Anyone who has opened an old book has encountered the distinctive smell of the archive: a complex, slightly sweet aroma with undertones of vanilla, almond, and grassy straw. Until the 2010s, the smell was treated as an incidental feature of ageing paper. A small research community, working at the intersection of heritage science and analytical chemistry, has since argued that the scent of a collection is a valuable source of information and, in some cases, a heritage object in its own right worth documenting and preserving.

A. The chemistry of book smell is now relatively well characterised. As paper ages, the cellulose and lignin it contains break down into hundreds of volatile organic compounds, each contributing a particular note to the overall aroma. Vanillin, a compound found in vanilla pods, is produced by the breakdown of lignin and contributes the sweet note. Furfural, derived from the sugars in cellulose, contributes an almond-like smell. Smaller quantities of acetic acid, benzaldehyde, and ethylhexanol all contribute as well. The relative proportions of these compounds depend on the materials from which the paper was made, on how it has been stored, and on its age.

B. Analytical chemists can measure the scent profile of an archive by drawing air samples through a device called a gas chromatograph, which separates the compounds in the sample and identifies each one by its chemical signature. The British heritage scientist Matilda Akhurst has used this technique in a systematic survey of British research libraries and has shown that each institution has a characteristic chemical signature, differing in subtle but consistent ways from those of comparable collections elsewhere. The factors that produce the variation include the types of binding used, the wood panelling of the rooms, and the heating and ventilation history of the building.

C. The applications of such measurements are partly practical. Certain breakdown compounds are associated with increased rates of paper deterioration, and measuring them can give curators an early warning of problems before they become visible. An elevated concentration of acetic acid, for example, indicates that the cellulose of the paper is breaking down and that intervention may be needed before the paper becomes too brittle to handle. Akhurst's laboratory has developed a portable sampling device that allows a conservator to survey an entire storage area in a single day, identifying the most vulnerable shelves for attention.

D. A more controversial application treats the scent itself as a feature of the archive. A library that digitises its collection and moves the originals to off-site storage may preserve the texts while losing an aspect of the experience of studying in the room with the books. Akhurst has collaborated with colleagues at the Slovenian National Library to document the scent profile of a historic reading room and to produce, in a laboratory, an approximate reconstruction that can be diffused through modified new buildings. The result is not a direct replica but a composed scent intended to evoke a similar experience. Whether such reconstructions are legitimate cultural preservation or a form of theatre is a matter of ongoing debate among heritage scholars.

E. The same techniques have begun to be applied to other kinds of archive. The scent of a traditional Japanese temple, accumulating over centuries from the incense burned inside and the wood of its structure, has been documented by a research group in Kyoto with the aim of providing a reference for conservators restoring similar buildings. The scent of a wartime bomb shelter, recorded by an English museum, has been used in a display that reproduces the conditions of the 1940s at the visitor's request. The scent of a historic tannery, although less immediately appealing, has been preserved in an analytical record that could support future understanding of industrial craft traditions.

F. Inevitably, the work has raised questions about what a scent actually is. A chemical profile measured by a gas chromatograph is not the same thing as a human olfactory experience. The brain processes smell in a distinctive way, integrating the chemical signal with memory, emotion, and context in a way that cannot be fully reconstructed from the chemistry alone. Akhurst has argued that the chemical description is therefore only a partial record, and that her work should be seen as documenting one aspect of a richer cultural phenomenon rather than the phenomenon itself. Attempts to reconstruct a historic scent will always involve some element of interpretation, and the best reconstructions will probably be those that acknowledge their own approximations.

G. The broader lesson of this research is that heritage has many dimensions that were not, until recently, considered part of what a museum or library should preserve. Smell, touch, and acoustic environment have all started to attract attention from researchers who argue that the full experience of a historic space cannot be captured in photographs and text alone. The scientific tools required to document these features have become much more accessible in the last decade, and many more collections are likely to be surveyed in the next. Akhurst has remarked that the archive of the future may include files that would have puzzled the curators of the past: chromatograms, microphone recordings of empty rooms, and thermal images of walls, alongside the paper documents that have traditionally been the core of the collection.