The Chemistry Nobel Went to Mirror Molecules. Textbooks Still Get Limonene Wrong

On 2 October we published three picks for this year’s chemistry Nobel and put our money on self-assembled monolayers. This morning in Stockholm the Royal Swedish Academy of Sciences picked mirror images instead. The prize went to Henri B. Kagan and Kenso Soai “for the discovery of non-linear effects and autocatalysis in asymmetric organic synthesis.” We got it wrong. The consolation is that one of the products on our own list already has the prize’s subject printed in its name, as a single capital letter: the D in D-limonene.

That letter has a story behind it. It runs from Louis Pasteur’s tartaric acid crystals, through a 1971 paper in Science, into most organic chemistry textbooks, and out the other side in a 2021 paper whose authors asked for those textbooks to be corrected. Along the way it says something useful about what a purity figure on a drum does and does not tell you.

This is a chemistry explainer built on the Nobel committee’s own background material and on published research. It is not a claim about the odor, purity or enantiomeric composition of any particular lot we ship.

What did the 2026 chemistry Nobel reward?

The 2026 Nobel Prize in Chemistry rewards two discoveries that let a reaction favour one mirror-image form of a molecule and then amplify it until almost nothing else is left. Many molecules, including the amino acids in proteins and the sugars in DNA, exist as two versions that are mirror images of each other, like a left and a right hand. Chemists call such molecules chiral and the two versions enantiomers. Life uses only one of each pair. In a test tube, ordinary reactions make both in equal amounts.

The committee frames the prize as the last two links in a chain of ideas. In 1953 the Bristol physicist Charles Frank wrote down three conditions under which a reaction could end up one-handed: a chiral catalyst, a mechanism that boosts one mirror image and suppresses the other, and a product that acts as its own catalyst. The first condition had been met early in the twentieth century. Kagan and Soai met the other two.

Kagan’s non-linear effect came from a simple question. Asymmetric catalysts usually pair a metal with a chiral molecule, and chemists assumed the product’s purity would track the catalyst’s in a straight line. Kagan reasoned that the metal holds at least two chiral molecules at once, so a mixed catalyst comes in three forms: right-right, left-left and left-right. The mixed form turned out to work far more slowly. In his 1986 paper in the Journal of the American Chemical Society, covering three different reactions, the plot of product purity against catalyst purity came out curved, not straight.

The Soai reaction takes that one step further. The product is itself the catalyst, so each molecule of the favoured form makes more of the favoured form. The committee’s background describes the 2003 result plainly: when Soai repeated the reaction, sometimes the other enantiomer won, depending on chance at the start. “Other than life itself, no one had ever achieved this feat.”

The committee also names who uses this. Kagan’s discovery, it writes, has been revolutionary “for chemists who develop reactions for the manufacture of pharmaceuticals, flavours, scents and new materials.” Flavours and scents are where limonene comes in.

We picked the wrong prize. Here is what we got right

Our 2 October predictions piece leaned on the Clarivate Citation Laureates list and named self-assembled monolayers as our first pick, because that is the one built on a bench with bulk acids and solvents. Chirality was not on our shortlist. We are not going to pretend otherwise or edit the old article; it stays up with its date on it.

What carries over is the habit that piece was built on: read the primary source, then read the label. The Nobel committee’s background spends paragraphs on how hard it is to tell two mirror-image molecules apart. One of them, as it happens, is a common industrial solvent.

What does the D in D-limonene mean?

The D in D-limonene means the liquid rotates plane-polarised light to the right, the same property Pasteur measured in tartaric acid in the 1850s. Limonene, C₁₀H₁₆, has one stereocentre, so it exists as two enantiomers. The one citrus fruit makes goes by several names, and they come from different labelling systems that happen to agree here:

So the full name of the citrus enantiomer is (R)-(+)-limonene, and the trade name D-limonene is shorthand for it. Its mirror image is (S)-(−)-limonene, sometimes sold as L-limonene.

Is a lemon a homochiral factory?

In practice, yes: citrus fruit makes (R)-limonene with an enantiomeric excess above 99.9 percent, which is the kind of one-handedness the Nobel committee describes as a signature of life. A team at the Norwegian University of Science and Technology ran chiral gas chromatography on orange oil and lemon oil for their 2021 paper in the Journal of Chemical Education and found an enantiomeric excess for (R)-(+)-limonene “of more than 99.9% in both oils, which agrees with previous findings.”

The two oils differ in how much limonene they contain and in what else is there. By non-chiral gas chromatography, orange oil was about 90.5 percent limonene. Lemon oil was about 68.0 percent, with roughly 16 percent β-pinene and 12 percent γ-terpinene. In the orange oil the authors counted about 120 peaks larger than 30 ppm of the limonene peak. That “forest” of minor components turns out to matter.

(S)-(−)-limonene, the mirror image, is “almost nonpresent in citrus fruits,” the authors write. It is the major limonene enantiomer in citronella and lemongrass oils, but at low levels, and the (S)-limonene one major laboratory supplier sells is, according to that supplier’s technical service scientist, made synthetically from α-pinene.

Commercially, (R)-limonene is recovered from citrus peel, which is a by-product of juice production: the peel oil is pressed or distilled out and the limonene separated from it by further distillation. The paper adds a detail any buyer should note. Limonene purified to 99.5 percent “will readily oxidize in air” if not properly stored, falling to approximately 95–96 percent.

Where did the orange-and-lemon story come from?

The claim that (R)-limonene smells of oranges and (S)-limonene of lemons traces back to a 1971 paper whose “(S)-limonene” came from lemons, which make (R). That year three groups independently tested how people perceive the two enantiomers of carvone, the classic spearmint-versus-caraway pair. One of them, by Friedman and Miller in Science under the title “Odor Incongruity and Chirality,” included a table linking (R)-(+)-limonene to orange and (S)-(−)-limonene to lemon.

The 2021 authors quote C. S. Sell’s 2014 book Chemistry and the Sense of Smell on what went wrong:

Sell’s explanation, as quoted in the paper, is that the lemon-smelling sample in the 1971 work had been extracted from lemons. Since lemons, like all citrus, produce (R)-(+)-limonene, he concludes the lemon odor came from contamination by traces of citral, the aldehyde pair (neral and geranial) usually credited with the smell of lemon. In the Norwegian chromatograms, citral showed up in both oils but at about ten times the concentration in lemon oil, at 1.4 and 2.3 percent for neral and geranial.

Not every textbook repeated the error. The authors note that McMurry associates (R)-limonene with both lemon and orange and (S)-limonene with pine trees, and that Vollhardt and colleagues link the (S) form to fir cones. Plenty of others kept the orange-lemon pairing for decades.

What did the 2021 smell test find?

The 2021 test found that the purer the (R)-limonene, the fewer people smelled orange in it, which means the orange note rides on the impurities. The authors ran two odor tests with 48 and 49 participants, a mix of students, staff and members of the public, using the same handedness of limonene at two purities. Both samples had an enantiomeric excess above 99.9 percent. What changed was everything else in the bottle:

The chromatograms explain the drop. Impurities in the technical-grade limonene were about ten times higher than in the analytical grade, and several of its peaks matched peaks in orange oil. Their conclusion, in their own words: “The cruder the (R)-(+)-limonene is, the more impurities it contains, and the more it will smell like orange oil.” For the analytical sample, the impurities fell below the testers’ orange-odor threshold.

The authors are candid about limits. Their panel was not trained, the two tests used different procedures, and odor perception depends on memory and language as well as chemistry. They report that the procedural change did not shift results for the oils or for (S)-limonene, and that 72 percent of test-2 participants correctly reported no odor from triacetin, an odorless control. Their abstract ends with a request:

What this means on a specification

A purity percentage and an enantiomeric purity are two different numbers, and the smell of a limonene lot is set mostly by the part that is not limonene. If odor, flavor or reproducibility matters to your process, these are the questions that move the result:

Two practical points follow. First, a figure on a label, including the 94 in the name of our D-Limonene 94% listing, is a purity figure. It does not tell you the handedness, and it does not tell you what the remaining few percent are. Second, if a formulation depends on a particular citrus character, the stable way to get it is to specify it and check it on the certificate, not to assume that “D” on the label means “orange” in the bottle. The difference between D-limonene and orange oil is exactly this forest of minor components.

A word on what this article does not say. It reports what the Nobel committee announced and what the 2021 study measured on its own samples. It does not state the enantiomeric excess, odor profile or impurity breakdown of any lot we sell, and it makes no claim about ours beyond what is printed on the product listing and the certificate.