Accuracy and Completeness of molecular linelists H2O, NH3 and CO2 CO, H3+, HCN Oleg L. Polyansky and Jonathan Tennyson 1Department of Physics and Astronomy, University College London, London WC1E 6BT, UK 2 Institute of Applied Physics, Russian Academy of Sciences, Uljanov Street 46, Nizhnii Novgorod, Russia 603950

Molecular Hydrogen H2 10-4 cM-1 v=0→1 J=0→1 α2 m 36 118.797 746 1(5) 4 161.164 070 3(1) 118.485 260 46(3) α4 m −0.531 8(3)a 0.023 41(1)c 0.002 580(1) α5 m −0.194 8(2)b −0.021 29(2)c −0.001 022(1) α6 m −0.002 065(6) −0.000 192 3(6) −0.000 008 9(1) α7 m 0.000 118(59) 0.000 012 0(60) 0.000 000 6(3) Er2p −0.000 031 −0.000 003 2 −0.000 000 2 Theory 36 118.069 1(6) 4 161.166 01(4) 118.486 810(4) [23–25] 36 118.069 62(37) 4 161.166 32(18) 118.486 84(10) M.Puchalski, J. Komasa PRL,117, 263002, (2016)

H2 Comparison Theory/Experiment (Theory: Pachucki, Komasa, Jeziorski et al.) QED

Realistic accuracy Frequencies and energies ~10 000cm-1 Up to dissociation Ab initio 0.1 cm-1. ~ 1 cm-1 Fitted 0.01 cm-1 ~ 0.1 cm-1 Intensities 0.1% - 1 %

5 FACTORS (6subfactors of BO) Born-Oppenheimer 1. a. MRCI , MOLPRO ~Full CI b. Number of points 200 -2000 3-5 cm-1 c. All electrons, CV 10 cm-1 d. Highest basis set aug-cc-pCV6z 5 cm-1 e. CBS 5 cm-1 f. Optimized (higher) CAS 10 cm-1 Corrections 2. Adiabatic (DBOC) 3 cm-1 3.Relativistic (MVD1, Gaunt (Breit), ) 10 cm-1 4. QED 0.5 cm-1 5. Vibration-rotation non-adiabatic 5 cm-1

Chemical Physics Letters Contents lists available at SciVerse ScienceDirect Chemical Physics Letters journal h omepage: www.else vier.com/locate/cplett Accurate bond dissociation energy of water determined by triple-resonance vibrational spectroscopy and ab initio calculations Oleg V. Boyarkin a, Maxim A. Koshelev a,b, Oleg Aseev a, Pavel Maksyutenko a, Thomas R. Rizzo a, Nikolay F. Zobov b, Lorenzo Lodi c, Jonathan Tennyson c,⇑, Oleg L. Polyansky b,c a Laboratoire de Chimie Physique Moléculaire, École Polytechnique Fédérale de Lausanne, CH-1015 Lausanne, Switzerland b Institute of Applied Physics, Russian Academy of Sciences, Uljanov Street 46, Nizhnii Novgorod 603950, Russia c Department of Physics and Astronomy, University College London, London WC1E 6BT, United Kingdom a r t i c l e i n f o a b s t r a c t Available online 21 March 2013 Article history: Triple-resonance vibrational spectroscopy is used to determine the lowest dissociation energy, D0, for the water isotopologue HD16O as 41 239.7 ± 0.2 cm 1 and to improve D0 for H2 O to 41 145.92 ± 0.12 cm . Ab initio calculations including systematic basis set and electron correlation convergence studies, relativ- istic and Lamb shift effects as well as corrections beyond the Born–Oppenheimer approximation, agree with the measured values to 1 and 2 cm 1 respectively. The improved treatment of high-order correlation terms is key to this high theoretical accuracy. Predicted values for D0 for the other five major water iso- topologues are expected to be correct within 1 cm 1. 2013 Elsevier B.V. All rights reserved. 16 1 1. Introduction 2. Experimental setup The OH bond dissociation energy of water isotopologues is used in a myriad of applications, ranging from astronomy and biology to combustion and atmospheric chemistry. While extreme accuracy in bond dissociation energies is not typically needed in practical applications, with only three nuclei and 10 electrons, water serves as a benchmark molecule for testing quantum–mechanical calcula- tions. Experimentally, it is a challenge to measure the dissociation energy of even a small polyatomic molecule accurately, as it usu- ally requires access to highly lying molecular levels around D0 with precisely known energy in a collision-free environment. Theoreti- cally, accurate computations must be performed using extremely sophisticated models with large basis sets and accounting for many effects and interactions that are routinely ignored. So far, the only highly accurate, sub-cm 1 calculations of D0 val- idated by accurate experimental data have been for diatomic hydrogen isotopologues [1–3] and their cations [4]. For the sim- plest polyatomic molecule, Hþ ; D0 has been calculated to a reported Several spectroscopic techniques have been employed to im- prove the accuracy of D0 for H2 O [7–9]. These experiments con- verged to an accurate value of D0 = 41 145.94 ± 0.15 cm 1, measured by direct observation of the H–OH dissociation contin- uum by detecting the appearance of OH as a function of the total excitation energy of the water molecule [9]. Here we report a com- plementary measurement of D0(H2O) using triple-resonance vibra- tional overtone excitation, but by detecting H-atom fragments. This removes any uncertainty that the detected ground state OH may result from hypothetically ultrafast collisional relaxation of the OH fragments, initially appearing in the f component of the ground state K-doublet, which is slightly higher in energy than the ground state. If this were the case, the measured D0 would be overestimated by the 0.06 cm 1 difference [10] between the two K-doublet components. We then employ state-selective over- tone excitation to access the dissociation continuum of H16OD on its electronic ground surface to measure D0 for the OH bond in HOD. The details of our approach and of the experimental apparatus have been reported elsewhere [9,11–13]; the excitation scheme is illustrated in Figure 1. We use three laser pulses (P1–P3) to ac- cess high-lying terminal vibrational levels through sequential exci- tation of three vibrational overtone transitions. For molecules in terminal levels lying below the dissociation threshold, a second photon of the third excitation laser (P3a) is required to promote them to the repulsive electronic surface. A subsequent laser pulse (P4) detects the appearing photofragments via laser-induced fluo- rescence. Manipulations with relative polarizations of laser beams 16 accuracy of 1 cm 1 [5] but experimentally confirmed to only about 3 60 cm 1 [6]. The simplicity of these systems means that they do not provide validated theoretical approaches for predicting D0 of more complex species. Here we report sub-cm 1 accuracy mea- surements of D0 for HD16O, an improved D0 for H16 O, and ab initio calculations of D0 with an unprecedented accuracy of a few cm 1 for seven isotopologues of this molecule. 2 ⇑ Corresponding author. E-mail addresses: oleg.boiarkin@epfl.ch (O.V. Boyarkin), j.tennyson@ucl.ac.uk (J. Tennyson). 0009-2614/$ - see front matter 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cplett.2013.03.007

0.1 cm-1 For energies Up to 15 000 cm-1

Realistic accuracy Frequencies and energies ~10 000cm-1 Up to dissociation Ab initio 0.1 cm-1. ~ 1 cm-1 Fitted 0.01 cm-1 ~ 0.1 cm-1 For water this accuracy is reached and we use PES and DMS to calculate LL. It brings us to the subject of completeness Intensities 0.1% - 1 %

Two types of completeness 1. All the molecules in the gas phase. In the first approximation – all molecules in HITRAN database. Realistically – H3+, H2O, CO, CO2, O3, NH3, CH4, HCN, N2O, C2H2 2. All the lines which include the levels up to dissociation

Transmission of main candidate molecules (H2O, CO2, CO, CH4,NH3)

UV spectrum of water using different Linelists (J. Lampel, D,Poehler,O UV spectrum of water using different Linelists (J.Lampel, D,Poehler,O.L.Polyansky et al.,Atmospheric Chemistry and Physics,v.17,1271,(2017) )

Upper bounds of the known experimental data H2O – 42 000 cm-1 NH3 - 18 000 cm-1 HCN – 23 000 cm-1 H3+ - 17 000 cm-1 the way to deliver completeness is to use the existing experimental data (or assign the new ones ) to fit PES or calibrate ab initio PES

Ab initio water spectrum at dissociation Csaszar AG,Matyus E,Szidarovsky T, Lodi L, Zobov NF, Shirin SV,Polyansky OL, Tennyson J, JQSRT, ,v.111,1043 (2010) Ab initio water spectrum at dissociation

We fitted this data to produce PES H216O and calculated LL POKAZATEL (PolyanskyO,KyuberisA,Zobov,TEnnyson,Lodi) which includes all bound Levels of water J=72, max energy -40 000 cm-1 More complete LL could be only If one includes resonances above dissociation

Global LL for H217,18O, up to 30 000 cm-1 Analogue of BT2 for H216O J Experimental Calculated Exp-Calc 27476,33 27476,24 0.09 1 27497,03 27496,92 0.11 27510,64 27510,31 0,33 27517,09 27517,44 -0.35 2 27537,12 27536,96 0.16 27546,82 27546,45 0.37 27550,03 27508,83 27509,55 27509,19 0.36 27545,66 27545,28 0.38 Global LL for H217,18O, up to 30 000 cm-1 Analogue of BT2 for H216O Is done – Polyansky OL et al.,MNRAS, v.466, 1363 (2017) Experimental data used for the fitted PES –up to18 000 cm-1 Prediction of data from Makarov D. et al,2015

At present, hot line lists are only published for H2 16 O and HD16 O. MNRAS 466, 1363–1371 (2017) Advance Access publication 2016 December 2 doi:10.1093/mnras/stw3125 ExoMol molecular line lists XIX: high-accuracy computed hot line lists for H218O and H217O Oleg L. Polyansky,1,2 Aleksandra A. Kyuberis,2 Lorenzo Lodi,1 Jonathan Tennyson,1‹ Sergei N. Yurchenko,1 Roman I. Ovsyannikov2 and Nikolai F. Zobov2 1 Department of Physics and Astronomy, University College London, London WC1E 6BT, UK 2 Institute of Applied Physics, Russian Academy of Sciences, Ulyanov Street 46, Nizhny Novgorod 603950, Russia Accepted 2016 November 29. Received 2016 November 23; in original form 2016 October 20 ABSTRACT Hot line lists for two isotopologues of water, H2 18 O and H2 17 O, are presented. The calculations employ newly constructed potential energy surfaces (PES), which take advantage of a novel method for using the large set of experimental energy levels for H2 16 O to give high-quality predictions for H2 18 O and H2 17 O. This procedure greatly extends the energy range for which a PES can be accurately determined, allowing an accurate prediction of higher lying energy levels than are currently known from direct laboratory measurements. This PES is combined with a high-accuracy, ab initio dipole moment surface of water in the computation of all energy levels, transition frequencies and associated Einstein A coefficients for states with rotational excitation up to J = 50 and energies up to 30 000 cm−1 . The resulting HotWat78 line lists complement the well-used BT2 H2 16 O line list. Full line lists are made available online as Supporting Information and at www.exomol.com. Key words: molecular data – opacity – astronomical data bases: miscellaneous – planets and satellites: atmospheres – brown dwarfs – stars: low-mass. 1 I NTR O DUCTION Water spectra can be observed from many different regimes in the Universe, several of which are discussed further below. The spectrum of water, particularly at elevated temperatures, is rich and complex. A few years ago Barber et al. (2006) presented a comprehensive line list, known as BT2, which used well-established theoretical procedures to compute all the transitions of H2 16 O of importance in objects with temperatures up to 3000 K. BT2 contains about 500 million lines. A similar line list for HD16 O, known as VTT, was subsequently computed by Voronin et al. (2010). The BT2 line list has been extensively used. It forms the basis of the most recent release of the HITEMP high-temperature spectro- scopic data base (Rothman et al. 2010) and for the BT-Settl model (Allard 2014) for stellar and substellar atmospheres covering the range from solar-mass stars to the latest type T and Y dwarfs. BT2 has been used to detect and analyse water spectra in objects as diverse as the Nova-like object V838 Mon (Banerjee et al. 2005), atmospheres of brown dwarfs (Rice et al. 2010) and M subdwarfs (Rajpurohit et al. 2014), and extensively for exoplanets (Tinetti et al. 2007; Birkby et al. 2013). Within the Solar system BT2 has been used to show an imbalance between nuclear spin and rotational temperatures in cometary comae (Dello Russo et al. 2004, 2005) and assign a new set of, as yet unexplained, high-energy water emissions in comets (Barber et al. 2009), as well as to model water spectra in the deep atmosphere of Venus (Bailey 2009). Although BT2 was developed for astrophysical use, it has been applied to a variety of other problems including the calculation of the refractive index of humid air in the infrared (Mathar 2007), high-speed thermometry and tomographic imaging in gas engines and burners (Kranendonk et al. 2007; Rein & Sanders 2010), as the basis for an improved theory of line-broadening (Bykov et al. 2008), and to validate the data used in models of the Earth’s atmosphere and in particular simulating the contribution of weak water transitions to the so-called water continuum (Chesnokova et al. 2009). There are several water line lists published in the literature (Partridge & Schwenke 1997; Viti, Tennyson & Polyansky 1997; Mikhailenko, Babikov & Golovko 2005; Barber et al. 2006). Two line lists have also been computed specifically for the isotopo- logues: Shirin et al. (2008) created the 3mol room-temperature line lists for H2 16 O, H2 17 O and H2 18 O based on the potential energy surfaces (PES) of Shirin et al. (2006); Tashkun created a number of line lists based on the work of Partridge & Schwenke (1997) (see Mikhailenko et al. 2005). These are considered further below. At present, hot line lists are only published for H2 16 O and HD16 O. However, isotopically substituted water containing 18 O or 17 O pro- vides important markers for a variety of astronomical problems (Nittler & Gaidos 2012). For example, Matsuura et al. (2014) re- cently detected H2 18 O in the emission-line spectrum of the luminous E-mail: j.tennyson@ucl.ac.uk C 2016 The Authors Published by Oxford University Press on behalf of the Royal Astronomical Society

The LL is ready up to J=50, 35 000 cm-1 HDO fitted PES The LL is ready up to J=50, 35 000 cm-1 v1 v2 v3 Obs. | Calc. | Obs.-Calc. 0 1 0 1403.4837 1403.4821 -0.0016 1 0 0 2723.6798 2723.6593 -0.0205 0 2 0 2782.0111 2781.9899 -0.0212 0 0 1 3707.4667 3707.4490 -0.0178 0 1 0 4099.9559 4099.9528 -0.0031 0 3 0 4145.4732 4145.4377 -0.0355 0 1 1 5089.5399 5089.5142 -0.0258 2 0 0 5363.8245 5363.8157 -0.0088 0 4 0 5420.0418 5420.0111 -0.0307 1 0 1 6415.4616 6415.4459 -0.0157 0 2 1 6451.9003 6451.8892 -0.0111 0 5 0 6690.4132 6690.4125 -0.0007 2 1 0 6746.9100 6746.8791 -0.0309 0 0 2 7250.5192 7250.5251 0.0059 0 3 1 7754.6055 7754.6069 0.0014 1 1 1 7808.7586 7808.7500 -0.0086 0 6 0 7914.3170 7914.2796 -0.0374 3 0 0 7918.1719 7918.1805 0.0086 0 1 2 8611.1020 8611.0890 -0.0130 2 0 1 9047.0685 9047.0671 -0.0014 1 2 1 9155.8178 9155.8364 0.0186 3 1 0 9293.0016 9293.0086 0.0070 1 5 0 9381.7861 9381.7816 -0.0045 2 3 0 9487.9153 9487.9278 0.0125 0 2 2 9934.7890 9934.7767 -0.0123 1 0 2 9967.0230 9967.0391 0.0161 ...................................................................... v1v2v3 obs calc obs-calc 1 4 2 15170.9510 15170.8933 -0.0577 0 2 4 16456.1903 16456.1407 -0.0496 1 0 4 16539.0400 16538.9848 -0.0552 0 0 5 16920.0240 16920.0104 -0.0136 0 1 5 18208.4465 18208.4434 -0.0031 0 0 6 19836.8828 19836.9359 0.0531 9 0 0 21464.4921 21464.3850 -0.1071 1 0 6 22454.4780 22454.6761 0.1981 0 0 7 22625.5285 22625.5501 0.0216 10 0 0 23431.7330 23431.7524 0.0194 1 0 7 25140.8500 25140.7857 -0.0643 11 0 0 25316.9660 25316.9669 0.0009 0 0 8 25332.7160 25332.6790 -0.0370 0 0 9 27970.1980 27970.2060 0.0080 0 0 0 30543.0170 30543.0179 0.0009 0 0 0 33007.8190 33007.8201 0.0011 0 0 0 33029.2240 33029.2236 -0.0004 0 0 0 33078.1860 33078.1852 -0.0008 0 0 0 35533.7440 35533.7434 -0.0006 0 0 0 35573.3740 35573.3746 0.0006 0 0 0 35612.2470 35612.2470 0.0000 0 0 0 35631.2200 35631.2198 -0.0002 0 0 0 37959.5590 37961.7785 2.2195

H+ has no known electronic spectrum and its ‘forbidden’ pure MNRAS 468, 1717–1725 (2017) Advance Access publication 2017 February 28 doi:10.1093/mnras/stx502 ExoMol molecular line lists – XX. A comprehensive line list for H+ 3 Irina I. Mizus,1 Alexander Alijah,2 Nikolai F. Zobov,1 Lorenzo Lodi,3 Aleksandra A. Kyuberis,1 Sergei N. Yurchenko,3 Jonathan Tennyson3‹ and Oleg L. Polyansky1,3 1 Institute of Applied Physics, Russian Academy of Sciences, Ulyanov Street 46, Nizhny Novgorod 603950, Russia 2 Groupe de Spectrome´trie Mole´culaire et Atmosphe´rique, GSMA, UMR CNRS F-7331, Universite´ de Reims Champagne-Ardenne, France 3 Department of Physics and Astronomy, University College London, London WC1E 6BT, UK Accepted 2017 February 24. Received 2017 February 23; in original form 2016 November 12 ABSTRACT H+ 3 is a ubiquitous and important astronomical species whose spectrum has been observed in the interstellar medium, planets and tentatively in the remnants of supernova SN1897a. Its role as a cooler is important for gas giant planets and exoplanets, and possibly the early Universe. All this makes the spectral properties, cooling function and partition function of H+ 3 key parameters for astronomical models and analysis. A new high-accuracy, very extensive line list for H+ called MiZATeP was computed as part of the ExoMol project alongside a 3 temperature-dependent cooling function and partition function as well as lifetimes for excited states. These data are made available in electronic form as supplementary data to this article and at www.exomol.com. Key words: molecular data – opacity – astronomical data bases: miscellaneous – planets and satellites: atmospheres. 1 I NTR O DUCTION The atomic composition of the Universe is dominated by hydro- gen which means that H+ , as the stable ionic form of molecular much of interstellar gas-phase chemistry (Watson 1973; Herbst & Klemperer 1973; Tennyson 1995; Oka 2013; Millar 2015). It pro- vides a unique means to monitor cosmic ray ionization rates in the interstellar medium (McCall et al. 2003; Indriolo & McCall 2012). Cooling by H+ is thought to be important for the stability of at- 3 hydrogen, is thought to be important in many diverse astronomical environments where it plays a variety of roles (McCall & Oka 2000; Oka 2006). So far H+ has been observed in the atmospheres of the 3 mospheres of giant extrasolar planets orbiting close to their stars (Koskinen, Aylward & Miller 2007; Khodachenko et al. 2015) and possibly in primordial gas (Glover & Savin 2006). Cooling is one of a number of functions performed by H+ in the ionospheres of 3 Solar system gas giants (Drossart et al. 1989; Geballe, Jagod & Oka 1993; Trafton et al. 1993; Miller, Lam & Tennyson 1994), dense molecular clouds (Geballe & Oka 1996; McCall et al. 1999), the diffuse interstellar medium (McCall et al. 1998, 2002) and ex- ternal galaxies (Geballe et al. 2006; Geballe, Mason & Oka 2015), and more tentatively in the remnants of supernova SN1897a (Miller et al. 1992). Observations of H+ provide a powerful tool for study- Second molecule to move towards completeness is H3+ LL up to 30 000 cm-1 3 Solar system gas giants (Miller et al. 2000) where observations of H+ have proved important for monitoring the ionospheric activity 3 (Miller et al. 1995, 2000; Lam et al. 1997a,b; Stallard et al. 2008a,b) and have, for example, been used to determine wind speeds (Rego et al. 1999). Elsewhere H+ is probably a key component of cool stars 3 ing the Galactic Centre (Goto et al. 2002, 2008; Oka et al. 2005), where it has been shown that lifetime effects in H+ lead to pop- 3 with low metallicity; for example it has been shown to play a cru- cial role in the chemical evolution of cool white dwarfs (Bergeron, Ruiz & Leggett 1997). H+ has no known electronic spectrum and its ‘forbidden’ pure 3 ulating long-lived meta-stable states. A similar mechanism is also important in laboratory studies of H+ (Kreckel et al. 2002, 2004). 3 So far, searches for H+ in the atmosphere of hot Jupiter exoplan- 3 rotational spectrum, although possibly observable (Pan & Oka 1986; Miller & Tennyson 1988b), is yet to be detected. This leaves its vibration–rotation spectrum as the means by which all spectroscopic studies are made. The laboratory spectroscopic data for H+ were 3 ets have proved negative (Shkolnik, Gaidos & Moskovitz 2006), while the claimed detection of H+ emission in a protoplanetary disc 3 (Brittain & Rettig 2002) was negated by Goto et al. (2005). 3 , which is rapidly formed from the collision of molecular hy- drogen and its ion (H+ ), has long been thought to be the initiator of 3 recently collected and reviewed by Furtenbacher et al. (2013) as part of their MARVEL, Measured Active Rotational–Vibrational Energy Levels (Furtenbacher, Csa´sza´r & Tennyson 2007; Furtenbacher & Csa´sza´r 2012), study of the system. This work replaced an earlier compilation and evaluation of the laboratory data by Lindsay & H+ 2 Email: j.tennyson@ucl.ac.uk C 2017 The Authors Published by Oxford University Press on behalf of the Royal Astronomical Society

The highest H3+ line. -3. 0 and +8. 5 cm-1 –previous predictions Phys The highest H3+ line. -3.0 and +8.5 cm-1 –previous predictions Phys.Rev.Lett.,v.108,023002 (2012) Pavanello, Adamovicz, Alijah, Zobov, Mizus and Polyansky et al.

H3+ Linelist up to 30 000 cm-1 Previous (NMT) was up to 15 000 cm-1 Figure 3. Comparison of MiZATeP line list with the NMT one (Neale et al. 1996) for the temperature value 2500 K. ExoMol molecular line lists – XX. A comprehensive line list for H3+ Mon Not R Astron Soc. 2017;468(2):1717-1725. doi:10.1093/mnras/stx502 Mon Not R Astron Soc | © 2017 The Authors Published by Oxford University Press on behalf of the Royal Astronomical Society 20

Next step towards completeness – 40 000 cm-1 and interpretation of Carrington –Kennedy - work in progress

Completeness: Absorption of ammonia (T=300 K) more than 70 years of research BYTe LL Less than 30,000 NH3 lines are known experimentally: our list contains 1.1 billion lines, or about 40,000 times as many! They represent all the allowed transitions between 1.2 million upper and lower ro-vibrational states, whose individual quantum numbers are detailed in the list. For comparison, it is worth noting that our earlier T=300 K NH3 line list comprises only 3.25 million transitions between 184,400 states. It has an upper energy cut-off of 12,000 cm-1 and a maximum rotational quantum number J=20. Less than 30,000 NH3 lines known experimentally: 22

Absorption spectra of ammonia near 1 μm Journal of Quantitative Spectroscopy & Radiative Transfer ∎ (∎∎∎∎) ∎∎∎–∎∎∎ Contents lists available at ScienceDirect Journal of Quantitative Spectroscopy & Radiative Transfer journal homepage: www.elsevier.com/locate/jqsrt Absorption spectra of ammonia near 1 μm Emma J. Barton a, Oleg L. Polyansky a, Sergei.N. Yurchenko a, Jonathan Tennyson a,n, S. Civiš b, M. Ferus b, R. Hargreaves c, R.I. Ovsyannikov d, A.A. Kyuberis d, N.F. Zobov d, S. Béguier e,f, A. Campargue e,f a Department of Physics and Astronomy, University College London, London WC1E 6BT, UK b Academy of Science Czech Republic, J Heyrovsky Inst Phys Chem, Dolejskova 3, CZ-18223 Prague 8, Czechia Republic c Department of Chemistry and Biochemistry, Old Dominion University, Norfolk, VA 23529 USA d Institute of Applied Physics, Russian Academy of Science, Uljanov Street 46, Nizny Novgorod 603950, Russia e Université Grenoble Alpes, LIPhy, F-38000 Grenoble, France f CNRS, LIPhy, F-38000 Grenoble, France a r t i c l e i n f o a b s t r a c t Article history: Received 2 January 2017 Received in revised form 30 March 2017 Accepted 30 March 2017 An ammonia absorption spectrum recorded at room temperature in the region 8800–10,400 cm 1 is analysed using a variational line list, BYTe, and ground state energies determined using the MARVEL procedure. BYTe is used as a starting point to initialise assignments by combination differences and the method of branches. Assignments are presented for the region 9400–9850 cm 1. 642 lines are assigned to 6 previously unobserved vibrational bands, (2v + 2v2)± , (2v + v1)± and (v + v1 + 2v2)±, leading to 428 new energy levels with 208 confirmed by combination differences. A recently calculated purely ab initio NH3 PES is also used to calculate rovibrational energy levels. Comparison with assigned levels shows better agreement between observed and calculated levels than for BYTe for higher vibrational bands. & 2017 Elsevier Ltd. All rights reserved. 1 4 1 3 1 3 4 Keywords: Room temperature Ammonia Absorption intensities FTIR spectroscopy Experimental energies BYTe Line assignments 1. Introduction variational line list of Yurchenko, Barber and Tennyson [15], en- ergy levels from the MARVEL study, and the method of branches [16] to assign 2474 lines in the 7400–8600 cm 1region; this is the first time any assignments had been made for ammonia spectra in this region. The success of this work, and the availability of unassigned, shorter-wavelength ammonia spectra in the Kitt Peak archive and from elsewhere (see below) motivated us to attempt to extend the analysis techniques employed by Barton et al. [12] to higher wa- venumbers. The results of this analysis are reported here. We note that the 2012 release of HITRAN contained no data on NH3 above 7000 cm 1. One reason for extending the range of assigned NH3 spectra is to help the construction of accurate potential energy surfaces (PES). Quite a number of PES are available for the ground states of NH3. The majority of these are the products of ab initio electronic structure calculations [17–23], although surfaces that use experi- mental data to improve their accuracy are also available [24–26]. Recently, Polyansky et al. [23] computed an ab initio surface with which they were able to make, for the first time, vibrational as- signments to the optical spectrum of ammonia recorded by Coy and Lehmann [27,28]. We consider results obtained with this PES further below. Ammonia is an atmospheric trace species which is frequently the by-product of human activity [1]. Ammonia is present in a variety of astronomical environments, including the interstellar medium, gas giant planets [2] and brown dwarfs [3]; indeed NH3 is thought to provide the signature of coolest brown dwarfs known as Y-dwarfs [4]. NH3 is also used in a number of industrial pro- cesses, such as the reduction of NOx emissions in smoke stacks [5] and the manufacture of hydrogen cyanide by the Andrussow process [6]. This has motivated a large number of experimental studies of ammonia spectra; those reported up to late 2014 are reviewed in the MARVEL (measured active rotation-vibration en- ergy levels) study of NH3 performed by Al-Derzi et al. [7], which is discussed further below. A number of new ammonia spectra have been reported in the last two years [8–14]. Of particular relevance to this work is the analysis by Barton et al. [12] of a near-infrared Fourier transform spectrum which was originally recorded by Dr Catherine de Bergh in 1980 at Kitt Peak. Barton et al. used a combination of the BYTe n Corresponding author. http://dx.doi.org/10.1016/j.jqsrt.2017.03.042 0022-4073/& 2017 Elsevier Ltd. All rights reserved. Please cite this article as: Barton EJ, et al. Absorption spectra of ammonia near 1 μm. J Quant Spectrosc Radiat Transfer (2017), http://dx. doi.org/10.1016/j.jqsrt.2017.03.042i

EJ Barton,OL Polyansky,SN Yurchenko, J Tennyson et al EJ Barton,OL Polyansky,SN Yurchenko, J Tennyson et al. JQSRT,in press (2017)

Analysis of optical spectrum of ammonia. Nikolay F. Zobov1, Peter F. Bernath2, R. Hargreaves2, Jonathan Tennyson3, Sergei N. Yurchenko3, Roman I. Ovsyannikov1, Philip A. Coles3, Oleg L. Polyansky1,3 1Institute of Applied Physics, Russian Academy of Sciences, 46 Uljanov Street, 603950 Nizhny Novgorod, Russia 2Department of Chemistry, Old Dominion University, Norfolk, VA 23529 USA 3Department of Physics and Astronomy, University College London, London WC1E 6BT, UK The analysis of regions of ammonia optical spectra with the strongest lines around 15000 cm-1 and 18000 cm-1 have been performed using both combination differences method and new variational linelists. Ammonia absorption spectra were recorded at room temperature in the 15200 – 15700 cm-1 and 17950 – 18250 cm-1 regions. The strongest lines belong to purely stretching excitations ν1 and ν3. Some of these lines had been observed 30 years ago in experiments of Lehmann and Coy with microwave detected microwave optical double resonance techni References: [1] S.L.Coy, K.K.Lehmann, Spectr. Acta, 45A (1989) 47. [2] O.L. Polyansky, R.I. Ovsyannikov, A.A. Kyuberis, L. Lodi, J. Tennyson, A. Yachmenev, S. N. Yurchenko, N. F. Zobov, J. Mol. Spectrosc. 327, (2016) 21. [3] S.N. Yurchenko, W Thiel, P. Jensen, J Mol Spectrosc. 245 (2007) 126.  

Spectrum of NH3 at 15 000 cm-1 P. Bernath and R. Hargreaves

Spectrum of NH3 at 18 000 cm-1 P. Bernath and R. Hargreaves

Observed energy levels of NH3 J K calc obs obs-calc delta cd 5 0 0 0 0 0 1 1 15478.33 15466.5198 -11.82 0.0015 2 0 15518.64 15506.0918 -12.55 0.0004 2 2 15505.97 15494.0029 -11.97 0.0006 3 0 15573.56 15561.9812 -11.58 -0.0012 3 2 15560.64 15548.7502 -11.90 -0.0015 3 3 15544.88 15533.9636 -10.92 0.0000 4 4 15597.58 15585.6611 -11.92 0.0006 5 3 15709.84 15414.9747 -11.92 0.0059 5 0 0 0 0 1 1 0 15480.7408 15466.9949 -13.75 0.0053 1 1 15477.3769 15463.4543 -13.92 0.0052 2 0 15517.0588 15502.4293 -14.63 -0.0003 3 0 15572.6798 15560.0324 -12.65 0.0000 3 3 15544.4086 15530.4653 -13.94 0.0013 4 0 15646.1915 15633.7890 -12.40 0.0006 4 1 15639.2451 15625.0227 -14.22 -0.0022 4 2 15631.8889 15618.6802 -13.21 0.0010 5 0 15738.9750 15724.3296 -14.65 -0.0096 5 3 15708.2297 15694.0690 -14.16 0.0019 7 3 15943.1415 15928.5263 -14.62 -0.0046 J K calc obs obs-calc delta cd 6 0 0 0 0 0 2 1 18184.35 18158.5039 -25.85 0.0002 2 2 18187.98 18164.4116 -23.57 0.0011 3 0 18242.65 18216.6530 -26.00 0.0007 3 1 18240.38 18216.3554 -24.03 0.0102 3 2 18229.36 18203.8817 -25.49 0.0080 4 4 18266.42 18239.5275 -26.90 0.0065 5 4 18358.66 18336.4739 -22.19 0.0010 6 6 18407.70 18381.9761 -25.73 0.0045 6 0 0 0 0 1 2 0 18200.04 18174.5032 -25.54 0.0074 2 1 18179.29 18157.5306 -21.76 0.0024 3 2 18243.76 18216.7658 -27.00 0.0034 5 4 18352.29 18322.8273 -29.47 0.0020 6 4 18481.50 18450.9003 -30.61 0.0075

HIGH ACCURACY POTENTIAL ENERGY SURFACE, DIPOLE MOMENT SURFACE, ROVIBRATIONAL ENERGIES AND LINE LIST CALCULATIONS FOR14NH3 PHILLIP COLES,Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, United Kingdom; SERGEI N. YURCHENKO, Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, United Kingdom; OLEG POLYANSKY, Departmentof Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, United Kingdom; ALEKSANDRA KYUBERIS, ROMAN I. OVSYANNIKOV, NIKOLAY ZOBOV, Microwave Spectroscopy, Institute of Applied Physics, Nizhny Novgorod, Russia ; JONATHAN TENNYSON,Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, United Kingdom . We present a new spectroscopic potential energy surface (PES) for14NH3, produced by refining a high accuracy ab initio PES a to experimental energy levels taken predominantly from MARVELb. The PES reproduces 1722 matched J=0-8 experimental energies with a root-mean-square error of 0.035 cm-1 under 6000 cmand 0.059 under 7200 cm -1 Inconjunction with a new DMS calculated using multi reference configuration interaction (MRCI) and H=aug-cc-pVQZ, N=aug-cc-pWCVQZ basis sets, an infrared (IR) line list has been computed which is suitable for use up to 2000 K. The line list is used to assign experimental lines in the 7500 - 10,500 cm region and previously unassigned lines in HITRAN in the 6000-7000 cm

HCN and HNC. The (ℓ = 0) effective frequency analysis for HCN and HNC. Shown are experimental data points (blue), Dunham polynomial expansion predictions using only experimental data (green), and the assigned ab initio data points (red) (26) (see supplementary text for details). The fitted ETS parameters using Eq. 4 (blue) are compared with the ab initio barrier heights (red). A one-dimensional cut through the potential energy surface is shown as a red dashed line. The unusual shapes of the HNC potential and ωeff plot near 5000 cm–1 result from interaction with a low-lying excited diabatic electronic state (44). Joshua H. Baraban et al. Science 2015;350:1338-1342 Published by AAAS

Room-temperature HCN/HNC line-list, part I: potential energy surface. Vladimir Yu. Makhnev1, Aleksandra A. Kyuberis1, Nikolay F. Zobov1, Jonathan Tennyson2, Oleg L. Polyansky1,2,* 1Institute of Applied Physics, Russian Academy of Science, Uljanov Street 46, Nizny Novgorod, Russia, 603950 2Department of Physics and Astronomy, University College London, London, WC1E 6BT, United Kingdom *o.polyansky@ucl.ac.uk A new high accuracy potential energy surface for the electronic ground state of HCN/HNC system is constructed by refining our new ab initio PES [1]. Comparisons with the results from reference [2] are given. We reproduce all the experimental energy levels of HCN up to 7600 cm-1 with about 0.05 cm-1. For HNC molecule states up to 7200 cm-1 there is 0.71 cm-1 standard deviation from experimental ones. This fitted PES was based on ab initio results from reference [1] where 0.35 cm-1 accuracy for energies up to 7600 cm-1 for HCN were obtained [2] A. J. C. Varandas and S. P. J. Rodrigues, Potential Energy Surface for Ground-State HCN J. Phys. Chem. A, Vol. 110, No. 2, (2

HCN

THIS IS THE END OF COMPLETENESS STORY THIS IS THE END OF COMPLETENESS STORY. A FEW WORDS ON THE ACCURACY OF INTENSITY

Line intensities of 16O12C16O 1.6 µm band: 30013 – 00001 O.L. Polyansky, Bielska K, Ghysels M,L. Lodi, NF Zobov, JT Hodges ,JTennyson. PRL 114, 233001 (2015)

Polyansky et al., PRL,v. 114,243001,2015 CO2 exp 20012->00001 from Gianfrani et al vs. Wuebbler et al

Highly-accurate intensity factors of pure CO2 lines near 2 µm T. A. Odintsova,1, a) E. Fasci,1 L. Moretti,1 E. Zack,2 O.L. Polyansky,2, a) J. Tennyson,2 L. Gianfrani,1, b) and A. Castrillo1 1) Dipartimento di Matematica e Fisica, Universit degli studi della Campania ”Luigi Vanvitelli”, Viale Lincoln 5, 81100 Caserta, Italia. 2) Department of Physics and Astronomy, University College London, London WC1E 6BT, United Kingdom (Dated: June 9, 2017) Line intensities for carbon dioxide are measured with a novel spectroscopic approach, assisted by an optical frequency comb synthesizer for frequency calibration purposes. The main feature of the spectrometer consists in the exploitation of optical feedback from a V-shaped high-finesse optical resonator to effectively narrow a distributed feedback diode laser at the wavelength of 2 µm. Laser- gas interaction takes place inside an isothermal cell, which is placed on the transmission from the cavity. High quality, self-calibrated, absorption spectra are observed in pure CO2 samples at dif- ferent gas pressures, in coincidence with three lines of the R-branch of the ν1 +2ν2+ν3 band. Line intensities are determined using a global fitting approach in which a man- ifold of spectra are simultaneously analyzed across the range of pressures between 5 and 100 Torr, sharing a restricted number of unknown parameters. Various sources of uncertainty have been identified and carefully quantified, thus leading to an overall uncertainty ranging between 0.17% and 0.23%. The measured values are in a very good agreement with recent ab-initio predictions. a) Also at Institute of Applied Physics, RAS, Uljanova 46, Nizhny Novgorod, Russia b) Electronic mail: Livio.Gianfrani@unicampania.it. 1

CO2 at 4900 cm-1 Joe Hodges will talk about his measurements of this band Confirmation of sub%

L. Lodi, J. Tennyson and O.L. Polyansky, JCP, v.135, 034113 (2011) 0.4% too strong and 1% scatter – pure ab initio (200) and (101) bands overtones

Manfred Birk, private communication: DLR vs ab initio, spectral ranges 1850-2280, 2390-4000 and 4165-4365 cm-1 fundamentals and lower overtones   mean deviation scatter n1 bands 010 <- 000 110 <- 010 -2.2 % -1.4 % 3.5 % 1.9 % n2 bands 020 <- 000 030 <- 010 -0.85 % -0.65 % -0.57 % 0.80 % 1.5 % 1.4 % n3 bands 001 <- 010 011 <- 010 +0.33 % +0.34 % 1.7 % 0.70 % combination bands 020 <- 001 100 <- 010 +0.02 % -0.51 % -1.04 % 0.60 % 0.73 %

Journal of Quantitative Spectroscopy & Radiative Transfer Contents lists available at ScienceDirect Journal of Quantitative Spectroscopy & Radiative Transfer journal homepage: www.elsevier.com/locate/jqsrt Accurate line intensities for water transitions in the infrared: Comparison of theory and experiment Manfred Birk a, Georg Wagner a, Joep Loos a, Lorenzo Lodi b, Oleg L. Polyansky b, Aleksandra A. Kyuberis c, Nikolai F. Zobov c, Jonathan Tennyson b,n a Remote Sensing Technology Institute, German Aerospace Center (DLR), D-82234 Wessling, Germany b Department of Physics and Astronomy, University College London, London WC1E 6BT, UK c Institute of Applied Physics, Russian Academy of Sciences, Ulyanov Street 46, Nizhny Novgorod 603950, Russia a r t i c l e i n f o a b s t r a c t Article history: Received 15 January 2017 Received in revised form 29 March 2017 Accepted 30 March 2017 Ab initio calculations of water intensities are becoming mature and are claimed to have 1% accuracy in many cases. Experimental intensities with 1% accuracy can be achieved with some care. An inter- comparison of ab initio against experimental water intensities is presented for a variety of infrared bands 16O and some for H 18O and H O. A new calculated H2 O line list is presented for which un- for H2 certainties in the ab initio line intensities are evaluated. Much of the data show agreement within 2% between ab initio and experiment, however, for some bands, notably those involving excitation of some stretching modes, there are larger offsets of up to 8% attributed to ab initio calculation errors but still within the uncertainty of the ab initio calculation. In the ν1 fundamental band differences of between +5% and −13% are found which show systematic dependence on wavenumber, ΔKa , and ΔJ , again at- tributable to ab initio calculation errors. In the ν2 band, intensity-dependent differences up to 2% ori- ginate from the analysis of the experimental data. At present experiments are important to validate ab initio calculations but ab initio predictions can be very useful in validating the experiment. As the two procedures display significantly different systematic errors, it is suggested that combining both gives the best results; this study will also facilitate further improvements of the theoretical methodology. & 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 2 2 17 16 Keywords: Water Absorption intensity FTIR spectroscopy Ab initio calculations 1. Introduction independent measurements, those of Hodges et al. [10] and the DLR (German Aerospace Center) results included in HITRAN2012 [1], show intensities in agreement to better than 1% and within their combined uncertainties. In some cases, such high-accuracy measurements comprise only a few selected lines and therefore serve as a benchmark rather than a dataset suitable for inclusion in standard databases. The intention of laboratory spectroscopy work at DLR is to provide spectroscopic data with small well-defined uncertainties. This is certainly true for H2O spectroscopy, as well. Laboratory infrastructure such as absorption cells have been con- tinuously improved together with analysis software. Special focus was placed on validation of data product accuracy, i.e. proving the error budget by redundancy, χ test and so forth [12]. More details are given in Section 2 which details our experimental studies on water line intensities. The second approach is to make use of the increasing accuracy of ab initio calculations. A number of studies have focussed on trying to determine the dipole moment surface (DMS) of the water molecule to high accuracy [13–16]. However, computed vibration- rotation transition intensities are also sensitive to the nuclear- motion wavefunctions of the initial and final state, and hence to the potential energy surface (PES). Lodi and Tennyson [17] Water is molecule number one in HITRAN [1], the major ab- sorber of incoming sunlight in Earth's atmosphere and its biggest greenhouse gas. Its spectrum is therefore very well studied, see Refs. [2–7] for systematic compilations of experimental spectra of the various water isotopologues. However the demands of atmo- spheric science in general and satellite instruments such as MIPAS (Michelson Interferometer for Passive Atmospheric Sounding) on ENVISAT [8] mean that the laboratory data on water spectra must be of high quality. A particular issue is the reliability of the avail- able transition intensities. There are two approaches to obtaining accurate intensities for individual transitions. Traditionally this is done experimentally and there are a number of studies [9–11] which have produced intensity measurements which are accurate to 1% or better. For these measurements an extensive error analysis of systematic and random errors was carried out. In case of the 1 μm region, two n Corresponding author. E-mail address: j.tennyson@ucl.ac.uk ( J. Tennyson). http://dx.doi.org/10.1016/j.jqsrt.2017.03.040 0022-4073/& 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Please cite this article as: Birk M, et al. Accurate line intensities for water transitions in the infrared: Comparison of theory and experiment. J Quant Spectrosc Radiat Transfer (2017), http://dx.doi.org/10.1016/j.jqsrt.2017.03.040i

UV spectrum of water using different Linelists (J. Lampel, D,Poehler,O UV spectrum of water using different Linelists (J.Lampel, D,Poehler,O.L.Polyansky et al.,Atmospheric Chemistry and Physics,v.17,1271,(2017) )

New highly-accurate Dipole Moment Surface of water molecule Aleksandra A. Kyuberis2, Nikolai F. Zobov2, Lorenzo Lodi1, Johannes Lampel3, Eamon Conway1, Jonathan Tennyson1 and Oleg L. Polyansky1,2* 1Department of Physics and Astronomy, University College London, London WC1E 6BT, United Kingdom 2Institute of Applied Physics, Russian Academy of Science, Ulyanov Street 46, Nizhny Novgorod, 603950 Russia 3Max Planck Institute for Chemistry, 55128 Mainz, Germany   * o.polyansky@ucl.ac.uk We present a new Dipole Moment Surface (DMS) created for water molecule which improves on the highly-accurate DMS for water molecule created in 2011, known as LTP2011 [1]. Using LTP2011 we calculated the most accurate and complete line list for H216O up until now, called POKAZATEL [2]. This line list includes all the transitions involving energies up to 40 000 cm−1 and J up to 72. The accuracy of representation of the energy levels is about 0.1 cm−1 for all the energies up to dissociation. For the well-studied region up to 25~000 cm−1, the accuracy is significantly better. According to the comparison from paper [3], LTP2011 provides desciption of the absorption in UV region much better than previous line lists, like BT2 [4]. BT2 shows significant absorption features in regions where no absorption is observed. And intensities calculated using LTP2011 are significantly weaker, as they should be according to the experimental measurements. A new DMS, which gives a correct description of the absorption was created. Here we present this new DMS, w

UV spectrum of water using different Linelists (J. Lampel, D,Poehler,O UV spectrum of water using different Linelists (J.Lampel, D,Poehler,O.L.Polyansky et al.,Atmospheric Chemistry and Physics,v.17,1271,(2017) )

CONCLUSIONS Global, H2O, H217,18O, HDO linelists, H3+ NH3 assignment at 10 000 cm-1, 15 000 cm-1, 18 000 cm-1 – steps towards completeness HCN accurate global ab initio calculations 0.3 % intensity calculation accuracy for CO2 is being confirmed H2O UV intensity calculations- significant improvement