Observation of hyperpositive non-linear effect in catalytic asymmetric organozinc additions to aldehydes

Article abstract of DOI:10.1002/chir.23271

Asymmetric amplification is a phenomenon that is believed to play a key role in the emergence of homochirality in life. In asymmetric catalysis, theoretical and experimental models have been investigated to provide an understanding of how chiral amplification is possible, in particular based on non-linear effects. Interestingly, it has been proposed a quarter century ago that chiral catalysts, when not enantiopure might even be more enantioselective than their enantiopure counterparts. We show here that such hyperpositive non-linear effect in asymmetric catalysis is indeed possible. An in-depth study into the underlying mechanism was carried out, and the scheme we derive differs from the previous proposed models.

Full text of DOI:10.1002/chir.23271

Received: 15 May 2020
DOI: 10.1002/chir.23271
Revised: 24 June 2020
Accepted: 6 July 2020
S P E C I A L I S S U E A R T I C L E
Observation of hyperpositive non-linear effect in catalytic
asymmetric organozinc additions to aldehydes
Yannick Geiger
| Thierry Achard
| Aline Maisse-François |
Stéphane Bellemin-Laponnaz
Institut de Physique et Chimie des
Matériaux de Strasbourg, Université de
Strasbourg-CNRS UMR 7504, Strasbourg,
France
Abstract
Asymmetric amplification is a phenomenon that is believed to play a key role
in the emergence of homochirality in life. In asymmetric catalysis, theoretical
and experimental models have been investigated to provide an understanding
of how chiral amplification is possible, in particular based on non-linear
effects. Interestingly, it has been proposed a quarter century ago that chiral
catalysts, when not enantiopure might even be more enantioselective than
their enantiopure counterparts. We show here that such hyperpositive non-
linear effect in asymmetric catalysis is indeed possible. An in-depth study into
the underlying mechanism was carried out, and the scheme we derive differs
from the previous proposed models.
Correspondence
Stéphane Bellemin-Laponnaz, Institut de
Physique et Chimie des Matériaux de
Strasbourg Université de Strasbourg-
CNRS UMR 7504, 23 Rue du Loess, BP43,
67034 Strasbourg cedex 2, France.
Email: bellemin@unistra.fr
Funding information
French National Research Agency (ANR),
Grant/Award Numbers: ANR-11-LABX-
0058_NIE, ANR-10-IDEX-0002-02
K E Y W O R D S
asymmetric catalysis, homogeneous catalysis, non-linear effect, zinc chemistry
1 | INTRODUCTION
particular, Kagan proposed a model where the formation
of catalytic species involving the combination of several
Non-linearity between the enantiomeric excess of a prod-
uct and of a chiral auxiliary in asymmetric synthesis or
catalysis is the result of interactions between the two
enantiomers of the chiral auxiliary (i.e., diastereomeric
perturbations).1–6 A non-linear effect that gives an asym-
metric amplification (positive non-linear effect, (+)-NLE)
is believed to play a key role in the emergence of biologi-
cal chirality.7 In addition, understanding these NLEs is of
great importance for asymmetric catalysis; it allows to
clarify the mechanism and the nature of the species that
are involved in the catalytic cycle.
chiral ligands is inducing the non-linearity: the model is
based on a metal M bearing n ligands and thus is called
MLn model.1 All these MLn complexes, which are homo-
chiral or heterochiral species—depending on the
enantiomers that are present around the metal—will
then react simultaneously to give the product.
Interestingly, it has been proposed for the ML3 model
that chiral catalysts, when not enantiopure, may be more
enantioselective than their enantiopure counterparts,
though such a case has never been observed experimen-
tally until very recently.9
Since the first comprehensive studies on asymmetric
synthesis using scalemic chiral catalysts by Kagan and
collaborators in 1986, several models have been proposed
and verified by mathematical argumentation.8 In
We discuss here the first example of such
hyperpositive NLE, which was observed while studying
the addition of dialkylzinc reagents to aromatic aldehyde
substrates with partially resolved chiral N-benzyl-
ephedrine ligands. A rationalization of the effect was pos-
sible by studying the concentration, temperature and
aggregation effect on the catalyst. The reaction scope was
Proceedings from the 31st International Symposium on Chirality.
CHIRALITY 2019 Bordeaux.
Chirality. 2020;1–7.
wileyonlinelibrary.com/journal/chir
© 2020 Wiley Periodicals LLC.
1

2
GEIGER ET AL.
also studied using various para-substituted substrates
with electron-donating and electron-attracting groups.
Overall, our results point towards a two-component catal-
ysis, where a mononuclear catalyst is in equilibrium with
a dinuclear catalyst, both systems being in competition to
give the chiral product.
3 | RESULTS AND DISCUSSION
3.1 | Alkylation of benzaldehyde
Because of the weak knowledge of NLEs with ephedrine-
based ligands, we investigated the behaviour of N-
benzylephedrine (NBE) derivatives in the zinc-catalysed
alkylation of benzaldehyde. The reaction conditions have
been fixed as displayed in Scheme 1. A key point to be
made is the catalyst loading that was deliberately fixed at
20 mol%, thus promoting potential catalyst aggregation.
The reaction temperature was fixed at 0ꢀC to minimize
the achiral background reaction11 and the formation of
benzyl alcohol as side product.12
Figure 1A displays the correlation between the enan-
tiomeric excess of the product (eeP) and the enantiomeric
excess of the ligand (eeL) with diethylzinc as reagent (red
triangles). When the enantiopure ligand was used, the
product was isolated in 76% ee. Interestingly, variation of
eeL lead to a very strong (+)-NLE: eeP stays at high level
down to an eeL of only 3%, which is an unprecedented
strong amplification in a nonautocatalysed reaction. Even
more interesting is the shape of the curve: starting at 76%
eeP with the enantiopure ligand, the enantiomeric excess
of the product eeP increased up to 81% in the presence of
(−)-NBE ligand of only 10% ee. Thus, the scalemic,
enantioimpure catalyst lead to a more stereoselective
reaction than the enantiopure one: this is a case of
hyperpositive NLE.
2 | MATERIALS AND METHODS
All reagents were purchased from commercial chemical
suppliers and used without further purification, except
for liquid benzaldehyde substrates that were distilled and
stored under argon over 4-Å molecular sieves prior to
use. (1R,2S)-N-Benzylephedrine was prepared according
to a literature procedure.10 The ligand ee was adjusted by
preparing appropriate mixtures of the (1R,2S)- and
(1S,2R)-enantiomers. Enantiomeric excesses were deter-
mined on a Varian 3900 gas chromatograph equipped
with a CP-8400 autosampler and a Chiraldex G-TA capil-
lary column (25 m long, 250 μm in diameter) with helium
as carrier gas.
2.1 | General procedure—N-
benzylephedrine-catalysed addition of
dialkylzinc to aromatic adehyde
The procedure is identical to what has been published
previously.9 In
a
N2-filled glovebox, (1R,2S)-N-
An even more significant hyperpositive NLE was
observed with dimethylzinc as reagent. Figure 1B dis-
plays the result of our investigations (red triangles).
While the enantiopure NBE induced a very low
enantioselectivity of only 16%, the progressive decrease of
eeL lead to a seemingly exponential growth of the eeP,
with a maximum of 53% in presence of only 5% ligand ee.
benzylephedrine (23.7 mg, 20 mol% and 92.6 μmol) and a
magnetic stirring bar were placed in an oven-dried vial,
which was then closed with
a septum-containing
screwcap. The vial was put out of the glovebox, and a
15% ZnEt2 solution in toluene (0.5 ml, 1.2 equiv. and
560 μmol) was added via syringe; gas evolution occurred.
After 10 min stirring, the mixture was cooled to 0ꢀC and
benzaldehyde (47 μl, 1 equiv. and 460 μmol) was added
via syringe. The yellow solution was stirred overnight at
0ꢀC and turned colourless, then was quenched carefully
with 3 M aqueous HCl in an ice-water bath under vigor-
ous stirring. The organic phase was isolated, dried over
Na2SO4 and analysed by chiral stationary phase GC.
ZnMe2 was used as a 1.0 M solution in toluene
(0.5 ml, 1.2 equiv. and 500 μmol) with adapted ligand
(21.3 mg, 20 mol% and 83.3 μmol) and benzaldehyde
quantities (1 equiv. and 420 μmol). Reaction media with
ZnMe2 stayed colourless and were let to stir for 4 days.
The procedures were identical for the use of para-
substituted benzaldehydes, except for those which are
solid (CN, I, Br and Cl). Those were weighed and added
into the vial inside of the glovebox, together with
the ligand.
3.2 | Mechanistic investigations
Given the importance of these experimental results, we
decided to conduct in-depth investigations to determine
the origin of such hyperpositive effect. The ML3 model
SCHEME 1 Enantioselective addition of dimethylzinc or
diethylzinc to benzaldehyde, catalysed by chiral (−)-N-
benzylephedrine (NBE)

GEIGER ET AL.
3
This implies also that decreasing the enantiomeric
excess of the ligand would thus generate more hetero-
chiral inactive aggregates and therefore less catalytically
active species. Reducing the eeL thus induces a decrease
of the active catalyst concentration. We assumed that the
observed hyperpositive NLE could be linked to the
effective catalyst concentration. Therefore, the
enantioselective reaction was evaluated by varying the
catalyst loading of the enantiopure NBE ligand with
diethylzinc (Figure 2A) or dimethylzinc (Figure 2B) as
reagent. The results revealed that a decrease of the
catalyst loading led to an increase of the enantiomeric
excess of the product (from 76% to 84% ee with ZnEt2
and from 16% to 49% ee with ZnMe2). Moreover, the
shape of the two curves is identical with the one seen in
our NLE investigations.
Thus, it would seem possible to link the observed
NLEs to the experiments from Figure 2. Indeed, if we
assume that the heterochiral aggregate is fully
FIGURE 1 Red triangles: Optical purity of the product as a
function of the ligand ee in the (−)-NBE-catalysed addition of
(A) ZnEt2 or (B) ZnMe2 to benzaldehyde.9 Green squares:
Superimposition of eeP as function of catalyst loading using either
(A) ZnEt2 or (B) ZnMe2, which was previously converted to eeP
vs. simulated ligand ee (from Figure 2). Each point is the mean of
three different experiments; the vertical bars depict standard
deviations. The second-order polynomial fits (dotted lines) serve as
visual guidelines
proposed by Kagan implies that the NLE curve can be
hyperpositive in the upper half of the ligand ee scale and
obviously, such a case cannot be considered here. There-
fore, the ML3 model is not applicable, and we had to sea-
rch for another model for this peculiar phenomenon.
A notable fact during our catalytic investigations is
that upon addition of the dialkylzinc reagent, a precipi-
tate was observed when the scalemic ligand was used,
whereas the solution stayed clear with the enantiopure
one. Further analyses of the hydrolysed precipitate
obtained from 50% ee ligand samples gave the free
ligand in its racemic form. This result suggests that the
precipitate is most likely a meso aggregate of the zinc
catalyst. Thus, the precipitation of an inactive hetero-
chiral aggregate is in accordance with the emergence of
a positive (+)-NLE, pushed by the presence of an active
enantioenriched Zn complex in solution. We can
reasonably apply here the ‘reservoir effect’ model from
Henri Kagan, which implies that part of the ligand is
stored in a reservoir.1
FIGURE 2 Optical purity of the product ee as a function of
the catalyst loading in the addition of (A) ZnEt2 or (B) ZnMe2 to
benzaldehyde, catalysed by enantiopure (−)-NBE.9 Each point is
the mean of three different experiments; the vertical bars depict
standard deviations. The second-order polynomial fits (dotted lines)
serve as visual guidelines

4
GEIGER ET AL.
precipitated, a catalytic system with 20 mol% catalyst
loading of a ligand of 50% ee should be equivalent to a
10 mol% catalyst loading of enantiopure catalyst. Thus,
we could correlate all data points from Figure 2 with a
virtual enantiomeric excess if a scalemic ligand was used.
The resulting graphs were superimposed with the hyp-
erpositive NLE curves and are displayed in Figure 1
(green squares). In both cases, an excellent agreement
was observed.
Based on these observations, it seems plausible that
the catalytic system is composed of two independent cat-
alysts that are in equilibrium and in competition: a
monomeric species and a possible dimeric species that
operate at a steady state. Alkylzinc aminoalkoxides are
known to form dimers (and even higher aggregates)13–16
which are at the origin of strong NLEs with the DAIB
ligand in dialkylzinc additions.17–19 Assuming that the
enantiomeric excess produced by the monomer catalyst
(eem) is higher than with the dimer (eed), then decreasing
the catalyst concentration leads to an increase of the
overall eeP as the equilibrium is shifted to the more
enantioselective catalyst (i.e., monomeric species). In the
same way, decreasing eeL would have the same effect as
it causes the lowering of the effective catalyst in solution
and therefore gives rise to a hyperpositive NLE.
FIGURE 3 Proposed enantiodivergent model for the NBE-
catalysed addition of dialkylzinc to benzaldehyde which explains
the hyperpositive non-linear effect
towards the more enantioselective monomeric species,
which induces an increase of the overall product ee. This
mechanistic proposal stands in contrast to the Noyori
model for DAIB-catalysed dialkylzinc additions, which
excludes the participation of dimers in the catalytic
step.17–19
Several additional experiments have been carried out
to support this hypothesis.9 These experiments include
the investigation of the impact of the reaction tempera-
ture on the eeP, where an unusual inverse temperature
dependence was observed. Indeed, the product eeP
increases with the reaction temperature, consistent with
a monomer/dimer equilibrium shifted towards the mono-
3.4 | Substrate screening
As for now, there is a lack of information about the struc-
ture of the dimeric catalyst and how it catalyses the reac-
tion. One way to obtain such information is to study the
catalytic reaction with substrates that differ in their elec-
tronic or steric properties. Upon a change in the sub-
strate, the monomeric and dimeric catalytic cycles may
react differently: the reaction rate, enantioselectivity and
affinity of the catalyst for the substrate—or for the
product—may vary to different extents for both catalysts.
This has been exploited, for example, by Noyori and col-
laborators in the case of DAIB-catalysed ZnMe2 reactions,
where they built Hammett plots using enantiopure and
racemic (−)-DAIB.20 As the plots turned out to be linear
and identical, the authors concluded that in both cases,
the reaction was catalysed by an identical species.
Because catalysis by homochiral dimeric [DAIB-ZnMe]2
had been ruled out, they deduced that the catalytic spe-
cies was a monomeric zinc complex, and catalysis by het-
erochiral dimers could also be excluded.
1
mer at higher temperature. Additionally, H DOSY NMR
experiments were conducted to evaluate the degree of
aggregation of the catalysts as function of the tempera-
ture. Finally, visual time-normalised analyses (VTNA) of
kinetic profiles resulted in an unusual U-shaped catalyst
order versus catalyst loading plot, which was found to be
consistent with a model in which both monomer and
homochiral dimer catalyse simultaneously.
3.3 | Proposed model
A mechanism that is coherent with the experimental
investigations is displayed in Figure 3. It describes a cata-
lytic system that operates with both a monomeric catalyst
and a dimeric catalyst that operate at a steady state. If we
assume that the enantiomeric excess produced by the
monomer catalyst is higher than with the dimer
(eem > eed), then decreasing the ee of the ligand induces
a lower effective concentration of catalyst, and therefore,
the equilibrium between monomer and dimer is shifting
This led us to investigate several para-substituted
benzaldehydes and study the influence of the electronic
effects on the substrate. Substitution in para position was

GEIGER ET AL.
5
chosen to avoid any steric influence of the substituent
onto the reaction mechanism. We performed a series of
catalytic runs using either enantiopure or 20% eeL-(−)-
NBE. The results are shown in Figure 4A for ZnMe2 and
Figure 4B for ZnEt2 as reagent (for conversions, see
Figure S1 and Table S1; for the data from Figure 4, see
Tables S2 and S3). With ZnMe2, all substrates showed a
hyperpositive non-linear relationship (i.e., eeP from scal-
emic ligand is higher than from the enantiopure one),
except those bearing an –OCF3 or –SF5 substituent. The
eeP difference between scalemic and enantiopure ligand
(ΔeeP) lies mostly between 18% and 25% (details can be
found in the Supporting Information). With the scalemic
ligand, eeP increases when going from electron-donating
(σP < 0) to neutral substituents (σP ≈ 0), then reaches a
plateau of approximately 40–45% eeP. The enantiopure
ligand roughly follows the same trend. –OMe, –OCF3 and
–SF5 stand out as exceptions: they show overall lower eeP
than observed in the global trend; the eeP yielded from
enantiopure para-OMe-benzaldehyde even broke down
to almost 0%.
Using ZnEt2 as the dialkylzinc reagent leads to overall
high eeP values of approximately 70–80%, with both scal-
emic and enantiopure (−)-NBE (Figure 4B). This leads to
small ΔeeP values of only few percent, and a hyp-
erpositive NLE is clearly achieved only using substituents
with σP ≈ 0 (H, F, 2-naphtaldehyde), except for electron-
donating –OiPr which also gives a hyperpositive NLE.
Otherwise, there are no general trends to read from these
results. –OMe again stands out as an exception, giving a
much lower eeP of approximately 50% with both 100 and
20% eeL; –OCF3 too, but only when using scalemic ligand
(29% eeP), the same goes for –CN and –SF5 in a less pro-
nounced way (approximately 60% eeP).
The fact that the reactions with ZnEt2 show only
small differences in eeP between the different substrates,
when compared their ZnMe2 counterparts, may indicate
that the dimeric catalyst plays a minor role with ZnEt2—
either because of a low dimer/monomer ratio or because
the monomer is significantly more active than the dimer.
This is further supported by the overall high eeP values
obtained with ZnEt2, which also suggest a stronger par-
ticipation of the monomeric catalyst. Another possible
factor might be the increase of the catalysts'
enantioselectivities eem and/or eed when switching from
ZnMe2 to ZnEt2. However in case of Noyori's system, the
monomeric DAIB yields similar eeP values with both dia-
lkylzinc reagents (ZnEt2: 98% at 0ꢀC; ZnMe2: 95% at
40ꢀC).17,18 This indicates that the nature of the alkyl
group has only minor influence on the enantioselectivity
of the catalytic step. Therefore, with NBE, a significant
change in enantioselectivity seems unlikely when
switching from ZnMe2 to ZnEt2, at least for the mono-
meric (−)-NBE-ZnMe complex.
In turn, it seems that the ZnMe2-system is more
adapted for this study as the overall lower eeP and higher
ΔeeP values indicate that both monomer and dimer par-
ticipate significantly in the reaction. Therefore, possible
changes in the catalytic system upon substrate variation
are more likely to be observed with ZnMe2, and the
remarkably low eeP values yielded by electron-rich sub-
strates might be caused by such a change. A decrease of
eem and/or eed or a faster achiral background reaction
(which may be accelerated by the coordinating –OMe
and –OiPr groups) are possible but not sufficient to
FIGURE 4 Product ee of NBE-catalysed additions of
(A) ZnMe2 and (B) ZnEt2 to para-substituted benzaldehydes using
(−)-NBE with 100% (blue dots) or 20% eeL (orange triangles). The
results are sorted according to the substituent's σP values. Outliers
(OMe, OCF3 and SF5) are depicted by hollow dots and triangles.
Each point is the mean of three different experiments; the vertical
bars depict standard deviations

6
GEIGER ET AL.
explain this behaviour, as the DAIB ligand gives only
slightly lower eeP with OMe-substituted benzaldehyde
(88%) than with the unsubstituted one (95%).20 Therefore,
an increased dimer/monomer ratio or dimer/monomer
relative reaction rate in presence of electron-rich sub-
strates seems to be more likely to lower eeP. It should be
noted that the electron-rich substrates react only slug-
gishly and gave only low conversion values after 4 days
(cf. Supporting Information for detailed conversion data).
We have reported before that eeP increases over time, at
least in the early stage of the NBE-catalysed reaction with
ZnMe2.9 However, the low conversion cannot account
alone for the low eeP because certain electron-rich sub-
strates do also reach only low conversions, although their
eeP stays high.
ORCID
Yannick Geiger https://orcid.org/0000-0003-2280-7107
Thierry Achard https://orcid.org/0000-0001-8271-8718
Stéphane Bellemin-Laponnaz https://orcid.org/0000-
0001-9462-5703
REFERENCES
1. Guillaneux D, Zhao SH, Samuel O, Rainford D, Kagan HB.
Nonlinear effects in asymmetric catalysis. J am Chem Soc.
1994;116(21):9430-9439.
2. Girard C, Kagan HB. Nonlinear effects in asymmetric synthesis
and stereoselective reactions: ten years of investigation. Angew
Chem Int Ed. 1998;37(21):2922-2959.
3. Satyanarayana T, Abraham S, Kagan HB. Nonlinear effects in
asymmetric catalysis. Angew Chem Int Ed. 2009;48(3):456-494.
4. Kagan HB, Luukas TO. Non-linear effects and autocatalysis.
In: Jacobsen EN, Pfaltz A, Yamamoto H, eds. Comprehensive
Asymmetric Catalysis. Berlin: Springer-Verlag; 1999, Chapter 4:
101-118.
4 | CONCLUSION
5. Girard C, Kagan HB. On diastereomeric perturbations. Can J
Chem. 2000;78(6):816-828.
6. Kagan HB. Nonlinear effects in enantioselective organometallic
catalysis. Oil Gas Sci Technol. 2007;62(6):731-738.
7. Blackmond DG. The origin of biological homochirality. Phil
Trans R Soc B. 2011;366(1580):2878-2884.
8. Puchot C, Samuel O, Dunach E, Zhao S, Agami C, Kagan HB.
Nonlinear effects in asymmetric synthesis. Examples in asym-
metric oxidations and aldolization reactions. J am Chem Soc.
1986;108(9):2353-2357.
9. Geiger Y, Achard T, Maisse-François A, Bellemin-Laponnaz S.
Hyperpositive nonlinear effects in asymmetric catalysis. Nat
Catal. 2020;3(5):422-426. https://doi.org/10.1038/s41929-020-
0441-1
10. Coote SJ, Davies SG, Fletcher AM, Roberts PM, Thomson JE.
Enantiospecific stereodivergent synthesis of trans- and cis-
N(2),3-dimethyl-4-phenyl-1,2,3,4-tetrahydroisoquinolines.
Chem Asian J. 2010;5(3):589-604.
11. Sung DWL, Hodge P, Stratford PW. Reactions of aldehydes
with diethylzinc catalysed by polymer-supported ephedrine
and camphor derivatives: comparisons of enantiomeric
excesses achieved with various supports: optimisation of sup-
port parameters to enable high enantiomeric excesses to be
obtained. J Chem Soc Perkin Trans 1. 1999;11:1463-1472.
12. Noyori R, Suga S, Kawai K, et al. Enantioselective addition of
diorganozincs to aldehydes catalyzed by β-amino alcohols.
J Organomet Chem. 1990;382(1-2):19-37.
The study of dialkylzinc addition to benzaldehyde,
catalysed by chiral (−)-NBE, has allowed us to find the
first example of a hyperpositive NLE. This fulfils Kagan's
25-year-old prediction of a scalemic catalyst being able to
be more enantioselective than its enantiopure counter-
part, even though the mechanism turned out to be differ-
ent from what Kagan proposed at that time. Instead of
the ML3 model, we have found (−)-NBE to follow a
model in which both monomeric and homochiral dimeric
(−)-NBE-ZnR (R = Me, Et) catalyse, the aggregate being
less enantioselective than the discrete complex. Precipita-
tion of an insoluble meso aggregate allows to trap the
minor enantiomer catalyst and to generate an enantio-
enriched monomer/dimer equilibrium that induces a
hyperpositive NLE. Evidence for this model has been
found by varying the catalyst loading, the reaction tem-
perature, by DOSY NMR investigations and by kinetic
analyses (catalyst order vs. catalyst loading and eeP
vs. time). Furthermore, we have started to study the
influence of para-substituents on the benzaldehyde sub-
strate on the overall catalytic outcome. Preliminary
results suggest that electronic effects from the substitu-
ents may induce changes in the relative reaction parame-
ters
(dimer/monomer
concentration
ratio
and
13. Coates GE, Ridley D. Reactions between some organozinc com-
dimer/monomer reaction rate ratio) and thus give infor-
mation about the distinct monomeric and dimeric cata-
lytic cycles. A more advanced study including the
analysis of Hammett plots is underway.
pounds
and
2-dimethylaminoethanol,
acetoxime,
phenylisocyanate, and benzophenone. Some observations on
the methylzinc methoxide tetramer. J Chem Soc A. 1966;1064-
1069.
14. Boersma J, Noltes JG. Investigations on organozinc compounds
X. Coordination chemistry of organozinc compounds RZnX:
organozinc-oxygen derivatives of potentially bidentate ligands.
J Organomet Chem. 1968;13(2):291-299.
15. van der Schaaf PA, Wissing E, Boersma J, Smeets WJJ,
Spek AL, van Koten G. Organozinc complexes with mono-
anionic chelating phenolates or 2-pyridylmethanolates.
ACKNOWLEDGEMENT
This work was funded by the French National Research
Agency (ANR) through the Programme d'Investissement
d'Avenir under contracts ANR-11-LABX-0058_NIE and
ANR-10-IDEX-0002-02.

GEIGER ET AL.
7
Molecular structure of [Z(CH2SiMe3){OCH2(2-Py)}]4. Organo-
amino alcohols. Evidence for the monomeric alkylzinc
metallics. 1993;12(9):3624-3629.
aminoalkoxide as catalyst. Tetrahedron. 1999;55(12):3605-3614.
16. Fitzpatrick K, Hulst R, Kellogg RM. Thiol and disulfide deriva-
tives of ephedra alkaloids 2: a mechanistic study of their effect
on the addition of diethyl zinc to benzaldehyde. Tetrahedron:
Asymmetry. 1995;6(8):1861-1864.
17. Kitamura M, Okada S, Suga S, Noyori R. Enantioselective addi-
tion of dialkylzincs to aldehydes promoted by chiral amino
alcohols. Mechanism and nonlinear effect. J Am Chem Soc.
1989;111(11):4028-4036.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of this
article.
18. Kitamura M, Suga S, Oka H, Noyori R. Quantitative analysis of
the chiral amplification in the amino alcohol-promoted asym-
metric alkylation of aldehydes with dialkylzincs. J Am Chem
Soc. 1998;120(38):9800-9809.
19. Noyori R, Suga S, Oka H, Kitamura M. Self and nonself recog-
nition of chiral catalysts: the origin of nonlinear effects in the
amino-alcohol catalyzed asymmetric addition of diorganozincs
to aldehydes. Chem Record. 2001;1(2):85-100.
How to cite this article: Geiger Y, Achard T,
Maisse-François A, Bellemin-Laponnaz S.
Observation of hyperpositive non-linear effect in
catalytic asymmetric organozinc additions to
aldehydes. Chirality. 2020;1–7. https://doi.org/10.
1002/chir.23271
20. Kitamura M, Oka H, Noyori R. Asymmetric addition of dia-
lkylzincs to benzaldehyde derivatives catalyzed by chiral β-