052337V
(20) For complete characterization of compound 10 and 11, please see
the Supporting Information.
J. Org. Chem, Vol. 71, No. 7, 2006 2895