A laboratory study reports the first measured infrared absorption spectrum of naphthalene in its lowest triplet state. It differs from the molecule’s singlet ground-state spectrum and is closely reproduced by anharmonic frequency calculations, meaning calculations that allow molecular vibrations to depart from a simple ideal pattern.
That is a laboratory result, not a direct astronomical detection. The manuscript is an arXiv version 1 preprint dated 25 Aug 2026. Its direct evidence comes from isolated naphthalene measured in two electronic-state conditions: S0, the singlet ground state, and T1, the lowest triplet state. They are states of the same molecule, not separate chemical species.
Timing the triplet-state signal
To make the comparison, naphthalene was expanded in argon at 4 bar into a supersonic molecular beam. The expansion cooled the rotational temperature to about 5 K and the vibrational temperature to about 20 K, while keeping the molecules isolated.
The experiment used IR-UV depletion spectroscopy. Ion-yield measurements were alternated with and without an infrared laser between excitation and ionization, so absorption was read as a depletion in the ion signal. The delayed ion signal contained a 0.27 ± 0.02 µs exponential component plus a constant component. Delays longer than 1 µs were considered suitable for measurements exclusively in T1.
The spectrum changes with electronic state
In the CH-stretch region, from 3.33 to 3.18 µm (3,000 to 3,140 cm−1), the T1 scan reached a maximum depletion of about 30%. The change involved more than signal strength: the T1 spectrum differed significantly from S0 in both band positions and intensity distribution.
Anharmonic calculations that included modes involving three quanta of vibration predicted T1 bands at 3,081.7 and 3,087.8 cm−1, both of which coincided with observed features. A cited S0 prediction at 3,113.7 cm−1 did not coincide with an observed band.
In the same 3-micrometre region, calculated band positions agreed for many common features. The reported mean absolute deviations were about 2–4 cm−1 for S0 and 1–6 cm−1 for T1, while several experimental features remained unassigned.
A mixed fingerprint yields candidate markers
The fingerprint scan covered 350–1,700 cm−1 and brought a different complication. Because of the laser configuration, the reported T1 spectrum contained contributions from both S0 and T1, and comparison with the S0 spectrum was used to decompose the mixture.
That decomposition identified a pronounced band at 667 cm−1 and weaker bands at 390 and 578 cm−1 as unique to T1 in naphthalene. The paper proposes them as markers for naphthalene in its triplet state. The fingerprint calculations reproduced all observed band positions, with a mean absolute deviation of about 3 cm−1 for S0 and about 10 cm−1 for T1.
A possible lead for astronomical searches
The calculation work also points to recurring regions where the reported triplet and singlet spectra differ: 600–900 cm−1 (16.7–11.1 µm) and 1,100–1,500 cm−1 (9.1–6.7 µm). For naphthalene, examples were the 667 cm−1 (14.99 µm) feature and a 1,430 cm−1 (6.99 µm) mode.
In a modeled comparison with an Orion Bar JWST/MIRI spectrum, triplet spectra showed a broad feature centered near 6.8 µm (1,480 cm−1), with intensity comparable to the 11–15 µm range. Singlet spectra lacked comparable structure. The comparison was a model, not a direct detection.
A laboratory result with a narrow reach
That distinction sets the study’s reach. Its direct evidence is the state-selective laboratory absorption spectrum of naphthalene and its comparison with S0; broader conclusions about PAH spectra and astronomical signatures come from calculations and modeled spectral comparisons. The proposed bands are therefore candidate markers, not a validated identification of triplet molecules in space.
The experimental scope is narrow: only naphthalene was measured, and the fingerprint scan contained both S0 and T1 contributions. The lower T1 concentration reduced signal-to-noise relative to S0, and several experimental features remained unassigned. The broader PAH and astronomical implications therefore remain prospective rather than directly measured.
The manuscript is identified as arXiv version 1 dated 25 Aug 2026. It reports support from the NWO Dutch Astrochemistry Network, COST Action NanoSpace, NWO support for the HFML-FELIX Institute and CPU time on the Dutch National Supercomputer Snellius.
Paper data and sources
Original title: Interstellar Aromatic Infrared Bands: IR absorption spectroscopy of the lowest triplet state of naphthalene
Authors: Priyanka Arvind Paunikar, Hugo Maurer, Lars Reems et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-25
DOI: Not available
Original paper · Full text