On October 7, the Royal Swedish Academy of Sciences announced that the 2026 Nobel Prize in Chemistry will be awarded to Henri B. Kagan of France and Kenso Soai, professor emeritus at Tokyo University of Science. The prize recognizes "the discovery of nonlinear effects and autocatalysis in asymmetric organic synthesis," honoring research on amplifying tiny left-right imbalances in molecules during chemical reactions.
The ability to preferentially make one of two mirror-image molecules links two very different questions: how to synthesize pharmaceuticals, and why life came to use only one form of certain molecules.
To understand what this research means, however, one has to distinguish two things. Amplifying a slight left-right bias in the laboratory is not the same as explaining how biological molecules on Earth acquired their handedness. Prize announcement
Molecular "Handedness" and the Nonlinear Effect Discovered by Henri B. Kagan
The chiral amino acids that make up proteins are, as a rule, the L form, while the sugars that make up DNA and RNA are the D form.
"Left and right" here is a metaphor for molecules that are mirror images of each other. Left and right hands have similar shapes, but no amount of rotation will make them overlap perfectly.
Molecules also have mirror-image forms, called enantiomers, which contain the same kinds of atoms connected in the same way but arranged differently in three dimensions.
The property whereby the molecules of life are all of one form is called homochirality.
Enzymes and receptors in the body also have three-dimensional structures, so whether a drug molecule is one form or the other can change how it binds and what it does. Asymmetric synthesis, which selectively produces the desired form, is an important technology for controlling these differences.
However, it was natural to assume that even if a catalyst favored one form, a catalyst containing a mixture of both forms would yield a product with a correspondingly smaller bias.
In the conventional simple model, if the bias of the chiral molecule used as the catalyst were halved, the bias of the product would also be half that obtained with a pure single form. In other words, a proportional relationship was assumed.
What Henri B. Kagan and colleagues showed in 1986 was a reaction that departed from this simple proportionality.
According to the Nobel Committee's scientific background document, the research at the time examined the relationship between the left-right bias of the catalyst and that of the product in reactions such as asymmetric oxidation and epoxidation.
In some reactions, the bias of the product exceeded the value predicted by simple proportionality. Because plotting the two against each other produces a curve rather than a straight line, this is called a "nonlinear effect."
A clue to understanding the phenomenon lies in what combinations the catalyst forms during the reaction.
For example, when several chiral molecules bind to a metal, combinations of the same form and combinations of opposite forms can both arise.
If the mixed left-right combination is less reactive, the minority molecules become tied up with molecules of the opposite form and are less able to take part in catalysis. As a result, the slightly more abundant form plays a relatively larger role, and the bias of the product becomes larger than expected.
A nonlinear effect does not always mean amplification of bias, however.
If the mixed left-right combination is more reactive, the bias of the product can be smaller than simple prediction.
Kagan's discovery showed that the bias of a product cannot be predicted from the optical purity of the catalyst alone; one must also consider how catalyst molecules combine with one another and how reactive each combination is.
The Soai Reaction: The Product Itself Promotes the Next Reaction
In a peer-reviewed paper published in Nature in 1995, Soai and colleagues reacted an aldehyde containing a pyrimidine ring with diisopropylzinc to produce a chiral alcohol.
The zinc alkoxide formed from the product acts as a catalyst, promoting the formation of more product of the same form.
The discovery combined autocatalysis, in which the substance produced by a reaction promotes the next reaction, with selective production of one handedness and with amplification of the bias. 1995 paper
As the product increases, so does the substance acting as catalyst.
The form that was slightly more abundant at the start further promotes production of the same form. Raw materials are supplied from outside, while the product itself accelerates the conversion.
By repeatedly using the product as the catalyst for the next reaction, a slight left-right bias can be enlarged step by step.
The three features realized by Soai are also organized in the Japan Academy's citation as "asymmetric discrimination," "autocatalysis," and "asymmetric amplification."
The reaction selectively makes the same form as itself, the product becomes the catalyst, and the proportion of that form rises further. The combination of these three functions is the major characteristic of the Soai reaction.
However, autocatalysis does not necessarily cause one form to increase.
According to the Nobel Committee's scientific background document, in a reaction using a different substrate that Soai reported in 1990, the product catalyzed its own formation, but the left-right bias became smaller than at the start.
What is important about the 1995 result is that it realized a reaction in which not just the amount of product increased through autocatalysis, but the left-right bias itself was amplified.
The conditions for such a reaction had been considered in theory well before.
In 1953, the British theoretical physicist Frederick Charles Frank proposed a mathematical model in which a product promotes its own formation, selects one form, and suppresses the activity of the opposite form.
If a very small initial difference is self-amplified, one form will ultimately become overwhelmingly dominant.
The work of Kagan and Soai advanced this theoretical idea of amplification to a stage where it could be investigated as an actual chemical reaction.
How Small Is a 0.00005% Bias?
Soai and colleagues continued to improve the reaction, and in 2003, using a 2-alkynylpyrimidine compound, succeeded in amplifying an extremely small initial bias to a state occupied almost entirely by one form.
The Nobel Committee's scientific background document describes an experiment that started from an enantiomeric excess of about 0.00005% and, through three consecutive reactions, amplified it to more than 99.5%.
Enantiomeric excess is abbreviated "ee."
It is the difference in the amounts of the two enantiomers divided by their sum.
A 50:50 split gives 0% ee, 51:49 gives 2% ee, and a single form alone gives 100% ee.
This is not a measure of the amount of substance produced or the reaction yield; it indicates how strongly the two forms are imbalanced.
In the 2003 amplification experiment, the starting ee of about 0.00005% corresponds to about 50.000025% for the dominant form. After three consecutive reactions it exceeded 99.5% ee, meaning the dominant form accounted for more than 99.75%.
| Stage of the same amplification experiment | ee | Share of dominant form | Share of minor form |
|---|---|---|---|
| Start | About 0.00005% | About 50.000025% | About 49.999975% |
| After 3 consecutive reactions | Over 99.5% | Over 99.75% | Under 0.25% |
The table converts the values by taking the total of the two forms as 100%, with the dominant form calculated as (100 + ee) ÷ 2 and the minor form as (100 − ee) ÷ 2.
The original figures are based on Figure 7 and its explanation in the scientific background document.
This means that an initial bias so small as to be nearly indistinguishable from 50:50 was amplified by repeated reactions to a state in which one form accounts for almost the whole.
However, the extremely fine numbers at the start need careful handling.
According to the supplementary material of the original paper, this initial composition was prepared by adding a solution of one form, with its concentration adjusted, to a mixture containing equal amounts of both forms.
The value of about 0.00005% ee is therefore an approximation calculated from the amounts mixed, not a figure obtained by directly measuring such a tiny bias to every digit.
What the experiment showed was that an extremely small left-right difference introduced artificially could be greatly amplified by an autocatalytic reaction.
But would the reaction still favor one form if no initial left-right difference were supplied from outside? That was examined in a separate experiment.
Bias Arising by Chance, and the Distance That Remains to the Origin of Life
Even in the Soai reaction without any externally added chiral catalyst, results have been obtained in which one form ultimately becomes dominant.
In a 2003 experiment described in the Nobel Committee's scientific background document, the R form dominated in 18 of 37 runs and the S form in 19.
The ee values obtained ranged from 15 to 91%, differing in both conditions and results from the amplification experiment above 99.5%, which began with a deliberately supplied tiny initial bias.
Even in a reaction that produces left and right forms in nearly equal amounts, molecule formation is subject to statistical fluctuation.
If one form happens to become very slightly more abundant and that difference is amplified by the autocatalytic reaction, one form could end up strongly dominant.
This is important as a model for considering experimentally the "symmetry breaking" by which one form is selected from a symmetric state.
At the same time, the fact that which of R or S dominated varied from experiment to experiment also shows that the same side is not always chosen without a direction being specified from outside.
In considering the homochirality of life, at least three problems need to be separated.
First, what produced the initial small left-right difference? Second, by what mechanism was that difference amplified? Third, did the same process actually hold for the molecules that make up life?
The Soai reaction is above all a powerful experimental system for investigating how a slight left-right difference can be amplified.
However, the organozinc compounds and specific aldehydes used in the reaction do not replicate the aqueous chemical environment in which the amino acids and sugars that form the materials of life are thought to have been produced.
The Nobel Committee, too, in its scientific background document, does not position Frank's model as a definitive answer to the origin of life's homochirality.
It treats the Soai reaction not as one directly applicable to the aqueous environment of life's origin, but as a proof-of-concept model showing a mechanism by which left-right bias can be self-amplified.
Showing that molecules of a single handedness can be made in the laboratory is not the same as proving that this reaction actually occurred during the emergence of life.
Making this difference clear lets us distinguish what the prize-winning research has actually elucidated from the questions about the origin of life that remain open.
Putting the Mechanism of Amplification to Use in Designing Future Reactions
Research continues on the structure of the catalyst actually at work in the Soai reaction.
In a peer-reviewed study published in the Journal of the American Chemical Society in 2020, Scott E. Denmark and colleagues combined structural comparison, reaction-rate measurements, and computational chemistry to examine a model of a tetramer formed by the assembly of product-derived zinc alkoxides. 2020 mechanistic study
In this model, a catalyst assembly made of molecules of the same form can bind the raw materials in an arrangement that favors reaction.
An assembly in which left and right forms are mixed, on the other hand, has difficulty achieving the same binding state.
The explanation for why left-right bias is amplified is thus not only that the product becomes the catalyst for the next reaction, but also that differences in catalytic activity arise from the structure of the assemblies the product forms.
However, it cannot be said that a molecular-level explanation of the reaction has been completely settled by a single model.
The Nobel Committee's scientific background document also describes an alternative model proposed by Oliver Trapp and colleagues, which proceeds via a zinc hemiacetal intermediate formed from the product and the raw materials.
Because the substrates studied differ, it is possible that different models each capture features of different reaction systems.
Research into what actually acts as the catalyst and at which stage of the reaction the left-right bias grows continues to feed into knowledge for designing new reactions.
The nonlinear effect found by Kagan became a clue for probing the molecular combinations formed during a reaction, by observing how the bias of the product responds when the left-right mixing ratio of the catalyst is changed.
Soai combined this with autocatalysis, establishing an experimental system in which a very small initial difference can be greatly amplified by the reaction itself.
What will matter in future research is how far the same principle can be extended to a wider variety of raw materials and reactions.
A further major challenge is whether similar amplification of bias can be demonstrated with molecules close to the materials of life, such as amino acids and sugars, and in environments containing water.
If such reactions are found, it would become possible to test hypotheses about why the molecules of life came to share a single form under conditions close to those of life's own materials.