Catalysis of the methanolysis of activated amides by divalent and trivalent metal ions. The effect of Zn2+, Co2+, and La3+ on the methanolysis of acetylimidazole and its (NH3)5CoIII complex

Article abstract of DOI:10.1021/ja0023940

The metal ions Zn²⁺, Co²⁺, and La³⁺ strongly catalyze the methanolysis of the activated amides acetylimidazole (1) and its ligand-exchange-inert Co^III complex, (NH₃)₅Co^III-AcIm (2). Studies of the kinetics of methanolysis are performed with _s/^spH measurement and control, and the metal ions are soluble in the medium throughout the _s/^spH regions where ionization of the M^x+(CH₃OH)_y occurs. Zn²⁺ and Co²⁺ act as Lewis acids toward 1, catalyzing attack of external methoxide on a 1:M²⁺ complex at values only 100-fold lower than the diffusion limit, the k_OR values being 5.6 × 10⁷ M^-1 s^-1 and 2.5 × 10⁷ M^-1 s^-1, while that for CH₃O^- attack on 2 is 4.69 × 10⁷ M^-1 s^-1. Since neither Zn²⁺ nor Co²⁺ promotes the methanolysis of 2, these metals appear to be acting through transient binding to the distal N of 1, which activates the C=O of the complex to external CH₃O^- attack. La³⁺ catalyzes the methanolysis of both 1 and 2, which occurs by a mechanism that is fundamentally different from that exhibited by Zn²⁺ and Co²⁺ in that the active species appears to be a bis-methoxy-bridged dimer (La³⁺)₂(CH₃O^-)₂ (CH₃OH)_x that interacts directly with the C=O unit of the substrate.

Full text of DOI:10.1021/ja0023940

210
J. Am. Chem. Soc. 2001, 123, 210-217
Catalysis of the Methanolysis of Activated Amides by Divalent and
Trivalent Metal Ions. The Effect of Zn2+, Co2+, and La3+ on the
Methanolysis of Acetylimidazole and Its (NH3)5CoIII Complex
Alexei A. Neverov, Pedro J. Montoya-Pelaez, and R. S. Brown*
Contribution from the Department of Chemistry, Queen’s UniVersity, Kingston, Ontario, K7L 3N6 Canada
ReceiVed July 3, 2000
Abstract: The metal ions Zn2+, Co2+, and La3+ strongly catalyze the methanolysis of the activated amides
acetylimidazole (1) and its ligand-exchange-inert CoIII complex, (NH3)5CoIII-AcIm (2). Studies of the kinetics
s
of methanolysis are performed with pH measurement and control, and the metal ions are soluble in the
s
s
s
medium throughout the pH regions where ionization of the Mx+(CH3OH)y occurs. Zn2+ and Co2+ act as
Lewis acids toward 1, catalyzing attack of external methoxide on a 1:M2+ complex at values only 100-fold
lower than the diffusion limit, the kOR values being 5.6 × 107 M-1 s-1 and 2.5 × 107 M-1 s-1, while that for
CH3O- attack on 2 is 4.69 × 107 M-1 s-1. Since neither Zn2+ nor Co2+ promotes the methanolysis of 2, these
metals appear to be acting through transient binding to the distal N of 1, which activates the CdO of the
complex to external CH3O- attack. La3+ catalyzes the methanolysis of both 1 and 2, which occurs by a
mechanism that is fundamentally different from that exhibited by Zn2+ and Co2+ in that the active species
appears to be a bis-methoxy-bridged dimer (La3+)2(CH3O-)2(CH3OH)x that interacts directly with the CdO
unit of the substrate.
w
w
above the Mx+(H2O)y pKa has, in many cases,1 been circum-
Introduction
vented through binding of the metal ions to complexing ligands,
but even so it is usually difficult to study the reactions under
highly basic conditions.
Metal ion catalysis of the hydrolysis1 of esters and amides
in aqueous solution has been well studied due to the biological
implications of these and the fact that many Zn2+-containing
enzymes promoting these reactions are known.2 One of the
major stumbling blocks in understanding metal ion catalysis of
We have embarked on an extensive investigation of metal
ion catalysis of methanolysis of carbon and phosphorus esters
and amides. Among organic solvents, methanol is the closest
to water in structure and properties, so in some ways metha-
nolysis reactions are anticipated to resemble hydrolysis reactions.
However, there can be important differences, particularly for
metal-ion-promoted reactions, attributable to a reduced dielectric
constant relative to water (31.5 vs 78.5 at 25 °C)4 that might
be relevant to the local dielectric constant of the enzyme
interior.5 Nevertheless, methanolysis reactions have not been
studied nearly so extensively as hydrolysis reactions, probably
due to difficulties in determining and controlling pH in
methanol. Recent reports by Bosch and co-workers6 have shown
that the pH in methanol spH can be measured simply by
hydrolyses is the fact that, above the wpKa’s of the metal-aquo
w
species, formation of metal hydroxide precipitates and gels often
complicates mechanistic analysis.3 The problem of precipitation
(1) Przystas, T. J.; Fife, T. H. J. Chem. Soc., Perkin Trans. 2 1990, 393
and references therein. (b) Suh, J. Acc. Chem. Res. 1992, 25, 273 and
references therein. (c) Tecilla, P.; Tonellato, U.; Veronese, F.; Fellugg, F.;
Scrimin, P. J. Org. Chem. 1997, 62, 7621. (d) Chin, J.; Jubian, V.; Mrejen,
K. J. Chem. Soc., Chem. Commun. 1990, 1326. (e) Meriwether, L.;
Westheimer, F. J. Am. Chem. Soc. 1956, 78, 5119. (f) Grant, T. J.; Hay, R.
W. Aust. J. Chem 1965, 18, 1189. (g) Groves, J. T.; Dias, R. M. J. Am.
Chem. Soc. 1979, 101, 1033. (h) Groves, J. T.; Chambers, R. R. J. Am.
Chem. Soc. 1984, 106, 630. (i) Groves, J. T.; Olson, J. R. Inorg. Chem.
1985, 24, 2715. (j) Sayre, L.; Reddy, K. V.; Jacobson, A. R.; Tang, W.
Inorg. Chem. 1992, 31, 935. (k) Fife, T. H. Acc. Chem. Res. 1993, 26, 325.
(l) Fife, T. H.; Bembi, R. J. Am. Chem. Soc. 1993, 115, 11358. (m) Parac,
T. N.; Kostic´, N. M. J. Am. Chem. Soc. 1996, 118, 51. (n) Singh, L.; Ram,
R. N. J. Org. Chem. 1994, 59, 710. (o) Sutton, P. A.; Buckingham, D. A.
Acc. Chem. Res. 1987, 20, 357. (p) Kimura, E.; Nakamura, I.; Koike, T.;
Shionoya, M.; Kodama, Y.; Ikeda, T.; Shiro, M. J. Am. Chem. Soc. 1994,
116, 4764. (q) Parac, T. N.; Ullman, G. M.; Kostic´, N. M. J. Am. Chem.
Soc. 1999, 121, 3127 and references therein. (r) Ridder, A. M.; Kellogg,
R. M. In Supramolecular ReactiVity and Transport: Bioorganic Systems;
Comprehensive Supramolecular Chemistry Vol. 4; Murakami, Y., Ed.;
Elsevier Science Ltd.: Oxford, 1996; Chapter 11. (s) Chin, J. Acc. Chem.
Res. 1991, 24, 145 and references therein.
(
s
)
using a glass electrode, and they have presented an extensive
list6b of spKa values for species that would be useful as
s
buffering agents in methanol. While originally investigated for
applications in HPLC separations, these contributions suggest
s
s
that it should now be possible to run detailed pH vs rate
studies for a variety of important processes in buffered methanol
as easily as in water. To our knowledge, this does not appear
to have been attempted even for simple systems.
(4) Harned, H. S.; Owen, B. B. The Physical Chemistry of Electrolytic
Solutions, 3rd ed.; ACS Monograph Series 137; Reinhold Publishing: New
York, 1957; p 161.
(2) Lipscomb, W.; Stra¨ter, N. Chem. ReV. 1996, 96, 2375. (b) Coleman,
J. E. Curr. Opin. Chem. Biol. 1998, 2, 222.
(3) Rustad, J. R.; Dixon, D. A.; Rosso, K. M.; Felmy, A. R. J. Am. Chem.
Soc. 1999, 121, 324. (b) Baes, C. F.; Mesmer, R. E. Hydrolysis of Cations;
R. E. Krieger Publishing Co.: Malabar, FL, 1976. (c) Smith, R. M.; Martell,
A. E. Inorganic Complexes; Critical Stability Constants Vol. 4; Plenum
Press: New York, 1976. (d) Richens, D. T. The Chemistry of Aquo Ions;
John Wiley and Sons: New York, 1997.
(5) Ferscht, A. Enzyme Structure and Function, 2nd ed.; W. H. Freeman
and Co.: New York, 1985; pp 64-69.
(6) Bosch, E.; Rived, F.; Roses, M.; Sales, J. J. Chem. Soc., Perkin Trans.
2, 1999, 1953. (b) Rived, F.; Rose´s, M.; Bosch, E. Anal. Chim. Acta 1998,
374, 309. (c) Bosch, E.; Bou, P.; Allemann, H.; Roses, M. Anal. Chem.
1996, 3651.
10.1021/ja0023940 CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/10/2001

Catalysis of the Methanolysis of ActiVated Amides
J. Am. Chem. Soc., Vol. 123, No. 2, 2001 211
(prepared from stock 0.5 M methoxide provided in a Sure Seal bottle)
were (1.00-1.33) × 10-3 and (1.88-2.00) × 10-3 M, respectively.
The latter was calibrated by titrating standardized HCl with the endpoint
taken to be sspH ) 8.38. The values of the dissociation constants
spK were calculated by using the computer program PKAS.10 The
In the following we report a detailed kinetic investigation of
the methanolysis of the activated amide, acetylimidazole (1),7
and its ligand-exchange-inert Co3+ complex, N-acetylimidazole-
pentaaminecobalt(III), (AcImCo(NH3)53+, 2).8 These substrates
(
)
s
a
wpKw in the program was replaced by the autoprotolysis constant for
w
methanol at 25 °C (spKMeOH ) 16.77).6 In the cases of Zn2+ and Co2+
,
s
the titration data were treated according to eq 3 (below), and values of
K
app (the product of the two dissociation constants for sequential removal
of CH3O- from M2+(CH3O-)2) were calculated by an NLLSQ fit of
the experimental data to the following equation:
were chosen in part because the mechanisms of their hydrolyses
are well understood,7,8 which allows ready comparison with their
methanolyses, and in part because, despite extensive study, it
has yet to be shown that 1 undergoes significant metal-ion-
promoted acceleration of its hydrolysis. Fife and co-workers
have shown1a,k,l that N-acylimidazoles and N-acylbenzimidazoles
sspH ) sspKMeOH - sspKapp/2 + 0.5 log(VbaseCbase/(2Vinit[M2+
]
-
total
VbaseCbase))
where V represents volume and C concentration. sKa values were
s
s
s
s
s
calculated as Ka ) KMeOH/sKapp
.
1/2
s
that incorporate proximal metal-binding sites do undergo M2+
-
Kinetic Measurements. The rate of disappearance of 1 was followed
by monitoring the decrease in absorbance of buffered methanol solutions
at 240 nm with an OLIS modified Cary 17 UV-vis spectrophotometer
or an Applied Photophysics SX-17MV stopped-flow reaction analyzer
at 25.0 ( 0.1 °C. Reactions were monitored under pseudo-first-order
catalyzed hydrolyses, but in the absence of such binding sites,
or when metal binding occurs away from the scissile (N)CdO
unit, catalysis is not observed. Indeed, such binding appears to
be a general requirement for strong metal ion catalysis of the
hydrolysis of more normal amides as well,1f-j,m-q since those
without proximal binding sites typically do not exhibit metal
ion catalysis of hydrolysis. As will be shown, strong metal ion
catalysis of the methanolyses of 1 and 2 by Zn2+, Co2+, and
La3+ is observed, so that a proximal metal-binding site appears
unnecessary in MeOH solution although amides containing such
may well be even more susceptible to metal-ion-promoted
methanolysis. Furthermore, our studies show that, in MeOH,
the La3+ and Zn2+ metal ions remain in solution throughout
the entire pH domain where Mx+(MeOH)y ionizes to form metal-
bound methoxides, thus avoiding the problems of insoluble
metal-containing precipitates inherent in analogous studies in
water.
conditions of excess metal ions in the range of (0.2-8.0) × 10-3
M
and, in the case of slow reactions using conventional UV analysis, were
initiated by addition of an aliquot of a 2.5 × 10-2 M stock solution of
1 in CH3CN to 2.5 mL of the buffered reaction mixture. The final
concentration of amide was in the range (0.4-1.0) × 10-4 M. Reactions
were followed for at least four half-lives and displayed good first-order
behavior. Pseudo-first-order rate constants (kobs) were determined by
NLLSQ fitting of the absorbance vs time traces to a standard
exponential model. N-Methylimidazole (spKa ) 7.60), collidine (spKa
s
s
) 7.72), N-methylmorpholine (spK a ) 8.28), trimethylamine (spK a )
s
s
9.806b), triethylamine (sspKa ) 10.786b), and cyclohexylamine (sspKa )
11.686b) buffers, partially neutralized with HClO4, were used to control
the sspH. Total buffer concentrations varied between 10-3 and 10-2 M.
Oxalic acid (sspKa ) 5.546b) was used to control sspH in the range 4-6
for methanolysis of 1 and 2. All other measurements for these
s
s
s
s
compounds between pH ) 1 and 4, and above pH ) 13, were done
Experimental Section
at constant [HClO4] and [NaOCH3]. sspH measurements were per-
formed before and after each experiment. To avoid any chloride ion
Materials. N-Acetylimidazolepentaaminecobalt(III) perchlorate,
[(NH3)5CoImCOCH3](ClO4)3 (2), was prepared as previously described.8
Methanol (99.8% anhydrous), sodium methoxide (0.5 M), HClO4 (70%,
BDH), acetonitrile (99.8% anhydrous), acetylimidazole, Zn(ClO4)2‚
6H2O, Co(ClO4)2‚6H2O, and La(CF3SO3)3 were all purchased from
Aldrich and used without any further purification.
contamination from the glass electrode which was found to affect the
metal ion reactions, duplicate solutions were prepared: one for pH
measurements, and the second for kinetics studies. In all cases, pH
values measured before and after reaction were consistent to within
0.05 unit.
s
s
s
s
s
+
spH Measurements. The CH3OH2 concentration was determined
Due to the high sensitivity of [(NH3)5CoImCOCH3](ClO4)3 to base-
catalyzed methanolysis, special precautions were used. Stock solutions
of [(NH3)5CoImCOCH3](ClO4)3 were prepared in methanol containing
10-2 M HClO4, sonicated until all solid had dissolved, and then stored
in an ice bath. For the La3+ reactions followed using the Applied
Photophysics SX-17MV stopped-flow reaction analyzer, one drive
syringe was charged with the slightly acidic (sspH ) 4) methanol
solution of [(NH3)5CoImCOCH3](ClO4)3. The second drive syringe
contained twice the desired concentration of La3+ salt and buffer. After
mixing, the final substrate concentrations were between 1 × 10-4 and
5 × 10-4 M. Reactions were monitored at 240 nm, and the pseudo-
first-order rate constants for methanolysis of 2 were determined in the
same fashion as for 1.
by using a Radiometer GK2322 combination (glass/calomel) electrode
calibrated with Fisher Certified standard aqueous buffers (pH ) 4.00
s
s
and 10.00). Values of pH were calculated by adding the correction
constant (2.24) to the experimental meter reading s pH . This method
(
)
w
was first described by Bates9 for a molality scale (correction constant
2.34), and later by Bosch et al.6 for a molar correction constant.
The spKa’s of buffers used for the present kinetic studies were
s
measured at half-neutralization of the bases with 70% HClO4 in MeOH.
spKa Determination. The potentiometric titrations of the above
s
metal salts in methanol were performed using a Radiometer Vit 90
autotitrator under anaerobic conditions (Ar) at ambient temperature.
The concentrations of metal salts and sodium methoxide titrant
(7) Jencks, W. P.; Carriuolo, J. J. Biol. Chem. 1959, 234, 1272. (b)
Wolfenden, R.; Jencks, W. P. J. Am. Chem. Soc. 1961, 83, 4390. (c)
Oakenfull, D. G.; Jencks, W. P. J. Am. Chem. Soc. 1971, 93, 178. (d)
Oakenfull, D. G.; Jencks, W. P. J. Am. Chem. Soc. 1971, 93, 188. (e) Fife,
T. H. J. Am. Chem. Soc. 1965, 87, 4597. (f) Palaitis, W.; Thornton, E. R.
J. Am. Chem. Soc. 1975, 97, 1193. (g) Hogg, J. L.; Phillips, M. K.; Jergens,
D. E. J. Org. Chem. 1977, 42, 2459. (h) Oakenful, P. G.; Jencks, W. P. J.
Am. Chem. Soc. 1971, 93, 178. (i) Huskey, W. P.; Hogg, J. L. J. Org. Chem.
1981, 46, 53-59.
Results
a. Methanolysis of 1 and 2 in the Absence of Metal Ion.
Shown in Figure 1 are logarithmic plots of the observed rate
constants (kobs) vs spH for methanolysis of 1 and 2 in buffered
s
s
s
methanol at 25 °C. The kobs values were obtained at each pH
using three to five buffer concentrations (in duplicate) between
10-3 and 10-2 M and extrapolating to zero buffer concentration.
(8) Harrowfield, J. MacB.; Norris, V.; Sargeson A. M. J. Am. Chem.
Soc. 1976, 98, 7282.
(9) Bates, R. Determination of pH. Theory and Practice; Wiley: New
York, 1973.
(10) Martell, A. E.; Motekaitis, R. Determination and Use of Stability
Constants; VCH: New York, 1988.

212 J. Am. Chem. Soc., Vol. 123, No. 2, 2001
NeVeroV et al.
Table 1. Rate Constants for Solvolysis of Acetylimidazole (1) and
[(NH3)5CoImCOCH3](ClO4)3 (2) at 25 °C in Water and Methanol at
Various Ionic Strengths
wpKa
wor
105 k0
kH
kOR
+
substrate
solvent
(s-1
8
)
(M-1 s-1
)
(M-1 s-1
)
spKa
s
1
H2O (I ) 0.2 M)a
1.87 × 102 e 320
3.6
(I ) 0.5 M)b
5.6 3.2 × 102 e
290
(I ) 1 M)c
4.6
(I ) 0 M)d
30
MeOH (I ) 0 M)
2.6 5.2 × 102 e 7900
4.2
2
H2O (I ) 1 M)c 110 5 × 10-4
6860
MeOH (I ) 0 M)
8.7 8.3 × 10-3 4.69 × 107
a Reference 7a. b Reference 7f. c Reference 8. d Deiko, S.; Neverov,
A.; Yatsimirskii, A. Kinet. Catal. 1989, 30, 694-698. e kH ) k1/sKa
+
s
or k1/wKa, depending on whether the solvent is methanol or water.
w
Figure 1. Logarithmic plot of observed rate constant of acetylimidazole
1 (b) and acetylimdazolepentaaminecobalt(III) perchlorate 2 (9)
s
s
methanolysis vs pH at 25 °C. Curve calculated by nonlinear fit to eq
1 for 1 and eq 2 for 2.
Scheme 1. Methanolysis of 1 (a) and 2 (b)
Methanolysis of 1 exhibited general-base-assisted buffer ca-
talysis similar to that in water7h (Bro¨nsted â constant in MeOH
) 0.89 ( 0.04), while that of 2 showed only minor buffer
retardation, so extrapolation to [buffer] ) 0 was not required
for the latter. We note that, for these measurements, the ionic
strength was not kept constant through the addition of salts but
varied in proportion to the [buffer]. The extrapolated values for
1 thus are at I ) 0, while those for 2 are at ionic strength I <
0.02. In addition, there is some small quantity of water present,
the amount of <0.1% stemming from that present in the supplied
MeOH and that derived from buffer neutralization with 70%
HClO4. The presence of water at this low concentration does
not influence the kinetics, vide infra.
s
The pH-rate constant plots for 1 and 2 can be accom-
s
Figure 2. (a) pH titration of Zn(ClO4)2 (b) and Co(ClO4)2 (9) in
modated by the mechanisms presented in Scheme 1, for which
s
methanol. (b) pH titration of 1.33 × 10-3 M La(OTf)3 in methanol.
s
can be derived the rate expression given in eq 1, where k1 and
1, and the constants given in Table 1, the autoprotolysis
kobs ) k [H+]/ sK (n) + [H+] +
(
(
))
s
1
s a
constant6 for MeOH being set at pKMeOH ) 10-16.77 M2.
s
(k0 + kOR sKMeOH/[H+]) sK (n)/ sK (n) + [H+] (1)
b. Metal Ion Titration. Potentiometric titration data for 1
× 10-4 M Zn(ClO4)2 and Co(ClO4)2 in methanol are shown in
Figure 2a. Both metals show a steeper than standard titration
curve with the consumption of 2 equiv of methoxide per metal.
Equation 3 describes the process for double dissociation of the
metal dimethoxide to form the probable hexakis(methanol)M2+
ions,11 although the subscripts x and y refer to an unspecified
s
(
(
))
s
a
s a
k0 represent the rate constants for attack of solvent on the
protonated and neutral species, and kOR that for attack of
methoxide on 1or 2. In the case of 2, since the kinetics are
determined at [H+] , Ka, the equation simplifies to the form
s
(2)
+
given in eq 2, where kH ) k1/sKa
.
s
(CH3OH)xM2+(CH3O-)2 ysKappz M2+(CH3OH)y + 2CH3O-
kobs ) kH+[H+] + k + kOR sKMeOH/[H+]
(2)
s
0
(3)
NLLSQ fitting of the kobs vs [H+] data for each amide to its
respective equation generates the lines through the data in Figure
s
s
number of solvent molecules on the M2+. The pK a’s calcu-

Catalysis of the Methanolysis of ActiVated Amides
J. Am. Chem. Soc., Vol. 123, No. 2, 2001 213
Table 2. Rate and Equilibrium Constants for Zn2+- and Co2+-Catalyzed Methanolysis of Acetylimidazole (1), T ) 25 °C
d
metal
Zn2+
ssKappkinetic (M2)
10-14.21(0.04
ssKapptitration (M2)a
ssKakinetic (M)b (sspKa)
ssKatitration (M)c (sspKa)
k′OR (M-2 s-1
)
10-14.04
10-9.67(0.04
(9.67)
10-9.75
(9.75)
(5.6 ( 0.6) × 108
(2.5 ( 1.3) × 108
Co2+
10-11.89 (0.08
10-10.84
10-10.82(0.08
(10.82)
10-11.35
(11.35)
s
s
s
s
a From titration fits to eq 3. sKa ) sKMeOH/sKapp1/2. Kappkinetic determined from fits of the kinetic data to eq 4; pKa refers to the pH at which 50%
b
s
s
s
of the M2+ is in the inactive M2+(CH3O-)2 form. sKa ) KMeOH/sKapp1/2. Kapp
determined from fits of titration data to eq 3. d k′OR ) kOR/KM,
s
s
c
titration
s
s
s
s
see text.
s
s
lated from the titration data (defined as the pH at which 50%
of the M2+ is in the inactive form, see Table 2) are 9.75 (Zn2+
)
and 11.35 (Co2+). In a separate set of experiments, the
absorption spectrum for Co2+ was also monitored as a function
s
s
of pH to investigate possible ionization-dependent transfor-
mations from octahedral (pink) to tetrahedral coordination or
pentacoordination (blue or violet). Progressing through the sspH
range of 4.2-10.4, the Co2+ solutions ((2.5-5.0) × 10-3 M)
turned yellow; no blue or violet was detected. At these
s
s
concentrations, precipitates are formed at higher pH.
Given in Figure 2b is the potentiometric titration curve for
La(OTf)3 (1.33 × 10-3 M) in methanol, showing two ionization
regions. The first, between sspH 6 and 9, results in consumption
of 1 equiv of CH3O- per metal ion and can be treated as a
s
s
s
s
normal pKa, (spKa ) 7.22 ( 0.15). The second, around pH
s
10, consumes 1.5 equiv of CH3O- per La3+, corresponding to
a total of 2.5 equiv of H+ per La3+ being released between sspH
6 and 12. This result can be explained by the formation of
dimeric species at high sspH comprising two La3+ ions and five
methoxide anions, analogous to what was proposed by Takasaki
and Chin for a lanthanide dimer bridged by 5 OH- in water.12
Nonlinear fitting of the sspH/titrant data for the proposed
dinuclear lanthanide complex with five ionizable protons using
the PKAS program10 shows very good correlation with experi-
mental data (Figure 2b).
Figure 3. Logarithmic plot of second-order rate constants (k2) of
acetylimidazole methanolysis in the presence of Zn2+ (b) and Co2+
s
s
(9) against pH. Curve calculated by nonlinear fit of data to eq 4.
Scheme 2
much lower [Co2+] (10-5-10-4 M), with the aliquot of the stock
Co2+ solution added just before measurements. Even so, for
s
the Co2+-catalyzed reactions at high pH, we were forced to
c. Metal Ion Catalysis of Methanolysis. The water content
present in the reaction mixture comes from three sources: (i)
water present in the methanol (0.002%), (ii) the six hydrates
s
use stopped-flow kinetics because of the speed of the reaction.
This in turn necessitates prolonged sample handling, with the
possible consequence that some oligomerization of the Co2+
-
present in the metal perchlorates, and (iii) the water present in
s
s
(OR-)x occurs, causing a shift of the kinetic sspKa toward lower
70% HClO4 used to control pH. The first source is insignifi-
s
spH values, explaining the slight difference of the pKa kinetic
cant compared to the other two. In a solution containing 8 ×
10-3 M buffer, the water present due to the acid varied from
0.005 to 0.02%, while the metal perchlorates contribute between
0.001 and 0.08%, depending on the concentration. Several
kinetic experiments were performed using Zn(ClO4)2‚6H2O, Zn-
(ClO4)2‚H2O (produced by drying the Zn(ClO4)2‚(H2O)6 under
vacuum over P2O5 for 3 days), and Zn(CF3SO3)2, and in all
s
s
and titrimetric values presented in Table 2.
We did not detect M2+ catalysis of the methanolysis of 2 (2
s
× 10-4 M) at pH ) 7.80 (N-methylimidazole buffer, 4 ×
s
10-3 M). In fact, both Zn2+ and Co2+ (up to 4 × 10-4 M) gave
linear decelerations of the kobs, probably due to an ionic strength
effect.
cases the kobs values obtained were statistically equivalent.
Second-order rate constants (k2) for metal ion catalysis of
methanolysis of 1 were calculated from the slopes of the linear
s
Furthermore, for the Co2+-catalyzed reactions at pH ) 9.74,
s
s
s
s
s
plots of kobs vs [M2+] at each pH. The pH vs k2 profiles for
the trimethylamine buffer concentration was varied from 4 ×
10-3 to 24 × 10-3 M with no effect on kobs. Thus, the M2+ ion
catalysis is insensitive to water concentrations up to 0.1% and
also to buffer catalysis.
Zn2+ and Co2+ shown in Figure 3 increase linearly with unit
s
s
slope from pH ) 7-9.5 and 8-10.5, respectively, reaching
maximum values and then dropping. Shown in Scheme 2 is a
simplified mechanism that accounts for the pH dependency,
Both Zn2+ and Co2+ catalyze the methanolysis of 1, and at
s
s
s
each pH the kobs of methanolysis of 1 vs [M2+] gave a linear
s
where P indicates the products imidazole and methyl acetate,
plot with no evidence of saturation up to a metal ion concentra-
with the metal dialkoxide being inactive, explaining the drop
tion of 8 × 10-3 M. With Co2+, precipitation occurred with
s
s
in rate at high pH. The expression given in eq 4 relates the
s
concentrations as low as 2 × 10-3 M at pH > 10.8 but with
s
second-order rate constants for metal ion catalysis of the attack
lag times of 30 min to 12 h, depending on sspH and concentra-
tion. Kinetic runs above these sspH’s were thus performed with
s
of methoxide (k2) to [H+], where Kapp is the apparent dis-
s
s
s
k2 ) k′OR sKMeOH sKapp([H+]/([H+]2 sKapp + ssKMeOH2)) (4)
s
(11) Sudbrake, C.; Mu¨ller, B.; Vahrenkamp, H. Eur. J. Inorg. Chem.
1999, 2009.
(12) Takasaki, B. K.; Chin, J. J. Am. Chem. Soc. 1995, 117, 8582.
sociation constant for M2+(CH3O-)2 as in eq 3, k′OR is a third-

214 J. Am. Chem. Soc., Vol. 123, No. 2, 2001
NeVeroV et al.
Figure 5. Plots of pseudo-first-order rate constants vs [La3+] for the
methanolysis of AcIm (9, right axis, sspH ) 6.7) and AcImCo-
s
s
3+
(NH3)5 (b, left axis, pH ) 7.6, I ) 0.2 (NaClO4)) at 25 °C.
Figure 4. (a) Plot of pseudo-first-order rate constants for the
s
s
methanolysis of AcIm vs [La(OTf)3] at 25 °C; pH ) 9.1 (9) and pH
) 7.33 (b). (b) Plot of pseudo-first-order rate constants for the
s
s
methanolysis of Co(NH3)5AcIm vs.[La(OTf)3] at 25 °C; pH ) 7.95
Figure 6. Plots of second-order rate constant (k2) vs pH for La(OTf)3-
catalyzed methanolysis of AcIm (9) and AcImCo(NH3)53+ (b). Curves
are the result of NLLSQ fitting of experimental data to eq 7.
s
(9) and pH ) 7.6 (b).
s
order rate constant for attack of methoxide on a M2+:1 complex
(k’OR ) kOR/KM, where KM is the dissociation constant for the
putative M2+:1 complex) or a kinetic equivalent (vide infra),
Scheme 3
and sKMeOH is the autoprotolysis constant of methanol. NLLSQ
s
fitting of the k2 vs [H+] data to eq 4 generates the constants
given in Table 2 and the lines through the data in Figure 2.
Pseudo-first-order rate constants, kobs, as a function of [La3+
]
s
obtained for the methanolysis of 1 and 2 at a given pH show
s
As described above, potentiometric titration of La(OTf)3 (1.33
mmol) in methanol shows ionization of one proton per metal
strong catalysis by La(OTf)3, with the effect being sspH-
sensitive and linearly proportional to [La(OTf)3] at higher
concentrations of metal salt (see Figure 4) but with significant
curvature at concentrations lower than ∼2 × 10-4 M (see Figure
5). This indicates the involvement of second-order terms in [La-
(OTf)3]. Unfortunately, direct kinetic analysis of all the experi-
mental kobs vs [La3+]t data does not provide reliable parameters,
particularly in the case of 2 due to the high rate of the
background reaction. However, the fact that the kobs vs [La-
(OTf)3]t dependencies look very similar for both substrates
s
s
ion in the pH region of interest (6-9), with pKa ) 7.22 (
s
s
0.15 (see Figure 2b). In combination with the sspH-rate
profiles (Figure 6) and the kobs vs [La(OTf)3] dependencies
(Figure 5), this indicates that the kinetically active species is a
dimeric complex consisting of two La3+ ions and two methoxide
anions with an unspecified number of solvent molecules as
s
s
shown in Scheme 3, where Ka(1,2) refers to the product of the
first and second H+ dissociation constants of the dimer. The
equations for the equilibrium constants and the conservation of
mass pertaining to the process in Scheme 3 provide the
expression for the rate constant given in eq 5. The linear part
eliminates any possibility that the higher reaction order in [La3+
]
is due to complexation of a La3+ ion by the distal nitrogen of
1, and rather indicates a concentration-dependent change in
speciation of La(OTf)3 in methanol solution such as formation
of soluble dimers. Analysis of the linear parts of the plots (Figure
4) gives the second-order rate constants (k2), which are then
kobs ) k0 + (k ssKa(1,2)[H+]2/K ) 1 + 8 sK
+
s a(1,2)
((
(
d
[H+]2 K [La3+] /[H+]2 0.5 - 1 2/ 4 sK
+ [H+]2 (5)
2
)
)
) ( (s
))
d
t
a(1,2)
plotted versus sspH as in Figure 6, revealing similar profiles for
s
s
both 1 and 2 with apparent kinetic pK a’s of 7-7.5.
of the plots of kobs vs [La3+]t (Figure 5) used to calculate the

Catalysis of the Methanolysis of ActiVated Amides
J. Am. Chem. Soc., Vol. 123, No. 2, 2001 215
second-order rate constant k2 describe the conditions where
for the formation of the bis-halide compared to the mono-halide
in DMSO14 and methanol.15 Doe and Kitagawa15a determined
K1 and K2 stability constants for Zn-Cl in methanol of (7.76
( 0.18) × 103 and (1.74 ( 0.10) × 104 M-1. This unusual
effect was attributed to a change in the Zn2+ complex structure
accompanying the binding of Cl-, leading to a difference in
charge density between the central metal as it transforms from
an octahedral structure (Zn2+(MeOH)6) to a tetrahedral one,
ZnX2(MeOH)2, and the entropy effect from the decrease in the
number of coordinating MeOH molecules substituted by Cl-.
A similar sort of explanation likely accounts for the preferential
formation of Zn2+ and Co2+ dimethoxides. Clearly, association
of CH3O- with the metal would liberate several of its solvating
methanols also, contributing to the entropy effect. In the case
of Co2+(CH3O-)2, we have checked for the appearance of
tetrahedral or five-coordinate Co2+ by UV/vis spectroscopy, but
we have found no evidence for blue colors, suggesting that the
coordination must still be octahedral.
s
δkobs/δ[La3+]t is constant (eq 6), where R ) Ka(1,2)[H+]2/
s
k2 ) δkobs/δ[La]t ) kRâ(1 - 1/(1 + â[La3+]t)0.5) (6)
(16Kd(sKa(1,2) + [H+]2)2), and â ) 8Kd(sKa(1,2) + [H+]2)/[H+]2.
s
s
To satisfy the condition that δkobs/δ[La3+]t is constant, the range
of [La3+]t for analysis must be chosen such that (1 + â[La3+]t)0.5
must be much bigger than 1, thereby providing an expression
s
s
s
s
relating k2 to Ka(1,2) (eq 7). NLLSQ fitting of pH-k2 data
k2 ) k ssKa(1,2)/ 2 sK
+ [H+]2
(7)
( (s
))
a(1,2)
(Figure 6) to eq 7 gives the following parameters: k ) (1.50 (
s
0.1) × 103 M-1 s-1, Ka(1,2) ) 10(-14.8(0.1) M2 (kinetic pK a )
s
s
7.40) for 1; and k ) (1.42 ( 0.02) × 102 M-1 s-1, Ka(1,2)
)
s
10(-14.30(0.02) M2 (kinetic spK a ) 7.15) for 2. The lines
s
The titration shown in Figure 2b for La3+ indicates that there
through the data in Figure 6 are those generated by the fits using
these values.
is a stepwise association of one CH3O- per La3+ (spK a
)
s
s
7.22) and a second event occurring between pH 10 and 11,
s
consuming 1.5CH3O-/La3+. These data, taken together with the
curved kobs vs [La3+] plots for methanolysis given in Figure 5,
Discussion
suggest that a dimeric species is formed which we show later
a. Methanolysis of 1 and 2. The pH-rate profiles for
methanolysis of 1 and 2 (Figure 1) bear a striking resemblance
to those determined in water,7,8 suggesting similar mechanisms
having acid, neutral, and base domains for both reactions. The
s
is kinetically active between pH 6 and 9. Dimeric structures
s
are not unknown for lanthanum ions and were previously
observed in methanol solutions of LaCl3 by using La139 and
Cl35 NMR.16
kinetic sspKa of 4.2 determined for 1-H+ in methanol is slightly
c. Metal Ion Catalysis of Methanolysis. (i) Zn2+ and Co2+
.
w
above its water pKa (3.6),7 generally agreeing with the rela-
w
Zn2+ and Co2+ catalyze methoxide addition to 1 by a process
that is first order in each of [M2+] and [CH3O-]. The proposed
mechanism involves formation of inactive M2+(CH3O-)2 at high
[CH3O-], explaining the drop in catalysis seen in Figure 3 above
sspH 10 and 11 for Zn2+ and Co2+, respectively. The values
provided in Table 2 from the fits of the k2 data for metal ion
catalysis vs [H+] to eq 4 indicate a large third-order rate constant
(k’OR) for the reaction of (CH3O- + M2+ + 1) or various kinetic
equivalents such as CH3O- attack on a transient M2+:1 complex,
or reaction of M2+:CH3O- + 1. In what follows we will provide
evidence limiting these possibilities.
tive trends in sspKa for nitrogen-protonated bases in methanol6b
w
w
that show slight increases (0.2-0.4) compared to their pKa’s
in water. The most notable differences between the water and
methanol kinetic parameters for solvolysis of 1 given in Table
1 concern the attack of HOR (k0) and -OR (kOR), where
methanol is about 10 times poorer a nucleophile than water (at
I ) 0), but methoxide in methanol is 26 times more reactive
than hydroxide in water. Comparison of the data for the ligand-
exchange-inert and positively charged Co3+ complex 2 provides
evidence for a much stronger solvent effect on the attack of
lyoxide. In water, attack of hydroxide on 2 increases only 24-
fold relative to -OH attack on 1, but attack of CH3O- in
methanol is accelerated 5900-fold. The accelerated attack of
anions on cationic 2 in MeOH relative to that in H2O is
anticipated on electrostatic grounds since the attractive forces
between oppositely charged ions is inversely proportional to
the dielectric constant of the medium.13
In aqueous solution, there are generally two mechanisms that
are proposed for metal-ion-promoted reaction of OH- on Cd
O units: the metal-bound hydroxide (M2+-OH-) mechanism
and the metal plus external hydroxide (M2+ + OH-) mecha-
nism.1 It is often difficult to distinguish these kinetically
equivalent possibilities, but one way is to rule out a mechanism
because the rate constant for the reaction would exceed diffusion
control. This approach was used by Fife and Przystas17 to show
that the Cu2+-catalyzed hydrolysis of N-(6-carboxypicolinyl)-
benzimidazole must proceed through a complexed Cu2+-OH-.
Involvement of metal alkoxide as an active species in the
b. Potentiometric Titration of Zn2+, Co2+, and La3+ in
Methanol. Potentiometric titration of both Zn2+ and Co2+
s
consumes 2 equiv of CH3O- per metal by pH ∼10 and ∼12.
s
The steep profile for consumption of CH3O- shown in Figure
2a cannot be accommodated by the stepwise process given in
eq 8, X ) CH3O-, unless there is cooperativity in association
of the second methoxide such that K2 > K1. While this
M
2+-promoted methanolysis of 1 can be ruled out by a
combination of titration data indicating sharp consumption of
2 equiv of CH3O- by M2+ (Figure 2a) and by the unit slope of
the log plots of the second-order rate constant for metal ion
catalysis vs sspH shown in Figure 3. Were the M2+-OCH3- the
active form, it would necessarily be depleted through equilibrium
K1
\
K2
\z
(CH3OH)nM2+ + 2X- y
z (CH3OH)xM2+(X-) + X- y
(CH3OH)yM2+(X-)2 (8)
(14) Ahrland, S.; Bjo¨rk, N.-O. Acta Chem. Scand. A 1976, 30, 265. (b)
Ahrland, S.; Bjo¨rk, N.-O.; Portanova, R. Acta Chem. Scand. A 1976, 30,
270.
(15) Doe, H.; Kitagawa, T. Inorg. Chem. 1982, 21, 2272. (b) Doe, H.;
Shibagaki, A.; Kitagawa, T. Inorg. Chem. 1983, 22, 1639.
(16) Smith, L. S.; McCain, D. C.; Wertz, D. L. J. Am. Chem. Soc. 1976,
98, 5125.
phenomenon has not yet been described for metal alkoxides, it
is of note that zinc halides (X ) Cl, Br and I) show a preference
(13) Harned, H. S.; Owen, B. B. The Physical Chemistry of Electrolytic
Solutions, 3rd ed.; ACS Monograph Series 137; Reinhold Publishing: New
York, 1957; pp 42-58.
(17) Fife, T. H.; Przystas, T. J. J. Am. Chem. Soc. 1986, 108, 4631.

216 J. Am. Chem. Soc., Vol. 123, No. 2, 2001
NeVeroV et al.
s
s
Scheme 4
a plateau, with apparent kinetic pKa’s of 7.40 and 7.15 for 1
and 2 determined by fitting the data to eq 7. These kinetic values
are in close agreement with what is determined for the apparent
spKa by potentiometric titration, lending strong credence to the
s
validity of the suggested model. The fits also provide maximal
second-order rate constants of 1.5 × 103 and 1.42 × 102 M-1
s-1 for methanolysis of 1 and 2 catalyzed by the dimer.
Similar double-bridge structures were suggested for La2-
formation of inactive M2+(CH3O-)2 at increasing pH, such that
the log plots of the rate constant for metal ion catalysis vs sspH
would not have a unit slope but would be curved downward.
Thus, the most reasonable mechanism involves attack of
external methoxide on a transiently formed M2+:1 complex as
in Scheme 4, which is a slightly expanded version of Scheme
2. Notably, the plots of kobs vs [M2+] are strictly linear up to at
least 8 × 10-3 M and show no evidence of saturation binding,
suggesting a lower limit for the KM value of ∼0.1 M. We
attempted to determine a better value for the binding constant
20
Cl6(CH3OH)10 (with bridging chloride ions) and a La3+
-
peroxide complex (with bridging peroxide dianions).12 From a
study of the hydrolysis of an RNA model dimer promoted by
La3+ in water, Hurst, Takasaki, and Chin21 proposed a similar
catalytically active structure having two lanthanide ions bridged
by five HO-.
Comparison of the second-order rate constants (k) for La3+
-
promoted methanolysis of 1 and 2 clearly indicates that
coordination of the (NH3)5Co3+ to the distal nitrogen does not
provide any enhancement of the catalytic effect of La3+. Even
by performing a competition experiment18 between H+ and Zn2+
though 2 has a much better leaving group than does 1 (wpKa )
w
s
s
10.0 for -ImCo(NH3)53+ 8 vs 14 for Im- 22), the rate of its La3+
-
for acetylimidazole at pH ) 3.94. At this value there is no
appreciable methoxide reaction, but the acid-promoted reaction
might be perturbed to lower pH values due to equilibrium
binding of the metal ion. At a [Zn2+] of 2 × 10-2 M, no change
in the rate constant for acid-catalyzed methanolysis is observed,
again supporting a lower limit of ∼0.1 M for KM. Combining
this value with the k′OR ) kOR/KM terms given in Table 2 gives
rate constants for methoxide attack on 1:M2+ (kOR) of 5.6 ×
107 and 2.5 × 107 M-1 s-1 for Zn2+ and Co2+, respectively.
These values compare favorably with the rate constant for attack
of CH3O- on the CoIII complex 2, given in Table 1 as 4.69 ×
107 M-1 s-1. Of course, it is difficult to assess the relative modes
promoted reaction is 10-fold slower. This contrasts our results
for CH3O--catalyzed methanolysis of 2 given in Table 1, where
the second-order rate constant kOR ) 4.69 × 107 M-1 s-1 is
5900-fold higher than that for CH3O- on 1 (kOR ) 7.9 × 103
M-1 s-1). Strong acceleration of methoxide attack on 2 relative
to that on 1 can be explained by a combination of leaving group
effects and electrostatic interaction between the positively
charged substrate (2) and a negatively charged nucleophile. The
fact that the La3+-promoted methanolysis reaction is slower with
2 probably obviates attack of free methoxide ion, since that
should result in much higher rate constants for 2 than for 1,
contrary to what is observed. The slower reaction of the
lanthanide reaction with 2 is best accommodated by a mecha-
nism where the reactive species is the dimeric complex
(La3+2(CH3O-)2(CH3OH)y) (3), having a total positive charge
of 4+, wherein electrostatic repulsion causes a decrease in the
rate for the positively charged substrate 2 relative to neutral 1.
of activation of methoxide attack by coordinated (NH3)5Co3+
-
and (CH3OH)xM2+-, but the results clearly indicate that
coordination leads to a reaction which is only 100 times less
than the diffusion limit.19 The most reasonable site of com-
plexation, based on relative basicities, is the distal N and not
the CdO unit.
(ii) La3+. As shown in Figure 5, plots of the methanolysis of
both 1 and 2 vs [La3+] are curved upward, suggesting a process
that is bimolecular in metal ion. This observation, coupled with
the titration data indicating the consumption of one CH3O- per
La3+ (sspK a ) 7.22), indicates that the catalytically active form
is a La dimer with two associated methoxides, formulated simply
as 3 with an unspecified number of coordinated molecules. At
It has been shown in the case of a structurally similar Co3+
-
hydroxide dimeric complex that bridging hydroxides are not
nucleophilic, being at least 1010 times less active than their
deprotonated oxo-bridged forms.23 The same should be true for
a bridging CH3O-, but clearly the proposed methoxy-bridged
La3+ dimer (3) cannot be further deprotonated, so the only way
to generate an active nucleophile within the dimeric complex
requires temporary breakage of a La3+-methoxy bond. This
creates a nucleophilic/electrophilic center where the La3+-bound
methoxy can attack the CdO carbon, with the second La3+
functioning as a Lewis acid to stabilize the developing negative
charge on carbonyl oxygen as in 4.
increased [La3+], the kinetic plots shown in Figure 4 for both
1 and 2 become linear, the slope at each sspH being the second-
order rate constant, k2, for dimer-promoted methanolysis. Plots
s
s
of these vs pH (Figure 6) show a sharp rise (slope ∼2, as
required for the involvement of two methoxides), followed by
(18) Neverov, A. A. Ph.D. Dissertation, Moscow State University, 1991.
Neverov uses the competition between metal ion and H+ to displace the
(20) Smith, L. S.; McCain, D. C.; Wertz, D. L. J. Am. Chem. Soc. 1976,
98, 5125.
(21) Hurst, P.; Takasaki, B. K.; Chin, J. J. Am. Chem. Soc. 1996, 118,
9982.
(22) Sundberg, R. J.; Martin, R. B. Chem. ReV. 1974, 74, 471.
(23) Wahnon, D.; Lebuis, A.-M.; Chin, J. Angew. Chem., Int. Ed. Engl.
1995, 34, 2412. (b) Williams, N. H.; Lebuis, A.-M.; Chin, J. J. Am. Chem.
Soc. 1999, 121, 3341.
normal acid-catalyzed reaction of acetyl imidazole to lower sspH values. He
reports that for the reaction of acetyl imidazole in water with Zn2+and Ni2+
,
the respective Km values for the dissociation AcIm:M2+ T AcIm + M2+
are 3.5 × 10-2M ([Zn2+] ) 0.1 M, ZnCl2) and 7.3 × 10-3 M ([Ni2+] )
0.2 M, NiCl2).
(19) Eigen, M.; De Maeyer, L. Z. Electrochem. 1955, 59, 986.

Catalysis of the Methanolysis of ActiVated Amides
J. Am. Chem. Soc., Vol. 123, No. 2, 2001 217
Conclusions
one La3+ to reveal an open form of the dimer where one La3+
acts as a Lewis acid to activate attack of the second La3+-bound
methoxide.
Recent seminal investigations by Bosch and co-workers6 have
shown that sspH measurement and control with buffers in
methanol solution can be done as simply as in water. This opens
up the possibility for detailed kinetic studies of various acyl
and phosphoryl transfer reactions in methanol which can then
be readily compared with hydrolyses. In the above we have
shown the following:
In the past decade, much attention was focused on the
remarkable acceleration provided by lanthanide cations (Lnx+
)
for the hydrolysis of phosphate esters12,24 and, to a lesser degree,
amides and peptides.25 However, mechanistic study of these
Lnx+-promoted hydrolyses is hindered by the fact that, in water,
formation of insoluble hydroxides and gels of poor definition
is a problem, particularly under basic conditions.26 The results
presented above indicate that metal ion reactions can now be
studied in methanol without the formation of precipitates during
the kinetic analysis. Furthermore, in the case of La3+-promoted
methanolyses, a soluble and kinetically active bis-methoxy-
bridged La dimer is present in solution. It appears that the
methanol medium stabilizes formation of this dimer without the
necessity of coordinating it to stabilizing ligands. Given that
the site of reaction seems to be the CdO group of the substrate,
work is currently underway in our laboratories to see whether
these, and related, La3+ dimers will promote the methanolyses
of more biologically relevant species such as nonactivated esters,
amides, and phosphates.
1. Metal ions such as Zn2+and La3+ are soluble in methanol
throughout the entire pH range of interest without the formation
of insoluble Mx+(OR-)n precipitiates during the kinetics which
so often accompany deprotonation of metal-aquo complexes
in water. Co2+ is less soluble at high pH values, but the
Co2+(CH3O-)2 species is sufficiently soluble for a limited time
that allows one to determine the kinetic effect on methanolysis
of 1.
2. Unlike hydrolysis, methanolysis of the activated amide
acetylimidazole (1) is subject to marked metal ion catalysis by
Zn2+ and Co2+. In the case of the divalent metal ions the
catalysis does not involve M2+(CH3O-)n forms, but rather
involves attack of external methoxide on a metal:1 complex.
The rate constants for methoxide attack on Zn2+:1, Co2+:1, and
(NH3)5Co3+:1 are 5.6 × 107, 2.5 × 107, and 4.69 × 107 M-1
s-1 respectively, roughly 104-fold greater than CH3O- attack
on 1 itself (7.9 × 103 M-1 s-1). It is tempting to suggest that
the reason that the divalent metal ions so strongly promote
methanolysis but not hydrolysis is that attack of anions on the
positively charged M2+:1 transient is promoted by the increase
in electrostatic interaction in methanol relative to that in water.
The fact that these metal ions do promote the methanolysis of
1 but not 2 suggests that the site of complexation is the distal
Acknowledgment. The authors gratefully acknowledge the
financial support of the Natural Sciences and Engineering
Research Council of Canada. We also acknowledge the helpful
suggestions of a reviewer.
JA0023940
(24) Blasko, A.; Bruice, T. C. Acc. Chem. Res. 1999, 32, 475 and
references therein. (b) Williams, N. H.; Tasaki, B.; Wall, M.; Chin, J. Acc.
Chem. Res. 1999, 32, 485 and references therein. (c) Moss, R. A.; Park, B.
D.; Scrimin, P.; Ghirlanda, G. J. Chem. Soc., Chem. Commun. 1995, 1627.
(d) Moss, R. A.; Zhang, J.; Bracken, K. J. Chem. Soc., Chem. Commun.
1997, 1639. (e) Sumaoka, J.; Miyama, S.; Komiyama, M. J. Chem. Soc.,
Chem. Commun. 1994, 1755. (f) Morrow, J. R.; Buttrey, L. A.; Berback,
K. A. Inorg. Chem. 1992, 31, 16. (g) Morrow, J. R.; Buttrey, L. A.; Shelton,
V. M.; Berback, K. A. J. Am. Chem. Soc. 1992, 114, 1903. (h) Breslow, R.
Zhang, B. J. Am. Chem. Soc. 1994, 116, 7893. (i) Takeda, N.; Irisawa, M.;
Komiyama, M. J. Chem. Soc., Chem. Commun. 1994, 2773.
(25) Takarada, T.; Takahashi, R.; Yashiro, M.; Komiyama, M. J. Phys.
Org. Chem. 1998, 11, 41. (b) Yashiro, M.; Takarada, T.; Miyama, S.;
Komiyama, M. J. Chem. Soc., Chem. Commun. 1994 1757.
(26) Go´mez-Tagle, P.; Yatsimirsky, A. K. J. Chem. Soc., Dalton Trans.
1998, 2957.
N and not the CdO group.
3. La3+ exists in methanol solution above pH 7.5 as a
s
s
(La3+(CH3O-))2-bridged dimer that promotes methanolysis of
both 1 and 2 with second-order rate constants of 1.5 × 103 and
1.42 × 102 M-1 s-1. The mechanism by which the methanolysis
occurs cannot involve coordination to the distal N with attack
of external CH3O-, as in the case of Zn2+, Co2+, or (NH3)5-
Co3+, but rather involves direct attack on the CdO group,
probably by temporarily releasing a bridging methoxide from