Poon 2006

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

Mix-and-Heat Benzylation of Alcohols Using a Bench-Stable

Pyridinium Salt
Kevin W. C. Poon and Gregory B. Dudley*
Department of Chemistry and Biochemistry, Florida State UniVersity, Tallahassee, Florida 32306-4390

[email protected]
ReceiVed February 9, 2006

2-Benzyloxy-1-methylpyridinium triflate (1) is a stable, neutral organic salt that converts alcohols into
benzyl ethers upon warming. The synthesis and reactivity of 1 are described herein. Benzylation of a
wide range of alcohols occurs in good to excellent yield.

Introduction
Benzyl ethers are among the most common and important
protecting groups in organic synthesis.1 Like other alkyl ethers,
they are advantageous for their stability to a wide range of
reaction conditions and for the minimal electronic impact that
they impart on the oxygen atom to which they are attached.
For example, benzyl ethers are often employed to establish
chelation control during addition to chiral aldehydes, which
provides selectivity opposite that predicted by the acyclic
Felkin-Anh model and observed with bulky silyl ethers.2
Similarly, benzyl-protected glycosyl donors are “armed” relative
to acylated analogues.3 Among alkyl ethers, benzyl (and
modified arylmethyl) ethers are perhaps the most versatile with FIGURE 1. Standard benzylation protocols and desired objective.
respect to modes of cleavage, which include hydrogenolysis,
oxidation, and acidic decomposition under a range of experi- most popular protocols being (1) the Williamson ether synthesis,
mental protocols (Figure 1).4 an SN2-type reaction between alkali metal alkoxides and benzyl
Relatively harsh conditions are typically required for generat- bromide, and (2) coupling using benzyl trichloroacetimidate,5
ing benzyl ethers from the corresponding alcohol, with the two which is generally promoted by trifluoromethanesulfonic acid
(triflic acid, TfOH).6 Typical benzylation reactions are thus
limited to substrates that tolerate either strongly acidic or basic
(1) (a) Greene, T. W., Wuts, P. G. M. ProtectiVe Groups in Organic
Synthesis, 3rd ed.; John Wiley and Sons: New York, 1999. (b) Kocienski, conditions.7 β-Hydroxy esters, for example, are subject to several
P. J. Protecting Groups, 3rd ed.; Thieme: Stuttgart, 2003. acid- or base-catalyzed reactions, including retro-Aldol, elimina-
(2) Gawley, R. E.; Aubé, J. In Principles of Asymmetric Synthesis; tion, and epimerization of stereogenic centers R- to the carbonyl
Baldwin, J. E., Magnus, P. D., Eds.; Tetrahedron Organic Chemistry Series
14; Pergamon: Tarrytown, NY, 1996; pp 121-134.
group. Benzylation of these ubiquitous intermediates in the
(3) (a) PreparatiVe Carbohydrate Chemistry; Hanessian, S., Ed.; Marcel
Dekker: New York, 1997. (b) Mootoo, D. R.; Konradsson, P.; Udodong, (5) (a) Iversen, T.; Bundle, D. R. J. Chem. Soc., Chem. Commun. 1981,
U.; Fraser-Reid, B. J. Am. Chem. Soc. 1988, 110, 5583-5584. (c) Paulsen; 1240-1241. (b) Wessel, H.-P.; Iversen, T.; Bundle, D. R. J. Chem. Soc.,
H.; Richter, A.; Sinnwell, V.; Stenzel, W. Carbohydr. Res. 1978, 64, 339- Perkin Trans. 1 1985, 2247-2250. (c) Eckenberg, P.; Groth, U.; Huhn, T.;
362. Richter, N.; Schmeck, C. Tetrahedron 1993, 49, 1619-1624.
(4) Recent arylmethyl protecting groups that are cleaved under mild (6) Boa, A. N.; Jenkins, P. R. Benzyl 2,2,2-Trichloroacetimidate. In
conditions: (a) Jobron, L.; Hindsgaul, O. J. Am. Chem. Soc. 1999, 121, Encyclopedia of Reagents for Organic Synthesis; Paquette, L. A., Ed.; John
5835-5836. (b) Plante, O.; Buchwald, S. L.; Seeberger, P. H. J. Am. Chem. Wiley and Sons: New York, 1995; Vol. 1, pp 374-375.
Soc. 2000, 122, 7148-7149. (c) Lam, H.; House, S. E.; Dudley, G. B. (7) p-Methoxybenzyl (PMB) ethers may be formed under mild conditions
Tetrahedron Lett. 2005, 46, 3283-3285. that do not extend to benzylation (ref 1).

10.1021/jo0602773 CCC: $33.50 © 2006 American Chemical Society


Published on Web 04/07/2006 J. Org. Chem. 2006, 71, 3923-3927 3923
Poon and Dudley

synthesis of polyketides and other important compounds can SCHEME 1. Synthesis of 2-Benzyloxy-1-methylpyridinium
be problematic. Selective protection of polyol systems (e.g., Triflate (1)
carbohydrates) can also be complicated by base-catalyzed
migration of esters and silyl ethers and by acid-catalyzed
cleavage of silyl ethers and acetal linkages.
Benzylation of alcohols under mild and nearly neutral
conditions would constitute a significant advance in synthetic
chemistry. A recent review addresses the myriad options for TABLE 1. Initial Optimization
protecting alcohols using mild, convenient, and environmentally
friendly conditions, but no methods for the formation of benzyl
ethers are discussed.8 Silylation and acylation of alcohols can
be accomplished under effectively neutral conditions using
activated reagents that react with the free alcohol.1 Imidazole
and DMAP are frequently employed to activate silyl and acyl
chlorides; conveniently, they are also capable of scavenging any entry acid scavenger equiv of 1 yieldb,c (%)
acid that is produced during the course of the reaction.
1 2,6-lutidine 1.0 43 (57)
Protonation of benzyl trichloroacetimidate provides an activated 2 Hünig’s base 1.0 29 (39)
reagent that reacts with free alcohols, but this mode of activation 3 K2CO3 1.0 68 (93)
precludes neutralization of free acid. In principle, irreversible 4 MgO 1.0 78 (93)
covalent activation (alkylation) of a trichloroacetimidate sur- 5 none 1.0 53 (87)
rogate would enable the formation of benzyl ethers in the 6 MgO 2.0 76 (85)
7 MgO 3.0 87
absence of external base or acid and in the presence of acid
a See the Supporting Information for details. b Values in parentheses refer
scavengers (if desired).
to the calculated yield based on recovered alcohol. c Estimated by 1H NMR
We envisioned that 2-benzyloxypyridine9 could serve as an spectroscopy.
imidate surrogate for benzylation of alcohols. Pyridinium salts
have been employed in esterification reactions, with Mukaiya- Results and Discussion
ma’s 2-chloro-1-methylpyridinium iodide being perhaps the 1. Synthesis and Isolation of Pyridinium Salt 1. The
most popular.10 Conversion of alcohols into thioesters and azides synthesis of 1 is illustrated in Scheme 1. Benzyl alcohol was
using 2-fluoro-1-methylpyridinium tosylate has also been coupled with 2-chloropyridine using a modification of a reported
demonstrated.11 The two pieces of prior knowledge that were procedure9a to afford 2-benzyloxypyridine (5) in high yield. We
most influential in guiding the current work are as follows: (1) then screened a range of alkylating agents and solvents in search
certain 2-alkoxypyridinium bromides decompose to bromoal- of optimal conditions for the irreversible covalent activation of
kanes and pyridones;12 (2) 2-alkoxypyridinium sulfonates do 5. The current best protocol is to add methyl triflate (bp 94-
not proceed spontaneously to alkyl sulfonates.13 We anticipated 99 °C) to an ice-cold solution of 5 in toluene and allow the
that decomposition of 2-alkoxypyridinium sulfonates in the mixture to warm to ambient temperature. A white microcrys-
presence of alcohols would give rise to alkyl ethers and talline solid (1) forms within minutes. Analytically pure 1 (mp
pyridones, and we reported preliminary data in support of this 82-86 °C) can be isolated by filtration or by evaporation of
hypothesis.14 the supernatant under reduced pressure.15 This salt (1) is
Herein we describe in detail our investigation into the remarkably stable. We store it under an argon atmosphere either
synthesis and reactivity of 2-benzyloxy-1-methyl-pyridinium in the refrigerator or on the laboratory benchtop, and the white
triflate (Bn-OPT, 1), which indeed provides benzyl ethers simply crystals of 1 are routinely handled open to the air. No differences
upon warming in the presence of a free alcohol. The overall have been observed between freshly prepared crystals and those
balanced equation for the benzylation of alcohols (2 f 3) is that were prepared 3 months prior.
shown in eq 1. Oxypyridinium triflate 1 may eventually supplant 2. Development and Analysis of the Optimal Benzylation
benzyl trichloroacetimidate for the synthesis of benzyl ethers Protocol. At room temperature, the title reagent is freely soluble
from alcohols. in chlorinated solvents (dichloromethane, chloroform, dichlo-
roethane), partially soluble in ethereal solvents (THF and ether),
and insoluble in aromatic hydrocarbons (benzene, toluene).
Solutions of 1 and 3-phenylpropanol (2a) provided the desired
benzyl ether upon heating. Because of its ability to solvate 1
and its convenient boiling point (83 °C), the initial screening
of reaction conditions was conducted in dichloroethane (DCE).
The first issue that we endeavored to address was the
(8) Sartori, G.; Ballini, R.; Bigi, F.; Bosica, G.; Maggi, R.; Righi, P.
presumed mild acidity of hydroxypyridinium triflate 4 (Table
Chem. ReV. 2004, 104, 199-250.
(9) (a) Serio Duggan, A. J.; Grabowski, E. J. J.; Russ, W. K. Synthesis (12) Joshi, R. A.; Ravindranathan, T. Ind. J. Chem. Sect. B 1984, 23,
1980, 573-575. (b) Cherng, Y.-J. Tetrahedron 2002, 58, 4931-4935. 260-262.
(10) (a) Armstrong, A. 2-Chloro-1-methylpyridinium Iodide. In Ency- (13) (a) Beattie, D. E.; Crossley, R.; Dickinson, K. H.; Dover: G. M.
clopedia of Reagents for Organic Synthesis; Paquette, L. A., Ed.; John Wiley Eur. J. Med. Chem. 1983, 18, 277-285. (b) Kornblum, N.; Coffey, G. P.
and Sons: New York, 1995; Vol. 2, pp 1174-1175. (b) Mukaiyama, T. J. Org. Chem. 1966, 31, 3449-3451.
Angew. Chem., Int. Ed. Engl. 1979, 18, 707-808. (14) Poon, K. W. C.; House, S. E.; Dudley, G. B. Synlett 2005, 3142-
(11) See ref 10b and: (a) Hojo, K.; Yoshino, H.; Mukaiyama, T. Chem. 3144.
Lett. 1977, 437-440. (b) Hojo, K.; Kobayashi, S.; Soai, K.; Ikeda, S.; (15) Recrystallization of 1 from THF has no discernible effect on its
Mukaiyama, T. Chem. Lett. 1977, 635-636. properties.

3924 J. Org. Chem., Vol. 71, No. 10, 2006


Mix-and-Heat Benzylation of Alcohols

SCHEME 2. SN1 vs SN2 Mechanistic Observations

FIGURE 2. Observed byproducts.


TABLE 2. Screening for Optimal Solvent

entry solvent yielda (%)


1 1,2-dichloroethane (DCE) 67
2 nitromethane low
3 acetonitrile -
4 N-methyl-2-pyrrolidinone (NMP) -
5 toluene 91
6 benzene 93
7 chlorobenzene >95
8 benzotrifluoride (PhCF3) >95
a Estimated by 1HNMR spectroscopy.

1). Among the various acid scavengers that we evaluated,


heterogeneous inorganic salts were most compatible with the
desired benzylation reaction (entries 3-5). Soluble amines epimerization detectable by chiral HPLC analysis.17 Benzyl ether
seemed to interfere with the coupling reaction (entries 1 and 3e was easily separated from Bn2O by chromatography on silica
2), and it was not clear if external amine bases would present gel. A series of primary and secondary alcohols were benzylated
any advantage in terms of moderating the potential acidity of under similar conditions and with similar efficiencies (70-76%
pyridinium 4. Based on these results and a quick cost analysis, yield), as described in our preliminary report.14
magnesium oxide (MgO) emerged as our preferred choice, and Despite limited solubility, mixtures of 1 in many solvents
so MgO was routinely included in all subsequent experiments. became homogeneous upon warming, especially as the tem-
In addition to the desired benzyl ether, two byproducts were peratures approached the melting point of 1 (82-86 °C).
observed in the crude product mixture: 1-methyl-2-pyridone Toluene emerged as a promising choice in small-scale explor-
(6) and dibenzyl ether (Bn2O, 7) (Figure 2). Pyridone 6, the atory experiments. Therefore, we screened a range of aromatic
conjugate base of hydroxypyridinium 4, is the expected byprod- solvents (2b f 3b, Table 2). Yields improved significantly in
uct of the benzylation reactions using 1. Pyridone 6 is freely aromatic hydrocarbon solvents relative to dichloroethane (>90%
water-soluble and easily removed by aqueous extraction. The vs 67%).
source of Bn2O is not clear. It may derive from reaction of 1 Reactions conducted in toluene (and, to a lesser extent,
with MgO, although small amounts of 7 were also observed benzene and chlorobenzene) gave rise to trace amounts of
during control experiments that did not include MgO. Adventi- benzylated solvent molecules (Scheme 2, vide infra). No such
tious moisture may be partly responsible for the formation of
7. Because dibenzyl ether is unlikely to interfere with most
benzylation reactions, we do not consider it to be a serious (16) Widmer, U. Synthesis 1987, 568-570.
(17) In response to a reviewer’s suggestion for documentation of the
concern. Nonetheless, it was difficult to separate 7 from many advantage of 1 over benzyl trichloroacetimidate, an additional experiment
of the alkyl benzyl ethers generated during the course of our was conducted using trimethylsilylethanol (13) as a test substrate. Note that
investigations. 13 is subject to Peterson elimination under acidic (or basic) conditions;
see: Ager, D. J. Org. React. 1990, 38, 1. Benzylation of trimethylsilylethanol
A crucial efficacy test for 1 was the benzylation of chiral (13 f 14) has not been reported previously. Reaction of 13 with 1 proceeded
β-hydroxy ester 2e (eq 2). Benzyl ethers derived from such to complete conversion with no evidence of decomposition, whereas a
similar experiment using benzyl trichloroacetimidate yielded no evidence
of the desired product (14). See the Supporting Information for 1H NMR
spectra of the crude product mixtures after aqueous workup.

chiral alcohols are difficult to obtain under Williamson ether


conditions because of the potential both for β-elimination and/
or for epimerization of the labile stereogenic center R- to the
ester. Attempts at effecting the benzylation of 2e under
Willamson ether conditions were unsuccessful.16 Benzylation
using 1 proceeded efficiently (2e f 3e) with no evidence of

J. Org. Chem, Vol. 71, No. 10, 2006 3925


Poon and Dudley

TABLE 3. Scope and Limitations

a Yields are estimated by 1H NMR spectroscopy, unless otherwise indicated. b Reagent 1 stored for 3 months at room temperature before use. c Not

determined; see ref 19. d Isolated yield of pure product. e Unreacted 2k also observed in the crude product mixture.

products were observed from reactions conducted in benzotri- Having identified our preferred solvent, acid scavenger, and
fluoride (R,R,R-trifluorotoluene, PhCF3). time and temperature, we were ready to probe the scope and
In addition to being an excellent solvent for the present limitations of what we consider to be mild and effectively neutral
benzylation reactions, benzotrifluoride is low-cost, moderately benzylation conditions. The tolerance of this protocol for
volatile (bp 100-103 °C), and highly regarded as an environ- sensitive functionality will be determined in due course, but
mentally friendly alternative to chlorinated solvents. Benzotri- for an initial data point we tested our benzylation reaction in
fluoride is our choice of solvent for the benzylation reactions, the presence of a primary silyl ether (eq 3).18 The desired benzyl
although Table 2 indicates that other aromatic hydrocarbons are ether (3a) was obtained in excellent yield, and silyl ether 8 was
also suitable. recovered unchanged.17
3926 J. Org. Chem., Vol. 71, No. 10, 2006
Mix-and-Heat Benzylation of Alcohols

mixture was then cooled to room temperature and partitioned


between ethyl acetate (20 mL) and water (10 mL). The organics
were washed (brine), dried (Na2SO4), filtered, concentrated under
vacuum, and purified on silica gel (elution with 100:1 hexane/
EtOAc) to provide 3.28 g of 5 (96% yield) as a yellow liquid.
2-Benzyloxy-1-methylpyridinium Triflate (1). To a cold (0 °C)
solution of 2-benzyloxypyridine (5) (100 mg, 0.54 mmol) in toluene
(0.540 mL) was added methyl trifluoromethanesulfonate (64 µL,
0.57 mmol). The mixture was allowed to warm to room temperature,
which resulted in the formation of a white crystalline precipitate.
After 40 min, the volatiles were removed in vacuo, providing 0.172
g (91% yield) of 1 as a white microcrystalline solid, mp 82-86
3. Mix-and-Heat Benzylation of Alcohols: Scope and °C. A similar large-scale experiment afforded 6.52 g (86% yield)
Limitations. Table 3 illustrates the benzylation reactions of of 1 as a white solid, which was collected by filtration of the crude
representative alcohols under our preferred conditions. Primary reaction mixture through a fritted glass funnel, followed by drying
(entries 1-6) and secondary (entries 7-9) alcohols all provided under vacuum: 1H NMR (300 MHz, CDCl3) δ 8.49 (d, J ) 7.8
the desired benzyl ethers (3a-h) in good to excellent yield. Hz, 1H), 8.34 (apparent t, J ) 8.3 Hz, 1H), 7.59 (d, J ) 9.0 Hz,
Among these substrates are an allylic alcohol (entry 4), a 1H), 7.53-7.42 (m, 6H), 5.58 (s, 2H), 4.13 (s, 3H); 13C NMR (75
homoallylic alcohol (entry 9), and a β-hydroxy ester (entry 6). MHz, CDCl3) δ 159.6, 148.0, 143.8, 132.5, 129.6, 129.1, 128.5,
We saw no difference between freshly prepared reagent and a 119.0, 112.1, 74.5, 42.0; HRMS (ESI+) found 200.1070 (M -
OTf)+ (calcd for C13H14NO+ 200.1075).
sample of 1 that had been aged for three months (cf. entries 1
Standard Procedure for Benzylation of Alcohols (2 f 3). A
and 2). mixture of pyridinium triflate 1 (100 mg, 0.29 mmol), benzotri-
Tertiary alcohols and phenols provided variable results fluoride (PhCF3, 0.29 mL), MgO (11.5 mg, 0.29 mmol, vacuum-
(entries 10-12). 1-Adamantanol (2i), which is not prone to dried), and alcohol 2 (0.14 mmol) was heated at 83 °C for 1 day.
elimination, afforded benzyl ether 3i in good yield. Tertiary The reaction mixture was cooled to room temperature and filtered
benzylic alcohol 2j, which is highly prone to elimination, through Celite. The filtrate was concentrated under vacuum and
provided only a moderate yield of ether 3j. These two substrates purified on silica gel to yield benzyl ether 3 (see Table 3), admixed
may approximate the upper and lower limits of benzylation with varying amounts of Bn2O.
efficiency for tertiary alcohol substrates using 1. Phenols (e.g., Benzylation of Diethylene Glycol Monomethyl Ether (Monogly-
2k, entry 12) reacted sluggishly in our study, possibly due to a me, 2d). A mixture of pyridinium triflate 1 (581 mg, 1.67 mmol),
benzotrifluoride (PhCF3, 1.7 mL), MgO (67 mg, 1.7 mmol), and
decrease in nucleophilicity relative to aliphatic alcohols. Because
2d (100 mg, 0.83 mmol) was subjected to the standard procedure
benzylation of phenols can be accomplished using Mitsunobu to afford 0.163 g (93%) of diethylene glycol benzyl methyl ether
conditions,20 this class of substrates was not investigated further. (3d) as a pale yellow liquid, which exhibited spectroscopic
4. Insights into the Potential Reaction Mechanism. The properties consistent with the reported data.23
mechanistic course of benzylation reactions using 1 undoubtedly Benzylation of 1-Adamantanol (2i). A mixture of pyridinium
falls along the continuum between SN1 and SN2 pathways triflate 1 (100 mg, 0.29 mmol), benzotrifluoride (PhCF3, 0.29 mL),
(Scheme 2). Although we have not performed detailed kinetic MgO (11.5 mg, 0.29 mmol), and 2i (21.8 mg, 0.14 mmol) was
studies, two key observations are more consistent with an SN1- subjected to the standard procedure to afford 0.0363 g of a yellow
type mechanism. Benzylation reactions conducted in toluene oil, which was determined by 1H NMR analysis to consist of 8.7
afforded trace amounts of o-12 and p-12. We assume that these mg of Bn2O and 0.0276 g (80%) of 1-benzyloxyadamantane (3i).
Spectroscopic analysis was consistent with the data reported
compounds derive from Friedel-Crafts alkylation of toluene,
previously for 3i.24
which suggests the presence of a highly electrophilic benzylating
species (e.g., benzyl cation 9) in the reaction mixture and argues Acknowledgment. We thank the FSU Department of
in favor of a more SN1-like pathway. Methoxypyridinium salt Chemistry and Biochemistry for support of this work, the NMR
1021 was completely inert under similar conditions, which argues Facility for spectroscopic support, the Krafft laboratory for the
against an SN2-type pathway. We therefore surmise that the use of their IR spectrometer and melting point apparatus, and
actual benzylation event using 1 is better approximated by the Dr. Umesh Goli for assistance with mass spectrometry.
SN1 mechanism. This conclusion is consistent with behavior Supporting Information Available: Characterization data and
observed in trichloroacetimidate reactions.22 NMR spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
Conclusion JO0602773
We report the synthesis and reactivity of 2-benzyloxy-1-methyl-
pyridinium triflate (1), a novel benzylation reagent for alcohols. (18) Silyl ether 8 was prepared in quantitative yield by treating a solution
of 4-(4-methoxyphenyl)butan-1-ol in CH2Cl2 with DMAP (0.10 equiv), Et3N
Salt 1 is easy to prepare, bench-stable, and preactivated. No (2.0 equiv), and TBSCl (1.1 equiv). See the Supporting Information for
acidic or basic promoters are needed for benzyl transfer, which characterization data.
occurs simply upon warming in the presence of the alcohol (19) The mass balance exceeded the theoretical yield of 3, and dibenzyl
substrate. Work on this and related reagents is in progress. ether (7) was observed by TLC and/or 1H NMR analysis. The amount of 7
could not be estimated with any precision based on the 1H NMR spectra
because the diagnostic benzylic singlets were coincident.
Experimental Section (20) Hughes, D. L. Org. React. 1992, 42, 335-656.
2-Benzyloxypyridine (5). The following is a modification of a (21) Methoxypyridinium triflate 10 was prepared in 85% yield (unop-
reported procedure.9a A mixture of benzyl alcohol (2.00 g, 18.5 timized) by a procedure similar to that used for the preparation of 1. See
mmol), 2-chloropyridine (3.46 g, 30.5 mmol), KOH (3.42 g, 61.0 the Supporting Information for characterization data.
(22) Cramer, F.; Hennrich, N. Chem. Ber. 1961, 94, 976-989.
mmol, ground with a mortar and pestle), toluene (37 mL), and 18- (23) Grobelny, Z.; Stolarzewicz, A.; Maercker, A.; Krompiec, S.;
crown-6 (24.4 mg, 0.925 mmol) was heated at reflux for 1 h with Kasperczyk, J.; Rzepa, J. J. Organomet. Chem. 2004, 689, 1580-1585.
azeotropic removal of water (Dean-Stark trap). The reaction (24) Hartz, N.; Prakash, G. K. S.; Olah, G. A. Synlett 1992, 569-572.

J. Org. Chem, Vol. 71, No. 10, 2006 3927

You might also like