An activated sulfonylating agent that undergoes general base-catalyzed hydrolysis by amines in preference to aminolysis

Article abstract of DOI:10.1021/jo800407x

(Chemical Equation Presented) Activated sulfonyl derivatives, similar to acyl ones, usually undergo aminolysis with amines in water as nucleophilic attack by the amine is preferred to hydrolysis. However, despite being active sulfonyl derivatives, four-membered heterocyclic sulfonamides, β-sultams, do not undergo aminolysis in aqueous solution but preferentially react to give hydrolysis products only. The rate of the reaction of β-sultams in buffered solutions of simple primary amines shows a first-order dependence on amine concentrations attributed to general base-catalyzed hydrolysis by the amine. Even N-benzyl-4,4-dimethyl-3-oxo-β-sultam, which is both a β-sultam and a β-lactam, undergoes hydrolysis at the sulfonyl center rather than aminolysis at either the sulfonyl or acyl center. The solvent kinetic isotope effects (SKIE, k^H2O/k^D2O) for the amine-catalyzed hydrolyses are 1.4 and 1.9 for the hydrolysis of N-benzoyl-β-sultam and N-benzyl-4,4-dimethyl-3-oxo-β-sultam, respectively, compatible with a general base-catalyzed mechanism. The amine-catalyzed hydrolysis gives a Bronsted β value of +0.9 for both N-benzoyl β-sultam and N-benzyl-4,4-dimethyl-3-oxo-β-sultam, indicating that the general base amine is almost fully protonated in the transition state. A general base-catalyzed mechanism for hydrolysis rather than nucleophilic attack was also deduced for the reaction of N-benzyl-4,4-dimethyl-3-oxo-β-sultam with carboxylate anions based on a SKIE of 1.7-1.9 and rate constants which fit the Bronsted plot for amines. In contrast to acyl transfer reactions, those for sulfonyl transfer appear to show an inverse reactivity-selectivity relationship - the most active compounds being the most selective. The lack of reactivity of β-sultams toward amine nucleophiles appears to be related to the mechanism of ring opening of β-sultams with a decreased reactivity toward amines relative to hydroxide ion, probably related to the expulsion of the relatively poor leaving group amide anion.

Full text of DOI:10.1021/jo800407x

An Activated Sulfonylating Agent That Undergoes General
Base-Catalyzed Hydrolysis by Amines in Preference to Aminolysis
Wing Y. Tsang, Naveed Ahmed, Karl Hemming, and Michael I. Page*
Department of Chemical and Biological Sciences, The UniVersity of Huddersfield,
Queensgate, Huddersfield, HD1 3DH, United Kingdom
m.i.page@hud.ac.uk
ReceiVed February 26, 2008
Activated sulfonyl derivatives, similar to acyl ones, usually undergo aminolysis with amines in water as
nucleophilic attack by the amine is preferred to hydrolysis. However, despite being active sulfonyl
derivatives, four-membered heterocyclic sulfonamides, ꢀ-sultams, do not undergo aminolysis in aqueous
solution but preferentially react to give hydrolysis products only. The rate of the reaction of ꢀ-sultams
in buffered solutions of simple primary amines shows a first-order dependence on amine concentrations
attributed to general base-catalyzed hydrolysis by the amine. Even N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam,
which is both a ꢀ-sultam and a ꢀ-lactam, undergoes hydrolysis at the sulfonyl center rather than aminolysis
at either the sulfonyl or acyl center. The solvent kinetic isotope effects (SKIE, kH O/kD O) for the amine-
catalyzed hydrolyses are 1.4 and 1.9 for the hydrolysis of N-benzoyl-ꢀ-sultam and N-benzyl-4,4-dimethyl-
3-oxo-ꢀ-sultam, respectively, compatible with a general base-catalyzed mechanism. The amine-catalyzed
hydrolysis gives a Bronsted ꢀ value of +0.9 for both N-benzoyl ꢀ-sultam and N-benzyl-4,4-dimethyl-
3-oxo-ꢀ-sultam, indicating that the general base amine is almost fully protonated in the transition state.
A general base-catalyzed mechanism for hydrolysis rather than nucleophilic attack was also deduced for
the reaction of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam with carboxylate anions based on a SKIE of 1.7-1.9
and rate constants which fit the Bronsted plot for amines. In contrast to acyl transfer reactions, those for
sulfonyl transfer appear to show an inverse reactivity-selectivity relationshipsthe most active compounds
being the most selective. The lack of reactivity of ꢀ-sultams toward amine nucleophiles appears to be
related to the mechanism of ring opening of ꢀ-sultams with a decreased reactivity toward amines
relative to hydroxide ion, probably related to the expulsion of the relatively poor leaving group
amide anion.
2
2
Introduction
by the reactivity-selectivity principle.1–3 For acylation, there
is generally a decrease in selectivity with increasing reactivity,
i.e., the more reactive the acylating agent the less selective and
The reaction of activated acyl or sulfonyl derivatives with
amines in water to give amides and sulfonamides, respectively,
is an interesting consequence of the relative rates of aminolysis
and hydrolysis. For aminolysis to occur in preference to
hydrolysis, not only do relative rates have to be considered,
but also absolute reactivity, which historically is encompassed
discriminatory it is toward nucleophiles and the ratio kOH/kNu
,
(1) (a) Stock, L. M.; Brown, H. C. AdV. Phys. Org. Chem. 1963, 1, 35154.
(b) Jencks, W. P. Chem. ReV. 1985, 85, 511–527. (c) Exner, O. J. Chem. Soc.,
Perkin Trans. 2 1993, 973–979.
(2) (a) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348–354. (b) Ritchie, C. D.
Can. J. Chem. 1986, 64, 2239–2250.
* Address correspondence to this author. Phone: + 44 (0) 1484 472531. Fax:
+ 44 (0) 1484 473075.
(3) (a) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66–77.
(b) Minegishi, S.; Mayr, H. J. Am. Chem. Soc. 2003, 125, 286–295.
4504 J. Org. Chem. 2008, 73, 4504–4512
10.1021/jo800407x CCC: $40.75 2008 American Chemical Society
Published on Web 05/15/2008

Hydrolysis of ActiVated Sulfonyl DeriVatiVes
representing the rate constants for the hydroxide ion promoted
hydrolysis and aminolysis, respectively, decreases with increas-
ing reactivity.4 By comparison, nucleophilic substitution reac-
tions at sulfonyl centers are not as well studied or understood
as the analogous acyl transfer reactions,5 although reactive
sulfonylating agents such as sulfonyl chlorides, reactive sul-
fonate esters, and sulfonyl imidazoles do undergo aminolysis
rather than hydrolysis with amines in aqueous solution.6 On
the other hand, although ꢀ-lactams (1) similarly undergo
aminolysis with amines in water, the corresponding ꢀ-sultams7
(2) appear to prefer hydrolysis under similar conditions.8 This
appears unusual because ꢀ-sultams are reactive compounds and,
for example, selectively sulfonylate some serine enzymes in
preference to hydrolysis.7 ꢀ-Sultams show greater reactivity,
as indicated by their rates of alkaline hydrolysis, than their acyl
analogues, the ꢀ-lactams: N-phenyl- and N-alkyl-ꢀ-sultams are
103 times more reactive toward hydroxide than their corre-
sponding ꢀ-lactams.9 However, despite the intrinsic reactivity
of ꢀ-sultams, there is no increase in the rate of reaction of
N-benzyl-ꢀ-sultam (3) in aqueous solutions of sodium hydroxide
in the presence of n-propylamine or hydrazine or in aqueous
buffers of these amines, and the only product is that from
hydrolysis of the ꢀ-sultam.8 This is in sharp contrast to N-aryl-
ꢀ-lactams10 and the ꢀ-lactams in penicillins and cephalosporins11
where aminolysis occurs readily in competition with hydrolysis.
Although this may be thought due to relative reactivities of the
acylating or sulfonylating agent, the rate of alkaline hydrolysis
of N-benzyl-ꢀ-sultam (3) is only about 10-fold less than that of
benzyl penicillin. Furthermore, there is no aminolysis or
significant change in the rate of hydrolysis of the more reactive
N-m-chlorophenyl-ꢀ-sultam (4) with increasing concentrations
of n-propylamine.8 Why are amine nucleophiles not able to
compete with HO- in attacking N-aryl-ꢀ-sultams and yet they
do with other similarly active sulfonyl and acyl derivatives?
(4) would appear to be a good sulfonylating agent as it is a
reactive compound8 with a kOH of 46.0 M-1 s-1 and less reactive
acyl systems such as ꢀ-lactams react with a variety of nucleo-
philes in water, showing both alcoholysis and aminolysis.11,12
The ꢀ-sultams do appear to be unusual sulfonyl compounds as
other sulfonyl derivatives such as benzenesulfonyl chloride (kOH
14
) 40.4 M-1 s-1 13 and tosylimidazole (kOH ) 3.16 M-1 s-1
) )
both undergo aminolysis in water. It is not clear whether this is
connected with the intrinsic reactivity of the sulfonyl center and
reactivity-selectivity relationships or due to some peculiar
aspect of the mechanism of ring opening of ꢀ-sultams.
SCHEME 1
Incorporating a carbonyl group within the ꢀ-sultam ring gives
the 3-oxo-ꢀ-sultams (5) which are both ꢀ-sultams and ꢀ-lactams
and so nucleophilic attack on them could involve either acylation
or sulfonylation resulting from substitution at the carbonyl center
and expulsion of the sulfonamide or from substitution at the
sulfonyl center and expulsion of the amide, respectively (Scheme
1). We have previously shown that the alkaline hydrolysis of
3-oxo-ꢀ-sultams (5) occurs by S-N fission and is only 10-fold
more reactive than N-benzoyl-ꢀ-sultam (6).15 However, N-ben-
zyl-4,4-dimethyl-3-oxo-ꢀ-sultam is 4000-fold more reactive than
N-m-chlorophenyl-ꢀ-sultam (4) toward hydroxide ion as a result
of the amide anion being a better leaving group than the
arylamine.8,15 It is therefore of interest to study if the moderate
increase in the reactivity toward alkaline hydrolysis may have
an effect on the selectivity of N-acyl-ꢀ-sultams toward ami-
nolysis rather than hydrolysis.
Results and Discussion
If the second order rate constant for alkaline hydrolysis, kOH
is taken as an indicator of reactivity, N-m-chlorophenyl-ꢀ-sultam
,
The Reaction of N-Benzoyl-ꢀ-sultam (6) with Nitrogen
Bases. The apparent lack of aminolysis of N-benzoyl-ꢀ-sultam
(4) Jencks, W. P.; Gilchrist, M. J. Am. Chem. Soc. 1968, 90, 2622–2637.
(5) (a) Gordon, I. M.; Maskill, H.; Ruasse, M. F. Chem. Soc. ReV. 1989, 18,
123–151. (b) Hoz, S.; Liu, P.; Buncel, E. Chem. Commun. 1996, 995–996. (c)
Kice, J. L. AdV. Phys. Org. Chem. 1980, 17, 65–181.
(11) (a) Page, M. I. AdV. Phys. Org. Chem. 1987, 23, 165–270. (b)
Gensmantel, N. P.; Page, M. I. J. Chem. Soc., Perkin Trans. 2 1979, 137–142.
(c) Morris, J. J.; Page, M. I. J. Chem. Soc., Perkin Trans. 2 1980, 212–219. (d)
Morris, J. J.; Page, M. I. J. Chem. Soc., Perkin Trans. 2 1980, 220–224. (e)
Page, M. I.; Proctor, P. J. Am. Chem. Soc. 1984, 106, 3820–3825.
(12) Davis, A. M.; Proctor, P.; Page, M. I. J. Chem. Soc., Perkin Trans. 2
1991, 1213–1217.
(6) (a) Davy, M. B.; Douglas, K. T.; Loran, J. S.; Steltner, A.; Williams, A.
J. Am. Chem. Soc. 1977, 99, 1196–1206. (b) Rogne, O. J. Chem. Soc., Perkin
Trans. 2 1972, 489.
(7) Page, M. I. Acc. Chem. Res. 2004, 37, 297–303.
(13) Rogne, O. J. Chem. Soc. B 1970, 1056.
(8) Wood, J. M.; Hinchliffe, P. S.; Laws, A. P.; Page, M. I. J. Chem. Soc.,
Perkin Trans. 2 2002, 938–946.
(14) (a) Monjoint, P.; Ruasse, M. F. Tetrahedron Lett. 1984, 25, 3183–3186.
(b) Monjoint, P.; Ruasse, M. F. Bull. Soc. Chim. Fr. 1988, 356–360.
(15) Ahmed, N.; Tsang, W. Y.; Page, M. I. Org. Lett. 2004, 6, 201–203.
(16) (a) Page, M. I.; Williams, A. Organic and bio-organic mechanisms;
Addison-Wesley Longman: Harlow, England, 1997. (b) Williams, A. Concerted
organic and bio-organic mechanisms; CRC Press: Boca Raton, FL, 1999. (c)
Colthurst, M. J.; Williams, A. J. Chem. Soc., Perkin Trans. 2 1997, 1493–1497.
(d) Hunter, A.; Renfrew, M.; Rettura, D.; Taylor, J. A.; Whitmore, J. M. J.;
Williams, A. J. Am. Chem. Soc. 1995, 117, 5484–5491. (e) Williams, A. Chem.
Soc. ReV. 1994, 23, 93–100. (f) Guthrie, J. P. J. Am. Chem. Soc. 1991, 113,
3941–3949.
(9) (a) Baxter, N. J.; Rigoreau, L. J. M.; Laws, A. P.; Page, M. I. J. Am.
Chem. Soc. 2000, 122, 3375–3385. (b) Hinchliffe, P. S.; Wood, J. M.; Davis,
A. M.; Austin, R. P.; Beckett, R. P.; Page, M. I. J. Chem. Soc., Perkin Trans. 2
2001, 1503–1505. (c) Baxter, N. J.; Laws, A. P.; Rigoreau, L. J. H.; Page, M. I.
Chem. Commun. 1999, 2401–2402. (d) Baxter, N. J.; Page, M. I. Chem. Commun.
1997, 2037–2038. (e) Baxter, N. J.; Laws, A. P.; Rigoreau, L.; Page, M. I.
J. Chem. Soc., Perkin Trans. 2 1996, 2245–2246.
(10) Blackburn, G. M.; Plackett, J. D. J. Chem. Soc., Perkin Trans. 2 1973,
981–985.
J. Org. Chem. Vol. 73, No. 12, 2008 4505

Tsang et al.
detectable dependence on the total amine concentration in a
buffered solution of ethylenediamine mono- and dication. The
reaction shows simple first-order kinetics and the observed
pseudo-first-order rate constants, kobs, were plotted against the
total concentration of ethylenediamine (Figure 1). Extrapolating
the rate constants to zero amine buffer concentration corresponds
to that calculated for the background hydrolysis at the given
pH and the slopes of these plots give the second-order rate
constants, kcat, for the reaction dependent on amine concentra-
tion, which when plotted against the fractions of monocationic
ethylenediamine R exhibit a sharp upward curvature (Figure 2).
The sharp upward curvature indicates that the amine-catalyzed
reaction either is hydroxide ion catalyzed or is the result of its
kinetic equivalent in the rate law. Since the second-order rate
constant, kcat, is indistinguishable from zero as the fraction of
the monocation form of ethylenediamine approaches zero, the
dication form is not catalytically important. The rate law can
thus be described as:
FIGURE 1. The dependence of the observed pseudo-first-order rate
constants, kobs, for the reaction of N-benzoyl-ꢀ-sultam (6) with
ethylenediamine against total amine concentration with use of the mono-
and dication as the buffer at various pH values in 1% acetonitrile at 30
°C with 1.0 M ionic strength (KCl).
rate
[ꢀ-sultam]
) kobs ) ko + k1R[EDAH+]tot
+
k2'R[EDAH+]tot[OH-] (1)
and the apparent second-order rate constant, kcat, is given
by:
(6) previously reported8 with propylamine in aqueous solution
is due to the more effective competing hydrolysis by hydroxide
ion. To determine whether aminolysis of ꢀ-sultams can actually
occur in aqueous solution, ethylenediamine was chosen both
as a potential nucleophile and a buffer. This is because
ethylenediamine has been shown to have enhanced reactivity
due to the second amine acting as an intramolecular general
base catalyst11c and buffered solutions of ethylenediamine mono-
and dication are around neutral pH thus reducing the rate of
alkaline hydrolysis. The rate of the reaction of N-benzoyl-ꢀ-
sultam (6) in aqueous buffers of ethylenediamine was studied
by UV-spectrophotometry in 1% acetonitrile-water (v/v) and
1.0 M ionic strength (KCl) at 30 °C. Interestingly, the rate of
the reaction of N-benzoyl-ꢀ-sultam does show a small but
kobs - ko
[EDAH+]tot
kcat
)
) k1R + k2′R[OH-]
(2)
where k1 is the second-order rate constant for the amine-
catalyzed reaction by the diamine monocation, EDAH+, and
k′2 is the apparent third-order rate constant for the reaction
catalyzed by EDAH+ and hydroxide ion. The rate constants
k1 and k′2 can be obtained from the intercept and slope,
respectively, of a plot of kcat/R against hydroxide ion
concentration (Figure 3). However, the term in the rate law,
k′2R[EDAH+]tot[OH-], is kinetically equivalent to the amine-
catalyzed reaction by neutral diamine with a term, k2[EDA]
where
[EDA][H+]
[EDAH+]
Kw
k2 ) k2'
and Ka )
Ka
and k2 is the second-order rate constant for the amine-catalyzed
reaction by neutral ethylenediamine, EDA. The rate law then
becomes:
rate
[ꢀ-sultam]
) kobs ) ko + k1[EDAH+] + k2[EDA] (3)
and the values of these rate constants are given in Table 1.
However, this kinetic analysis does not distinguish between the
amine acting as a nucleophile giving an aminolysis reaction or
it being a general base catalyst (7) leading to hydrolysis of the
ꢀ-sultam (6). A product analysis with HPLC was undertaken
with the reaction of N-benzoyl-ꢀ-sultam with 1.0 M ethylene-
diamine at pH 8.03 with 1% acetonitrile-water (v/v). Under
the kinetic experimental conditions the background hydrolysis
only accounts for 10% and the amine-catalyzed reaction
accounts for 90% of the reaction. The HPLC chromatogram of
the product reaction mixture is similar to that obtained for the
alkaline hydrolysis reaction with the same concentration of
N-benzoyl-ꢀ-sultam. Although the peak area of the signal for
the hydrolysis product from the reaction mixture with ethyl-
enediamine is slightly less than that obtained from alkaline
FIGURE 2. The dependence of the second-order rate constant, kcat
,
on the fraction of free base (the monocation), R, of ethylenediamine in
the reaction with N-benzoyl-ꢀ-sultam (6) in 1% acetonitrile at 30 °C
with 1.0 M ionic strength (KCl). The theoretical curve was generated
by using eq 2 and the rate constant in Table 1.
4506 J. Org. Chem. Vol. 73, No. 12, 2008

Hydrolysis of ActiVated Sulfonyl DeriVatiVes
TABLE 1. The Rate Constants for the Reaction of ꢀ-Sultams with Nitrogen and Oxygen Bases in Water and Deuterium Oxide
kH O/M-1 s-1
kD O/M-1 s-1
kH O/kD O
2
2
2
2
pKa
N-benzoyl ꢀ-sultam (6)
aminoacetonitrile
5.46a
7.32a
10.1b
15.6
1.02 × 10-3
1.02 × 10-1
11.9
ethylenediamine monocation (EDAH+)
ethylenediamine (EDA)
hydroxide-ion
5.44 × 10-2
2.39 × 10-2
1.88
1.46 × 104
N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8)
aminoacetonitrile
trifluoroethylamine
ethylenediamine monocation (EDAH+)
2-cyanoethylamine
taurine
formate
acetate
hexafluoroacetone
hydroxide-ion
5.46a
5.65a
7.32a
7.91a
8.85a
3.58a
4.57a
6.38a
3.91 × 10-2
3.34 × 10-2
1.30
1.40
4.90
33.3
2.10 × 10-3
6.95 × 10-3
1.02 × 10-1
1.83 × 105
4.14 × 10-3
5.33 × 10-2
1.68
1.91
15.6
N-benzyl-3,3-dimethyl-2-oxo-ꢀ-lactam (10)
taurine
9.05b
7.48 × 10-2
a ionization constants measured potentiometrically at 30 °C with 1.0 M ionic strength (KCl) in water. b ionization constants obtained from ref 11c.
N-methylimidazole17 but solvent kinetic isotope effects as small
as 1.1 have been observed in some general base-catalyzed
reactions.18 It is interesting to note that a very large SKIE kH O
/
2
kD O of 4.4 was observed for the pH-independent hydrolysis of
2
the ꢀ-sultam (6) consistent with rate limiting proton transfer
presumably involving general base or acid catalysis by a second
molecule of water.8
The observation that a reaction due to EDA occurs in buffers
of EDAH+/EDAH2 despite its very low concentration indi-
2+
cates the amine-catalyzed reaction must show a large Brønsted
ꢀ value, i.e., a large dependence of the rate constant on the
basicity of the amine. To confirm this, the reaction was also
studied in aqueous buffers of the weakly basic amine aminoac-
etonitrile. Similar to that observed with ethylenediamine,
aminoacetonitrile also shows a small, but detectable, catalysis
of the hydrolysis reaction at low fractions of free base of the
amine, i.e., at low pH. As the fraction of free base of the amine
and the pH of the solution increases, catalysis by the amine
becomes undetectable, due to the increasing significance of
alkaline hydrolysis. The observed pseudo-first-order rate con-
stant, kobs, for the reaction plotted against the total concentrations
of aminoacetonitrile and extrapolated to zero amine buffer
concentration corresponds to that calculated for the background
hydrolysis at the given pH. As before, the slopes of these plots
give the second-order rate constants, kcat, which as a function
of different fractions of free base, R, of the amine give the
contributions to the rate from the protonated and the unproto-
nated form of the amine. However, unlike that observed with
ethylenediamine, a straight line passing through the origin was
obtained from the plot of kcat against R, indicating that the
reactive form of the amine is the free base form and there is no
reaction by the protonated form or catalysis by hydroxide ion.
This also supports the interpretation of general base catalysis
by the amine rather than hydroxide ion-catalyzed aminolysis,
as is commonly observed with reactive acylating agents.10,11
The rate law that adequately describes the reaction is thus
simply:
FIGURE 3. The dependence of the second-order rate constants, kcat
/
R, for the reaction of ethylenediamine monocation with N-benzoyl
ꢀ-sultam (6) upon the activity of hydroxide ion. The rate constants
were determined in 1% acetonitrile at 30 °C with 1.0 M ionic strength
(KCl).
hydrolysis, there are no other peaks observed and it appears
that the reaction catalyzed by the amine is a hydrolysis reaction.
Another experimental distinction between nucleophilic and
general base catalysis is their different solvent kinetic isotope
effects.16a The corresponding rate constants for the reaction of
the ꢀ-sultam (6) with EDAH+ were determined as before but
in deuterium oxide. The solvent kinetic isotope effect for
rate
[ꢀ-sultam]
) kobs ) ko + kBR[aminoacetonitrile]tot
)
EDAH+, k1H O/k1D O, is 1.88 (Table 1), which is compatible with
general base catalysis as the reaction is slower in deuterium
oxide than it is in water.16a For comparison, a solvent kinetic
isotope effect as high as 4 was determined for the general base-
catalyzed hydrolysis of N-acetyl-N′-methylimidazolium ion by
2
2
ko + kB[aminoacetonitrile] (4)
(17) Wolfenden, R.; Jencks, W. P. J. Am. Chem. Soc. 1961, 83, 4390–4393.
(18) Jencks, W. P.; Carriuolo, J. J. Am. Chem. Soc. 1960, 82, 675.
J. Org. Chem. Vol. 73, No. 12, 2008 4507

Tsang et al.
aminolysis of reactive acyl and sulfonyl centers in water is
well-known.13,18 It is interesting that ꢀ-sultams do not undergo
aminolysis whereas their acyl analogues, ꢀ-lactams, react with
amines readily in preference to hydrolysis.10,11,21 It has been
shown that, in the nucleophilic-catalyzed hydrolysis of car-
boxylic acid esters,22 the mechanism of a nucleophilic displace-
ment reaction depends on the relative nucleophilicity of the
nucleophile and the nucleofugacity of the leaving group. With
a good leaving group the nucleophilic reaction undergoes an
uncatalyzed pathway but it becomes general base or hydroxide
ion catalyzed with decreasing nucleofugacity of the leaving
group, and no nucleophilic reaction is detected at all for a poor
leaving group. The nucleophilicity of the nucleophile and the
nucleofugacity of the leaving group in the aminolysis of N-aroyl-
ꢀ-sultams and N-aroyl-ꢀ-lactams are the same, and yet only the
latter readily undergoes aminolysis in aqueous solution.21
Sulfonylating agents such as sulfonyl chlorides which are 2
orders of magnitude less reactive toward hydroxide ion com-
pared with N-benzoyl-ꢀ-sultam undergo aminolysis in preference
to hydrolysis, so the lack of reactivity of the ꢀ-sultams toward
amine nucleophiles appears not to be related to absolute
reactivity or to the fact that the leaving group is an amide.
N-Benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8). (i) Reactions
with N-Nucleophiles. Similar to other active acylating agents,
ꢀ-lactams are susceptible to nucleophilic displacement reactions
with various nucleophiles, including amines, in water in
preference to hydrolysis.10–12,23 Although ꢀ-sultams are gener-
ally 103 more reactive than their acyl analogues, ꢀ-lactams,
toward hydroxide ion, they do not undergo aminolysis in water.
3-Oxo-ꢀ-sultams (5) are both ꢀ-sultams and ꢀ-lactams and
nucleophilic attack on them could involve either acylation or
sulfonylation resulting from substitution at the carbonyl center
and expulsion of the sulfonamide or from substitution at the
sulfonyl center and expulsion of the amide, respectively (Scheme
1). The alkaline hydrolysis of 3-oxo-ꢀ-sultams (5) occurs by
S-N fission as a result of the preferential attack of hydroxide
ion on the sulfonyl center,15 but given the greater susceptibility
of ꢀ-lactams to aminolysis, it is of interest to determine if amines
selectively attack the carbonyl center in 5.
The rate of reaction of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam
(8) was studied in aqueous amine buffers with 1% acetonitrile-
water (v/v) and 1.0 M ionic strength (KCl) at 30 °C. Similar to
the amine-catalyzed reaction with N-benzoyl ꢀ-sultam (6), a
small but measurable increase in rate with increasing amine
concentration was observed. The observed pseudo-first-order
rate constant, kobs, for the reaction of 8 in aqueous solution
buffered with 2-cyanoethylamine is linearly dependent on the
total amine concentration and extrapolation to zero amine
concentration corresponds to that calculated for the background
hydrolysis. The slopes of these plots give the second-order rate
constants, kcat, for the reaction by both the protonated and the
deprotonated form of the amine and a plot of kcat against R, the
fraction of free base, gives a straight line that passes through
the origin for all amines studied. This indicates the basic form
but not the protonated from of the amines is catalytically active.
The second-order rate constants, kB, for the amine-catalyzed
reaction are obtained as the right-hand intercept at R ) 1.0 and
FIGURE 4. The dependence of the second-order rate constant for the
reaction of amines with N-benzoyl-ꢀ-sultam (6) as a function of the
basicity of the amine in 1% acetonitrile at 30 °C with 1.0 M ionic
strength (KCl), where ACN ) aminoacetonitrile, EDAH+ ) ethylene-
diamine monocation, and EDA ) ethylenediamine.
where ko is the rate constant for the background hydrolysis and
kB is the second-order rate constant for the amine-catalyzed
reaction by aminoacetonitrile (Table 1).
The second-order rate constants for the general base-catalyzed
hydrolysis of N-benzoyl-ꢀ-sultam (6) are correlated with the
ionization constants of the conjugate acids of the amines to give
a Brønsted ꢀ value of +0.87 (Figure 4); although this is based
on only three points, the range of basicity covered is nearly 5
pKa units (a Brønsted ꢀ value of +0.80 is obtained with
statistical correction). The value of ꢀ indicates the degree of
charge development of the catalyst in the transition state of the
rate-limiting step of the reaction.16a The Brønsted ꢀ value of
+0.87 indicates that the amine, which is acting as a general
base in removing a proton from the attacking water molecule
(7), is almost fully protonated in the transition state. This is an
unusually large degree of charge development on the base for
a general based-catalyzed reaction. For comparison, the Brønsted
ꢀ values for the general base-catalyzed hydrolysis of acetyl-
imidazole, acetylimidazolium ion, and ethyl dichloroacetate are
+0.55, +0.34, and +0.47, respectively.19,20 The later transition
state for the hydrolysis of the ꢀ-sultam is indicative of a much
greater selectivity on the general base than that observed with
acyl transfer.
The Brønsted plot (Figure 4) can also be used to explain why
there was no reaction observed in a previous report8 with
propylamine and N-benzoyl-ꢀ-sultam (6). From the Brønsted
plot, the second-order rate constant for the reaction with propy-
lamine is calculated to be 90 M-1 s-1 so by using 0.1 M
propylamine at pH 11.62 (0.84 fraction of free base), the
contribution to the total rate of hydrolysis from the amine
catalyzed reaction would be about 8 s-1. The rate of the
background hydrolysis at this pH is 80 s-1, which is 10-fold
greater than that of the amine-catalyzed reaction, so the detection
of this contribution is experimentally difficult.
In summary, although a reaction involving amine and the
N-benzoyl-ꢀ-sultam (6) has now been demonstrated it involves
general base-catalyzed hydrolysis and not aminolysis. The
(21) Tsang, W. Y.; Ahmed, N.; Page, M. I. Org. Biomol. Chem. 2007, 5,
485–493.
(22) (a) Kirsch, J. F.; Jencks, W. P. J. Am. Chem. Soc. 1964, 86, 833–837.
(b) Kirsch, J. F.; Jencks, W. P. J. Am. Chem. Soc. 1964, 86, 837–846.
(23) Llinas, A.; Vilanova, B.; Page, M. I. J. Phys. Org. Chem. 2004, 17,
521–528.
(19) Jencks, W. P.; Oakenfull, D. G. J. Am. Chem. Soc. 1971, 93, 178–188.
(20) Jencks, W. P.; Carriuolo, J. J. Am. Chem. Soc. 1961, 83, 1743–1750.
4508 J. Org. Chem. Vol. 73, No. 12, 2008

Hydrolysis of ActiVated Sulfonyl DeriVatiVes
SCHEME 2
similar to that of +0.87 for the amine-catalyzed reaction with
N-benzoyl-ꢀ-sultam, and is indicative of an almost fully
protonated amine in the transition state of the rate-limiting step
(9). A solvent kinetic isotope effect, kBH O/kBD O, of 1.40 was
determined for the reaction with trifluoroethylamine (Table 1)
which, although low, is compatible with a mechanism involving
proton transfer in the transition state of the rate-limiting step
(9). Further confirmation for amine general base-catalyzed
hydrolysis of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8) comes
from the similar magnitude of the rate constants to those for
the hydrolysis of N-benzoyl-ꢀ-sultam (6) (Table 1) indicative
of both reactions involving sulfonyl transfer and opening of the
four-membered ring by S-N fission.
2
2
are given in Table 1. The rate law that adequately describes the
reaction is therefore the same as that for the amine-catalyzed
hydrolysis of N-benzoyl-ꢀ-sultam with aminoacetonitrile (eq 4)
and unlike the aminolysis of ꢀ-lactams and other active acylating
agents shows no second order term in amine- or base-catalyzed
aminolysis.10,11
A product analysis by HPLC was undertaken with the
reaction solution containing N-benzyl-4,4-dimethyl-3-oxo-
ꢀ-sultam (0.01 M) and taurine (1.0 M) at pH 8.87 in 5%
acetonitrile-water (v/v), as these conditions, based on the
kinetic data, correspond to the amine-dependent reaction
accounting for 95% of the reaction with the remaining 5%
due to the background hydrolysis. The HPLC chromatogram
of the product reaction mixture is identical with that obtained
from the alkaline hydrolysis reaction with the same concen-
tration of the 3-oxo-ꢀ-sultam. The fraction was collected and
submitted for ESI-MS (negative mode) analysis, which shows
an observed mass peak of 256 corresponding to the mass of
the hydrolysis product. As the hydrolysis product formed in
aqueous amine buffer has the same retention time as that
formed in water, it is the ꢀ-amido sulfonic acid (Scheme 2)
as a result of S-N bond fission. Therefore, the amine-
catalyzed reaction represents general base-catalyzed hydroly-
sis, as found for N-benzoyl ꢀ-sultam (6).
The expulsion of the amide anion as a leaving group is
presumably facilitated by the energy released in ring-opening
and has been previously suggested.21 For example, the ami-
nolysis of N-aroyl-ꢀ-lactams has a high sensitivity to both the
basicity of the attacking amine and that of the amide leaving
group. Electron-withdrawing groups in the amide leaving group
increase the rate of reaction and the aminolysis of N-aroyl-ꢀ-
lactams appears to occur by a concerted pathway in which bond
formation and fission occur simultaneously.21
N-Benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam thus reacts as a
ꢀ-sultam with water preferentially attacking the sulfonyl
center resulting in S-N bond fission and expulsion of the
amide rather than as a ꢀ-lactam undergoing amine nucleo-
philic attack on the acyl center resulting in C-N bond fission
and expulsion of the sulfonamide. The poorer susceptibility
of the sulfonyl center to N nucleophiles relative to the acyl
center has also been observed in the reaction of aniline with
acyclic acyl sulfonates.24
The second-order rate constants, kB, for the general base-
catalyzed hydrolysis of 8 can be correlated with the ionization
constants of the conjugate acids of the amines to give a Brønsted
ꢀ value of +0.90 (Figure 5). The Brønsted ꢀ value is very
It was of interest to investigate the site of nucleophilic
attack on 8 with amines in a nonaqueous system to determine
whether solvent water plays an important role in determining
the site of nucleophilic attack. N-Benzyl-4,4-dimethyl-3-oxo-
ꢀ-sultam (50 mg, 209 µmol) was left stirring in neat
2-methoxyethylamine (5 mL) and after 2 h the reaction
mixture was purified by column chromatography and the main
product (20 mg, 64 µmol) was analyzed by NMR and MS.
ESI-MS (positive mode) showed mass peaks of 315.0, 337.1,
and 354.0 corresponding to those of the aminolysis product
complexed with one proton, sodium ion, and potassium ion,
respectively. The site of nucleophilic attack was demonstrated
by 1H, 13C HMBC NMR spectra which show a coupling
between the methylene protons of the attached amine to the
carbonyl carbon of the 3-oxo-ꢀ-sultam. This shows that the
amine preferentially attacks the acyl center to form the amide
aminolysis product in a nonaqueous system and that solvent
water does have a role in determining selectivity.
FIGURE 5. The dependence of the second-order rate constants for
the reaction of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8) with amines
(b) and carboxylate anions (O) as a function of the pKa of their
conjugate acids. The second-order rate constant for the alkaline
hydrolysis of the 3-oxo-ꢀ-sultam is also shown (2). The rate constants
were determined in 1% acetonitrile at 30 °C with 1.0 M ionic strength
(KCl), where ACN ) aminoacetonitrile, TFE ) trifluoroethylamine,
EDAH+ ) ethylenediamine monocation, CE ) 2-cyanoethylamine, T
) taurine, and HFA ) hexafluoroacetone.
To estimate the rate constant for amine nucleophilic attack
on the ꢀ-lactam of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8)
the aminolysis of the analogous 2-oxo-ꢀ-lactam (10) was
studied. This system may also demonstrate that the lack of
(24) Laird, R. M.; Spence, M. J. J. Chem. Soc. B 1971, 1434.
J. Org. Chem. Vol. 73, No. 12, 2008 4509

Tsang et al.
SCHEME 3
CHART 1
aminolysis of 3-oxo-ꢀ-sultam is not due to an intrinsic
property of the four-membered ring containing two electro-
philic centers. The aminolysis of the ꢀ-lactam (10) was
demonstrated by a product analysis, using HPLC with the
reaction solution of N-benzyl-3,3-dimethyl-2-oxo-ꢀ-lactam
(0.01M) in taurine (0.2 M) at pH 9.75, 5% acetonitrile-water
(v/v). There was no signal corresponding to the hydrolysis
product (retention time ) 15.7 min) but a new signal at 8.35
min was detected. The fraction was collected and analyzed
by ESI-MS (negative mode), which gave a peak of 328
corresponding to the mass of the aminolysis product (11)
(Scheme 3). The rate of the aminolysis of 10 with taurine
was studied by UV spectroscopy in 1% acetonitrile-water
(v/v) and 1.0 M ionic strength at 30 °C, using TAPS, CHES,
and CAPS as buffers. Similar to the aminolysis of N-aroyl-
ꢀ-lactams,21 the observed pseudo-first-order rate constants,
kobs, are linearly dependent on the total amine concentration
the protonated and the deprotonated form of acetate. Plotting
kcat against the fraction of free base, R, gives a straight line
that passes through the origin for all carboxylate buffers
studied. This indicates the basic form of the carboxylates is
catalytically active and there is no reaction with the undis-
sociated acid, as is observed with N-benzyl-ꢀ-sultam.9c The
second-order rate constants, kB, for the reaction with the basic
form of carboxylates are obtained as the right intercept at R
) 1.0 and are given in Table 1.
A solvent kinetic isotope effect, kBH O/kBD O, of 1.68 and 1.91
was determined for the reaction of acetate and the anion of
hexafluoroacetone hydrate with the 3-oxo-ꢀ-sultam (8), respec-
tively (Table 1), indicative of a general base-catalyzed reaction.
By contrast, the reaction of N-benzoyl-ꢀ-sultam (6) with more
basic oxygen anions does, however, appear to involve nucleo-
philic catalyzed hydrolysis.8 The second-order rate constants,
kB, for the reaction of the 3-oxo-ꢀ-sultam (8) with carboxylate
anions are correlated with the ionization constant of the
carboxylate buffer and are of a similar magnitude to those
determined for amines of similar basicity. Although by them-
selves they would give a lower Brønsted ꢀ value, they do give
a reasonable fit to the plot generated for the amine general base-
catalyzed hydrolysis (Figure 5).
The lack of reactivity of ꢀ-sultams toward amine nucleophiles
appears to be the intrinsic reactivity of ꢀ-sultams and not related
to their absolute reactivity as indicated by their rates of alkaline
hydrolysis. Presumably this is related to the mechanism of
sulfonyl transfer in four-membered rings compared with that
in analogous acyclic derivatives. However, a fundamental
difference between ꢀ-sultams and ꢀ-lactams is the enhanced
reactivity of greater than 106 toward hydroxide ion of the
sulfonyl derivative compared with acyclic analogues, whereas
ꢀ-lactams show little or no enhanced reactivity compared with
their acyclic analogues.7–9
The difference in behavior between the aminolysis of acyl
and acyclic sulfonyl derivatives on one hand and the hydrolysis
of ꢀ-sultams on the other in aqueous solutions of amines is
unlikely to be the result of different reactivity-selectivity
relationships. Although sulfonyl transfer reactions do appear to
show opposite behavior to acyl transfer, ꢀ-sultams show the
same unusual pattern. A simple selectivity parameter is the ratio
kOH/kw, representing the rate constants for the hydroxide ion
promoted and pH independent water hydrolysis, respectively.
We have noticed an inverse selectivity-reactivity relationship
with respect to hydrolysis and a plot of log kOH/kw(selectivity)
against log kOH(reactivity) (Figure 6) shows a good correlation
but, surprisingly, the more reactive sulfonylating agent is more
selective! However, ꢀ-sultams show a similar inverse behavior,
indicating that this is a general property of all sulfonyl transfer
reactions. The structure around the sulfonamide group in
2
2
from which the second-order rate constant, kRNH , was
2
determined to be 7.48 × 10-2 M-1 s-1 (Table 1). The rate
constant for the aminolysis of 10 by taurine is 375-fold slower
than that for the general base-catalyzed hydrolysis of the
ꢀ-sultam of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8) by the
same amine. Nucleophilic attack at the acyl center of 8 would
displace a sulfonamide compared with an amide in 10, but
despite their difference in acidity, sulfonamides and amides
appear to behave similarly as leaving groups. For example,
the second-order rate constants for alkaline hydrolysis for
their displacement from acyl center are similar (Chart 1).9b,25
Although a weaker nucleophile than hydroxide ion, such as
an amine, may have a stronger dependence on the basicity
of the leaving group, it seems reasonable to assume that the
rate constants for the aminolysis of the acyl centers in 8 and
10 would be similar. It is therefore self-consistent with the
observed faster rate of the amine general base-catalyzed
hydroylsis of 8 by sulfonyl transfer rather than aminolysis
by acyl transfer.
(ii) Reactions with O-Nucleophiles. Having established that
hydrolysis of the active sulfonylating ꢀ-sultam occurs at a
faster rate than aminolysis, it was necessary to demonstrate
whether this was simply a preference for sulfonyl transfer
to O rather than N, and so the reaction in the presence of
O-nucleophiles/bases was studied. The rate of the reaction
of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8) in the presence
of carboxylate buffers was measured by UV-spectrophotom-
etry in 1% acetonitrile-water (v/v) and 1.0 M ionic strength
(KCl) at 30 °C. The dependence of the observed pseudo-
first-order rate constants, kobs, for the reaction of the 3-oxo-
ꢀ-sultam (8) in aqueous solution buffered with acetate on
total buffer concentration is linearly dependent on the total
acetate concentration. Extrapolation of the reaction rate to
zero buffer concentration corresponds to that calculated for
the background hydrolysis. The slopes of these plots give
the second-order rate constants, kcat, for the reaction by both
(25) Tsang, W. Y.; Ahmed, N.; Hemming, K.; Page, M. I. Can. J. Chem.
2005, 83, 1432–1439.
4510 J. Org. Chem. Vol. 73, No. 12, 2008

Hydrolysis of ActiVated Sulfonyl DeriVatiVes
SCHEME 4
concerted mechanism then aminolysis probably requires general
base catalysis to avoid formation of a N-protonated sulfonamide
product and proton removal from the attacking amine by base
necessitates a termolecular reaction. In a stepwise mechanism
this entropically unfavorable step can be avoided. The enhanced
reactivity of ꢀ-sultams toward hydroxide ion maybe the result
of rate-limiting formation of the trigonal bipyramidal intermedi-
ate, whereas nucleophilic attack of an amine on the sulfonyl
center is reversible and does not lead to reaction unless general
base catalysis occurs to remove a proton from the attacking
amine and/or the leaving group is expelled in a facile step. In
activated acyclic sulfonyl derivatives there is often a good
leaving group, whereas in ꢀ-sultams it is a poorer onesamide
anion or aminespresumably leading to a higher activation
energy and a later transition state with significant S-N bond
fission.
FIGURE 6. The relationship between selectivity (log kOH/kw) and
reactivity (log kOH) for the hydrolysis of sulfonyl derivatives (b) and
sulfate esters (o). Data are from: King, J. F.; Rathore, R.; Lam, J. Y. L.;
Guo, Z. R.; Klassen, D. F. J. Am. Chem. Soc. 1992, 114, 3028-3033.
ꢀ-sultams is very different from that in acyclic derivatives for
which the stability is very dependent on the hybridization of
the sulfonamide nitrogen.26 Furthermore, the mechanism of bond
fission in four-membered rings may not be a stretching motion
but rather a bond rotational mode that may lead to differences
in behavior.27
The mechanisms for sulfonyl transfer reactions involving the
displacement of a leaving group L by a nucleophile Nu may
occur in an associative or dissociative process (Scheme 4).7 The
dissociative SN1(S) process would generate a sulfonylium ion
that is then subsequently attacked by a nucleophile. Although
the evidence for this mechanism on sulfonyl transfer is ambigu-
ous,5 it has been proposed to occur in the acid-catalyzed
hydrolysis of ꢀ-sultams as a result of the release of ring strain
upon ring-opening.9d However, in general it appears that
sulfonylium ions are more difficult to generate than acylium
ions,28 and this is unlikely to be relevant to the present
discussion. Associative, stepwise mechanisms involving forma-
tion of unstable intermediates are the norm for acyl transfer
whereas the concerted process remains controversial.16 By
contrast, sulfonyl transfer is usually discussed in terms of a
concerted displacement and it is the evidence for a stepwise
process that is questioned.29 Most observations can be inter-
preted in terms of either a stepwise or a concerted mechanism.
Although there is no clear evidence for the formation of an
intermediate, the transition state and/or the intermediate in the
associative mechanism adopt a trigonal bipyramidal geometry
and it appears that there are some requirements for pseudoro-
tation30 and apical displacement of the leaving group by a
nucleophile.9b
Experimental Section
The methods of preparation for N-benzyol-ꢀ-sultam (6), N-ben-
zyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8), and N-benzyl-3,3-dimethyl-
2-oxo-ꢀ-lactam (10) have been given previously.9a,31,32
Standard UV spectroscopy was carried out on UV-visible
spectrophotometers in double beam mode with cell holders ther-
mostated by using a peltier system. In all experiments temperatures
were maintained at 30 °C and ionic strength at 1.0 M with AnalaR
grade KCl unless otherwise stated.
The pH of solutions of an amine was controlled by the use of
e0.2 M buffer solutions of the amine. Buffer sulutions were
prepared by partial neutralization of solutions of their neutral form
to the required pH. Hydroxide ion concentrations were calculated
by using pKw(H2O) ) 13.83 at 30 °C. AnalaR grade reagents and
deionized water were used throughout. Organic solvents were glass
distilled prior to use and stored under nitrogen.
Reactions studied by UV spectrophotometry were commenced
by injections (20 µL) of acetonitrile stock solutions (1 × 10-2 M)
of the substrate into the cells containing preincubated buffer (2.0
mL). Final reaction cells contained e1% acetonitrile v/v. The pH
of the reaction cells was measured before and after each kinetic
run at 30 °C, and kinetic runs experiencing a change >0.05 units
were rejected. Reactant disappearance or product appearance were
followed at absorbance change maxima for individual compounds.
The lack of aminolysis is due to either an enhanced reactivity
of ꢀ-sultams toward oxygen nucleophiles and hydroxide ion in
particular or to a relatively reduced activity of amine nucleo-
philes to the sulfonyl center. If sulfonyl transfer occurs by a
(29) (a) Ciuffari, E.; Senatore, L.; Isola, M. J. Chem. Soc., Perkin Trans. 2
1972, 468. (b) Deacon, T.; Farrar, C. R.; Sikkel, B. J.; Williams, A. J. Am.
Chem. Soc. 1978, 100, 2525–2534. (c) Rogne, O. J. Chem. Soc., Perkin Trans.
2 1975, 1486–1490. (d) King, J. F.; Gill, M. S.; Klassen, D. F. Pure Appl. Chem.
1996, 68, 825–830.
(26) Breneman, C. M.; Weber, L. W. Can. J. Chem. 1996, 74, 1271–1282.
(27) Page, M. I. Philos. Trans. R. Soc. London, Ser. B 1991, 332, 149–156.
(28) (a) Bentley, T. W.; Jones, R. O. J. Phys. Org. Chem. 2007, 20, 1093–
1101. (b) Olah, G. A.; Kobayash, S.; Nishimur, J. J. Am. Chem. Soc. 1973, 95,
564-569. (c) Kice, J. L. Prog. Inorg. Chem. 1972, 17, 147–206.
(30) Kaiser, E. T. Acc. Chem. Res. 1970, 3, 145.
(31) Tsang, W. Y.; Ahmed, N.; Hemming, K.; Page, M. I. Org. Biomol.
Chem. 2007, 5, 3993.
(32) Mulchande, J.; Guedes, R. C.; Tsang, W.-Y.; Page, M. I.; Moreira, R.;
Iley, J. J. Med. Chem. 2008, 51, 1783.
J. Org. Chem. Vol. 73, No. 12, 2008 4511

Tsang et al.
The solubility of compounds was ensured by working within the
linear range of absorbance in corresponding Beer-Lambert plots.
Reaction concentrations were generally within the range of 2 ×
10-5 to 2 × 10-4 M. Pseudo-first-order rate constants from
exponential plots of absorbance against time or gradients of initial
slopes were obtained by using the Cary Win UV kinetics application
(Version 02.00(26)).
Supporting Information Available: Plots of the kinetic data
for the reaction of N-benzyl-4,4-dimethyl-3-oxo-ꢀ-sultam (8)
with 2-cyanoethylamine and acetate ion. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO800407X
4512 J. Org. Chem. Vol. 73, No. 12, 2008