Structural contributions to autocatalysis and asymmetric amplification in the soai reaction

Article abstract of DOI:10.1021/jacs.0c05994

Diisopropylzinc alkylation of pyrimidine aldehydes-the Soai reaction, with its astonishing attribute of amplifying asymmetric autocatalysis-occupies a unique position in organic chemistry and stands as an eminent challenge for mechanistic elucidation. A n

Full text of DOI:10.1021/jacs.0c05994

pubs.acs.org/JACS
Article
Structural Contributions to Autocatalysis and Asymmetric
Amplification in the Soai Reaction
Soumitra V. Athavale, Adam Simon, K. N. Houk, and Scott E. Denmark*
Cite This: https://dx.doi.org/10.1021/jacs.0c05994
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Diisopropylzinc alkylation of pyrimidine aldehydesthe
Soai reaction, with its astonishing attribute of amplifying asymmetric
autocatalysisoccupies a unique position in organic chemistry and
stands as an eminent challenge for mechanistic elucidation. A new
paradigm of “mixed catalyst−substrate” experiments with pyrimidine and
pyridine systems allows a disconnection of catalysis from autocatalysis,
providing insights into the role played by reactant and alkoxide structure.
The alkynyl substituent favorably tunes catalyst solubility, aggregation,
and conformation while modulating substrate reactivity and selectivity.
The alkyl groups and the heteroaromatic core play further comple-
mentary roles in catalyst aggregation and substrate binding. In the study
of these structure−activity relationships, novel pyridine substrates
demonstrating amplifying autocatalysis were identified. Comparison of
three autocatalytic systems representing a continuum of nitrogen Lewis basicity strength suggests how the strength of N−Zn binding
events is a predominant contributor toward the rate of autocatalytic progression.
Curiously, replacement of the t-Bu group in 1d by a
triisopropylsilyl (TIPS) group (1c) is detrimental to
asymmetric amplification.4
1. INTRODUCTION
1.1. Asymmetric Autocatalysis and the Soai Reaction.
The diisopropylzinc alkylation of pyrimidine-carbaldehydes,
crowned the Soai reaction (Figure 1a), represents an
astonishing and fortuitous confluence of three themes in
catalysis: (1) autocatalysis, (2) enantioselective catalysis, and
(3) positive non-linear effects. Each product enantiomer
catalyzes its own formation (selective autocatalysis) while
also inhibiting catalysis by its counterpart (a positive non-linear
effect)resulting in a continuous amplification of the excess
enantiomer with reaction progression. Soai’s seminal 1995
report describes how the inclusion of scalemic product carbinol
2a at the beginning of the reaction with 1a results in the newly
formed product being more enantioenriched than the initial
additive (Figure 1b).1,2 Near racemic autocatalyst with only
51:49 e.r. provides a higher enantioenriched product with
55:45 e.r. The product from one reaction is used as the catalyst
for a subsequent reaction, and in successive reaction cycles, the
enantioenrichment of 2a increases to up to 94.5:5.5 e.r.
Following this discovery, it was demonstrated that, in general,
the reaction of diisopropylzinc with 2-substituted pyrimidinyl
aldehydes displays autocatalytic, asymmetric amplification.3−8
In particular, a bulky alkynyl substituent (Figure 1c) drastically
increases the performance of these reactions in terms of
enantioenrichment achieved, reaction times, and isolated
yields. For example, the asymmetric autocatalytic reaction
with 1d affords the product in an enantiopure form in
quantitative yield. Following this report, 1d became the
workhorse substrate for future studies by the Soai group.
Although the theoretical framework for amplifying auto-
catalysis was articulated as far back as 1953 (the term
“asymmetric autocatalysis” itself was coined by Wynberg in
1989), the Soai reaction remains the only example of an
autocatalytic, non-equilibrium, irreversible chemical trans-
formation capable of robust asymmetric amplification.9,10
With substrate 1d, the reaction provides enantiopure products
within three reaction cycles, even when the autocatalyst used
has a calculated e.e. as low as 5 × 10−5 %.11 Investigations by
Soai12−14 and Singleton15,16 have demonstrated that the Soai
reaction is capable of producing a non-racemic product even in
the absence of any added catalyst. This symmetry breaking
arises from a statistical imbalance in the formation of the
enantiomeric products in the early evolution of the reaction,
which is subsequently propagated and amplified by asymmetric
autocatalysis.9 The Soai reaction is thus a successful example of
absolute asymmetric synthesis.17 A large variety of chiral
additives ranging from amino acids18 to enantiomorphic
Received: June 2, 2020
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 1. (a) The Soai reaction, with the general scheme of amplifying autocatalysis and salient features. (b) Seminal report with substrate 1a. (c)
Improved performance with alkynyl-substituted substrates, except in the case of 1c.
crystals,19 helical hydrocarbons,20 and even cryptochiral
molecules21 can influence the outcome of the reaction by
biasing an initial imbalance of one of the enantiomers.22
Remarkably, even circularly polarized light23 and isotopic
chirality24−26 have been demonstrated to result in a non-
racemic product. One may state that in multiple cycles of the
Soai reaction, symmetry breaking and enantioenrichment are
inevitable. The Soai reaction hence qualifies as a chemical
transformation with a predisposition to evolve toward
homochirality. Soai’s seminal discoveries have received wide-
spread attention in diverse chemical fields and have revived
discussions regarding absolute asymmetric synthesis, symmetry
breaking, and the origin of biological homochirality.27−31 The
interested reader is directed to a number of reviews for further
details, accounts, and descriptions of the remarkable Soai
system.17,22,32−35
1.2. The Soai Reaction as a Mechanistic Challenge.
More than two decades since Soai’s initial discovery,
fundamental questions regarding its mechanism have remained
unanswered. The most obvious outstanding issues are the
precise identity of the alkoxide autocatalyst, its genesis and
propagation, the origin of its non-linear behavior, and a
compelling transition state model accounting for generation of
the homochiral product from the reactants.
Empirically, it is clear that the structural constraints on
reactants are severe, but the basis for this narrow substrate
scope remains enigmatic. In the dialkylzinc alkylations of
pyridine-3-carbaldehyde (3a), inclusion of the scalemic
B
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 2. Crystal structures for the square−macrocycle−square (SMS) tetrameric isopropylzinc alkoxide derived from 2d. Adapted with permission
from ref 60. Copyright 2015 John Wiley and Sons.
product carbinol provides newly formed product that is not
racemic but has a lower enantioenrichment than the additive.36
The reports with pyrimidine-5-carbaldehydes without excep-
tion describe the use of only diisopropylzinc as the
nucleophile. The incompetence of other dialkylzinc reagents
is presumed, although systematic studies are lacking.37,38 Soai
et al. have described isolated reports wherein amplifying
autocatalysis with diisopropylzinc was demonstrated with other
substituted quinoline-3-carbaldehydes and 5-carbamoyl-
pyridine-3-carbaldehyde.39−42 However, pyrimidine-5-carb-
aldehydes remain the workhorse substrates for the reaction,
and among these, 2-alkynyl-substituted aldehydes display
superior autocatalysis and selectivities. TIPS-alkynyl substrate
1c presents a remarkable exception to this trend and highlights
the puzzling idiosyncrasies typical of the substrate scope of this
transformation. The rationalization of these structural con-
straints poses a test to any mechanistic proposal.
The reaction is extremely sensitive to initial chiral
imbalances, including the extraordinary susceptibility to
additives with isotopic chirality and other cryptochiral
additives. The fundamental chemical interactions between
such additives and the reactants that ultimately trigger
symmetry breaking remain poorly understood. Similarly,
rationalization of the structural and kinetic boundary
conditions assuring symmetry breaking in uncatalyzed
reactions remains a key issue.43,44
The study of the Soai reaction also presents technical
challenges including: (1) the dialkylzinc reagent is highly
pyrophoric and moisture sensitive, complicating experiment
design and execution; (2) the autocatalytic nature of the
reaction demands non-trivial strategies for kinetic analysis; and
(3) structural investigations may be impeded by the stability
and solubility of the zinc alkoxide intermediates. Thus, overall,
the Soai reaction, with its sui generis characteristics, presents
unique challenges to mechanistic investigation. A holistic
understanding of the reaction raises hopes of developing new
transformations that may display amplifying asymmetric
autocatalysis and potentially contribute to the understanding
of the origin of homochirality in nature.
predominantly focused on probing the aggregation states of the
product zinc alkoxide. Early kinetic studies with the 3-formyl-
6-methylpyrimidine-carbaldehyde system included a compar-
ison between enantiopure and racemic autocatalysts and
indicated a dimeric alkoxide aggregate as the catalytic species,
while subsequent investigations with an enantiopure catalyst
suggested two aldehyde substrates in a tetrameric com-
plex.45,48,58 The first structural studies in 2004, deduced from
1H NMR spectroscopy with the 2b-derived isopropylzinc
alkoxide, are consistent with a dimeric model.49,50 Note that
these experiments cannot conclusively distinguish between
dimers and higher order aggregates. Likewise, modeling studies
indicated that the extraordinarily high experimentally observed
amplification efficiency (especially for the alkynyl-substituted
systems) was not possible with dimeric catalysis and
necessitated the invocation of higher order aggregates in the
catalytic cycle.35,47,55 Theoretical suggestions for possible
higher order aggregate structures, including the square−
macrocycle−square (SMS) tetramer, were made for the first
time by Gridnev and Brown.37,62 Subsequent investigations
provided further momentum for a tetrameric catalyst model.
Reaction progress kinetic analysis (RPKA) experiments57 with
substrate 1e indicated the reaction rate to be zeroth order in
diisopropylzinc, first order in catalyst (product alkoxide), and
1.6 order in aldehyde. An intriguing inverse temperature
dependence of the reaction rate was also observed. On the
basis of kinetic and spectroscopic clues, the suggestion that the
SMS tetrameric alkoxide might be the active catalyst was raised
for the first time. A DFT study to model these tetramers was
subsequently reported by Gridnev in 2012.59
Finally, in 2015, the first crystal structure of the isopropyl-
zinc alkoxide derived from 2d was published by Soai et
al.60,61,63 Depending on the conditions for crystallization, a
tetrameric or oligomeric aggregate is obtained. The homochiral
and heterochiral tetramers possess a 12-member macrocycle
with a connectivity resembling the SMS tetrameric structure
proposed earlier (Figure 2). The assembly can be considered
as a concatamer of two Zn−O square dimers, ligated through
pyrimidine-Zn coordination. In the homochiral tetramer, the
unbound pyrimidine units (referred to as the tetramer “arms”)
are oriented on the same face of the macrocycle. The racemic,
heterochiral tetramer possesses a similar overall connectivity,
but the arms are placed on opposite sides of the macrocycle. In
both the homochiral (enantiopure) and heterochiral (racemic)
1.3. Current Mechanistic Consensus. The striking
properties of the Soai system have motivated numerous
studies to elucidate the mechanism of this iconic trans-
formation.34,37,38,43,45−−61 Because catalyst aggregation is a
central contributor to non-linear effects, investigations have
C
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. (a) Amplifying autocatalysis with substrate 3b. (b) The cube-escape model to rationalize the formation of the (auto)catalytically
competent SMS tetramer. (c) (From left) DFT calculated structures and relative energies (kcal/mol) for the ground-state heterochiral and
homochiral SMS tetramers. The stereodetermining transition-state structure on the right arises after a floor-to-floor binding of the substrate
aldehyde 3b to the homochiral tetramer. The location of the concerned zinc centers (yellow) in the heterochiral tetramer precludes such a binding
mode. Adapted with permission from ref 64. Copyright 2020 Springer Nature.
tetramers, a pair of three-coordinate, unsaturated (alkoxy)zinc
atoms are present as part of the pyrimidinyl arms. The
orientation of these unsaturated zinc centers follows the
pattern of the arms. Finally, for both the tetramers, the crystal
structure also showed that, except for the two macrocycle-
residing nitrogen atoms, all the remaining six pyrimidinyl
nitrogen atoms are bound by diisopropylzinc molecules (these
are not shown in the structure in Figure 2). Soai’s crystal
structure elucidation is a landmark achievement in providing a
definitive understanding of the solid-state SMS tetramer and
revealing the differences in the homochiral and heterochiral
tetrameric assemblies.
D
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. Structural constituents of the three components (catalyst, reagent, and substrate) in the Soai reaction.
Two decades of investigations into the Soai reaction
culminated in the identification of alkoxide tetramers as the
likely autocatalysts. However, this knowledge had not provided
insights into the mode of action of the catalyst nor a
compelling transition-state model or an explanation of the
structural constraints on the reactants. Recently, we addressed
this central issue regarding the modus operandi of the
autocatalyst.64 These investigations, summarized in Figure 3,
revealed the structural logic behind the assembly and working
of the SMS tetrameric (auto)catalyst on the basis of the
following key findings: (1) The discovery of amplifying
autocatalysis in the diisopropylzinc alkylation of the related
pyridine substrate 3b proved that only one nitrogen in the
aromatic core is necessary for the Soai phenomenon (Figure
3a). (2) Whereas related structures like PyEE, PyEI, and PyIE
predominantly form catalytically inactive, Zn−O cubic
tetramers, the steric bulk of the product zinc alkoxide PyII,
in combination with pyridyl-nitrogen coordination, enables a
“cube-escape” pathway to an alternative, catalytically active
tetrameric cluster sharing identical connectivity with Soai’s
crystal structure (the SMS tetramer) (Figure 3b). (3) This
cluster engages the substrate aldehyde through a two-point
coordination (via the carbonyl oxygen and the aromatic
nitrogen) to the unsaturated (alkoxy)zinc centers on the
catalyst floor (highlighted in yellow in the ground-state
structures in Figure 3c), termed as the “floor-to-floor” model.
(4) The resulting activated aldehyde is alkylated by an
activated diisopropylzinc molecule, delivered preferentially
from only one of the catalyst arms, leading to the
homomorphic product alkoxide (TS-4b) in Figure 3c (the
competing transition-state structure which leads to the
heteromorphic product and is disfavored by 4.2 kcal/mol is
not shown), comporting with an enantioselective, autocatalytic
process. (5) Such a floor-to-floor binding is structurally
impossible for the thermodynamically preferred heterochiral
PyII SMS tetramer (Figure 3c, Grel = 0), and the alternative
single-point binding pathway results in an energetic penalty for
catalysis, hence providing a basis for the signature non-linear
effect in the reaction. Overall, the cube-escape, floor-to-floor
catalysis model rationalizes both the genesis of the SMS
tetramer as well as its specific, non-linear, autocatalytic activity.
The current work complements and builds on these findings.
Experiments providing specific insights into the structural
contributions that affect asymmetric autocatalysis in the Soai
reaction are described. Further subtle effects of the nitrogen
Lewis basicity in controlling autocatalytic progression are
uncovered by a comparison of the pyrimidine and pyridine
systems.
1.4. Research Plan and Rationale. The asymmetric
autocatalytic reaction of 1b involves three components: the
zinc alkoxide “catalyst”, the dialkylzinc “reagent”, and the
aldehyde “substrate”. Each component may be divided
intuitively into identifiable constituents (Figure 4). For
example, at least four constituents of the catalyst are (1) the
alkylzinc counterion, (2) the carbinol group (isopropyl in this
case), (3) the heteroaromatic core, and (4) the 2-alkynyl
substituent. The aldehyde possesses fewer constituents, only
the pyrimidine ring and the 2-substituent. No such
deconstruction of the dialkylzinc component is possible. An
overarching objective as the mechanistic investigations were
inaugurated was to holistically delineate the role played by
these three components, and their structural constituents, in
affecting catalysis, stereoselectivity, and non-linearity in the
reaction. Rate enhancement, enantioselectivity, and a non-
linear effect are the most relevant properties of the alkoxide
catalyst in combination with the dialkylzinc reagent. However,
E
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 5. Compound numbering for substrates and zinc alkoxide catalysts.
Figure 6. In situ IR spectroscopic monitoring of the Soai reaction (wavenumber at ∼1710 cm−1).
which structural constituents of the alkoxide affect these
properties was, at the outset, unclear. Likewise, the
contributions from the aldehyde structural constituents that
make it an ideal substrate for the catalyst were unknown.
Extracting meaningful structure−activity relationships from
a set of analogous autocatalytic reactions is challenging because
a single structural change cannot be performedboth the
(auto)catalyst and the substrate structures have to be modified
F
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 7. PmII-catalyzed, “mixed catalyst−substrate” diisopropylzinc alkylations monitored by in situ IR spectroscopy.
at the same time. However, autocatalysis can be considered as
a special case of catalysis wherein the substrate structure is
“matched” to generate a product that is catalytically active.
From such a perspective, it is seen that catalysis,
enantioselectivity, and non-linearity arise as a result of the
catalyst structure and can manifest in other reactions where the
catalyst is operational, whereas autocatalysis is an incidental
property emerging from a matched substrate structure. We
hypothesized that individual contributions from the catalyst
and the substrate can be studied independently from each
other in a set of reactions where the (auto)catalyst and
substrate structures are not matched but the substrates are
nonetheless competent (in other words, the reaction is catalytic
but not autocatalytic). Thus, experiments combining various
zinc alkoxides and aldehyde substrates were planned and are
described in the upcoming sections. Figure 5 depicts the
compound numbering scheme in these experiments.
analytical method, the reactant combination of diisopropylzinc
and pyrimidine-5-carbaldehyde 1b was chosen as a model
system for investigations. In situ IR spectroscopic analysis was
considered a convenient technique to monitor the trans-
formation. A distinct advantage of this method over previously
used in situ monitoring strategies (calorimetry or NMR
spectroscopy) is that the experimental setup allows for
simultaneous manual sampling of the reaction mixture to
determine product e.r. at any chosen time point. The aldehyde
is added as the final component to the reaction mixture (this
time is assigned as t = 0), and the resulting carbonyl IR
absorbance peak is tracked over time. Figure 6 depicts a typical
data set obtained for such experiments. Combining 1b with
diisopropylzinc in the absence of catalyst led to a sigmoidal
aldehyde decay with an initial induction period followed by
rapid aldehyde consumption (entry 1). Inclusion of scalemic
product carbinol as a catalyst in the reaction eliminates the
induction period and results in a rapid reaction with higher
final product e.r. (entry 2). Such behavior is typical for a
reaction showing amplifying autocatalysis. In contrast, the
same reaction with diethylzinc shows slow consumption in the
2. RESULTS AND DISCUSSION
2.1. Structure−Activity Relationships in the Soai
Reaction. 2.1.1. In Situ IR Monitoring. To establish a reliable
G
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 8. PmIE-catalyzed “mixed catalyst−substrate” diethylzinc alkylations monitored by in situ IR spectroscopy.
absence of added autocatalyst (entry 3). Inclusion of scalemic
product results in enantiomeric erosion with no rate
enhancement over the uncatalyzed reaction (entry 4). Thus,
the reaction of 1b with diethylzinc demonstrates no autocatalysis
or chiral amplification. The surprising failure of the TIPS
analog, 1c, to display amplifying autocatalysis in reaction with
diisopropylzinc was also examined. Indeed, in a series of in situ
IR-monitored experiments with inclusion of the scalemic
product carbinol as a catalyst, neither asymmetric amplification
nor appreciable rate enhancement was observed over the
sluggish background alkylation reaction, thus confirming Soai’s
previous findings.65 In situ monitoring in combination with
manual sampling emerged as a convenient tool to interrogate
catalysis and selectivity in the Soai reaction.
2.1.2. “Mixed Catalyst−Substrate” Experiments and the
Discovery of Amplifying Autocatalysis with 5-(Trimethyl-
silylethynyl)pyridine-3-carbaldehyde. The investigation of
the role of the aldehyde substrate constituents began with a
study of the reaction of diisopropylzinc and various aldehydes
with the isopropylzinc alkoxide of 2b, henceforth denoted as
PmII (nomenclature: Pm/Py/Ph = pyrimidine/pyridine/
phenyl indicating the aromatic core, I/E = isopropyl/ethyl
indicating the carbinol alkyl group, and I/E = isopropyl/ethyl
indicating the alkylzinc group, see Figure 5), as the catalyst
(Figure 7). The aim of this survey was to determine if the
catalyst−reagent combination of PmII and diisopropylzinc
could successfully affect rate enhancement and positive non-
linear enantioselectivity in the alkylation of aldehydes distinct
from its “natural” substrate 1b and to simultaneously evaluate
the effect of substrate structure on such an activity. The runs
are “mixed catalyst−substrate” experiments because the
alkylation of the substrate produces a species distinct from
the initial catalyst. Note that under these conditions, the
uncatalyzed addition of diisopropylzinc to the test substrates is
slow with low conversion. In Figure 7, the substrates are
arranged in decreasing order of their rate of reaction, and this
convention continues in the subsequent figures.
For aldehydes 3b, 3a, 1c, and 3c, diisopropylzinc alkylation
with scalemic PmII afforded products with e.r. significantly
higher than that of the added catalyst (Figure 7). The relative
rates of reaction were qualitatively in the order 3b ≫ 3a > 1c >
3c. In contrast, no product was detected with the analogous phenyl
substrates 5a and 5b. The striking rate observed with 3b
motivated further detailed control experiments which revealed
H
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
that, in fact, the reaction of this substrate with diisopropylzinc
displayed robust, amplifying, asymmetric autocatalysis! Thus,
the high rate is attributable to an autocatalytic progression
supported by the newly formed, catalytically competent
product alkoxide. Such a possibility was rigorously ruled out
for the other substrates. Thus, such mixed-catalyst experiments
led to the discovery of the competent pyridine system, 3b, and
the subsequent cube-escape, floor-to-floor model summarized
earlier.
latter is apparently high enough to affect catalysis). Rate and
selectivity vary considerably with aldehyde structure (entries
2−5), and a positive non-linear effect is observed with all
substrates except 3a (entry 2). Only marginal catalysis was
observed with the challenging TIPS-substituted substrates, 1c
and 3c (entries 4, 5). As was the case in reactions with PmII,
the phenyl substrates 5a and 5b were unreactive.65 Although
3b reacts more slowly than 3a, the substituted pyrimidine 1b is
more reactive than either of these, suggesting that a balance of
two opposing factors determines aldehyde reactivity: (1) the
activating pyrimidine ring and (2) the deactivating 2-alkynyl
substitution. Likewise, comparison of entries 2, 3, and 5
suggests that the 2-alkynyl substituent improves substrate-
controlled selectivity, and among them, an optimum balance of
reactivity and selectivity is achieved for aldehyde 3b.
These conclusions strengthen and extend the trends
deduced from Figure 7. In terms of rate and selectivity,
comparison of analogous reactions in Figures 7 and 8 suggests
that PmII−diisopropylzinc as a catalyst−reagent combination
is superior to PmIE−diethylzinc. The reduction in activity of
the latter allows identification of substrate effects, which are
otherwise masked in the case of highly efficient catalysis with
the former.
Poor catalysis and asymmetric erosion were observed with
PmEE-catalyzed “mixed catalyst−substrate” diethylzinc alkyla-
tions.65 This behavior is consistent with the prediction that
PmEE will predominantly exist as a catalytically inactive cubic
tetramer. Finally, the activity of PmEI could not be
unambiguously established because insolubility of this alkoxide
precluded reliable estimation of catalyst concentration in
solution.
2.1.3. An Empirical Structure−Activity Correlation. A
qualitative assignment of the contribution of structural
constituents in the Soai reaction emerges from the results of
the mixed catalyst−substrate experiments. Before elaborating
on these assignments, it is instructive to restate the inferences
gained through the PyII floor-to-floor transition-state model
described in our previous work:
Clearly, the signature characteristic of PmII, namely,
catalyzed, positive non-linear, enantioselective addition of
diisopropylzinc, is also conserved in reactions with some
“unnatural” substrates (3b, 3a, 1c, and 3c). However, it
appears that, for such a substrate to be competent, the minimal
structural requirement is a 3-azaaryl group. In hindsight, these
constraints are identical to the ones described for the PyII
tetramer autocatalyst,64 a further confirmation of the striking
similarity between the activities and mode of action of PyII
and PmII. Note that the Gridnev transition-state model,59
which discounts any interactions between the substrate
nitrogen atoms and the unsaturated zinc atoms in PmII,
cannot be reconciled with the incompetence of 5a and 5b.
These results indicated that the alkyl transfer property of the
Soai autocatalyst PmII could be studied in a manner which is
disconnected from autocatalysis by providing a suitable
surrogate substrate that does not produce an autocatalytically
competent product. Comparison of 3c with 3a and 1c,
respectively, suggests preliminary contributions of the substrate
structure independent of the (auto)catalyst attributes, as (1)
increasing the bulk of the alkynyl substituent reduces the
catalytic rate, (2) a pyrimidine substrate is more active than a
pyridine, and (3) these changes have a minimal effect on
enantioselectivity of the catalytic transformation. Note that
these conclusions specifically delineate substrate contributions to
the overall catalysis by PmII.
In continuing the investigation of structural changes, the
effect of the zinc alkoxide substituent was examined. The
catalytic activity of alkoxide PmIE can be tested by including
2b as an additive (resulting in the rapid formation of PmIE
under reaction conditions) in the diethylzinc alkylation of
various substrates (Figure 8). This species represents a
combination in which the carbinol alkyl group (isopropyl) is
distinct from the (alkoxy)zinc alkyl group (ethyl). The
resultant mixed-catalyst experiments can hence show only
conventional catalysis without a possibility of autocatalysis.
The results of these catalyzed reactions proved to be extremely
informative. First, the control reactions of aldehydes 1b, 3a,
3b, 1c, and 3c showed no autocatalysis with diethylzinc (for
example, in the case of 1b, see Figure 7). With inclusion of
PmIE, varying efficiencies of catalytic diethylzinc alkylations
were observed (Figure 8). With aldehyde 1b, a significant rate
enhancement and a positive non-linear effect are seen (entry
1). The striking difference in the behavior of 1b here, in
comparison to results in entry 4, Figure 6 (which shows the
lack of autocatalysis in the diethylzinc alkylation of 1b, that is,
the catalytic incompetence of PmEE), demonstrates that a
subtle change in the alkoxide structure from PmEE to PmIE
turns on amplifying asymmetric catalysis even with diethylzinc.
It thus appears that unlike PmEE, PmIE can potentially
construct, at least to some meaningful degree, a catalytic
structure similar to PmII (note that partial cube escape seen in
the case of PyIE must be also operative in the case of PmIE
with the difference that the activity of the SMS tetramer in the
(1) The bulky isopropyl groups enable cube escape from the
catalytically inactive cubic tetramer.
(2) Only a single nitrogen atom is necessary in the aromatic
core to assemble the autocatalytically active SMS
tetramer.
(3) The SMS tetramer can process substrates belonging to
the pyridine-3-carbaldehyde scaffold; the aromatic
nitrogen is indispensable for two-point binding.
(4) The substrate alkynyl substituent increases alkyl transfer
enantioselectivity by disfavoring the minor transition
state through an unfavorable steric interaction.
Results of the mixed catalyst−substrate experiments with PmII
and PmIE, while being consistent with all the above
conclusions, provide additional inferences to this list:
(5) The substrate alkynyl substituent decreases reactivity
(while increasing selectivity).
(6) The pyrimidine substrates are more reactive, presumably
because of their increased electrophilicity in comparison
to the pyridine substrates.
(7) Diethylzinc, in principle, can be accepted as a reagent by
the SMS tetramer to affect carbonyl alkylation, albeit
with lower enantioselectivity as compared to diisopro-
pylzinc.
I
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 9. Transition-state structures for NII-catalyzed alkyl transfer to 3a. Calculated at the M06-2X/def2-TZVPP-SMD (toluene)//B3LYP/6-
31G(d) level of theory. Relative energies reported in kcal/mol. Purple, zinc; yellow, silicon; red, oxygen; blue, nitrogen; gray, carbon. Hydrogens
are hidden for clarity.
A case for the role of the alkoxide alkynyl substituent can be
made from the following observations:65 (1) The reaction of
pyridine-3-carbaldehyde (3a) with diisopropylzinc does not
show amplifying autocatalysis, and the product alkoxide is
poorly soluble. (2) In our hands, the reaction of pyrimidine-5-
carbaldehyde (1a) with diisopropylzinc shows asymmetric
amplification but with a lower potency than the alkynyl
analogs, with the product alkoxide being poorly soluble. (3)
The alkylation reactions of 3a and 1a are heterogeneous;
precipitation of the product alkoxide is observed early in the
reaction. In situ monitoring of the diisopropylzinc alkylation of
1a did not give a sigmoidal profile despite being autocatalytic,
perhaps because alkoxide precipitation results in a constant
solution-state catalyst concentration. The alkynyl substituent
acts favorably by increasing the solubility of the product
alkoxide as well as by minimizing formation of unproductive
network aggregates by providing a steric shield on one side of
the molecule. Thus, 3a is autocatalytically incompetent owing
to formation of insoluble, unproductive aggregates of the
product alkoxide. Such an aggregation is presumably mitigated
in the case of 1a because of weaker binding interactions of the
pyrimidine nitrogen, allowing the maintenance in solution of a
critical concentration of active catalyst necessary to promote
asymmetric amplification. It is important to note that
precipitation is not the fundamental cause of autocatalytic
amplification in the Soai reaction (reactions with the stalwart,
highly efficient alkynyl-substituted substrates, for example, 1b
and 3b, are homogeneous). However, precipitation effects may
play an additional role in some casesfor example, with 3a,
our observations (and comparisons with reaction of a 1a)
suggest that poor solubility of the product alkoxide contributes
unfavorably.
heterochiral (R)-4a product is slightly favored (0.2 kcal/mol)
in comparison to TS2, which leads to the homochiral (S)-4a
product. In contrast to the PyII system, these transition
structures are almost isoenergetic, implying a poorly
enantioselective alkyl transfer by the NII tetramer.
Hence, the diisopropylzinc alkylation of 3a does not show
amplifying autocatalysis for at least two reasons: (1) the poor
solubility of the product alkoxide and (2) the poor
enantioselectivity of the NII SMS tetramer (the solution
concentration of which is already depleted due to poor
solubility)-catalyzed alkyl transfer pathway. On the basis of
these results, the structure−activity inference list is extended
further:
(8) From the perspective of the autocatalyst, the alkynyl
substituent contributes to efficient, enantioselective
activity by (a) improving solubility characteristics, (b)
directing the alkoxide away from unproductive aggre-
gates by providing a steric shield on one side of the
molecule, and (c) favoring a specific conformation of the
SMS tetramer that increases enantioselectivity to
meaningful levels.
The incompetence of substrates 1c and 3c (containing the
triisopropylsilylalkynyl substituent; both these systems are
highly soluble) in the Soai reaction is among the most
bewildering idiosyncrasies of the Soai system. However, the
mixed catalyst−substrate experiments demonstrated that both
of these were competent substrates for PmII (Figure 7), PmIE
(Figure 8), and PyII.65 Clearly, these aldehydes can undergo
positive, non-linear enantioselective alkylation with a viable
SMS tetramer. Thus, it can be concluded that the TIPS
substituent maintains substrate competence but renders the
corresponding SMS tetramer (derived from isopropylzinc
alkoxides of 2c and 4c) inactive, hence impeding autocatalysis.
Spectroscopic studies indicate that the isopropylzinc alkoxide
of 4c exists as a tetramer.65 It is believed that this tetramer may
exist in a highly hindered SMS conformation, which precludes
substrate approach and binding.
To gain further insight into the incompetence of 3a to
undergo amplifying autocatalysis with diisopropylzinc, floor-to-
floor transition states for alkyl transfer to 3a by the expected
SMS tetramer arising from the isopropylzinc alkoxide of 4a
(termed NII, equivalent to PyII minus the trimethylsilylalkynyl
substituentthus representing the theoretical autocatalytic
alkylation of 3a by diisopropylzinc) were computationally
modeled (Figure 9). Transition structure TS1 leading to the
Taken together, these conclusions paint a qualitative picture
of the contributions of various structural components in the
J
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 10. Summary of the roles played by various structural constituents in effecting catalysis and selectivity in the Soai system.
Chart 1. Substituted Nicotinaldehydes Tested for Asymmetric Autocatalysis with Diisopropylzinc
Soai reaction toward effecting amplifying asymmetric auto-
catalysis and are summarized in Figure 10.
1). Compounds with substituents at either the 5- or 6-position,
as well as 5,6-disubstituted substrates were tested.
2.2. Investigating Asymmetric Autocatalysis with
Pyridine-3-carbaldehyde Derivatives. The observation of
amplifying, asymmetric autocatalysis with 3b stimulated a
survey of diisopropylzinc alkylations of other substituted
pyridine-3-carbaldehyde (nicotinaldehyde) derivatives (Chart
Aldehydes 7, 8, 9, 10, and 14 were obtained from
commercial sources whereas 11, 12, 13, and 15 were prepared
by Suzuki−Miyaura or Sonogashira coupling reactions of the
parent 6-bromonicotinaldehyde.6 Aldehydes 16, 17, and 18
were accessed by routes originating from 5- bromonicotin-
K
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 1. Synthesis of 5,6-Disubstituted Nicotinaldehydes
aldehyde (14) and involved the common intermediate 5-
bromo-6- iodonicotinaldehyde (21) (Scheme 1). Deprotona-
tion of the known dioxolane derivative 20 with tetramethylpi-
peridinylmagnesium chloride-lithium chloride66 followed by
trapping with iodine afforded the key intermediate 21.
Sonogashira coupling of 21 with 1.0 equiv of ethynyltrime-
thylsilane afforded 22 in high yield which, after dioxolane
deprotection provided target aldehyde 16. Likewise, Sonoga-
shira coupling of 21 with excess ethynyltrimethylsilane and
subsequent dioxolane deprotection in a single pot allows direct
access to 17. The 5-methyl substituent was introduced by a
Kumada coupling of 22 and methylmagnesium chloride and
was followed by dioxolane deprotection to afford 18 in good
yield. Buchwald (t-Bu3P)2Pd-G3 precatalyst was uniquely
successful in the Kumada coupling whereas other palladium−
ligand combinations gave poor yields.
A similar strategy to access 19 was unsuccessful because
treatment of the protected dioxolane obtained from 23 (3-
(1,3-dioxolan-2-yl)-5-fluoropyridine) with a variety of bases
followed by trapping with iodine predominantly yielded
iodination at the 4-position. Thus, the TIPS ether 24 was
utilized for iodination with tetramethylpiperidinylmagnesium
chloride, lithium chloride, and iodine. Carbinol 25 obtained in
this way was alkynylated under Sonogashira coupling
conditions to yield 26. Finally, Swern oxidation of 26 afforded
the target aldehyde 19 in high yield.
In a series of in situ IR-monitored reactions, each aldehyde
was tested for asymmetric autocatalysis by inclusion of the
corresponding scalemic product carbinol in the diisopropylzinc
alkylation reaction. These reactions were compared to
background reactions without any added product. In general,
most of the substrates in Chart 1 displayed poor autocatalytic
rate enhancement compared to the background reaction.65
Apart from 3b, amplifying asymmetric autocatalysis was
observed only with 12, 16, 18, and 19.
Reaction of substrate 12 without added catalyst (entry 1,
Figure 11) shows sigmoidal aldehyde consumption. Inclusion
of scalemic product carbinol (entry 2), results in drastic rate
acceleration with the product carbinol being more enantioen-
riched (91:9 e.r.) than the added catalyst (71:29 e.r.). These
results contradict a report by Amedjkouh et al. wherein this
transformation was reported to proceed with enantiomeric
erosion.67 From our perspective, Amedjkouh et al.’s findings of
enantiomeric amplification in the diisopropylzinc alkylation of
12, with 2b as the catalyst are consistent with 12 itself being
autocatalytically competent.
Reactions of 5,6-disubstituted nicotinaldehyde substrates 16
and 18 with diisopropylzinc show asymmetric amplification
under autocatalytic conditions, however the rate enhancement
compared to the background reaction is only modest (Figure
12, entries 1−4). In case of 16, a number of unidentified side
products were observed. Qualitatively, the reaction of 18
appears to be more efficient, with cleaner conversion and a
L
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 11. Amplifying asymmetric autocatalysis in the reaction of 12 with diisopropylzinc.
higher final product e.r. Similar experiments with substrate 17
led to poor conversion and multiple side products without any
asymmetric amplification in the target carbinol.65 Conversely,
autocatalyzed diisopropylzinc alkylation of 19 proceeded with
robust asymmetric amplification and high rate enhancement
compared to the background reaction (Figure 12, entries 5, 6),
mirroring the behavior seen with substrate 3b. This outcome
demonstrates that in the diisopropylzinc alkylations of 6-
trimethylsilylethynyl-nicotinaldehydes, increasing the bulk of
the 5-substituent adversely affects asymmetric autocatalytic
behavior. The autocatalytic isopropylzinc alkoxide generated in
the reaction with substrate 19 is labeled FPyII (FPy =
fluoropyridine, I = isopropyl) and is discussed below.
2.3. A Comparison of PmII, PyII, and FPyII Systems.
2.3.1. Comparison of Initial Rates with Nicotinaldehyde.
The autocatalytic systems arising from substrates 1b, 3b, and
19 differ only in the heteroaromatic core. The pyrimidine
nitrogen, with lower Lewis basicity than the pyridine,68,69 is
expected to display weaker binding to the Lewis acidic
diisopropylzinc reagent as well as to the O−Zn moiety.
Likewise, the FPyII nitrogen should be less Lewis basic than
the PyII nitrogen (pKa: pyridine = 5.23, 3-fluoropyridine = 3.0,
pyrimidine = 1.30).68,70 The comparison of these three systems
thus provides an interesting opportunity to study the effects of
such electronic changes on the autocatalytic behavior. To
compare the inherent activities of the three alkoxide catalysts
(PmII, PyII, and FPyII) derived from these substrates, a
comparison of initial rates with a common substrate, pyridine-
3-carbaldehyde (3a), was undertaken.
alkoxide catalyst, with in situ IR monitoring of aldehyde
consumption. Initial rates were calculated and are reported as
averages of three replicates (Figure 13). The three catalysts
provide the product carbinol in comparable enantioenrich-
ment. However, initial rates of FPyII and PmII, while
comparable to each other, are at least 2-fold lower than the
rate with PyII. This observation may be rationalized by taking
into account the floor-to-floor TS model for the alkyl transfer
step. The transferring diisopropylzinc in the PmII and FPyII
tetramer is deactivated (as compared to PyII) because of the
weaker Lewis basicity of the coordinating nitrogen, resulting in
a slower elementary alkyl transfer step. Thus, for an identical
substrate, the activities of the three catalysts are in the order
PyII > PmII ∼ FPyII.
2.3.2. Comparison of Autocatalytic Behavior. Autocata-
lytic reactions of substrates 1b (PmII system), 3b (PyII
system), and 19 (FPyII system) were compared with inclusion
of 8 mol% of initial autocatalyst. Under these conditions, rate
of reaction in the PyII system is faster than the FPyII system
(Figure 14) but significantly slower than the PmII system.
While the PmII system reaction is nearly complete within one
minute, less than 10% conversion is seen for the other two at
this time point. In these experiments, each autocatalyst
interrogates a different substrate, and thus, the alkyl transfer
rate is also affected by changes in the electrophilicity of the
substrate aldehyde. On the basis of the increased electron-
withdrawing effect of a pyrimidine ring and a fluoropyridine
group in comparison to an unsubstituted pyridine core, 1b and
19 are expected to be more electrophilic than 3b. This
conflicting balance of catalyst Lewis basicity and substrate
electrophilicity must determine the overall rate of alkyl transfer
Diisopropylzinc alkylation reactions under identical con-
ditions were evaluated in the presence of 10 mol% of the
M
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 12. Amplifying asymmetric autocatalysis in the reaction of 5,6-disubstituted nicotinaldehyde with diisopropylzinc.
Figure 13. Comparison of alkyl transfer rates of PyII, FPyII, and PmII. TS model for the alkyl transfer with PyII highlighting the transferring
diisopropylzinc moiety (blue arrow) bound to the activating arm nitrogen (green arrow).
in the three systems. With this perspective, the relative rates of
the PyII and FPyII system appear reasonable. However, the
extraordinarily high rate of the PmII system compelled further
investigations into factors affecting autocatalytic progression.
2.3.3. Inhibition of Autocatalysis by Excess Diisopropyl-
zinc. The PyII system resembles many aspects of the
autocatalytic PmII system.64 However, studies probing the
effect of excess dialkylzinc reagent led to the discovery of an
N
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 14. Comparison of autocatalytic progression in PyII, FPyII, and PmII systems.
unexpected divergence in the behavior of the two systems.
Consistent with earlier reports,57 autocatalytic progression of
the PmII system was found to be insensitive to diisopropylzinc
concentration, but similar experiments with the PyII system
resulted in inhibition of reaction progression with increasing
reagent concentration. This phenomenon was also noted for
the FPyII system. Increasing diisopropylzinc concentration
also adversely affected the selectivity of the reaction,
presumably by accelerating the uncatalyzed background
addition while concomitantly compromising the catalytic
pathway.
Figure 15 presents the results of these comparative
experiments (all runs are reproducible and represent one
among duplicate experiments). The reactions of the three
systems were carried out under identical conditions, with
varying concentration of diisopropylzinc. The yellow curves
represent reactions with the PmII system, which are extremely
rapid, are invariant in rate with respect to diisopropylzinc
concentration, and provide the product with >99:1 e.r. in all
cases. Reaction profiles with the PyII and FPyII system are
represented by solid and dashed curves, respectively. The
autocatalytic reaction of PyII was inherently faster than that of
FPyII at lower concentrations of zinc reagent (compare entries
1 and 5, also see Figure 14). However, with increasing
diisopropylzinc concentration, reactions of the former were
inhibited more significantly than the latter (entries 1−8).
Thus, at diisopropylzinc concentrations upward of 2.4 equiv,
the FPyII reactions were faster than those of PyII (compare
entries 2 and 6, 3 and 7, 4 and 8). These results demonstrate
that the extent of inhibition by excess diisopropylzinc is greater
for PyII than for FPyII and is entirely absent for PmII (i.e.,
PyII > FPII > PmII).
Qualitatively, the correlation of this trend with decreasing
Lewis basicities of the aromatic cores in these systems provides
suggestions for the cause of this inhibitory behavior (Figure
16). Spectroscopic investigations prove that the substrate
aldehyde is involved in dynamic association with the
dialkylzinc reagent by coordination of the ring nitrogen (red
equilibrium).64 However, such a binding precludes the floor-to-
floor association to the tetrameric catalyst, which is necessary
for catalyst activity. Thus, it is hypothesized that inhibition
results from competitive binding of the reagent to the substrate
ring nitrogen, making the substrate unavailable for association
to the O−Zn center in the catalyst. Consistent with
observations, the extent of inhibition correlates with the
strength of the N−Zn association, which in turn must depend
on the Lewis basicity of the azaaryl moieties. This hypothesis
was further explored by reaction modeling of kinetic data to an
autocatalytic mechanism and model simulations.
2.3.4. Insights from Kinetic ModelingSteps Affecting
Reaction Progression. On the basis of the mechanistic
knowledge gained so far, a minimal set of elementary steps
can be compiled for the definition of the autocatalytic reaction
pathway (the simplified floor-to-floor “SF” model, Figure 17).
These steps are (1) equilibrium binding of substrate aldehyde
(A) and diisopropylzinc (Z) to yield the N−Zn bound adduct
(AZ); (2) equilibrium binding of the catalyst (C) to
diisopropylzinc (Z) at the pyridyl arm (binding by only one
O
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 15. Inhibition of autocatalytic progression in PyII and FPyII systems with increasing diisopropylzinc.
Figure 16. Schematic of the autocatalytic process (represented for the PyII system) and hypothesis for the origin of inhibition in the three systems.
P
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 17. SF model of autocatalysis. Steps affecting rate of reaction progression are highlighted in red. The consequences of these steps in the
PyII, FPyII, and PmII systems are indicated.
arm is considered for simplicity) to yield the active species
(CZ); (3) equilibrium binding of substrate aldehyde (A) to the
catalyst−diisopropylzinc adduct (CZ) to yield the activated
complex (ACZ); (4) alkyl transfer from the SMS tetramer
active complex (ACZ) to yield the catalyst-bound product
alkoxide adduct (PC); (5) equilibrium dissociation of the
product monomer alkoxide (P) to regenerate the SMS
tetramer (C); (6) dimerization of the monomeric alkoxide
(P) product to a dimeric unit (2P); and (7) dimerization of
the dimeric unit (2P) to yield the newly constituted SMS
tetramer (C). Rate constants for each step are designated.
Arguably, this is a simplified picture of the actual catalytic
process, and additional steps can be envisaged. For example,
the binding of two molecules of diisopropylzinc to the catalyst
is ignored; the order of binding of the reagent and aldehyde to
the SMS tetramer may be different than the one suggested
here; reagent binding to the product alkoxide may exist; the
precise mechanism of tetramer reformation is reasonable but
nonetheless speculative; the contribution of an uncatalyzed
background alkylation is ignored; etc. However, the scheme
incorporates all fundamental binding events necessary and
sufficient to achieve compatibility with the “SMS tetramer−
floor to floor (SF)” autocatalysis model.
This model could qualitatively reproduce the autocatalytic
reaction profiles for the PyII system, including the inhibitory
effect of diisopropylzinc, and with chemically realistic kinetic
parameters.65 Model simulations reveal that among the seven
proposed elementary steps, the rate of reaction progression is
sensitive only to the kinetic parameters of steps 1, 3, and 4 (Figure
17, red) in the following manner:65 (1) decreasing the
Q
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
association constant K1a results in faster reactions and
decreasing extent of diisopropylzinc inhibition, mimicking
the transitions from the PyII (to FPyII) to PmII systems; (2)
an increase in the value of K3a results in faster reactions; and
(3) an increase in the value of kc results in faster reactions.
To a first approximation, the relative rates of the PyII,
FPyII, and PmII systems and the differences in severity of
diisopropylzinc inhibition originate from divergent contribu-
tions from the three autocatalytic-progression-determining
steps in the SF model (Figure 17). Considering the Lewis
basicity of the aromatic nitrogen atoms, the K1a values for the
three systems are expected to be in the order PmII ≪ FPyII <
PyII.68−70 Moreover, the adduct for the pyrimidine system
(red box) is, in principle, a competent species because floor-to-
floor binding is still feasible with the second pyrimidine
nitrogen atom. The binding constant for reagent coordination
to this second nitrogen (necessary to render the adduct
incompetent) is expected to be very low. Taken together, these
factors must contribute to a very low effective value of K1a in
the pyrimidine system, thus accounting for its extraordinary
rate and absence of inhibitory behavior (this is reproduced in
the model simulations). The qualitative prediction of K3a and kc
is non-trivial. In the three systems, the Lewis basicity of the
substrate nitrogen atoms and the Lewis acidities of the
unsaturated tetramer zinc atoms follow opposite trends, thus
confounding a comparative estimate of the association
constant for step 3. Similarly, the nucleophilicity of the
catalyst-arm-bound diisopropylzinc molecule and the electro-
philicity of the bound substrate follow opposite trends in the
three systems, which complicates a comparative estimate of the
alkyl transfer rate constant. Ultimately, an interplay of
contributions from these steps must dictate the overall trend
of reaction progression. Kinetic simulations are consistent with
this hypothesis; modifying the values of K1a, K3a, and kc to
mimic the three systems can indeed reproduce their relative
rate profiles.65 Taken together, these preliminary modeling
studies reveal how the substrate−reagent binding equilibrium,
the catalyst−substrate binding equilibrium, and the alkyl
transfer rate contribute to (1) rate of autocatalytic progression,
(2) the origin of the extraordinarily rapid reactions in the PmII
system, and (3) the origin of the relative extents of inhibition
in the PyII and FPyII systems.
of the pyrimidine nitrogen abrogates inhibition effects and
results in superlative performance in the Soai reaction. These
structural insights are completely consistent with the floor-to-
floor transition-state model presented earlier.64
The mechanistic landscape of the Soai reaction can now be
summarized as follows: (1) The diisopropylzinc alkylation of
Soai substrates affords isopropylzinc alkoxides that can access
the SMS-tetramer by a combination of steric effects and
nitrogen coordination (cube-escape model). (2) In general, this
tetramer can enantioselectively alkylate a variety of aldehydes
that belong to the pyridine-3-carbaldehyde scaffold (which
includes pyrimidine-5-carbaldehydes) via a two-point floor-to-
floor modelan autocatalytic reaction is simply a special case
occurring for a “matched” substrate. (3) Structural effects play
a critical role in the assembly of the SMS tetramer, its
solubility, active conformation, and substrate accessibility, as
well as the reactivity and selectivity imparted by the
substratethe rationalization of these effects largely explains
the idiosyncratic substrate scope of this transformation. (4)
Reaction progression is dictated by the strength of N−Zn
binding eventssome of which are inhibitorythe low Lewis
basicity of the pyrimidine nitrogen allows this system to
overcome such inhibitory effects and attain superior auto-
catalytic rates. (5) The striking non-linear behavior of the
reaction is attributed to the poor catalytic competence of the
heterochiral SMS tetramer that cannot engage the substrate in
the floor-to-floor binding mode.
The precise manner in which the autocatalytic cycle is
completed (dissociation of the product alkoxide and its re-
assembly into the SMS tetramer) remains to be elucidated.
Likewise, for a scalemic SMS tetramer, multiple inter-
converting isomeric forms (due to alternative configurations
of the chiral, tetracoordinate alkoxyzinc atoms) are theoret-
ically possible. A rigorous study on the distribution and relative
catalytic activities of these isomers has not yet been performed.
In our opinion, these outstanding investigations, anticipated to
be predominantly computational in enquiry, will provide
further mechanistic understanding of the remarkable Soai
reaction.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c05994.
3. CONCLUSIONS AND OUTLOOK
The strategy of mixed catalyst−substrate experiments allowed
extensive study of structure−activity relationships in the
pyrimidine and pyridine systems. The results from these
studies enabled a formulation of the roles played by structural
constituents in effecting catalysis and selectivity in the Soai
reaction. Contributions of the alkynyl substituent toward
catalyst solubility, autocatalyst conformation, aggregate for-
mation, and enhancing substrate-directed reactivity and
enantioselectivity were elucidated. Similarly, subtle correlations
of the alkyl groups and the aromatic core were identified.
An exploration of substituted nicotinaldehyde derivatives led
to the determination of novel, competent substrates. Among
these, the 5-fluoro-6-(trimethylsilylethynyl)-nicotinaldehyde
(FPyII) system provided an opportunity to carry out a careful
comparison between three autocatalytic systems in regard to
the Lewis basicity of the aromatic nitrogen. Model kinetic
simulations revealed how this factor critically influences the
interplay of three steps in the autocatalytic pathway that are
determinants of reaction progression. The low Lewis basicity
Full experimental details and characterization data along
with descriptions of ancillary experiments (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Scott E. Denmark − Roger Adams Laboratory, Department of
Chemistry, University of Illinois, Urbana, Illinois 61801, United
States; orcid.org/0000-0002-1099-9765;
Email: sdenmark@illinois.edu
Authors
Soumitra V. Athavale − Roger Adams Laboratory, Department
of Chemistry, University of Illinois, Urbana, Illinois 61801,
United States; orcid.org/0000-0001-5620-6997
Adam Simon − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095, United
States
R
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
to Pyrimidine-5-Carbaldehyde in Conjunction with Asymmetric
Autocatalysis. Tetrahedron: Asymmetry 2003, 14 (2), 185−188.
(14) Kaimori, Y.; Hiyoshi, Y.; Kawasaki, T.; Matsumoto, A.; Soai, K.
Formation of Enantioenriched Alkanol with Stochastic Distribution of
Enantiomers in the Absolute Asymmetric Synthesis under Heteroge-
neous Solid−Vapor Phase Conditions. Chem. Commun. 2019, 55
(36), 5223−5226.
(15) Singleton, D. A.; Vo, L. K. Enantioselective Synthesis without
Discrete Optically Active Additives. J. Am. Chem. Soc. 2002, 124 (34),
10010−10011.
K. N. Houk − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095, United
States; orcid.org/0000-0002-8387-5261
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c05994
Notes
The authors declare no competing financial interest.
(16) Singleton, D. A.; Vo, L. K. A Few Molecules Can Control the
Enantiomeric Outcome. Evidence Supporting Absolute Asymmetric
Synthesis Using the Soai Asymmetric Autocatalysis. Org. Lett. 2003, 5
(23), 4337−4339.
ACKNOWLEDGMENTS
■
We are grateful for generous financial support from the
University of Illinois. S.V.A. is grateful to the University of
Illinois for Graduate Fellowships. A.S. thanks the NIH
Chemistry-Biology Interface Research Training Grant
(T32GM008496). We are also grateful for the support services
of the NMR, mass spectrometry, and microanalytical
laboratories of the University of Illinois at Urbana−
Champaign.
(17) Mislow, K. Absolute Asymmetric Synthesis: A Commentary.
Collect. Czech. Chem. Commun. 2003, 68, 849−864.
(18) Sato, I.; Ohgo, Y.; Igarashi, H.; Nishiyama, D.; Kawasaki, T.;
Soai, K. Determination of Absolute Configurations of Amino Acids by
Asymmetric Autocatalysis of 2-Alkynylpyrimidyl Alkanol as a Chiral
Sensor. J. Organomet. Chem. 2007, 692 (9), 1783−1787.
(19) Kawasaki, T.; Jo, K.; Igarashi, H.; Sato, I.; Nagano, M.;
Koshima, H.; Soai, K. Asymmetric Amplification Using Chiral
Cocrystals Formed from Achiral Organic Molecules by Asymmetric
Autocatalysis. Angew. Chem., Int. Ed. 2005, 44 (18), 2774−2777.
(20) Sato, I.; Yamashima, R.; Kadowaki, K.; Yamamoto, J.; Shibata,
T.; Soai, K. Asymmetric Induction by Helical Hydrocarbons: [6]- and
[5]Helicenes. Angew. Chem., Int. Ed. 2001, 40 (6), 1096−1098.
(21) Kawasaki, T.; Tanaka, H.; Tsutsumi, T.; Kasahara, T.; Sato, I.;
Soai, K. Chiral Discrimination of Cryptochiral Saturated Quaternary
and Tertiary Hydrocarbons by Asymmetric Autocatalysis. J. Am.
Chem. Soc. 2006, 128 (18), 6032−6033.
(22) Soai, K.; Kawasaki, T.; Matsumoto, A. Asymmetric
Autocatalysis of Pyrimidyl Alkanol and Its Application to the Study
on the Origin of Homochirality. Acc. Chem. Res. 2014, 47 (12), 3643−
3654.
(23) Kawasaki, T.; Sato, M.; Ishiguro, S.; Saito, T.; Morishita, Y.;
Sato, I.; Nishino, H.; Inoue, Y.; Soai, K. Enantioselective Synthesis of
Near Enantiopure Compound by Asymmetric Autocatalysis Triggered
by Asymmetric Photolysis with Circularly Polarized Light. J. Am.
Chem. Soc. 2005, 127 (10), 3274−3275.
(24) Kawasaki, T.; Matsumura, Y.; Tsutsumi, T.; Suzuki, K.; Ito, M.;
Soai, K. Asymmetric Autocatalysis Triggered by Carbon Isotope
(13C/12C) Chirality. Science 2009, 324 (5926), 492−495.
(25) Kawasaki, T.; Okano, Y.; Suzuki, E.; Takano, S.; Oji, S.; Soai, K.
Asymmetric Autocatalysis: Triggered by Chiral Isotopomer Arising
from Oxygen Isotope Substitution. Angew. Chem., Int. Ed. 2011, 50
(35), 8131−8133.
(26) Matsumoto, A.; Ozaki, H.; Harada, S.; Tada, K.; Ayugase, T.;
Ozawa, H.; Kawasaki, T.; Soai, K. Asymmetric Induction by a
Nitrogen 14N/15N Isotopomer in Conjunction with Asymmetric
Autocatalysis. Angew. Chem., Int. Ed. 2016, 55, 15246−15249.
(27) Blackmond, D. G. The Origin of Biological Homochirality. Cold
Spring Harbor Perspect. Biol. 2010, 2 (5), a002147.
REFERENCES
■
(1) Soai, K.; Shibata, T.; Morioka, H.; Choji, K. Asymmetric
Autocatalysis and Amplification of Enantiomeric Excess of a Chiral
Molecule. Nature 1995, 378 (6559), 767−768.
(2) Although the active catalyst is certainly derived from the in situ
generated zinc alkoxide, the term “catalyst”, for simplicity throughout
this paper, is used interchangeably for the carbinol or the alkoxide.
Similarly, the term “product” is used interchangeably for the product
zinc alkoxide or the product carbinol obtained after workup.
(3) Shibata, T.; Morioka, H.; Hayase, T.; Choji, K.; Soai, K. Highly
Enantioselective Catalytic Asymmetric Automultiplication of Chiral
Pyrimidyl Alcohol. J. Am. Chem. Soc. 1996, 118 (2), 471−472.
(4) Shibata, T.; Yonekubo, S.; Soai, K. Practically Perfect
Asymmetric Autocatalysis with (2-Alkynyl-5-Pyrimidyl)Alkanols.
Angew. Chem., Int. Ed. 1999, 38 (5), 659−661.
(5) Sato, I.; Yanagi, T.; Soai, K. Highly Enantioselective Asymmetric
Autocatalysis of 2-Alkenyl- and 2-Vinyl-5-Pyrimidyl Alkanols with
Significant Amplification of Enantiomeric Excess. Chirality 2002, 14
(2−3), 166−168.
(6) Lutz, F.; Kawasaki, T.; Soai, K. Asymmetric Autocatalysis of a
Ferrocene-Containing Chiral Compound with Amplification of
Chirality. Tetrahedron: Asymmetry 2006, 17 (4), 486−490.
(7) Busch, M.; Schlageter, M.; Weingand, D.; Gehring, T. Systematic
Studies Using 2-(1-Adamantylethynyl)Pyrimidine-5-Carbaldehyde as
a Starting Material in Soai’s Asymmetric Autocatalysis. Chem. - Eur. J.
2009, 15 (33), 8251−8258.
(8) Kawasaki, T.; Kamimura, S.; Amihara, A.; Suzuki, K.; Soai, K.
Enantioselective C-C Bond Formation as a Result of the Oriented
Prochirality of an Achiral Aldehyde at the Single-Crystal Face upon
Treatment with a Dialkyl Zinc Vapor. Angew. Chem., Int. Ed. 2011, 50
(30), 6796−6798.
(9) Frank, F. C. On Spontaneous Asymmetric Synthesis. Biochim.
Biophys. Acta 1953, 11, 459−463.
(10) Wynberg, H. Asymmetric Autocatalysis: Facts and Fancy. J.
Macromol. Sci., Chem. 1989, 26 (8), 1033−1041.
(11) Sato, I.; Urabe, H.; Ishiguro, S.; Shibata, T.; Soai, K.
Amplification of Chirality from Extremely Low to Greater than
99.5% ee by Asymmetric Autocatalysis. Angew. Chem., Int. Ed. 2003,
42 (3), 315−317.
(12) Soai, K.; Shibata, T.; Kowata, Y. Asymmetric Synthesis of
Enantioenriched Alkanol by Spontaneous Asymmetric Synthesis.
Japan Kokai Tokkyo Koho JP 9-268179, 1997.
(28) Flugel, R. M. Chirality and Life: A Short Introduction to the Early
̈
Phases of Chemical Evolution, 1st ed.; Springer-Verlag: Berlin
Heidelberg, 2011.
(29) Weissbuch, I.; Lahav, M. Crystalline Architectures as Templates
of Relevance to the Origins of Homochirality. Chem. Rev. 2011, 111
(5), 3236−3267.
(30) Cintas, P. Biochirality: Origins, Evolution and Molecular
Recognition, 1st ed.; Cintas, P., Ed.; Topics in Current Chemistry;
Springer-Verlag: Berlin Heidelberg, 2013.
́
̈
(31) Barabas, B.; Fulop, O. Graph Theoretical and Statistical
̈
Analysis of the Impact of Soai Reaction on Natural Sciences. In
Advances in Asymmetric Autocatalysis and Related Topics; Palyi, G.,
Kurdi, R., Zucchi, C., Eds.; Academic Press, 2017; Chap. 3.
(32) Gehring, T.; Busch, M.; Schlageter, M.; Weingand, D. A
Concise Summary of Experimental Facts about the Soai Reaction.
Chirality 2010, 22 (1E), E173−E182.
(13) Soai, K.; Sato, I.; Shibata, T.; Komiya, S.; Hayashi, M.;
Matsueda, Y.; Imamura, H.; Hayase, T.; Morioka, H.; Tabira, H.;
Yamamoto, J.; Kowata, Y. Asymmetric Synthesis of Pyrimidyl Alkanol
without Adding Chiral Substances by the Addition of Diisopropylzinc
S
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(33) Soai, K.; Matsumoto, A.; Kawasaki, T. Asymmetric
Autocatalysis and the Origins of Homochirality of Organic
Compounds. In Advances in Asymmetric Autocatalysis and Related
Topics; Palyi, G., Kurdi, R., Zucchi, C., Eds.; Academic Press, 2017;
Chap. 1.
in the Soai Reaction: A Kinetic Understanding. Proc. Natl. Acad. Sci.
U. S. A. 2005, 102 (39), 13743−13748.
́
́
(52) Micskei, K.; Pota, G.; Caglioti, L.; Palyi, G. Empirical
Description of Chiral Autocatalysis. J. Phys. Chem. A 2006, 110
(18), 5982−5984.
(53) Schiaffino, L.; Ercolani, G. Unraveling the Mechanism of the
Soai Asymmetric Autocatalytic Reaction by First-Principles Calcu-
lations: Induction and Amplification of Chirality by Self-Assembly of
Hexamolecular Complexes. Angew. Chem., Int. Ed. 2008, 47 (36),
6832−6835.
́
(34) Buhse, T.; Noble-Teran, M. E.; Cruz, J.-M.; Micheau, J.-C.;
Coudret, C. Kinetic and Structural Aspects of Mirror-Image
Symmetry Breaking in the Soai Reaction. In Advances in Asymmetric
Autocatalysis and Related Topics Palyi, G., Kurdi, R., Zucchi, C., Eds.;
Academic Press, 2017; Chap. 4.
́
́
́
(54) Micskei, K.; Rabai, G.; Gal, E.; Caglioti, L.; Palyi, G. Oscillatory
Symmetry Breaking in the Soai Reaction. J. Phys. Chem. B 2008, 112
(30), 9196−9200.
(35) Blackmond, D. G. Autocatalytic Models for the Origin of
Biological Homochirality. Chem. Rev. 2020, 120, 4831−4847.
(36) Soai, K.; Niwa, S.; Hori, H. Asymmetric Self-Catalytic Reaction.
Self-Production of Chiral 1-(3-Pyridyl)Alkanols as Chiral Self-
Catalysts in the Enantioselective Addition of Dialkylzinc Reagents
to Pyridine-3-Carbaldehyde. J. Chem. Soc., Chem. Commun. 1990,
No. 14, 982−983.
(37) Klankermayer, J.; Gridnev, I. D.; Brown, J. M. Role of the
Isopropyl Group in Asymmetric Autocatalytic Zinc Alkylations. Chem.
Commun. 2007, No. 30, 3151−3153.
(38) Gridnev, I. D.; Serafimov, J. M.; Quiney, H.; Brown, J. M.
Reflections on Spontaneous Asymmetric Synthesis by Amplifying
Autocatalysis. Org. Biomol. Chem. 2003, 1 (21), 3811−3819.
(39) Shibata, T.; Choji, K.; Hayase, T.; Aizu, Y.; Soai, K. Asymmetric
Autocatalytic Reaction of 3-Quinolylalkanol with Amplification of
Enantiomeric Excess. Chem. Commun. 1996, No. 10, 1235−1236.
(40) Shibata, T.; Choji, K.; Morioka, H.; Hayase, T.; Soai, K. Highly
Enantioselective Synthesis of a Chiral 3-Quinolylalkanol by an
Asymmetric Autocatalytic Reaction. Chem. Commun. 1996, No. 6,
751−752.
(41) Sato, I.; Nakao, T.; Sugie, R.; Kawasaki, T.; Soai, K.
Enantioselective Synthesis of Substituted 3-Quinolyl Alkanols and
Their Application to Asymmetric Autocatalysis. Synthesis 2004, 9,
1419−1428.
(42) Tanji, S.; Kodaka, Y.; Ohno, A.; Shibata, T.; Sato, I.; Soai, K.
Asymmetric Autocatalysis of 5-Carbamoyl-3-Pyridyl Alkanols with
Amplification of Enantiomeric Excess. Tetrahedron: Asymmetry 2000,
11 (21), 4249−4253.
(43) Hawbaker, N. A.; Blackmond, D. G. Rationalization of
Asymmetric Amplification via Autocatalysis Triggered by Isotopically
Chiral Molecules. ACS Cent. Sci. 2018, 4 (6), 776−780.
(44) Hawbaker, N. A.; Blackmond, D. G. Energy Threshold for
Chiral Symmetry Breaking in Molecular Self-Replication. Nat. Chem.
2019, 11 (10), 957−962.
(45) Blackmond, D. G.; McMillan, C. R.; Ramdeehul, S.; Schorm,
A.; Brown, J. M. Origins of Asymmetric Amplification in Autocatalytic
Alkylzinc Additions. J. Am. Chem. Soc. 2001, 123 (41), 10103−10104.
(46) Buhse, T. A Tentative Kinetic Model for Chiral Amplification
in Autocatalytic Alkylzinc Additions. Tetrahedron: Asymmetry 2003,
14, 1055−1061.
(47) Sato, I.; Omiya, D.; Igarashi, H.; Kato, K.; Ogi, Y.; Tsukiyama,
K.; Soai, K. Relationship between the Time, Yield, and Enantiomeric
Excess of Asymmetric Autocatalysis of Chiral 2-Alkynyl-5-Pyrimidyl
Alkanol with Amplification of Enantiomeric Excess. Tetrahedron:
Asymmetry 2003, 14 (8), 975−979.
(55) Schiaffino, L.; Ercolani, G. Amplification of Chirality and
Enantioselectivity in the Asymmetric Autocatalytic Soai Reaction.
ChemPhysChem 2009, 10 (14), 2508−2515.
(56) Micheau, J.-C.; Cruz, J.-M.; Coudret, C.; Buhse, T. An
Autocatalytic Cycle Model of Asymmetric Amplification and Mirror-
Symmetry Breaking in the Soai Reaction. ChemPhysChem 2010, 11
(16), 3417−3419.
(57) Quaranta, M.; Gehring, T.; Odell, B.; Brown, J. M.; Blackmond,
D. G. Unusual Inverse Temperature Dependence on Reaction Rate in
the Asymmetric Autocatalytic Alkylation of Pyrimidyl Aldehydes. J.
Am. Chem. Soc. 2010, 132 (43), 15104−15107.
(58) Ercolani, G.; Schiaffino, L. Putting the Mechanism of the Soai
Reaction to the Test: DFT Study of the Role of Aldehyde and
Dialkylzinc Structure. J. Org. Chem. 2011, 76 (8), 2619−2626.
(59) Gridnev, I. D.; Vorobiev, A. K. Quantification of Sophisticated
Equilibria in the Reaction Pool and Amplifying Catalytic Cycle of the
Soai Reaction. ACS Catal. 2012, 2 (10), 2137−2149.
(60) Matsumoto, A.; Abe, T.; Hara, A.; Tobita, T.; Sasagawa, T.;
Kawasaki, T.; Soai, K. Crystal Structure of the Isopropylzinc Alkoxide
of Pyrimidyl Alkanol: Mechanistic Insights for Asymmetric
Autocatalysis with Amplification of Enantiomeric Excess. Angew.
Chem., Int. Ed. 2015, 54 (50), 15218−15221.
(61) Matsumoto, A.; Kawasaki, T.; Soai, K. Structural Study of
Asymmetric Autocatalysis by X-Ray Crystallography. In Advances in
Asymmetric Autocatalysis and Related Topics; Palyi, G., Kurdi, R.,
Zucchi, C., Eds.; Academic Press, 2017; Chap. 10.
(62) Gridnev, I. D.; Brown, J. M. Asymmetric Catalysis Special
Feature Part II: Asymmetric Autocatalysis: Novel Structures, Novel
Mechanism? Proc. Natl. Acad. Sci. U. S. A. 2004, 101 (16), 5727−
5731.
(63) Matsumoto, A.; Fujiwara, S.; Abe, T.; Hara, A.; Tobita, T.;
Sasagawa, T.; Kawasaki, T.; Soai, K. Elucidation of the Structures of
Asymmetric Autocatalyst Based on X-Ray Crystallography. Bull.
Chem. Soc. Jpn. 2016, 89 (10), 1170−1177.
(64) Athavale, S. V.; Simon, A.; Houk, K. N.; Denmark, S. E.
Demystifying the Asymmetry-Amplifying, Autocatalytic Behaviour of
the Soai Reaction through Structural, Mechanistic and Computational
Studies. Nat. Chem. 2020, 12 (4), 412−423.
(65) Refer to the Supporting Information.
(66) Krasovskiy, A.; Krasovskaya, V.; Knochel, P. Mixed Mg/Li
Amides of the Type R2NMgCl·LiCl as Highly Efficient Bases for the
Regioselective Generation of Functionalized Aryl and Heteroaryl
Magnesium Compounds. Angew. Chem., Int. Ed. 2006, 45, 2958−
2961.
(67) Romagnoli, C.; Sieng, B.; Amedjkouh, M. Asymmetric
Amplification Coupling Enantioselective Autocatalysis and Asym-
metric Induction for Alkylation of Azaaryl Aldehydes. Eur. J. Org.
Chem. 2015, 2015 (19), 4087−4092.
(68) Albert, A.; Goldacre, R.; Phillips, J. 455. The Strength of
Heterocyclic Bases. J. Chem. Soc. 1948, 2240−2249.
(69) Laurence, C.; Gal, J.-F. Lewis Basicity and Affinity Scales: Data
and Measurement; John Wiley & Sons Ltd., 2010.
(70) Clarke, K.; Rothwell, K. 377. A Kinetic Study of the Effect of
Substituents on the Rate of Formation of Alkylpyridinium Halides in
Nitromethane Solution. J. Chem. Soc. 1960, 1885.
(48) Buono, F. G.; Blackmond, D. G. Kinetic Evidence for a
Tetrameric Transition State in the Asymmetric Autocatalytic
Alkylation of Pyrimidyl Aldehydes. J. Am. Chem. Soc. 2003, 125
(30), 8978−8979.
(49) Gridnev, I. D.; Serafimov, J. M.; Brown, J. M. Solution
Structure and Reagent Binding of the Zinc Alkoxide Catalyst in the
Soai Asymmetric Autocatalytic Reaction. Angew. Chem., Int. Ed. 2004,
43 (37), 4884−4887.
(50) Brown, J. M.; Gridnev, I.; Klankermayer, J. Asymmetric
Autocatalysis with Organozinc Complexes; Elucidation of the
Reaction Pathway. In Amplification of Chirality; Soai, K., Ed.;
Springer: Berlin Heidelberg, 2008; pp 35−65.
(51) Islas, J. R.; Lavabre, D.; Grevy, J.-M.; Lamoneda, R. H.;
Cabrera, H. R.; Micheau, J.-C.; Buhse, T. Mirror-Symmetry Breaking
T
https://dx.doi.org/10.1021/jacs.0c05994
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX