Skip to main content
ACS Omega logoLink to ACS Omega
. 2022 Nov 18;7(48):44078–44084. doi: 10.1021/acsomega.2c05494

Neglected Importance of Anharmonicity in Quantifying the Renner–Teller Effect

Qing Lu 1,*
PMCID: PMC9730495  PMID: 36506164

Abstract

graphic file with name ao2c05494_0006.jpg

Linear molecules in degenerate electronic states are influenced by the Renner–Teller (RT) effect. Currently the formula to quantify this vibronic interaction only considers the harmonic term. However, such a harmonic formula cannot well describe molecules that have double-well potential along the bending coordinate. In this work, we propose a new formula to quantify the RT effect by explicitly including the anharmonic term. A sign function and a delta function are additionally included to distinguish different scenarios. It is mathematically proved that the new formula is capable of differentiating different degrees of RT splitting. Representative molecules of different types are calculated. According to the new formula, molecules experiencing a weak RT effect have a positive parameter smaller than 1. For those experiencing a medium RT effect, the parameter will be negative. For others experiencing a large RT effect, the parameter will be larger than 1.

Introduction

The Renner–Teller (RT) effect describes the energy splitting of degenerate states for linear molecules during bending motion.1 It influences chemical behaviors of linear molecules in many aspects. In the field of direct laser cooling of polyatomic molecules, quite a few studied molecules possess a linear configuration during photon excitations.2 For a molecule to be laser-coolable, it should stay at its ground vibronic state, which requires that it should have its vibrational branching ratio close to one.3 If the molecule happens to be linear in the degenerate state, the RT effect influences the vibrational branching ratios, which then dictates the efficiencies of optical transition cycles. These imply that the linear laser-coolable molecules should experience a weak RT effect and thus preferably remain linear at relevant electronic states.4 In addition to its significance in ultracold chemistry, the RT effect is important in the interpretation of complicated spectra.57 The most famous example might be the spectral interpretation of the NH2 radical.8,9 By taking account of the RT effect, Pople and co-workers obtained a very good agreement between observed frequencies and calculated frequencies of the NH2 radical. This in turn confirms Dressler and Ramsay’s attributing of the absorption spectrum to transitions between two real components of the Π state (Figure 1c). Other illustrations, where the RT effect plays an important role, include enhancement of magnetism of luminescence properties by suppressing RT vibronic coupling or restriction of RT distortion,10 bond stretch isomerism,11 chemistry involving continuum processes,12 or anomaly dynamic behaviors.13

Figure 1.

Figure 1

Different types of RT adiabatic potential energy curves along the bending coordinate. (a) both potential energy curves of a parabola shape; (b) upper potential energy curve of a parabola shape and lower potential energy curve of a double-well shape; (c) both potential energy curves of a double-well shape.

Intuitively, different molecules would exhibit different behaviors when experiencing the RT distortion. As shown in Herzberg’s famous book14 as well as other’s work,9,15 the RT splitting can be classified into different types, although their criteria for categorization are not exactly the same. In our view, the RT splitting can be categorized into three types (Figure 1). The first type has both curves in a parabola form; the second type has the upper curve in a parabola form while the lower curve is in a double-well form; and the third type has both curves in a double-well form. The other categorizations have finer criteria such as the order of symmetries for the upper and lower curves or the curvature of the curves.

To describe the RT splitting, the RT parameter ε is introduced, so the upper (V+) and lower (V) component curves are connected by ε

graphic file with name ao2c05494_m001.jpg 1

where Vm is the mean potential or the zeroth-order potential. The upper and lower components are of different spatial symmetries, and different molecules may have different symmetries for the upper or lower components. Taking the CCP radical as an example, the upper and lower states are A′ and A″, respectively. However, for other molecules, the state symmetries might be the opposite. The RT parameter ε is defined as16

graphic file with name ao2c05494_m002.jpg 2

As the molecule is symmetric with respect to the molecular axis, the bending potential must be an even function along the bending coordinate (q), so that

graphic file with name ao2c05494_m003.jpg 3
graphic file with name ao2c05494_m004.jpg 4

where a and α are quadratic coefficients, while b and β are quartic coefficients. They, respectively, represent the force constant and the anharmonic correction. As the anharmonic terms are usually much smaller than the harmonic terms, the higher-order terms are usually neglected.14,15 Therefore, the RT parameter ε becomes

graphic file with name ao2c05494_m005.jpg 5

where ω+ and ω are, respectively, the harmonic frequencies associated with the V+ and V potentials.

This definition of ε has been widely used in characterizing the RT effect, even till today.6,1719 For an extension, very recently, it was shown that the NO molecule encapsulated in a fullerene can also experience a vibronic coupling effect very similar to the RT effect regarding the NO translation.20 Thus, the RT parameter may not only characterize the distortion of linear molecules but also quantify energy splitting of more complex systems.

However, there is a flaw in the harmonic formula to quantify the RT effect. If only harmonic terms are used, combining eqs 3 and 4, the lower component of the potential energy curve (V) would be

graphic file with name ao2c05494_m006.jpg 6

It is obvious that V is simply a parabola equation, which has one and only one minimum. Consequently, the double-well potential energy curve shown in Figure 1c can never be achieved. Therefore, one cannot compare the RT parameter ε1 for the case corresponding to Figure 1a and the RT parameter ε2 for the case corresponding to Figure 1c since the RT parameter ε cannot transit from case (a) to case (c). A consequence of this incapability is that the trend for molecular properties is not well recognized. For instance, in recent years, it has received significant attention for exploring polyatomic candidates suitable for direct laser cooling.3,21 Our recent efforts try to understand why certain linear molecules are good laser-coolable candidates while others are not and try to setup a trend to make predictions. This leads us to compare the RT parameters. Nevertheless, as discussed, the harmonic RT parameter cannot characterize the double-well potential; therefore, the trend based on it cannot reflect the true physics.

Therefore, in this work, we propose a new formula to calculate the RT parameter εa. The subscript “a” stands for the inclusion of “anharmonic” terms into the formula to differentiate itself from the conventional RT parameter ε. This proposed parameter εa can well distinguish the three types of molecules. To examine the performance of the new formula, three types of molecules are studied. They include 2Π CCN, 2Π CCP, 2Π HCSi, 2Π OCS+, and 2Π NCO (type a), 2A1 BH2, 2A1 AlH2, 2A1 SNO, and 2A1 NO2, (type b), and 1A1 CH2, 1A1 SiH2, 1A1 GeH2, and 2B1 NH2 (type c). As will be shown later, the new-defined parameter can well describe molecules for all types and shows a transition between different scenarios, which remedies the deficiency of the previous definition of the RT parameter.

Methods

In this work, the CCN, CCP, HCSi, OCS+, and NCO (type a), BH2, AlH2, SNO, and NO2, (type b), and singlet CH2, singlet SiH2, singlet GeH2, and NH2 (type c) were calculated for their RT parameters εa. Unless explicitly stated, such as in cases of singlet CH2, SiH2, and GeH2, all molecules are studied for their lowest spin state. Since the molecules have different term symbols for linear and bent configurations, symbols for the orbital angular momentum part of the term are not listed to avoid any confusion.

The local potential energy surfaces (PESs) were established for these molecules around their ground equilibrium geometry. The sampling of grid points and the fitting of PESs have been described in previous studies.7,22 For a brief summary, the range for the bond length spans from Re-0.3 to Re +0.5 in the unit of Bohr, where Re is chosen close to the experimental value. The range for the bond angle spans from 180 to 100°. The PESs were obtained by the non-linear least-square fitting to the fifth-order polynomial in the displacement coordinates. The geometry minimum is obtained using the differential evolution algorithm.23 The fitting was carried out using an in-house Python-based script.

For each grid point, the electronic energies were calculated by the state-averaged multi-reference configuration interaction method with the Davidson correction.24 The Cs symmetry was used during the calculation. The full valence active space was used together with the cc-pVQZ basis sets.25

After obtaining the PES, the upper and lower bending curves are directly fitted to the second or fourth even order polynomial in the unit of radian with bond lengths equal to the equilibrium values

graphic file with name ao2c05494_m007.jpg
graphic file with name ao2c05494_m008.jpg 7

Neglecting the constant term, each bending energy curve can be described using the second- and fourth-order coefficients as the features of the potential energy curves. The feature length can be defined as

graphic file with name ao2c05494_m009.jpg 8

The RT parameter in eq 2 then becomes

graphic file with name ao2c05494_m010.jpg 9

Notably, when the fourth-order coefficients are neglected, eq 9 reduces to eq 2.

Having extended the canonical definition of the RT parameter formula, the bending potential is fitted according to eq 7 to obtain the quadratic (a+, a) and quartic (b+, b) coefficients for upper and lower states. For the harmonic curve, the quartic coefficient is set to 0 and the second even order polynomial is used for fitting.

To distinguish different types of molecules, a sign function of Inline graphic is first introduced, where the Inline graphic returns 1 if the ratio of Inline graphic is positive, otherwise −1. Additionally, a delta function is introduced to exhibit the quartic effect. Eventually, the proposed RT parameter equation reads as

graphic file with name ao2c05494_m014.jpg 10

The physical interpretations and applications of eq 10 are illustrated in detail in the next section.

Results and Discussion

As discussed above, the harmonic RT parameter cannot characterize the double-well of potential V. Therefore, the anharmonic terms need to be included. Two features can be summarized for εa calculated using eq 10. First, the coefficients a and b for double-well curve fitting must possess different signs. Otherwise, the bending potential will be monotonical along the bending coordinate when the molecular configuration deviates from linearity. Moreover, the sign of a must be negative in order to have a downward parabola when the bending angle is small. As a result, the quadratic coefficients a± for type (a) molecules will both be positive; the coefficient, a+ and a for type (b) molecules will be, respectively, positive and negative; and the coefficients a± for type (c) molecules will both be negative. Benefiting from this property, a sign function is included in the εa formula so that only type (b) molecules will have a negative RT parameter.

Second, if no quartic term is present in the bending expansion, the delta function in eq 10 will be 0. This corresponds to the type (a) situation. As discussed above, the sign function for type (a) molecules returns 1. Therefore, the proposed RT formula becomes similar to the canonical formula. The difference is that the canonical ε equation is based on the harmonic frequencies, while the proposed formula depends on the fitted quadratic coefficient, which is equivalent to the force constant. On a related note, having determined the sign function and the delta functions, the remaining part of eq 10 is a pure fractional number, namely, smaller than 1. On the other hand, when the quartic term is present, as for type (c) molecules, the delta functions remain in the bracket, and the final RT parameter will be greater than 1.

Overall, the RT parameter calculated using the proposed formula is robust to distinguish the three types of molecules, as is mathematically proved. In addition, the new formula only needs to fit a set of energies along the bending curve, which is computationally less expensive than the frequency calculations.

To illustrate the application of the new RT parameters, three types of molecules are studied. Figure 2 shows the bending curves of CCP, BH2, and NH2 molecules as examples for the three types of molecules. The bond lengths are kept fixed at the ground equilibrium values, which are obtained either from the National Institute of Standards and Technology computational chemistry comparison and benchmark database26 or direct fitting of the local PES. The potential energy curves for other molecules and the corresponding bond lengths for fitting can be found in the Supporting Information.

Figure 2.

Figure 2

Different types of molecules defined in this work.

Figure 3 shows fitting of the upper and lower bending curves of CCP, BH2, and NH2. It becomes more evident that the bending curves follow the previous categorization. For type (a) molecules, as the RT parameter only depends on the quadratic terms.

graphic file with name ao2c05494_m015.jpg 11

Figure 3.

Figure 3

Fitting of lower (left) and upper (right) bending potential curves for the three types of molecules.

Assigning t = Inline graphic and taking advantage of the fact that V+ is steeper than V, one will obtain that a+ > a and t > 1. The εa in eq 11 then becomes

graphic file with name ao2c05494_m017.jpg 12

It can be easily shown that eq 12 increases monotonically with respect to t. Therefore, a larger RT parameter εa indicates a larger t, which in turn generally (but not always) indicates a larger splitting of the component curves. This remark is also valid for the RT parameter calculated using the canonical equation as eq 10 of εa for type (a) molecules is reduced to the canonical equation form.

For type (b) molecules, the RT parameters εa differentiate themselves from those of type (a) and type (c) molecules in term of the sign. To compare molecules within type (b), a similar analysis can be carried out. As the quartic coefficients are for higher-order terms, they are generally smaller in absolute values than the quadratic term. Thus, a larger absolute RT parameter suggests larger splitting, but caution should be given, and other characteristic parameters will become useful. For type (c) molecules, their RT parameters εa are greater than 1 and thus larger than those for type (a) molecules.

Table 1 shows the RT parameters calculated by both the proposed and canonical formulae for different types of molecules. It needs to be mentioned that the negative sign for the new-defined RT parameter εa has a different meaning from the canonical one. In the canonical case, the negative sign means that the A′ curve is lower in energy than the A” curve. In our framework, we do not distinguish whether the A′ curve is lower or higher in energy. Therefore, for type (a) molecules, the absolute values of canonical RT parameters are present for comparison. For type (b) and type (c) molecules, one or both of the component energy curves has a double-well shape; thus, the frequency is an imaginary number at the linear configuration. The RT parameters for these cases are consequently ill-defined.

Table 1. RT Parameter (εa) Calculated Using the Proposed Formula and the RT Parameter Calculated Using Canonical Formula.

  type εa(this work) |ε|a
CCN a 0.35 0.43b
CCP a 0.36 0.57c
HCSi a 0.18 0.11d
OCS+ a 0.15 0.10e
NCO a 0.14 0.12–0.18f
BH2 b –0.22  
AlH2 b –0.16  
SNO b –0.26  
NO2 b –0.25  
NH2 c 1.77  
CH2 c 1.65  
SiH2 c 1.42  
GeH2 c 1.85  
a

Absolute value of the RT parameter from canonical formula.

b

Adapted from ref (6).

c

Adapted from ref (7).

d

Adapted from ref (17).

e

Adapted from ref (18).

f

Adapted from ref (27).

Comparing the RT parameters for type (a) molecules, the two formulae generally have the same trend. The correlation coefficient reaches 0.96. The difference might result from the use of frequency or the “force constant”.

The new definition of the RT parameter is important in exploring laser-coolable molecules. For polyatomic molecules capable of being cooled to the ultracold regime, the excited state should retain nearly identical geometries to that of the ground state. One of the well-known laser-coolable molecules is the alkaline-earth metal monohydroxide, such as the CaOH molecule.4 The ground state of the CaOH molecule is the Σ state, while its first excited state is the Π state. The Σ state is a non-degenerate state, so it will not experience the RT splitting. On the contrary, the Π state is capable of RT splitting. As the geometry change needs to be minimal for being laser-coolable, the molecule needs to have a small RT parameter (εa) so that the linear configuration remains as the equilibrium. In fact, the CaOH molecule does have a parabola bending potential (Figure 4), and its εa is calculated to be 0.24.

Figure 4.

Figure 4

Bending potential curves of the CaOH2Π state.

A further conjecture can be made based on the current framework that the type (c) molecule also has the potential to be laser-coolable as long as the bond lengths and bond angles are similar between the upper and lower states. In fact, very recent studies28 indeed propose that bent molecules such as SOH and CaSH can also be good laser-coolable molecules. The current work helps rationalize their hypothesis. On the contrary, if a molecule belongs to a type (b) molecule, then it would not be a good laser-coolable molecule, and it would feature a negative RT parameter. Lastly, it should be noted that the above analysis is a necessary but not sufficient condition to find laser-coolable molecules.

Conclusions

In this work, a new formula to quantify the RT effect is proposed by explicitly including the anharmonic term. The formula is capable of distinguishing three types of RT splitting with distinct RT parameters. For type (a) molecules, which experience a weak RT effect, they have a positive RT parameter smaller than 1. For type (b) molecules experiencing a medium RT effect, the RT parameter will be negative. For molecules experiencing a large RT effect, the RT parameter will be larger than 1. The proposed formula extends the conventional one and has a wider range of applications, such as helping rationalize recent laser-cooling studies and helping predict novel laser-coolable molecules.

Acknowledgments

This work was supported by the Beijing Natural Science Foundation (no. 2214065) and National Natural Science Foundation of China (no. 22003068).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c05494.

  • Bending potential energy curves for the studied molecules along with the corresponding bond length and the fitted quadratic and quartic coefficients for the studied molecules (PDF)

The author declares no competing financial interest.

Supplementary Material

ao2c05494_si_001.pdf (444.7KB, pdf)

References

  1. Jungen C.; Merer A. J. Orbital angular momentum in triatomic molecules. Mol. Phys. 1980, 40, 1–23. 10.1080/00268978000101291. [DOI] [Google Scholar]
  2. Baum L.; Vilas N. B.; Hallas C.; Augenbraun B. L.; Raval S.; Mitra D.; Doyle J. M. Establishing a nearly closed cycling transition in a polyatomic molecule. Phys. Rev. A 2021, 103, 043111. 10.1103/PhysRevA.103.043111. [DOI] [Google Scholar]; a Owens A.; Clark V. H. J.; Mitrushchenkov A.; Yurchenko S. N.; Tennyson J. Theoretical rovibronic spectroscopy of the calcium monohydroxide radical (CaOH). J. Chem. Phys. 2021, 154, 234302. 10.1063/5.0052958. [DOI] [PubMed] [Google Scholar]
  3. Fitch N. J.; Tarbutt M. R.. Laser-cooled molecules. In Advances In Atomic, Molecular, and Optical Physics; Dimauro L. F., Perrin H., Yelin S. F., Eds.; Academic Press, 2021; Vol. 70, Chapter 3, pp 157–262. [Google Scholar]
  4. Isaev T. A. Direct laser cooling of molecules. Phys. Usp. 2020, 63, 289–302. 10.3367/ufne.2018.12.038509. [DOI] [Google Scholar]
  5. Brünken S.; Lipparini F.; Stoffels A.; Jusko P.; Redlich B.; Gauss J.; Schlemmer S. Gas-phase vibrational spectroscopy of the hydrocarbon cations l-C3H+, HC3H+, and c-C3H2+: structures, isomers, and the influence of Ne-tagging. J. Phys. Chem. A 2019, 123, 8053–8062. 10.1021/acs.jpca.9b06176. [DOI] [PMC free article] [PubMed] [Google Scholar]; a Lyle J.; Jagau T.-C.; Mabbs R. Spectroscopy of temporary anion states: Renner–Teller coupling and electronic autodetachment in copper difluoride anion. Faraday Discuss. 2019, 217, 533–546. 10.1039/C8FD00224J. [DOI] [PubMed] [Google Scholar]; b Coudert L. H.; Gans B.; Garcia G. A.; Loison J.-C. Renner-Teller effects in the photoelectron spectra of CNC, CCN, and HCCN. J. Chem. Phys. 2018, 148, 054302. 10.1063/1.5011152. [DOI] [PubMed] [Google Scholar]; c Freund J.; Kempf S. C. G.; Jensen P.; Nagashima U.; Hirano T. Computational spectroscopy of NCS in the Renner-degenerate electronic state X2Π. J. Mol. Spectrosc. 2018, 345, 31–38. 10.1016/j.jms.2017.11.010. [DOI] [Google Scholar]
  6. Hill J. G.; Mitrushchenkov A.; Yousaf K. E.; Peterson K. A. Accurate ab initio ro-vibronic spectroscopy of the X2Π CCN radical using explicitly correlated methods. J. Chem. Phys. 2011, 135, 144309. 10.1063/1.3647311. [DOI] [PubMed] [Google Scholar]
  7. Lu Q. Accurate ab initio vibronic spectroscopy of the CCP and CCAs radicals. Chem. Phys. Lett. 2020, 739, 137017. 10.1016/j.cplett.2019.137017. [DOI] [Google Scholar]
  8. Pople J. A.; Longuet-Higgins H. C. Theory of the Renner effect in the NH2 radical. Mol. Phys. 1958, 1, 372–383. 10.1080/00268975800100441. [DOI] [Google Scholar]
  9. Dressler K.; Ramsay D. A. The electronic absorption spectra of NH2 and ND2. Phil. Trans. Roy. Soc. Lond. Math. Phys. Sci. 1959, 251, 553–602. 10.1098/rsta.1959.0011. [DOI] [Google Scholar]
  10. Huzan M. S.; Fix M.; Aramini M.; Bencok P.; Mosselmans J. F. W.; Hayama S.; Breitner F. A.; Gee L. B.; Titus C. J.; Arrio M.-A.; et al. Single-ion magnetism in the extended solid-state: insights from X-ray absorption and emission spectroscopy. Chem. Sci. 2020, 11, 11801–11810. 10.1039/D0SC03787G. [DOI] [PMC free article] [PubMed] [Google Scholar]; a Li L.; Ravinson D. S. M.; Haiges R.; Djurovich P. I.; Thompson M. E. Enhancement of the luminescent efficiency in carbene-Au[I];-aryl complexes by the restriction of Renner–Teller distortion and bond rotation. J. Am. Chem. Soc. 2020, 142, 6158–6172. 10.1021/jacs.9b13755. [DOI] [PubMed] [Google Scholar]
  11. Homray M.; Mondal S.; Misra A.; Chattaraj P. K. Bond stretch isomerism in Be32– driven by the Renner–Teller effect. Phys. Chem. Chem. Phys. 2019, 21, 7996–8003. 10.1039/C9CP00643E. [DOI] [PubMed] [Google Scholar]
  12. Jungen C. The Renner-Teller effect revisited 40 years later. J. Mol. Spectrosc. 2019, 363, 111172. 10.1016/j.jms.2019.07.003. [DOI] [Google Scholar]
  13. Han S.; Sun G.; Zheng X.; Song Y.; Dawes R.; Xie D.; Zhang J.; Guo H. Rotational modulation of Ã2A″-state photodissociation of HCO via Renner–Teller nonadiabatic transitions. J. Phys. Chem. Lett. 2021, 12, 6582–6588. 10.1021/acs.jpclett.1c01932. [DOI] [PubMed] [Google Scholar]; a Gamallo P.; González M.; Petrongolo C. Quantum Dynamics of Nonadiabatic Renner–Teller Effects in Atom + Diatom Collisions. J. Phys. Chem. A 2021, 125, 6637–6652. 10.1021/acs.jpca.1c04654. [DOI] [PubMed] [Google Scholar]
  14. Herzberg G.Molecular spectra and molecular structure. Electronic spectra and electronic structure of polyatomic molecules; Van Nostrand Reinhold, 1966; Vol. 3. [Google Scholar]
  15. Lee T. J.; Fox D. J.; Schaefer H. F.; Pitzer R. M. Analytic second derivatives for Renner–Teller potential energy surfaces. Examples of the five distinct cases. J. Chem. Phys. 1984, 81, 356–361. 10.1063/1.447313. [DOI] [Google Scholar]
  16. Renner R. Zur Theorie der Wechselwirkung zwischen Elektronen- und Kernbewegung bei dreiatomigen, stabförmigen Molekülen. Z. Phys. 1934, 92, 172–193. 10.1007/BF01350054. [DOI] [Google Scholar]
  17. Sari L.; Gonzales J. M.; Yamaguchi Y.; Schaefer H. F. The X 2Π and à 2Σ+ electronic states of the HCSi radical: Characterization of the Renner–Teller effect in the ground state. J. Chem. Phys. 2001, 114, 4472–4478. 10.1063/1.1345512. [DOI] [Google Scholar]
  18. Wheeler S. E.; Simmonett A. C.; Schaefer H. F. Renner–Teller Bending Frequencies of the à 2Π State of OCS+. J. Phys. Chem. A 2007, 111, 4551–4555. 10.1021/jp0712046. [DOI] [PubMed] [Google Scholar]
  19. Finney B.; Mitrushchenkov A. O.; Francisco J. S.; Peterson K. A. Ab initio ro-vibronic spectroscopy of the Π2 PCS radical and Σ+ 1PCS anion. J. Chem. Phys. 2016, 145, 224303. 10.1063/1.4971183. [DOI] [PubMed] [Google Scholar]; a Gharaibeh M. A.; Clouthier D. J. A laser-induced fluorescence study of the jet-cooled nitrous oxide cation (N2O+). J. Chem. Phys. 2012, 136, 044318. 10.1063/1.3679744. [DOI] [PubMed] [Google Scholar]; b Pernpointner M.; Salopiata F. A four-component quadratic vibronic coupling approach to the Renner–Teller effect in linear triatomic molecules. The 2Pi1/2 2Pi3/2 manifold of BrCN+and ClCN+. J. Phys. B: At., Mol. Opt. Phys. 2013, 46, 125101. 10.1088/0953-4075/46/12/125101. [DOI] [Google Scholar]; c Tarroni R.; Clouthier D. J. Giant Renner–Teller vibronic coupling in the BF2 radical: An ab initio study of the X A21 and A Π2 electronic states. J. Chem. Phys. 2010, 133, 064304. 10.1063/1.3477765. [DOI] [PubMed] [Google Scholar]; d Wei J.; Grimminger R. A.; Sunahori F. X.; Clouthier D. J. Electronic spectroscopy of the jet-cooled arsenic dicarbide (C2As) free radical. J. Chem. Phys. 2008, 129, 134307. 10.1063/1.2988311. [DOI] [PubMed] [Google Scholar]
  20. Hauser A. W.; Pototschnig J. V. Vibronic Coupling in Spherically Encapsulated, Diatomic Molecules: Prediction of a Renner–Teller-like Effect for Endofullerenes. J. Phys. Chem. A 2022, 126, 1674–1680. 10.1021/acs.jpca.1c10970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Tarbutt M. R. Laser cooling of molecules. Contemp. Phys. 2018, 59, 356–376. 10.1080/00107514.2018.1576338. [DOI] [Google Scholar]
  22. Lu Q.; Peterson K. A. Coupled cluster spectroscopic properties of the coinage metal nitrosyls, M–NO (M = Cu, Ag, Au). Theor. Chem. Acc. 2020, 139, 81. 10.1007/s00214-020-02597-w. [DOI] [Google Scholar]
  23. Bilal; Pant M.; Zaheer H.; Garcia-Hernandez L.; Abraham A. Differential Evolution: A review of more than two decades of research. Eng. Appl. Artif. Intell. 2020, 90, 103479. 10.1016/j.engappai.2020.103479. [DOI] [Google Scholar]
  24. Knowles P. J.; Werner H.-J. Internally contracted multiconfiguration-reference configuration interaction calculations for excited states. Theor. Chim. Acta 1992, 84, 95–103. 10.1007/BF01117405. [DOI] [Google Scholar]; a Werner H. J.; Knowles P. J. An efficient internally contracted multiconfiguration–reference configuration interaction method. J. Chem. Phys. 1988, 89, 5803–5814. 10.1063/1.455556. [DOI] [Google Scholar]
  25. Dunning T. H. Gaussian basis sets for use in correlated molecular calculations. I. The atoms boron through neon and hydrogen. J. Chem. Phys. 1989, 90, 1007–1023. 10.1063/1.456153. [DOI] [Google Scholar]
  26. Johnson R. D., IIINIST Computational Chemistry Comparison and Benchmark Database, 2022.
  27. Kashinski D. O.; Radziewicz T. J.; Suarez M. G.; Stephens C. C.; Byrd E. F. C. Density functional theory calculation of the Renner–Teller effect in NCO: Preliminary assessment of exact exchange energy on the accuracy of the Renner coefficient. Int. J. Quantum Chem. 2021, 121, e26804 10.1002/qua.26804. [DOI] [Google Scholar]
  28. Liu L.; Yang C.-L.; Sun Z.-P.; Wang M.-S.; Ma X.-G. Direct laser cooling schemes for the triatomic SOH and SeOH molecules based on ab initio electronic properties. Phys. Chem. Chem. Phys. 2021, 23, 2392–2397. 10.1039/D0CP04963H. [DOI] [PubMed] [Google Scholar]; a Augenbraun B. L.; Doyle J. M.; Zelevinsky T.; Kozyryev I. Molecular asymmetry and optical cycling: laser cooling asymmetric top molecules. Phys. Rev. X 2020, 10, 031022. 10.1103/PhysRevX.10.031022. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao2c05494_si_001.pdf (444.7KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES