Rotational Spectroscopy

Most molecules identified in the interstellar medium have been found through their rotational spectra: patterns of radiation emitted or absorbed as molecules rotate. Those patterns depend on molecular structure, making them a powerful way to distinguish species, even when the molecules have the same chemical formula. But the frequencies have to be known accurately enough to match narrow astronomical lines. For the molecules we study, calculations usually provide a starting point; laboratory measurements provide the precision needed for a reliable search.

Our laboratory builds and uses chirped-pulse Fourier-transform microwave (CP-FTMW) spectrometers to measure broad stretches of a spectrum at once. Our jet-cooled centimeter-wave instrument operates from 6.5 to 18.5 GHz, resolving nuclear hyperfine structure and detecting rare isotopic species at natural abundance. These measurements support searches for aromatic molecules, heterocycles, and oxygen-bearing organics in space. Read about the instrument.

Spectrum of the source TMC-1, a cold dark molecular cloud in the constellation of Taurus, taken with the Green Bank Telescope (black). Spectrum of pure CH₂CN as measured in the laboratory on Earth (red). The exact match of the frequency pattern confirms the identification in the ISM.

Different molecules require different ways of getting them into the gas phase and measuring their spectra. Heated sources let us study less volatile aromatic species, while laser ablation and sensitive cavity-enhanced instruments, used with our collaborators, have made measurements of larger PAHs possible. We also combine centimeter-wave work with millimeter- and submillimeter-wave measurements when a search requires predictions across a wider frequency range.

This combination enabled the laboratory study and ALMA detection of 2-methoxyethanol in NGC 6334I. For the cyanopyrenes and cyanocoronene, synthesis and laboratory spectroscopy supplied the frequencies needed to identify their much colder emission in TMC-1. The choice of experiment follows the molecule and the astronomical question.

Example of complementary broadband and cavity measurements. A complex mixture was measured with a CP-FTMW instrument (black). The signals identified were then re-measured at high-resolution with a cavity FTMW instrument (blue). Adapted from McGuire et al. 2018 PCCP 20, 13870.

Reaction/Mixture Screening

A reaction mixture can contain many molecular species, each contributing its own set of rotational lines. Sorting those lines into individual carriers is often the slowest part of the experiment. Broadband spectroscopy, targeted follow-up measurements, and quantum chemical calculations provide complementary constraints, but interpreting the full mixture also requires a way to decide which molecular assignments make chemical sense.

We developed Automated Mixture Analysis via Structural Evaluation (AMASE) to bring that chemical context into the assignment. The method represents molecules numerically in a chemical space and uses a graph-based ranking to evaluate candidate species alongside their spectral matches. It has reproduced laboratory mixture assignments with high accuracy in benchmark tests. We have since adapted this approach to astronomical line surveys, where it can also suggest molecules for further searches. Those suggestions are candidates to test, rather than detections on their own.

Gas-Phase Molecular Structures

Rotational constants describe a molecule's moments of inertia: how its mass is distributed about its principal axes. Three constants alone do not uniquely determine a molecular geometry. By measuring many isotopically substituted versions of a molecule, however, we obtain additional constraints on the positions of individual atoms and can derive detailed experimental gas-phase structures.

Our study of ethynylbenzonitrile isomers uses these measurements to ask how attaching nitrile and ethynyl groups changes the benzene ring. Comparing the experimental bond lengths with those of benzene reveals a pattern of shorter and longer bonds that can be interpreted in terms of electronic structure. The same measurements also enable astronomical searches: none of the three isomers was detected in TMC-1 in that study, placing upper limits on their abundances.

Bond-length changes in para-ethynylbenzonitrile relative to benzene: red bonds are longer and blue bonds are shorter. The resonance structures provide a way to interpret the measured pattern. Figure adapted from Stewart et al. (2025).