Intramolecular kinetic isotope effect in hydride transfer from dihydroacridine to a quinolinium ion. Rejection of a proposed two-step mechanism with a kinetically significant intermediate

Article abstract of DOI:10.1039/b806869k

The intramolecular kinetic isotope effect (KIE) for hydride transfer from 10-methyl-9,10-dihydroacridine to 1-benzyl-3-cyanoquinolinium ion has been found to be 5-6 by both ¹H NMR and mass spectrometry. This KIE is consistent with other hydride transfers. It is inconsistent with the high intermolecular KIEs derived by fitting to a two-step mechanism with a kinetically significant intermediate complex, and it is inconsistent with the strong temperature dependence of those KIEs. We therefore reject the two-step mechanism for this reaction, and we suggest that other cases proposed to follow this mechanism are in error.

Full text of DOI:10.1039/b806869k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Intramolecular kinetic isotope effect in hydride transfer from dihydroacridine
to a quinolinium ion. Rejection of a proposed two-step mechanism with a
kinetically significant intermediate†
Charles L. Perrin* and Chen Zhao
Received 23rd April 2008, Accepted 28th May 2008
First published as an Advance Article on the web 28th July 2008
DOI: 10.1039/b806869k
The intramolecular kinetic isotope effect (KIE) for hydride transfer from
10-methyl-9,10-dihydroacridine to 1-benzyl-3-cyanoquinolinium ion has been found to be 5–6 by both
1H NMR and mass spectrometry. This KIE is consistent with other hydride transfers. It is inconsistent
with the high intermolecular KIEs derived by fitting to a two-step mechanism with a kinetically
significant intermediate complex, and it is inconsistent with the strong temperature dependence of those
KIEs. We therefore reject the two-step mechanism for this reaction, and we suggest that other cases
proposed to follow this mechanism are in error.
base, or that photosolvolysis intrudes.11 These explanations were
Introduction
rejected because neither added salts nor CO2-containing water
In recent years, Vernon D. Parker and his coworkers have
affects the kinetics, and because the same results were obtained
from solutions stored in the dark.12 Moreover, other researchers
have tended to accept Parker’s conclusions.13
published a series of papers that reinterpret some fundamental
reactions of organic chemistry. The data from stopped-flow or
cyclic-voltammetry experiments deviate from the simple second-
order mechanism that has long been accepted [eqn (1)]. The
data could be fitted much better to a two-step mechanism with
a kinetically significant intermediate complex, A·B [eqn (2)]. A
cogent summary of his analysis is based on a plenary lecture
presented at the 2004 IUPAC Conference on Physical-Organic
Chemistry.1 Among the reactions that show this mechanistic
behavior are nucleophilic substitutions,2 E2 eliminations,3 a [4 + 2]
cycloaddition,4 proton-transfers from radical cations,5 hydrogen-
atom abstractions,6 hydrolysis of p-nitrophenyl acetate,7 and
nucleophilic capture of (p-CH3OC6H4)3C+ by acetate.8
Of particular interest is the hydride transfer from 10-methyl-
9,10-dihydroacridine (1) to 1-benzyl-3-cyanoquinolinium ion (2).
KIEs generally between 4 and 6, depending on temperature,
solvent, and overall equilibrium constant, were observed for this
and related reactions by Kreevoy and coworkers.14 Similarly,
Table 1 lists rate constants and KIEs calculated according to
simple second-order kinetics [eqn (1)], as obtained by Parker
and coworkers.15 Table 2 lists Parker’s rate constants and KIEs,
derived from fitting to the two-step kinetic scheme of eqn (2).
The KIEs in Table 1 are quite ordinary, and in agreement with
Kreevoy’s, but those in Table 2 are unusually large, especially at
lower temperatures. Such large KIEs have been taken as evidence
for quantum-mechanical tunneling, which is consistent with the
activation parameters associated with the data of Table 2.
k 2
A + B ææÆ product
(1)
(2)
kf
kp
A ∑ B ææÆproduct
ꢀꢀꢀꢁ
ꢂꢀꢀꢀ
A + B
kb
This analysis has significant consequences for kinetic isotope
effects (KIEs), because the observed KIE is reduced if the first
step, which is isotope-independent, is partially rate-limiting. The
KIE for the hydrogen transfer itself, in the second step, is then
considerably larger than the apparent KIE, measured on the
assumption of a simple second-order mechanism. Unusually large
KIEs were thus deduced for deprotonation of radical cations by
pyridines,9 and also for proton transfers from nitroalkanes to
hydroxide.10
These conclusions have been met with some scepticism. Alter-
native explanations for the deviation from simple second-order
kinetics were that reaction occurs via both ion pairs and free ions,
or that there are acidic impurities, such as CO2, that consume
For reasons that are presented below, we too were sceptical
about the two-step mechanism [eqn (2)] and the derived KIEs. In
Table 1 Second-order rate constants (M-1 s-1) and KIEs for hydride
transfer from 1-h2 and 1-d2 to 2a
H2
D2
D2
T/K
k2
k2
k2H2/k2
291
299
308
316
325
0.00908
0.0165
0.0327
0.0526
0.0766
0.0016
0.00339
0.0066
0.0112
0.0186
5.675
4.87
4.95
4.70
4.12
Department of Chemistry, University of California, San Diego La Jolla,
California 92093-0358, USA. E-mail: cperrin@ucsd.edu
a From Ref. 15.
† Electronic supplementary information (ESI) available: Further experi-
mental details. See DOI: 10.1039/b806869k
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3349–3353 | 3349
©

View Article Online
Table 2 Rate constants and KIEs for hydride transfer from 1-h2 and 1-d2
to 2, analyzed according to eqn (2)a
Table 3 Deuterium content of 3 from reaction of 1-d with 2
T/K
t
elapsed/min [1-d]init/M [2]init/M
[3-h2]/[3-d]a [3-h2]/[3-d]b
D2
T/K
kf/M-1 s-1 kb/s-1
kpH2/s-1
kpD2/s-1
kpH2/kp
273
273
273
299
299
299
299
180
240
300
31
32
53
0.061
0.062
0.056
0.056
0.059
0.059
0.038
0.053
0.048
0.051
0.05
0.049
0.043
0.042
6.4
6.1
5.8
5.35
5.2
5.02
1.12
6.1
5.9
6.0c
5.3
5.4
5.0
—
291
299
308
316
325
0.01
0.0031
0.0041
0.0194
0.0172
0.033
0.023
0.028
0.03
0.035
0.044
0.00058
0.0009
0.0021
0.0035
0.0053
40
31
14
10
8.3
0.019
0.047
0.09
0.135
1440
a From Ref. 15.
a By NMR analysis. b By MS analysis. c From [M + NH4+] and [M +
NH4++1].
order to test this, we have measured the intramolecular KIE for
hydride transfer from 10-methyl-9,10-dihydroacridine-9-d (1-d) to
1-benzyl-3-cyanoquinolinium ion (2). Regardless of mechanism,
this KIE arises solely from the hydride-transfer step. Even if the
first step is partially rate-limiting, the second step is product-
determining. Therefore, even if eqn (2) is the mechanism, the
KIE measured is kpH/kpD, without the necessity of extracting
rate constants from two-step kinetics. Moreover, this KIE can
be obtained simply by measuring the deuterium content of the
1-benzyl-3-cyano-1,4-dihydroquinoline product (3), according to
eqn (3). We now report that this KIE is 5–6, consistent with
previous KIEs measured on the assumption of a simple second-
order reaction [eqn (1)] and providing no evidence for a two-step
mechanism with a kinetically significant intermediate complex
[eqn (2)].
intensity of the CHD signal is immediate evidence that deuterium
is transferred, and that the intramolecular KIE is not large.
Moreover, the 10.35 : 1 integration ratio corresponds to a [3-h2] :
[3-d] ratio of 5.2 : 1.
The deuterium content of 1-benzyl-3-cyano-1,4-dihydro-
quinoline (3) obtained from reaction of 10-methyl-9,10-
dihydroacridine-9-d (1-d) with 1-benzyl-3-cyanoquinolinium ion
(2) under various conditions is compiled in Table 3. There is good
agreement between the values obtained by NMR and those by
mass spectrometry, indicating that NMR integration is not overly
sensitive to the range of chemical shifts assigned to the CHD
signal.
H
kP
kp
[3-h2 ]
[3-d]
=
(3)
D
The average ratio [3-h2]/[3-d] across the first six runs in Table 1
is 5.6 0.5. There are small variations with time and temperature,
which are discussed below. The immediate conclusion though is
that this ratio, which measures kpH/kpD, agrees reasonably well
with the KIEs in Table 1 but is much smaller than the KIEs in
Table 2.
The ratio of 1.12 after 1440 minutes of reaction represents
complete scrambling of deuterium. After so many half-lives
both H and D have been transferred among 1, 2, 3, and 10-
methylacridinium ion.
Results
Fig. 1 shows the 1H NMR spectrum of 1-benzyl-3-cyano-
1,4-dihydroquinoline (3) obtained from hydride transfer
from 10-methyl-9,10-dihydroacridine-9-d (1-d) to 1-benzyl-3-
cyanoquinolinium ion (2). The spectrum shows that the CH2 and
CHD signals are well enough resolved to be integrated separately
in order to evaluate the deuterium content. The significant
According to the data in Table 3, the average ratio at 299 K is
5.2 0.2. The average ratio at 273 K is 6.05 0.2. The increase
at lower temperature is expected for KIEs, as seen in Tables 1 and
2, but the variation is small and barely beyond the experimental
error.
Discussion
Doubts about the two-step mechanism
Our scepticism about the two-step mechanism [eqn (2)] and the
derived KIEs was prompted by a perceived inconsistency regard-
ing the energetics of the intermediate complex. A representative
example is the reaction of 1 with 2. According to Parker’s data
in Table 2, kf at 299 K is 0.019 M-1s-1 and kb is 0.0041 s-1,
corresponding to an equilibrium constant, kf/kb, for formation of
the complex of 4.6 M-1. This is certainly weak binding, with a free
energy of complex formation of only -0.9 kcal mol-1. Nevertheless,
the rate constant kb for dissociation of the complex is small,
Fig. 1 500-MHz 1H NMR spectrum of 3 in CD3CN, from reaction of 1-d
with 2. The inset is an expansion, with integration, of the CH2 and CHD
signals near d 3.7.
3350 | Org. Biomol. Chem., 2008, 6, 3349–3353
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
corresponding to a DG‡ of 21 kcal mol-1. This is an unbelievably
high activation energy for the dissociation of a complex with such
weak binding. If there is hardly any energy holding the complex
together, why is so much activation energy required to take it apart?
Fig. 2 is an energy diagram representing these rate constants,
including both the intermediate complex and a weak encounter
complex or charge-transfer complex, but omitting such a complex
involving the products.
Comparison of intermolecular and intramolecular KIEs
A standard test of multistep reactions is the comparison of
intermolecular and intramolecular KIEs.18 Equality of the two
KIEs is good evidence that the rate-limiting step is also the
product-determining step. In contrast, inequality of the KIEs is
good evidence that these steps are distinct, thus indicating the
involvement of an intermediate following the rate-limiting step.
Examples of the first case are the ene reaction of methylenecyclo-
hexane with dimethyl dioxosuccinate and the reaction of ArNMe2
with diphenylpicrylhydrazyl.19 These are one-step hydrogen-atom
transfers. Unequal KIEs are seen in the dimerization of allene, the
Swern oxidation of benzyl alcohols with Me2SCl+, the reaction of
1O2 with tetramethylethylene, and the oxidation of ArNMe2 by an
iron porphyrin + PhIO, all of which proceed via an intermediate
that then partitions subject to an intramolecular KIE.20 One
cautionary exception that is similar to the study reported here
is the observation of an intermolecular KIE different from the
intramolecular KIE in the hydride transfer from a dihydropyridine
to PhCOCF3, but this is not due to an intermediate but rather to
the reversible formation of an adduct.21
To use this test, it is necessary to measure both intermolecular
and intramolecular KIEs. Parker measured only the intermolecu-
lar KIE, as in Table 1.15 His evidence for a kinetically significant
intermediate was the inadequacy of the fit of the data to simple
second-order kinetics. Thus the KIEs in Table 2 were not measured
directly, but only by adjusting the observed KIEs for the kinetic
complexity of eqn (2). We now have measured the intramolecular
KIE, to test whether it is the same as the KIEs in Table 2.
Fig. 2 Energy diagram, with energies to scale, corresponding to the
mechanism of eqn (2) with the rate constants at 299 K from Table 2
and including a weak encounter complex or charge-transfer complex.
Evidence against the two-step mechanism
The proposed intermediate of eqn (2) is recognized as dis-
tinct from the encounter complex that must be formed in any
bimolecular reaction.16 It is also distinct from the charge-transfer
complex that is often formed between electron-rich and electron-
poor reactants.17 Both of these complexes share the feature of
weak binding, but they are formed at nearly a diffusion-controlled
rate and they dissociate very quickly, with little activation bar-
rier. These aspects of the encounter complex or charge-transfer
complex are also illustrated in Fig. 2
Our key result is that the intramolecular KIE, from the deuterium
content of the 1-benzyl-3-cyano-1,4-dihydroquinoline product (3)
is 5.6 0.5. This value is inconsistent with the rate constants and
KIEs in Table 2, obtained by fitting to the two-step mechanism
[eqn (2)]. We therefore reject this mechanism, and the energy
diagram of Fig. 2
Attempts to salvage the two-step mechanism
Parker’s rationalization for a weakly bound intermediate com-
plex with high activation barrier to dissociation is not convincing.
He proposed that the intermediate complex differs from a charge-
transfer complex “by a shortening of the distances between the
reaction centers and by the extrusion of solvent, giving rise to a
significant reaction barrier”.1 Similarly, he proposed ion–dipole
complexes in proton transfers from nitroalkanes to hydroxide,
nucleophilic substitutions, and E2 eliminations. In no case did he
address the dilemma of weak binding and slow dissociation, except
to assert that the activation barrier to dissociation is high. For the
nucleophilic capture of (p-CH3OC6H4)3C+ by acetate he expanded
his rationalization by suggesting that the intermediate complex
is analogous to an intimate ion pair in solvolysis.8 However,
Winstein’s ion-pair intermediates are all very short-lived. Indeed,
Parker’s energy diagrams show the intermediate complex at high
energy and with a fairly low activation barrier to reaction. This is
inconsistent with the energetics of Fig. 2, which was constructed in
accord with the rate constants in Table 2. The unlikelihood of such
a complex is what led us to reinvestigate the KIE in the hydride
transfer from 1 to 2.
Might reversibility of the reaction, equilibrating products with
reactants, lead to a lower apparent KIE than in Table 2? The
reaction is indeed reversible, with an equilibrium constant of
only 6.7 or 11.7 favoring products.14,15 Reversibility would then
equilibrate H and D among all species and increase the D content
of 3. Indeed, when the reaction was allowed to proceed for 24 h,
the ratio of [3-h2] to [3-d] was found to be 1.12, corresponding to
an apparent KIE near 1. It is necessary to measure the KIE for the
hydrogen transfer, rather than an apparent KIE that is reduced
by isotopic scrambling. Therefore the reaction must be carried
out only to low conversion. The reaction times in Table 3 were
chosen as a compromise, to permit sufficient product for isolation
and isotopic analysis while minimizing equilibration. Those times
correspond to approximately one or two half-lives. Simulation of
the kinetics, with the rate and equilibrium constants of Kreevoy
and Kotchevar,14 shows that even at two half-lives the observed
KIE is reduced by only 6% by isotopic scrambling. The lower
KIEs observed after longer reaction times is a reflection of this
scrambling. Correcting the values in Table 3 for the incursion of
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3349–3353 | 3351
©

View Article Online
isotopic scrambling then leads to a KIE for hydrogen transfer
of 6.3 0.2 at 273 K and 5.5 0.2 at 296 K. The temperature
dependence is real but small, and the KIE of 5.9 0.5 averaged
over both temperatures is less dependable than the 5.5 0.2 at
296 K.
certainly not observed. Therefore a high barrier to rotational
diffusion within an intermediate complex cannot account for the
discrepancy between the low intramolecular KIE that we measure
and the high KIEs of Table 2.
The substantial discrepancy between our observed intramolec-
ular KIE of 5.2 at 296 K, or a corrected KIE of 6.1, and KIEs
as high as 40 from analysis of stopped-flow data thus lead us to
reject the two-step mechanism of eqn (2). Measurement of the
intramolecular KIE is an independent method that avoids the
necessity for analysis of the deviations from simple second-order
kinetics in stopped-flow data. We therefore conclude that there
must be an error in that analysis, even though we admit that
we are unable to find it, and we are grateful to Vernon Parker
for his patience and cooperation in providing details and raw
data. Nevertheless, the key result is that the intramolecular KIE is
inconsistent with the KIEs derived from the two-step mechanism
of eqn (2), which corresponds to the energy diagram of Fig. 2
Whether there are errors in the other studies1–10 is uncertain. We
have no independent measurements to test those rate constants,
but all of those studies entailed a high activation energy for the
dissociation of a weakly bound complex, with an energy diagram
like Fig. 2. We therefore suggest that the error that must be present
in the analysis of stopped-flow kinetic data for the reaction of 1
with 2 may also be present in those other cases, which might
warrant reinvestigation.
A further correction that must be applied is due to the 2.2%
undeuterated 10-methyl-9,10-dihydroacridine in the sample of
1-d. This leads to a greater proportion of [3-h2], so that kpH/kpD is
D
overestimated. To account for this, kpH/kp from eqn (3) must be
reduced by 4.2%.
In comparing this intramolecular KIE that we measure with
Parker’s intermolecular KIE, it is further necessary to consider
secondary KIEs due to the hydrogen that is not transferred.
D2
The observed intermolecular KIE in Table 1 or 2 is k2H2/k2
,
the product of a primary and a secondary KIE. The observed
intramolecular KIE of eqn (3), kpH/kpD, is the ratio of the primary
and the secondary KIEs. The secondary KIE for hydride transfer
from NADH is 1.15.22 Therefore values in Tables 1 and 2 should
be decreased by 15% to obtain the primary KIE, and values in
Table 3 should be increased by 15%.
These three corrections are not all in the same direction.
Cumulatively they contribute an average increase of ~16% that
should be applied to the values in Table 3. Therefore we conclude
that the average KIE for hydride transfer from 1 to 2 is 6.6
0.5. A more reliable measure is the KIE at 296 K of 6.1 0.2.
However, the corrections probably generate more uncertainty than
the statistical errors. Nevertheless, this intramolecular KIE is in
reasonable agreement with the intermolecular KIEs measured
by Kreevoy and coworkers,14 and with the values in Table 1 for
the simple second-order mechanism [eqn (1)]. This KIE is very
much lower than the values in Table 2, from fitting to a two-step
mechanism with a kinetically significant intermediate complex
[eqn (2)]. Moreover, the KIE at 273 K is 7.0 0.2, which is
very different from the extremely high KIE > 40 expected by
extrapolating the values in Table 2. No correction will bring the
intramolecular and intermolecular KIEs into agreement.
Experimental
Instrumentation
1H NMR spectra were obtained on a 400-MHz Varian Mercury
or 500-MHz JEOL ECA spectrometer. Mass-spectral analyses
were obtained with electrospray ionization on a Thermo-Finnigan
LCQDECA instrument. Samples were injected as solutions in
methanol, and the total HPLC intensity at each of m/z 245, 246,
247, and 248 was integrated.23
Significance of discrepancy between intermolecular and
intramolecular KIEs
Synthesis
10-Methylacridinium iodide was prepared by
a standard
Finally, we must address the question of whether the low in-
tramolecular KIE that we measure can be reconciled with the high
intermolecular KIEs that Parker derived.15 One possibility is that a
barrier to rotational diffusion within the postulated intermediate
complex decreases the intramolecular KIE. If that intermediate
complex has a high activation barrier for dissociation to separated
components, perhaps it also has a high activation barrier for one
component of the complex to rotate relative to the other. This
might be due to an attractive interaction between the p faces
of 1 and 2. Above we have disputed a high activation energy
for the dissociation of a complex with such weak binding, and
the same objection applies to the activation energy for internal
rotation. Despite those objections, we can consider the case of
slow rotation, where the selectivity for H or D transfer would be
partially determined by which face of 1-d is in proximity to 2.
However, according to a simulation of these kinetics, with the
rate constants for 299K in Table 2 and with a 21 kcal mol-1
barrier to internal rotation, equal to the barrier for dissociation,
the intramolecular KIE would be reduced only to 10. This is
method and converted to 10-methyl-9,10-dihydroacridine-9-d
with NaBD4. Its 1H NMR spectrum shows 97.7% deuteration. 1-
Benzyl-3-cyanoquinolinium bromide was prepared by a standard
method and converted to the perchlorate salt. Details of all
procedures are provided in the Supporting Information†.
Measurement of kinetic isotope effect
The hydride transfer reaction between 10-methyl-9,10-dihydro-
acridine-9-d (1-d) and 1-benzyl-3-cyanoquinolinium (2) perchlo-
rate was carried out in acetonitrile under a narrow variety of condi-
tions. A temperature between 291 K and 299 K was chosen to test
the largest KIEs in Table 2, as well as a lower temperature, at which
the KIE would be extrapolated to even larger values. A 1.2–1.3-
fold excess of dihydroacridine over quinolinium ion was generally
used. The reaction was quenched after approximately one or two
half-lives. The product 1-benzyl-3-cyano-1,4-dihydroquinoline (3)
was purified by extraction, flash column chromatography, and
recrystallization.
3352 | Org. Biomol. Chem., 2008, 6, 3349–3353
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Isotopic analysis
References
Isotopic analyses of product (3) were performed by both 1H-NMR
in CD3CN or CDCl3 and mass spectrometry. An isotope shift leads
to distinct C4 signals for CH2 and CHD. The former appears as a
singlet at d 3.732, and the latter as a broadened 1 : 1 : 1 triplet at d
3.711, owing to spin–spin coupling to the D. The CHD intensity
can be compared to that of CH2 by integration. The deuterium
content of 3 was also obtained from the summed intensities of
[M + H+] = 247 and [M + H+ - 2] = 245 peaks, compared to
[M + H+ + 1] = 248 and [M + H+ - 1] = 246. In one case the
1 V. D. Parker, Pure Appl. Chem., 2005, 77, 1823–1833.
2 Y. Lu, K. L. Handoo and V. D. Parker, Org. Biomol. Chem., 2003, 1,
36–38.
3 K. L. Handoo, Y. Lu, Y. Zhao and V. D. Parker, Org. Biomol. Chem.,
2003, 1, 24–26.
4 K. L. Handoo, Y. Lu, Y. Zhao and V. D. Parker, J. Am. Chem. Soc.,
2003, 125, 9381–9387.
5 V. D. Parker, Y. Zhao, Y. Lu and G. Zheng, J. Am. Chem. Soc., 1998,
120, 12720–12727.
6 K. L. Handoo, J. P. Cheng and V. D. Parker, J. Chem. Soc., Perkin
Trans. 2, 2001, 1476–1480.
7 V. D. Parker, J. Phys. Org. Chem., 2006, 19, 714–724.
8 W. F. Hao and V. D. Parker, J. Org. Chem., 2008, 73, 48–55.
9 Y. Lu, Y. X. Zhao and V. D. Parker, J. Am. Chem. Soc., 2001, 123,
5900–5907; Y. X. Zhao, Y. Lu and V. D. Parker, J. Chem. Soc., Perkin
Trans. 2, 2001, 1481–1488.
+
+
[M + NH4 ] and [M + NH4 + 1] intensities were found to be
more reliable. In order to correct for the natural abundance of
13C the mass spectrum of the 3-d sample was compared with that
of 3-h2 alone. From replicates the mass-spectroscopic [3-h2]/[3-d]
ratio has a precision of 0.15, but there are slight disagreements
with the NMR analyses.
10 Y. X. Zhao, Y. Lu and V. D. Parker, J. Am. Chem. Soc., 2001, 123,
1579–1586.
11 E. Humeres and T. W. Bentley, Org. Biomol. Chem., 2003, 1, 1969–
1971.
12 V. D. Parker and Y. Lu, Org. Biomol. Chem., 2003, 1, 2621–2623.
13 X. Q. Zhu, J. Y. Zhang and J.-P. Cheng, J. Org. Chem., 2006, 71, 7007–
7015; K. J. Stanger, J. J. Lee and B. D. Smith, J. Org. Chem., 2007, 72,
9663–9668.
14 D. Ostovic, R. M. G. Roberts and M. M. Kreevoy, J. Am. Chem. Soc.,
1983, 105, 7629–7631; M. M. Kreevoy and A. T. Kotchevar, J. Am.
Chem. Soc., 1990, 112, 3579–3583; I.-S. H. Lee, E. H. Jeoung and
M. M. Kreevoy, J. Am. Chem. Soc., 2001, 123, 7492–7496.
15 Y. Lu, Y. Zhao, K. L. Handoo and V. D. Parker, Org. Biomol. Chem.,
2003, 1, 173–181.
16 M. Eigen, Angew. Chem., Int. Ed. Engl., 1964, 3, 1–19.
17 V. D. Kiselev and J. G. Miller, J. Am. Chem. Soc., 1975, 97, 4036–
4039.
18 B. K. Carpenter, Determination of Organic Reaction Mechanisms,
Wiley-Interscience, New York, 1984, pp. 101–104.
19 Z. Song and P. Beak, J. Am. Chem. Soc., 1990, 112, 8126–8134; E.
Baciocchi, A. Calcagni and O. Lanzalunga, J. Org. Chem., 2008, 73,
DOI: 10.1021/jo8001672.
Conclusions
We have unambiguously measured an intramolecular KIE of 5–
6 for hydride transfer from 10-methyl-9,10-dihydroacridine (1)
to 1-benzyl-3-cyanoquinolinium ion (2). This KIE is consistent
with other hydride transfers. It is inconsistent with the high
intermolecular KIEs derived by fitting to the mechanism of eqn
(2). We therefore reject the two-step mechanism for the reaction
between 1 and 2, via a kinetically significant intermediate. Besides,
that intermediate is implausible, because it is bound only weakly
but has a high activation energy for dissociation, and an energy
diagram like Fig. 2. These results cast doubt on other cases
of the two-step mechanism of eqn (2), involving intermediates
with these same implausibilities, and those results might warrant
reinvestigation.
20 S.-H. Dai and W. R. Dolbier, Jr., J. Am. Chem. Soc., 1972, 94, 3946–
3952; M. Marx and T. T. Tidwell, J. Org. Chem., 1984, 49, 788–93;
L. M. Stephenson, M. J. Grdina and M. Orfanopoulos, Acc. Chem.
Res., 1980, 13, 419–125; E. Baciocchi, A. Calcagni and O. Lanzalunga,
J. Am. Chem. Soc., 1998, 120, 5783–5787.
Acknowledgements
21 J. J. Steffens and D. Chipman, J. Am. Chem. Soc., 1971, 93, 6694–6696;
D. M. Chipman, R. Yaniv and P. van Eikeren, J. Am. Chem. Soc., 1980,
102, 3244–3246.
22 L. C. Kurz and C. Frieden, J. Am. Chem. Soc., 1980, 102, 4198–4203.
23 M. B. Goshe and V. E. Anderson, Anal. Biochem., 1995, 231, 387–392.
This research was supported by NSF Grants CHE03-53091 and
CHE07-42801. Purchase of the NMR spectrometers was made
possible by grants from NIH and NSF. We are grateful to Dr
Yongxuan Su for help with the mass-spectrometric analysis.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3349–3353 | 3353
©