Bayesian Analysis of Theoretical Rotational Constants from Low-Cost Electronic Structure Methods

Laboratory Rotational Spectra of Cyanocyclohexane and Its Siblings (1- and 4-Cyanocyclohexene) Using a Compact CP-FTMW Spectrometer for Interstellar Detection

Chirped-pulse Fourier transform microwave (CP-FTMW) spectroscopy is a versatile technique to record broadband gas-phase rotational spectra, enabling detailed investigations of molecular structure, dynamics, and hyperfine interactions. Here, we present the development and application of a CP-FTMW spectrometer operating in the 6.5-18 GHz frequency range, studying cyanocyclohexane, 1-cyanocyclohexene, and 4-cyanocyclohexene using a heated pulsed supersonic expansion source. The dynamic range, experimental resolution, and high sensitivity enable observation of multiple conformers, precise measurements of hyperfine splitting arising from nuclear quadrupole coupling due to the nitrogen atom in the cyano group, as well as the observation of singly 13C- and 15N-substituted isotopic isomers in natural abundance. Using the latter, precise structures for the molecules are derived. The accurate rotational spectra enabled a search for these species toward the dark, cold molecular cloud TMC-1; no signals are found, and we discuss the implications of derived upper limits on the interstellar chemistry of the cyanocyclohexane family.

Detection of interstellar 1-cyanopyrene: A four-ring polycyclic aromatic hydrocarbon

Polycyclic aromatic hydrocarbons (PAHs) are organic molecules containing adjacent aromatic rings. Infrared emission bands show that PAHs are abundant in space, but only a few specific PAHs have been detected in the interstellar medium. We detected 1-cyanopyrene, a cyano-substituted derivative of the related four-ring PAH pyrene, in radio observations of the dense cloud TMC-1, using the Green Bank Telescope. The measured column density of 1-cyanopyrene is [Formula: see text] cm-2, from which we estimate that pyrene contains up to 0.1% of the carbon in TMC-1. This abundance indicates that interstellar PAH chemistry favors the production of pyrene. We suggest that some of the carbon supplied to young planetary systems is carried by PAHs that originate in cold molecular clouds.

Unveiling the Spectroscopy of Complex Organic Radicals by Exploiting Faraday Rotation at (Sub-)millimeter Wavelengths. Illustration with the Acetonyl Radical

Radical species constitute the main reactants of numerous chemical reaction networks occurring in diverse environments. Rotationally resolved laboratory data, essential to undertake the detection of these highly reactive species, remain difficult to obtain using conventional high-resolution spectroscopy techniques. In the present work, we exploit a new experimental setup based on the Faraday rotation detection technique which allows us to study the gas phase spectra of relatively large radicals, such as dehydrogenated complex organic molecules (COMs). We recorded 2086 pure rotational transitions of the acetonyl radical (CH3COCH2) in the 150-450 GHz range, for which no rotational information was previously available. The radical exhibits relatively complex couplings of angular momenta, involving the overall rotation angular momentum, the spin of the unpaired electron, and two large amplitude motions. The data set has been fit using a semirigid Hamiltonian and shows the need for the development of specific theoretical models.

Twins in rotational spectroscopy: Does a rotational spectrum uniquely identify a molecule?

Rotational spectroscopy is the most accurate method for determining structures of molecules in the gas phase. It is often assumed that a rotational spectrum is a unique "fingerprint" of a molecule. The availability of large molecular databases and the development of artificial intelligence methods for spectroscopy make the testing of this assumption timely. In this paper, we pose the determination of molecular structures from rotational spectra as an inverse problem. Within this framework, we adopt a funnel-based approach to search for molecular twins, which are two or more molecules, which have similar rotational spectra but distinctly different molecular structures. We demonstrate that there are twins within standard levels of computational accuracy by generating rotational constants for many molecules from several large molecular databases, indicating that the inverse problem is ill-posed. However, some twins can be distinguished by increasing the accuracy of the theoretical methods or by performing additional experiments.

Unbiased Comparison between Theoretical and Experimental Molecular Structures and Properties: Toward an Accurate Reduced-Cost Evaluation of Vibrational Contributions

The tremendous development of hardware and software is constantly increasing the role of quantum chemical (QC) computations in the assignment and interpretation of experimental results. However, an unbiased comparison between theory and experiment requires the proper account of vibrational averaging effects. In particular, high-resolution spectra in the gas phase are now available for molecules containing up to about 50 atoms, which are too large for a brute-force approach with the available QC methods of sufficient accuracy. In the present paper, we introduce hybrid approaches, which allow the accurate evaluation of vibrational averaging effects for molecules of this size beyond the harmonic approximation, with special attention being devoted to rotational constants. After the validation of new tools for relatively small molecules, the β-estradiol hormone and a prototypical molecular motor have been considered to witness the feasibility of accurate computations for large molecules.

High-Resolution Spectroscopic Investigation of the CH2CHO Radical in the Sub-Millimeter Region

In this work, the pure rotational spectrum of the vinoxy radical (CH2CHO) has been studied at millimeter and sub-millimeter wavelengths (110-860 GHz). CH2CHO was produced by H-abstraction from acetaldehyde (CH3CHO) using atomic fluorine in a double-pass absorption cell at room temperature. A Zeeman-modulation spectrometer, in which an external magnetic field generated inside the absorption cell is amplitude-modulated, was used to record the pure rotational transitions of the radical. The recorded spectra are devoid of signals from closed-shell species, allowing for relatively fast acquisitions over wide spectral windows. Transitions involving values of the rotational quantum numbers N″ and Ka″ up to 41 and 18, respectively, were measured and combined with all available high-resolution literature data (both pure rotation and ground-state combination differences from ro-vibration) to greatly improve the modeling of the CH2CHO spectrum. The combined experimental line list is fit using a semirigid rotor Hamiltonian, and the results are compared to quantum chemical calculations. This laboratory study provides the spectroscopic information needed to search for CH2CHO in various interstellar environments, from cold (e.g., typically 10 K for dense molecular clouds) to warm (e.g., ∼200 K for hot corinos) objects.

Searches for bridged bicyclic molecules in space-norbornadiene and its cyano derivatives

The norbornadiene (NBD) molecule, C7H8, owes its fame to its remarkable photoswitching properties that are promising for molecular solar-thermal energy storage systems. Besides this photochemical interest, NBD is a rather unreactive species within astrophysical conditions and it should exhibit high photostability, properties that might also position this molecule as an important constituent of the interstellar medium (ISM)-especially in environments that are well shielded from short-wavelength radiation, such as dense molecular clouds. It is thus conceivable that, once formed, NBD can survive in dense molecular clouds and act as a carbon sink. Following the recent interstellar detections of large hydrocarbons, including several cyano-containing ones, in the dense molecular cloud TMC-1, it is thus logical to consider searching for NBD-which presents a shallow but non-zero permanent electric dipole moment (0.06 D)-as well as for its mono- and dicyano-substituted compounds, referred to as CN-NBD and DCN-NBD, respectively. The pure rotational spectra of NBD, CN-NBD, and DCN-NBD have been measured at 300 K in the 75-110 GHz range using a chirped-pulse Fourier-transform millimetre-wave spectrometer. Of the three species, only NBD was previously studied at high resolution in the microwave domain. From the present measurements, the derived spectroscopic constants enable prediction of the spectra of all three species at various rotational temperatures (up to 300 K) in the spectral range mapped at high resolution by current radio observatories. Unsuccessful searches for these molecules were conducted toward TMC-1 using the QUIJOTE survey, carried out at the Yebes telescope, allowing derivation of the upper limits to the column densities of 1.6 × 1014 cm-2, 4.9 × 1010 cm-2, and 2.9 × 1010 cm-2 for NBD, CN-NBD, and DCN-NBD, respectively. Using CN-NBD and cyano-indene as proxies for the corresponding bare hydrocarbons, this indicates that-if present in TMC-1-NBD would be at least four times less abundant than indene.

A rotational investigation of the three isomeric forms of cyanoethynylbenzene (HCC-C6H4-CN): benchmarking experiments and calculations using the "Lego brick" approach

We report the study of three structural isomers of phenylpropiolonitrile (3-phenyl-2-propynenitrile, C₆H₅-C₃N) containing an alkyne function and a cyano group, namely ortho-, meta-, and para-cyanoethynylbenzene (HCC-C₆H₄-CN). The pure rotational spectra of these species have been recorded at room temperature in the millimeter-wave domain using a chirped-pulse spectrometer (75-110 GHz) and a source-frequency modulation spectrometer (140-220 GHz). Assignments of transitions in the vibrational ground state and several vibrationally excited states were supported by quantum chemical calculations using the so-called "Lego brick" approach [A. Melli, F. Tonolo, V. Barone and C. Puzzarini, J. Phys. Chem. A, 2021, 125, 9904-9916]. From these assignments, accurate spectroscopic (rotational and centrifugal distortion) constants have been derived: for all species and all observed vibrational states, predicted rotational constants show relative accuracy better than 0.1%, and often of the order of 0.01%, compared to the experimental values. The present work hence further validates the use of the "Lego brick" approach for predicting spectroscopic constants with high precision.

Methyl Internal Rotation in Fruit Esters: Chain-Length Effect Observed in the Microwave Spectrum of Methyl Hexanoate

The gas-phase structures of the fruit ester methyl hexanoate, CH₃-O-(C=O)-C₅H₁₁, have been determined using a combination of molecular jet Fourier-transform microwave spectroscopy and quantum chemistry. The microwave spectrum was measured in the frequency range of 3 to 23 GHz. Two conformers were assigned, one with C_s symmetry and the other with C₁ symmetry where the γ-carbon atom of the hexyl chain is in a gauche orientation in relation to the carbonyl bond. Splittings of all rotational lines into doublets were observed due to internal rotation of the methoxy methyl group CH₃-O, from which torsional barriers of 417 cm⁻¹ and 415 cm⁻¹, respectively, could be deduced. Rotational constants obtained from geometry optimizations at various levels of theory were compared to the experimental values, confirming the soft degree of freedom of the (C=O)-C bond observed for the C₁ conformer of shorter methyl alkynoates like methyl butyrate and methyl valerate. Comparison of the barriers to methyl internal rotation of methyl hexanoate to those of other CH₃-O-(C=O)-R molecules leads to the conclusion that though the barrier height is relatively constant at about 420 cm⁻¹, it decreases in molecules with longer R.

Large Amplitude Motions in 2,3-Dimethylfluorobenzene: Steric Effects Failing to Interpret Hindered Methyl Torsion

The microwave spectrum of 2,3-dimethylfluorobezene, one of the six isomers of dimethylfluorobenzene, was recorded using a pulsed molecular jet Fourier transform microwave spectrometer operating in the frequency range from 2 to 26.5 GHz. The internal rotations of two inequivalent methyl groups, causing splittings of up to several hundred MHz of all rotational energy levels into quintets, were analyzed and modeled. The torsional barriers of the methyl groups at the ortho and the meta positions were determined to be 215.5740(56) cm^-1 and 488.53(11) cm^-1. A comparison with the barrier heights observed for the two isomers 2,6-dimethylfluorobenzene and 3,4-dimethylfluorobenzene has shown that the methyl group at the meta position seems to be invisible to its neighboring ortho-methyl group, while the meta-methyl group clearly senses the ortho one. Steric effects are not able to explain this observation, and electrostatic effects are most probably the reason. Highly accurate molecular parameters determined experimentally were compared with those obtained from quantum chemical calculations at different levels of theory.

Microwave Spectrum of Two-Top Molecule: 2-Acetyl-3-Methylthiophene

The microwave spectrum of 2-acetyl-3-methylthiophene (2A3MT) was recorded in the frequency range from 2 to 26.5 GHz using a molecular jet Fourier transform microwave spectrometer and could be fully assigned to the anti-conformer of the molecule, while the syn-conformer was not observable. Torsional splittings of all rotational transitions in quintets due to internal rotations of the acetyl methyl and the ring methyl groups were resolved and analyzed, yielding barriers to internal rotation of 306.184(46) cm^-1 and 321.813(64) cm^-1 , respectively. The rotational and centrifugal distortion constants were determined with high accuracy, and the experimental values are compared to those derived from quantum chemical calculations. The experimentally determined inertial defect supports the conclusion that anti-2A3MT is planar, even though a number of MP2 calculations predicted the contrary.

Copper-Catalyzed Transfer Hydrodeuteration of Aryl Alkenes with Quantitative Isotopomer Purity Analysis by Molecular Rotational Resonance Spectroscopy

A copper-catalyzed alkene transfer hydrodeuteration reaction that selectively incorporates one hydrogen and one deuterium atom across an aryl alkene is described. The transfer hydrodeuteration protocol is selective across a variety of internal and terminal alkenes and is also demonstrated on an alkene-containing complex natural product analog. Beyond using ¹H, ²H, and ¹³C NMR analysis to measure reaction selectivity, six transfer hydrodeuteration products were analyzed by molecular rotational resonance (MRR) spectroscopy. The application of MRR spectroscopy to the analysis of isotopic impurities in deuteration chemistry is further explored through a measurement methodology that is compatible with high-throughput sample analysis. In the first step, the MRR spectroscopy signatures of all isotopic variants accessible in the reaction chemistry are analyzed using a broadband chirped-pulse Fourier transform microwave spectrometer. With the signatures in hand, measurement scripts are created to quantitatively analyze the sample composition using a commercial cavity enhanced MRR spectrometer. The sample consumption is below 10 mg with analysis times on the order of 10 min using this instrument-both representing order-of-magnitude reduction compared to broadband MRR spectroscopy. To date, these measurements represent the most precise spectroscopic determination of selectivity in a transfer hydrodeuteration reaction and confirm that product regioselectivity ratios of >140:1 are achievable under this mild protocol.

Seed-Adsorbate Interactions as the Key of Heterogeneous Butanol and Diethylene Glycol Nucleation on Ammonium Bisulfate and Tetramethylammonium Bromide

Condensation particle counter (CPC) instruments are commonly used to detect atmospheric nanoparticles. They operate on the basis of condensing an organic working fluid on the nanoparticle seeds to grow the particles to a detectable size, and at the size of few nanometers, their efficiency depends on how well the working fluid interacts with the seeds under the measurement conditions. This study models the first steps of heterogeneous nucleation of two working fluids commonly used in CPCs (diethylene glycol (DEG) and n-butanol) onto two positively charged seeds, ammonium bisulfate and tetramethylammonium bromide. The nucleation process is modeled on a molecular level using a combination of systematic configurational sampling and density functional theory (DFT). We take into account the conformational flexibility of DEG and n-butanol and determine the key factors that can improve the efficiency of nanoparticle measurements by CPCs. The results show that hydrogen bonding between the seed and the working fluid molecules is central to the adsorption of the first DEG/n-butanol molecules onto the seeds. However, intermolecular hydrogen bonding between the adsorbed molecules can also enhance the nucleation process for the weakly adsorbing vapor molecules. Accordingly, the heterogeneous nucleation probability is higher for working fluid-nanoparticle combinations with a higher potential for hydrogen bonding; in this case, DEG and ammonium bisulfate. Moreover, conformational analysis and methodology evaluations indicate that the consideration of adsorbate conformers and step-wise addition of the vapor molecules to the seeds is not essential for qualitative modeling of heterogeneous nucleation systems, at least for systems where the adsorbate and seed chemical properties are clearly different. This is the first molecular-level modeling study reporting detailed chemical reasons for experimentally observed seed and working fluid preferences in CPCs and reproducing the experimental observations. Our presented approach can be likely used for predicting preferences in similar nucleating systems.

Structure, Stability, and Spectroscopic Properties of Small Acetonitrile Cation Clusters

Ionization and fragmentation pathways induced by ionizing agents are key to understanding the formation of complex molecules in astrophysical environments. Acetonitrile (CH₃CN), the simplest organic nitrile, is an important molecule present in the interstellar medium. In this work, DFT and MP2 calculations were performed in order to obtain the low energy structures of the most relevant cations formed from electron-stimulated ion desorption of CH₃CN ices. Selected reaction pathways and spectroscopic properties were also calculated. Our results indicate that the most stable acetonitrile cation structure is CH₂CNH⁺ and that hydrogenation can occur successively without isomerization steps until its complete saturation. Moreover, the stability of distinct cluster families formed from the interaction of acetonitrile with small fragments, such as CH_n⁺, C₂H_n⁺, and CH_nCNH⁺, is discussed in terms of their respective binding energies. Some of these molecular clusters are stabilized by hydrogen bonds, leading to species whose infrared features are characterized by a strong redshift of the N-H stretching mode. Finally, the rotational spectra of CH₃CN and protonated acetonitrile, CH₃CNH⁺, were simulated using distinct computational protocols based on DFT, MP2, and CCSD(T) considering centrifugal distortion, vibrational-rotational coupling, and vibrational anharmonicity corrections. By adopting an empirical scaling procedure for calculating spectroscopic parameters, we were able to estimate the rotational frequencies of CH₃CNH⁺ with an expected average error below 1 MHz for J values up to 10.

Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning

A proof-of-concept framework for identifying molecules of unknown elemental composition and structure using experimental rotational data and probabilistic deep learning is presented. Using a minimal set of input data determined experimentally, we describe four neural network architectures that yield information to assist in the identification of an unknown molecule. The first architecture translates spectroscopic parameters into Coulomb matrix eigenspectra as a method of recovering chemical and structural information encoded in the rotational spectrum. The eigenspectrum is subsequently used by three deep learning networks to constrain the range of stoichiometries, generate SMILES strings, and predict the most likely functional groups present in the molecule. In each model, we utilize dropout layers as an approximation to Bayesian sampling, which subsequently generates probabilistic predictions from otherwise deterministic models. These models are trained on a modestly sized theoretical dataset comprising ∼83 000 unique organic molecules (between 18 and 180 amu) optimized at the ωB97X-D/6-31+G(d) level of theory, where the theoretical uncertainties of the spectoscopic constants are well-understood and used to further augment training. Since chemical and structural properties depend strongly on molecular composition, we divided the dataset into four groups corresponding to pure hydrocarbons, oxygen-bearing species, nitrogen-bearing species, and both oxygen- and nitrogen-bearing species, training each type of network with one of these categories, thus creating "experts" within each domain of molecules. We demonstrate how these models can then be used for practical inference on four molecules and discuss both the strengths and shortcomings of our approach and the future directions these architectures can take.