Mechanistic implications of pseudo zero order kinetics in kinetic resolutions

Article abstract of DOI:10.1021/ja062173f

This work provides simulations as well as experimental results from kinetic resolutions to demonstrate that a constant product enantioselectivity versus conversion profile in kinetic resolution is not a general consequence of pseudo zero order kinetics in

Full text of DOI:10.1021/ja062173f

Published on Web 05/18/2006
Mechanistic Implications of Pseudo Zero Order Kinetics in Kinetic
Resolutions
Donna G. Blackmond,*,† Neil S. Hodnett, and Guy C. Lloyd-Jones*
‡
,‡
Department of Chemistry and Department of Chemical Engineering and Chemical Technology,
Imperial College, London SW7 2AZ, United Kingdom, and School of Chemistry, UniVersity of Bristol,
Cantock’s Close, Bristol BS8 1TS, United Kingdom
Received March 30, 2006; E-mail: d.blackmond@imperial.ac.uk; guy.lloyd-jones@bris.ac.uk
Scheme 1. First Order, Noncatalytic Kinetic Resolution3
Kinetic resolutions offer a practical route to enantiopure com-
pounds via asymmetric catalysis.1 Such multistep competitive
reaction systems present a more complex kinetic scenario than does
a single-substrate catalytic network, but in most cases, kinetic
resolutions have been treated as the simplified first order system
shown in Scheme 1, with the efficiency of the resolution character-
Scheme 2. Generic Regimes for Catalytic Kinetic Resolution; klim
is the Rate-Limiting Step Rate Constant in Each Case
2
ized by the selectivity factor, krel. Results are most commonly
presented by consideration of how enantiomeric excess of substrate
(eesm) or product (eeprod) evolves as a function of substrate
conversion (c). Temporal concentration or rate profiles are not
typically reported.
A number of recent studies of kinetic resolutions have treated
more complex catalytic reaction mechanisms. In particular, the
consequences of pseudo zero order dependence on substrate
concentration in reactions using enantioimpure catalysts have been
4
5
exploited by Blackmond and by Lloyd-Jones.
Most recently, a report showed that kinetic resolution of
azlactones via Cu-catalyzed alcoholysis (eq 1) can follow a
relationship for krel derived for a true mathematical zero order
dependency on substrate concentration rather than the first order
The observation of pseudo zero order substrate dependency in
catalytic kinetic resolution reactions typically arises due to one of
the two generic mechanisms outlined in Scheme 2. Either a strong
pre-equilibrium binding resulting in a “saturated” intermediate
species precedes rate-limiting product formation (Scheme 2a) or
reaction of the substrate occurs after a rate-limiting step involving
the catalyst and another reagent (Scheme 2b). The alcoholysis
reaction of ref 6 (eq 1) was suggested to follow pathway 2a, while
an example of pathway 2b is provided by the (salen)Mn-catalyzed
epoxidation of allylic alcohols.7
6
expression of Scheme 1. Graphical analysis was used to demon-
strate that in this case the product enantioselectivity (ee) remains
constant with conversion until the matched enantiomer is consumed,
affording an apparent enhancement of selectivity over a true
mathematical first order dependency, where product ee erodes with
conversion. Citing the many observations of zero order kinetics in
catalysis, these authors asserted that higher selectivity will be easier
to achieve in such cases compared to first order reactions.
For reactions obeying Scheme 2a, the impact of enantiomer
binding strength (K ) on the efficiency of kinetic resolution is
eq
illustrated in Figure 1 for several limiting cases (see Supporting
Information for mathematical and modeling details). When neither
enantiomer binds strongly (Case i, KR,eq ≈ KS,eq ) low), the product
ee versus conversion profile corresponds to that for simple first
order kinetics given in Scheme 1.
Similarly, we show that under conditions of pseudo zero order
dependency on substrate, when both enantiomers bind strongly,
We report herein reaction simulations and experimental data
illustrating that a noneroding product ee versus conversion profile,
as reported in ref 6, is not a general consequence of pseudo zero
order dependency of the global reaction rate on the substrate
concentration. Instead, such behavior only occurs under specific
mechanistic constraints that cannot be expected to apply in many
observed cases of pseudo zero order behavior. Furthermore, we
describe how the analysis of rate behavior in addition to selectivity
can yield significant additional mechanistic information about
kinetic resolutions.
Figure 1. Simulations of kinetic resolutions for a racemic mixture of R
and S according to the mechanism of Scheme 2a. See text for conditions of
Cases i-v and the Supporting Information for details of the simulations.
†
‡
Imperial College.
University of Bristol.
7450
9
J. AM. CHEM. SOC. 2006, 128, 7450-7451
10.1021/ja062173f CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
the plot of product ee versus conversion curve retains the form
observed for first order kinetics, regardless of whether the binding
is equal (Case ii, KR,eq ≈ KS,eq ) high) or unequal (Case iii, KR,eq
,
K
S,eq ) high; KR,eq * KS,eq). Figure 1 shows that product ee does
not erode prior to 50% conversion only when binding is sufficiently
differentiated such that only one of the two parallel pathways
exhibits zero order kinetics in [substrate] (Cases iv and v).
Under the conditions of Cases iv and v, differential binding
allows the more stable intermediate to occupy a large fraction of
the catalyst, preventing the more weakly binding enantiomer from
competing effectively for the catalyst until complete conversion of
the matched enantiomer. This phenomenon may be observed both
in the case where the more stable intermediate reacts faster, as in
Case iv (KR,eq ) high; KS,eq ) low; kR,lim > kS,lim), and in the case
where it reacts more slowly, as in Case v (KR,eq ) high; KS,eq
)
low; kR,lim < kS,lim). Studying the global kinetics in examples
showing Case iv or Case v behavior will allow differentiation
between these possibilities.
Figure 2. Experimental conversion and enantioselectivity of starting
material from the reaction of eq 2. (a) Expression for first order kinetics
according to the equation in Scheme 1; (b) expression for zero order kinetics
according to eq 1 of ref 6. The first order fit in part (a) gives a krel ≈ 6.0.
It is also important to note that this noneroding product ee profile
necessarily leads to near-perfect selectivity until full consumption
of the strong binding enantiomer. The nonperfect but noneroding
product ee profiles produced in ref 6 by simulation of true
mathematical zero order kinetics are thus not chemically meaning-
ful, suggesting that the nonperfect, noneroding product ee profiles
observed experimentally in ref 6 require further rationalization. One
possibility is that two separate catalyst species, both exhibiting Case
iv or v behavior but with different net reactivities and opposite
selectivities, may operate in that system. The suggestion by the
authors of ref 6 that the slow- and fast-reacting enantiomers exhibit
zero order kinetics in [substrate]. This behavior arises only from a
case of saturation kinetics in one enantiomeric substrate on only
one of the two parallel pathways in a kinetic resolution following
a mechanism such as that given in Scheme 2a. For a resolution
involving a single catalyst species, this will be limited to reactions
of near-perfect selectivity (cf. Cases iv and v, Figure 1).
In conclusion, these findings demonstrate that observation of a
noneroding product ee versus conversion profile can provide
additional mechanistic information concerning relative enantiomer
binding strengths and reactivity of intermediate species and can
hint at the possibility of turnover by multiple catalyst species. The
combined analysis of global kinetics and ee versus conversion
profiles can provide significant mechanistic detail as well as help
distinguish between proposed mechanisms in kinetic resolution.
different coordination modes may in fact signify that R and S
substrates each bind strongly to separate catalyst species.8
Scheme 2b shows a further mechanistic possibility for pseudo
zero order kinetics in substrate concentration that holds for the
7
example of kinetic resolution in the (salen)Mn-catalyzed epoxi-
dation of alkenes shown in eq 2, where MndO generation is rate
limiting.7 While the global rate has a zero order dependency on
substrate (alkene 1) concentration, the local dependency at the stage
b,c
Acknowledgment. D.G.B. and G.C.L.-J. thank AstraZeneca for
research support.
of enantioselection remains first order (Scheme 2b, k
R
[R] versus
Supporting Information Available: Mathematical details and
simulation results for the five cases in Figure 1 for Scheme 2a;
simulations for reactions following Scheme 2b; details of the experi-
ments in Figure 2 (9 pages, print/PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
k
S
[S]). The experimental data (ee and conversion of 1) should and
does give a linear correlation plotting the standard first order
equation in Scheme 1, the selectivity factor (s ) krel) being given
by the slope of the plot in Figure 2a. In contrast, the “zero order”
expression developed in ref 6 is inappropriate, giving a curved
relationship (Figure 2b) devoid of chemical meaning. Thus the
observation of zero order dependency on substrate concentration
does not result in an apparent enhancement of selectivity in kinetic
resolutions following the mechanism presented in Scheme 2b (see
Supporting Information).
References
(
1) Kagan, H. B.; Fiaud, J. C. In Topics in Stereochemistry; Eliel, E. L., Wilen,
S. H., Eds.; Wiley & Sons: New York, 1988; Vol. 18, p 249.
(2) Reference 1 also treats reactions that are second order in the enantiomeric
substrate concentration.
(3) This equation for krel holds for experiments initiated with a strictly racemic
mixture of enantiomers.
(
(
4) Blackmond, D. G. J. Am. Chem. Soc. 2001, 123, 545.
5) Dominguez, B.; Hodnett, N. S.; Lloyd-Jones, G. C. Angew. Chem., Int.
Ed. 2001, 40, 4289.
(
6) Tokunaga, M.; Kiyosu, J.; Obora, Y.; Tsuji, Y. J. Am. Chem. Soc. 2006,
128, 4481.
(
7) (a) Adam, W.; Humpf, H. U.; Roschmann, K. J.; Saha-Moller, C. R. J.
Org. Chem. 2001, 66, 5796. (b) Finney, N. S.; Pospisil, P. J.; Chang, S.;
Palucki, M.; Konsler, R. G.; Hansen, K. B.; Jacobsen, E. N. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1720. (c) Hughes, D. L.; Smith, G. B.; Dezeny,
G. C.; Senanayake, C. H.; Larsen, R. D.; Verhoeven, T. R.; Reider, P. J.
J. Org. Chem. 1997, 62, 2222. (d) Linde, C.; Arnold, M.; Norrby, P.-O.;
Akermark, B. Angew. Chem., Int. Ed. Engl. 1997, 36, 1723. (e) Linker,
T. Angew. Chem., Int. Ed. Engl. 1997, 36, 2060.
(
8) See Supporting Information for simulations of two-catalyst cases that
reproduce the experimental results of ref 6.
The work presented herein provides simulations and experimental
results to demonstrate that a constant product ee versus conversion
profile in kinetic resolution is not a general consequence of pseudo
JA062173F
J. AM. CHEM. SOC.
9
VOL. 128, NO. 23, 2006 7451