Isotope Effects on the Decarboxylation of 4-Pyridylacetic Acid
J . Org. Chem., Vol. 66, No. 16, 2001 5535
Ta ble 1. Com p a r ison of th e Exp er im en ta l a n d Th eor etica l Nitr ogen KIEs on th e Deca r boxyla tion of 4-P yr id yla cetic
Acid a t Room Tem p er a tu r e
water
2
5%
50%
75%
3 explicit
dioxane SM5.42
0.9825a
water
water
water
SM5.42
1.0082
COSMO
1.0079
water molecules
theory
experiment
1.0066
0.994 ( 0.002b
0.997 ( 0.001
1.0019 ( 0.0004
a
All calculations at HF/6-31G(d) level. b Standard deviation is reported.
isolated from the reaction mixture and subjected to
isotope-ratio mass spectrometric analysis. Natural iso-
topic composition was used; thus, no isotopic synthesis
was necessary. KIEs were calculated from the equation
Ta ble 2. P r op er ties of th e NH Moiety Ca lcu la ted on a t
th e SM5.42/HF /6-31G(d ) Level
dioxane
water
A
Z
T
A
Z
T
d(N)a
d(H)
r(N-H) c
n(N-H)
n(N-C)
-0.37 -0.26
0.37
-0.33
0.35
1.003
0.76
-0.37 -0.26
0.37
-0.34
0.35
1.002
0.77
ln(1 - f)
k /k )
15
(1)
1
4
ln[1 - f(1000 + δ )/(1000 + δ )]
0.998
0.74
1.009
0.74
f
∞
1.40
1.13
1.06
1.40
1.13
1.05
where f is the fraction of reaction and δ values are
isotopic ratios, relative to a standard, obtained from the
mass-spectrometric measurements of the product after
a
CM2 atomic partial charges.14 Bond lengths in Å. c Mayer
b
bond order.15
full conversion (δ
∞
) and after the fraction of reaction f
calculations for the mixed solvents because current
solvation models are not ready to address problems
connected with the preferential solvation, which has been
postulated for dioxane-water mixtures.11 Interestingly,
this nitrogen KIE in water is modeled equally well by
the COSMO solvent continuum model,12 as well as a
model that contains three explicit water molecules: two
in the vicinity of the carboxyl group and one close to
nitrogen atom. This is not the case for the oxygen EIE
on the same reaction, which seems to indicate that
modeling of nitrogen isotope effects is less demanding
than modeling of oxygen isotope effects. The value
obtained for the reaction in dioxane can be further split
into its components. We have calculated EIE ) 0.9756
f
(δ ). Air nitrogen was used as the local standard; its
isotopic composition does not affect the isotope effect (cf.
eq 1). For reactions carried out in solvents containing 25%
and 50% water, duplicate independent measurements
were made at the fraction of reaction of about 0.1. The
smaller the fraction of reaction the larger the difference
between δ
f
and δ
∞
and, thus, the more reliable the
determination of KIEs. For the reaction carried out in
the solvent containing 75% of water more thorough
studies were performed. These included nine independent
measurements with the fraction of reaction ranging from
0
.05 to 0.30. Obtained values for all three solvents are
listed in Table 1. Nitrogen KIEs show increasing trend;
at low water content the nitrogen KIE is inverse (smaller
than unity, indicating that the heavier isotope is reacting
faster than the light one). At 50% dioxane-water mixture
the nitrogen KIE is still inverse but closer to unity and
becomes normal at water rich composition. These values
are within the range predicted theoretically by computing
nitrogen KIEs for pure dioxane and the aqueous solution.
Also the observed trend in the change of the nitrogen KIE
agrees with the theoretical prediction. Analysis of the
calculations performed with the SM5.42 continuum
and KIE ) 1.0071. As can be seen, it is the equilibrium
2
isotope effect that makes the overall KIE so different in
dioxane and in water. The EIE calculated here is in
excellent agreement with the experimental values of
0
.978, 0.979, and 0.981 recalculated from the nitrogen
EIEs on the deprotonation of 3-acetyl-, 4-methyl-, and
pyridine, respectively.13
The comparison of the main properties of the nitrogen
atom and the hydrogen atom bonded to it for all the
reacting species in both solvents is given in Table 2. As
can be seen, all calculated properties are very similar in
both solvents, the only exception being the length of the
N-H bond, which is stretched in aqueous solution due
to hydrogen bonding.
8
solvent model of dioxane and aqueous solutions indicates
that the mechanism changes with the change of the
9
solvent polarity. In aqueous solution, the zwitterionic
form is more stable than the acid and thus the reaction
proceeds from the zwitterion Z to the transition state T.
In pure dioxane, however, the zwitterion is less stable
than the acid. Thus, the reaction can be viewed as a
stepwise process and the nitrogen KIE can be calculated
as a product of the isotope effect on the equilibrium
Exp er im en ta l Section
Ma ter ia ls. 4-Pyridylacetic acid hydrochloride (98%) was
obtained from Aldrich. Chloroform (pure p.a.), 1,4-dioxane
(9) Actually, the SM5.42 model includes several factors connected
between acid A and zwitterion Z (EIE) and KIE on the
conversion of the zwitterion to product:10
with the solvation, including cavitation energy and hydrogen bonding.
The detailed influence of these factors on the rate of the decarboxy-
lation of 4-pyridylacetic acid will be discussed elsewhere. We make a
point here that of all factors, electrostatic effects play the dominant
role.
KIE ) (KIE /KIE )KIE ) EIE‚KIE
2
(2)
1
-1
2
(
10) Note that, from the thermodynamic point of view, the overall
In Table 1, we have reported values calculated theo-
retically at the SM5.42/HF/6-31G(d) level for pure diox-
ane and for aqueous solution. We did not attempt
isotope effect can be calculated directly from the reaction of A going
to transition state T. Analysis of the isotopic fractionation assuming
intervention of the zwitterion, and using eq 2, is, however, more
chemically intuitive and informative.
(
11) (a) Kinart, W. J .; Kinart, C. M.; Skulski, L. Pol. J . Chem. 1989,
(
8) (a) Li, J .; Hawkins, G. D.; Cramer, C. J .; Truhlar, D. G. Chem.
63, 581. (b) Skulski, L.; Kinart, C. M. Pol. J . Chem. 1992, 66, 287.
(12) Klamt, A.; Sch u¨ u¨ rmann, G. J . Chem. Soc., Perkin Trans. 2 1993,
799.
Phys. Lett. 1998, 288, 293. (b) Li, J .; Zhu, T.; Hawkins, G. D.; Winget,
P.; Liotard, D. A.; Cramer, C. J .; Truhlar, D. G. Theor. Chem. Acc.
1
999, 103, 9. (c) Zhu, T.; Li, J .; Liotard, D. A.; Cramer, C. J . J . Chem.
(13) Kurz, J . L.; Pantano, J . E.; Wright, D. R.; Nasr, M. M. J . Phys.
Chem. 1986, 90, 5360.
Phys. 1999, 110, 5503.