- Email: [email protected]
REMPI-TOF studies of the HF dimer
Journal of Molecular Structure 790 (2006) 27–30 www.elsevier.com/locate/molstruc
REMPI-TOF studies of the HF dimer ´ gu´st Kvaran *, O ´ mar Freyr Sigurbjo¨rnsson, Huasheng Wang A Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavı´k, Iceland Received 29 September 2005; received in revised form 17 October 2005; accepted 26 October 2005 Available online 19 December 2005
Abstract Resonance-enhanced multiphoton ionisation and time-of-flight mass spectra analysis of jet-cooled HF–argon mixtures are presented. Twophoton resonance transitions in the HF dimer, (HF)2, are observed for the first time. Simulation calculations reveal transitions to a Rydberg state and allow determination of energy parameters for the excited state dimer. The process of excitation is discussed. q 2005 Elsevier B.V. All rights reserved. Keywords: Hydrogen bonds; REMPI; dimer; multiphoton; ionization; Laser spectroscopy; jet cooling; simulations
1. Introduction Hydrogen fluoride forms the strongest hydrogen bond (1038 cmK1)[1] in the series of the hydrogen halides. Therefore, the system has become a popular model for investigations of hydrogen bonds. The HF dimer has been extensively investigated both experimentally[1–7] and theoretically [4,8,9,10], whereas properties of larger HF oligomers mostly have been predicted theoretically [4,11,12] but to a much lesser extent experimentally [13]. Microwave [2] and infrared [3,1,4,6] spectra analysis as well as laser-induced fluorescence studies [5,7] have revealed accurate spectroscopic and structural parameters for the dimer. In studies of the HF, dimer emphasis has been laid on energy and structure properties of the ground electronic state whereas no studies have been performed on excited states. As a matter of fact even very limited experimental studies have been performed on electronically excited states of the monomer [14]. This could partly be due to the fact that absorption due to electronic transitions appears in the VUV region [15,16] making experimental studies unfavourable. Furthermore, several early reports on high-energy interactions of HF have revealed highly anomalous and aggressive behaviour of the gas [17] suggesting that a great care needs to be taken in its handling. For some time, REMPI studies of the hydrogen halide monomers, HX; XZCl, Br and I, have been performed in our * Corresponding author. Tel.: C354 525 4694/4800; fax: C354 552 8911. E-mail address: [email protected] (A Kvaran). URL: http://www.hi.is/wagust/.
0022-2860/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.molstruc.2005.10.032
laboratory emphasizing characterization of Rydberg and ion-pair excited states based on the simulation calculations for two- and three-photon resonance excitation processes [18– 25]. In this paper, we report resonance-enhanced multiphoton ionisation of jet-cooled hydrogen fluoride gas diluted in argon followed by TOF mass analysis. Simulation analysis of recorded spectra revealed two-photon absorption in the HF dimer due to transitions to a Rydberg state and allowed characterization of energy and spectroscopic parameters. 2. Experimental Resonance-enhanced multiphoton ionisation (REMPI) of jet-cooled HF gas in argon was performed in the middle of an ionisation chamber. Ions were directed into a time-of flight tube and detected by MCP plates to record ion yield as a function of flight time and/or as a function of laser radiation wavenumber. Tunable excitation radiation was generated by an Excimer laser-pumped dye laser system, using a Lambda Physik COMPex 205 Excimer laser and a Lumonics Hyperdye 300 laser. A C-460 dye was used and the energy per pulse was about 5 mJ. The repetition rate was 10 Hz. The bandwidth of the dye laser beam was about 0.05 cmK1. The radiation was focused into an ionisation chamber between a repeller and an extractor separated by 19 mm. Gas samples, made by mixing HF gas from Schuchardt and argon on a vacuum line, typically in the ratio 1:3, were pumped through a 200 mm pulsed nozzle from a typical total backing pressure of about 3 bar into the ionisation chamber, which was held at lower than about 5! 10K6 mbar pressure during experiments. The distance between the nozzle and the centre between the repeller and the extractor was about 5 cm. The nozzle was held open for about 180 ms
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
28
and the LASER beam was typically fired about 450 ms after opening the nozzle. Ions were extracted into a 70-cm long time-of-flight tube and focused with an electric lens onto a MCP plate detector. Voltage outputs as a function of flight time were fed into a LeCroy 9310A, 400 MHz storage oscilloscope. Average voltage outputs for a fixed number of laser pulses were evaluated and recorded on a computer to get mass spectra. Either mass peak heights or integrals were measured and averaged for fixed number of laser pulses as a function of laser radiation wavenumbers to obtain (2C1)REMPI-TOF spectra. Typically spectral points were obtained by averaging over 100 pulses. Care was taken to prevent power broadening due to ac-Stark effects by minimizing the laser power. Wavelength calibration was achieved by recording iodine atomic lines or by measurements and comparison of the strongest hydrogen chloride rotational lines with those reported by Green et al. [26]. The accuracy of the calibration was found to be about G 1.0 cmK1 on a two-photon wavenumber scale. Care was taken to correct for possible drifts in signal intensities during long scans. Furthermore, the effect of varying laser power was corrected by dividing the measured intensity by the power squared. 3. Results and analysis The wavenumber range, where (2C1)REMPI spectra of HF will be observed, is determined by the molecular ionisation energy (IEZ16.039 eV/IEZ129,360 cmK1) [14]. The minimum and maximum laser wavenumbers ( n~laser ) for (2C1)REMPI correspond to the photon energies which equal one-third and half of the ionisation energy, respectively. Therefore, the excitation wavenumbers range (n~exc Z 2n~laser )
for (2C1)REMPI is 86241! n~exc =cmK1 ! 129360 Fig. 1 shows mass spectra as ion yields vs flight time for the two-photon resonance excitation wavenumber range 86,612– 86,626 cmK1. These spectra were recorded after flushing the system with HF–argon mixtures for several hours to eliminate signals from possible impurities on the line, which is known to be important, due to the aggressive nature of HF(g) [17]. Only one mass at about 11.5 ms flight time, which corresponds to a mass of about 19–21 amu, was detected. This we believe that is HFC or alternatively H2FC which is isoelectronic to H2O and lower in energy than HFC [27]. Experimental REMPI spectrum obtained by recording the integrated mass spectra as a function of excitation wavenumbers is shown at the top of Fig. 2. This must be a rovibrational band due to a two-photon resonance excitation followed by a one-photon ionisation in hydrogen fluoride. The observed, high-frequency, partly resolved rotational structure of the band could not be reproduced by simulation calculations assuming rotational structure of the ground state of HF (Bv 00 Z0Z20.5567) [14] whereas it could easily be simulated by assuming the ground state to be the dimer (HF)2, with two orders of magnitude smaller rotational constant (B ð0;0Þ Z 0.21666 cmK1) [3] as seen in Fig. 2. Furthermore, the signal enhanced profoundly with jet cooling, hence increasing cluster formations, as the backing pressure was enlarged. We, therefore, conclude that the signal originates from resonance excitation in the HF dimer. Since only HFC or H2FC ions but no (HF)C ions were detected, dissociation of a neutral 2 intermediate state or the dimer ion must occur in the excitation process.
Ion yield
ν
86622 cm -1
86617 cm -1 86612 cm -1 2
4
6
8
10 12 14 16 18 20 TOF
µs
Fig. 1. Jet-cooled HF(g)–argon (1:3) sample mass spectra recorded in a REMPI-TOF mass spectrometer (ion yield vs time-of-flight (ms)) as a function of two-photon laser wavenumber excitation (86,612–86,626 cmK1).
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
29
0
ν = 86616.3 cm–1
Relative Intensity
B = 0.21666 cm –1 * B = 0.2416 cm–1 T = 24 K
-6
D = 2*10 cm–1 * -6 D = 2*10 cm –1 ∆Ω = 0 Bandwidth = 0.36 cm –1
Q Exp.
Calc.
S
O
86.614
86.616
86.618
86.620
86.622
86.624
3
86.628x10
2hν / cm–1 Fig. 2. REMPI spectrum (ion yield vs two-photon laser wavenumber excitation) for a jet-cooled HF(g)–argon (1:3) sample (top) and simulation (underneath). Calculated spectrum (middle) for parameters indicated in the figure and corresponding rotational lines (bottom) are shown (see text).
The simulation calculation shown in Fig. 2 is based on a twophoton resonance absorption in a diatomic approximation [19,21,23,28]. This is based on the assumption that the electron transition is largely localized in either of the two HF units of the dimer for excitation of electrons in highest occupied molecular orbitals, corresponding to non-bonding, nZ2, p-orbital fluorine atom electrons. Rotational constants for the lowest vibrational state (v1Z0, v2Z0), sublevels KZ0 of the ground electronic state of (HF)2, were used [3]. The quantum numbers v1 and v2 correspond essentially to the free hydrogen stretch (n1) and the bonded hydrogen stretch (n2) of the dimer, respectively [3]. The total electronic angular momentum (spinCorbital angular momentum; U) needs to be set equal to zero and unaltered (DUZ0) for the overall transition. Thus, the major observed rotational structure could be interpreted as being due to Q rotational lines with transition strengths determined mostly by the zero rank component of the transition tensor [23,28] and weighted by a Boltzmann population distribution corresponding to a rotational temperature close to 24 K. Parameters used and derived from the simulation are listed in Fig. 2. The rotational parameter for the excited state (B Z 0.2416 cmK1) is close to that of the ground state (BZ 0.21666 cmK1). The Bs are related to B and C rotational constants corresponding to the two principal axes b and c as
experimental spectrum is stronger than any other observed features nearby on the wavenumber scale suggesting that the transition is Franck–Condon factor favoured. We therefore conclude that the excited state, most probably, is a ground vibrational state (v1 Z 0, v2 Z 0) of a Rydberg state. Furthermore, for the following reasons, most probably the Rydberg state correlates with the (H*(nZ2, p1) and F(2p5) atomic species. The ground state dimer correlates with the ground atomic species (H(nZ1) and F(2p5)) with a dissociation energy close to that of the monomer (47,340 cmK1)14, since the bond dissociation for the HF dimer is only about 1038 cmK1 [27]. The dissociation energy of the Rydberg state, for which E(v1 Z 0, v2 Z 0)Z86,616.3 cmK1, to form H*(nZ2) and F(2p5) would be close to that of the ground state (i.e. about 43,000 cmK1) whereas the dissociation energy to form H(nZ1)CF*(nZ3) would be much larger (about 63,220 cmK1 for F*(3s1) formation and about 76,720 cmK1 or larger for F*(3p)).
B Z ðB C CÞ=2 for a prolate near-symmetric rotor. In a near-symmetric top approximation, BZC, and B Z Z2 =ð2hcIt Þ Hence, the moments of inertia vertically to the top axis for the two states are ItZ2.06!10K46 kg m2 for the ground state and It Z1.84!10K46 kg m2 for the excited state, indicating that the geometries of the two states are comparable. The
Fig. 3. Schematic figure showing electronic-, vibrational- and rotationalresonance transitions in the (2C1)REMPI, 86,616.3 cmK1 system of (HF)2.
30
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
Furthermore, conservation of electronic angular momentum by two-photon excitation of an electron from a non-bonding 2pF orbital to a Rydberg state, as observed, requires a transition to the non-bonding p orbital (i.e. H(2p1)). Fig. 3 summarizes the nature of two-photon resonance excitation observed in the HF dimer as discussed above. For rotational temperature near 24 K, close to 10% population might be expected in sublevels KZ1 which are about 36.5 cmK1 higher in energy than the KZ0 levels [29]. Therefore, we cannot exclude some contribution from KZ1 transitions. However, alterations of the jet cooling by use of different backing pressures and gas mixture ratios did not show significant structure changes in the spectrum, which makes us believe that the structure is mostly due to KZ0, DKZ0 transitions.
4. Conclusion REMPI-TOF spectra recorded for jet-cooled HF(g)–argon mixtures in the laser wavenumber region 43,306–43,313 cmK1 (two-photon excitation region 86,612–86,626 cmK1) are interpreted as being due to two-photon resonance excitation in the HF dimer followed by one-photon ionisation, forming HFC or H2FC ions. Its partly resolved rotational structure was analysed by approximation simulation calculations to allow assignment and characterization of the electronically excited state involved. The excited state is assigned as a Rydberg state, with zero total electronic angular momentum, lowest vibrational state (v1 Z 0, v2 Z 0), which correlates with H*(2p1)CFCHF. Rotational parameters (B* and D*) as well as band origin (n~0 ) values are determined. Acknowledgements The financial support of the University Research Fund, University of Iceland and the Icelandic Science Foundation is gratefully acknowledged.
References [1] A. Pine, B. Howard, J. Chem. Phys. 84 (1986) 590–596. [2] T. Dyke, B. Howard, W. Klemperer, J. Chem. Phys. 56 (1972) 2442– 2454. [3] A. Pine, W. Lafferty, J. Chem. Phys. 78 (1983) 2154–2162. [4] H. Sun, R. Watts, U. Buck, J. Chem. Phys. 96 (1992) 1810–1821. [5] H. Chang, W. Klemperer, J. Chem. Phys. 100 (1994) 1–14. [6] D. Anderson, S. Davis, D. Nesbitt, J. Chem. Phys. 105 (1996) 4488–4503. [7] H. Chang, W. Klemperer, J. Chem. Phys. 98 (1993) 9266–9278. [8] M. Quack, M. Suhm, J. Chem. Phys. 95 (1991) 28–59. [9] G. Vissers, G. Groenenboom, A.V.D. Avoird, J. Chem. Phys. 119 (2003) 277–285. [10] G. Vissers, G. Groenenboom, A.V.D. Avoird, J. Chem. Phys. 119 (2003) 286–292. [11] M. Quack, J. Stohner, M. Suhm, J. Mol. Struct. 599 (2001) 381–425. [12] G. Chaban, R. Gerber, Spectrochim. Acta 58 (2002) 887–898. [13] T. Blake, S. Sharpe, S. Xantheas, J. Chem. Phys. 113 (2000) 707–718. [14] K.P. Hu¨ber, G. Herzberg, Constants of Diatomic Molecules, Van Nostrand-Reinhold, New York, 1979. [15] F. Carnovale, R. Tseng, C. Brion, J. Phys. B At. Mol. Phys. 14 (1981) 4771–4785. [16] J. Nee, M. Suto, L. Lee, J. Phys. B At. Mol. Phys. 18 (1985) L293–L294. [17] C. Brion, F. Carnovale, J. Phys. B At. Mol. Phys. 16 (1983) 499–504. ´ . Kvaran, H. Wang, A ´ . Logado´ttir,, in: Recent Research Development in [18] A Physical Chemistry, vol. 2, Transworld Research Network, Kerala, India, 1998, pp. 233–244. ´ . Kvaran, A ´ . Logado´ttir, H. Wang, J. Chem. Phys. 109 (1998) [19] A 5856–5867. ´ . Kvaran, B.G. Waage, H. Wang, J. Chem. Phys. 113 (2000) [20] A 1755–1761. ´ . Kvaran, H. Wang, A ´ . Logado´ttir, J. Chem. Phys. 112 (2000) [21] A 10811–10820. ´ . Kvaran, J. Mol. Struct. 563–564 (2001) 235–239. [22] H. Wang, A ´ . Kvaran, H. Wang, B.G. Waage, Can. J. Phys. 79 (2001) 197–210. [23] A ´ . Kvaran, H. Wang, Mol. Phys. 100 (2002) 3513–3519. [24] A ´ . Kvaran, H. Wang, J. Mol. Spectrosc. 228 (2004) 143–151. [25] A [26] D.S. Green, G.A. Bickel, S.C. Wallace, J. Mol. Spectrosc. 150 (1991) 388–469. ´ . Kvaran, To be published. [27] A. Bodi, K. Matthı´asson, H. Wang, A [28] R.G. Bray, R.M. Hochstrasser, Mol. Phys. 31 (1976) 1199–1211. [29] Z. Yu, E. Hammam, W. Klemperer, J. Chem. Phys. 122 (2005) 194318,1–194318,8.
REMPI-TOF studies of the HF dimer ´ gu´st Kvaran *, O ´ mar Freyr Sigurbjo¨rnsson, Huasheng Wang A Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavı´k, Iceland Received 29 September 2005; received in revised form 17 October 2005; accepted 26 October 2005 Available online 19 December 2005
Abstract Resonance-enhanced multiphoton ionisation and time-of-flight mass spectra analysis of jet-cooled HF–argon mixtures are presented. Twophoton resonance transitions in the HF dimer, (HF)2, are observed for the first time. Simulation calculations reveal transitions to a Rydberg state and allow determination of energy parameters for the excited state dimer. The process of excitation is discussed. q 2005 Elsevier B.V. All rights reserved. Keywords: Hydrogen bonds; REMPI; dimer; multiphoton; ionization; Laser spectroscopy; jet cooling; simulations
1. Introduction Hydrogen fluoride forms the strongest hydrogen bond (1038 cmK1)[1] in the series of the hydrogen halides. Therefore, the system has become a popular model for investigations of hydrogen bonds. The HF dimer has been extensively investigated both experimentally[1–7] and theoretically [4,8,9,10], whereas properties of larger HF oligomers mostly have been predicted theoretically [4,11,12] but to a much lesser extent experimentally [13]. Microwave [2] and infrared [3,1,4,6] spectra analysis as well as laser-induced fluorescence studies [5,7] have revealed accurate spectroscopic and structural parameters for the dimer. In studies of the HF, dimer emphasis has been laid on energy and structure properties of the ground electronic state whereas no studies have been performed on excited states. As a matter of fact even very limited experimental studies have been performed on electronically excited states of the monomer [14]. This could partly be due to the fact that absorption due to electronic transitions appears in the VUV region [15,16] making experimental studies unfavourable. Furthermore, several early reports on high-energy interactions of HF have revealed highly anomalous and aggressive behaviour of the gas [17] suggesting that a great care needs to be taken in its handling. For some time, REMPI studies of the hydrogen halide monomers, HX; XZCl, Br and I, have been performed in our * Corresponding author. Tel.: C354 525 4694/4800; fax: C354 552 8911. E-mail address: [email protected] (A Kvaran). URL: http://www.hi.is/wagust/.
0022-2860/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.molstruc.2005.10.032
laboratory emphasizing characterization of Rydberg and ion-pair excited states based on the simulation calculations for two- and three-photon resonance excitation processes [18– 25]. In this paper, we report resonance-enhanced multiphoton ionisation of jet-cooled hydrogen fluoride gas diluted in argon followed by TOF mass analysis. Simulation analysis of recorded spectra revealed two-photon absorption in the HF dimer due to transitions to a Rydberg state and allowed characterization of energy and spectroscopic parameters. 2. Experimental Resonance-enhanced multiphoton ionisation (REMPI) of jet-cooled HF gas in argon was performed in the middle of an ionisation chamber. Ions were directed into a time-of flight tube and detected by MCP plates to record ion yield as a function of flight time and/or as a function of laser radiation wavenumber. Tunable excitation radiation was generated by an Excimer laser-pumped dye laser system, using a Lambda Physik COMPex 205 Excimer laser and a Lumonics Hyperdye 300 laser. A C-460 dye was used and the energy per pulse was about 5 mJ. The repetition rate was 10 Hz. The bandwidth of the dye laser beam was about 0.05 cmK1. The radiation was focused into an ionisation chamber between a repeller and an extractor separated by 19 mm. Gas samples, made by mixing HF gas from Schuchardt and argon on a vacuum line, typically in the ratio 1:3, were pumped through a 200 mm pulsed nozzle from a typical total backing pressure of about 3 bar into the ionisation chamber, which was held at lower than about 5! 10K6 mbar pressure during experiments. The distance between the nozzle and the centre between the repeller and the extractor was about 5 cm. The nozzle was held open for about 180 ms
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
28
and the LASER beam was typically fired about 450 ms after opening the nozzle. Ions were extracted into a 70-cm long time-of-flight tube and focused with an electric lens onto a MCP plate detector. Voltage outputs as a function of flight time were fed into a LeCroy 9310A, 400 MHz storage oscilloscope. Average voltage outputs for a fixed number of laser pulses were evaluated and recorded on a computer to get mass spectra. Either mass peak heights or integrals were measured and averaged for fixed number of laser pulses as a function of laser radiation wavenumbers to obtain (2C1)REMPI-TOF spectra. Typically spectral points were obtained by averaging over 100 pulses. Care was taken to prevent power broadening due to ac-Stark effects by minimizing the laser power. Wavelength calibration was achieved by recording iodine atomic lines or by measurements and comparison of the strongest hydrogen chloride rotational lines with those reported by Green et al. [26]. The accuracy of the calibration was found to be about G 1.0 cmK1 on a two-photon wavenumber scale. Care was taken to correct for possible drifts in signal intensities during long scans. Furthermore, the effect of varying laser power was corrected by dividing the measured intensity by the power squared. 3. Results and analysis The wavenumber range, where (2C1)REMPI spectra of HF will be observed, is determined by the molecular ionisation energy (IEZ16.039 eV/IEZ129,360 cmK1) [14]. The minimum and maximum laser wavenumbers ( n~laser ) for (2C1)REMPI correspond to the photon energies which equal one-third and half of the ionisation energy, respectively. Therefore, the excitation wavenumbers range (n~exc Z 2n~laser )
for (2C1)REMPI is 86241! n~exc =cmK1 ! 129360 Fig. 1 shows mass spectra as ion yields vs flight time for the two-photon resonance excitation wavenumber range 86,612– 86,626 cmK1. These spectra were recorded after flushing the system with HF–argon mixtures for several hours to eliminate signals from possible impurities on the line, which is known to be important, due to the aggressive nature of HF(g) [17]. Only one mass at about 11.5 ms flight time, which corresponds to a mass of about 19–21 amu, was detected. This we believe that is HFC or alternatively H2FC which is isoelectronic to H2O and lower in energy than HFC [27]. Experimental REMPI spectrum obtained by recording the integrated mass spectra as a function of excitation wavenumbers is shown at the top of Fig. 2. This must be a rovibrational band due to a two-photon resonance excitation followed by a one-photon ionisation in hydrogen fluoride. The observed, high-frequency, partly resolved rotational structure of the band could not be reproduced by simulation calculations assuming rotational structure of the ground state of HF (Bv 00 Z0Z20.5567) [14] whereas it could easily be simulated by assuming the ground state to be the dimer (HF)2, with two orders of magnitude smaller rotational constant (B ð0;0Þ Z 0.21666 cmK1) [3] as seen in Fig. 2. Furthermore, the signal enhanced profoundly with jet cooling, hence increasing cluster formations, as the backing pressure was enlarged. We, therefore, conclude that the signal originates from resonance excitation in the HF dimer. Since only HFC or H2FC ions but no (HF)C ions were detected, dissociation of a neutral 2 intermediate state or the dimer ion must occur in the excitation process.
Ion yield
ν
86622 cm -1
86617 cm -1 86612 cm -1 2
4
6
8
10 12 14 16 18 20 TOF
µs
Fig. 1. Jet-cooled HF(g)–argon (1:3) sample mass spectra recorded in a REMPI-TOF mass spectrometer (ion yield vs time-of-flight (ms)) as a function of two-photon laser wavenumber excitation (86,612–86,626 cmK1).
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
29
0
ν = 86616.3 cm–1
Relative Intensity
B = 0.21666 cm –1 * B = 0.2416 cm–1 T = 24 K
-6
D = 2*10 cm–1 * -6 D = 2*10 cm –1 ∆Ω = 0 Bandwidth = 0.36 cm –1
Q Exp.
Calc.
S
O
86.614
86.616
86.618
86.620
86.622
86.624
3
86.628x10
2hν / cm–1 Fig. 2. REMPI spectrum (ion yield vs two-photon laser wavenumber excitation) for a jet-cooled HF(g)–argon (1:3) sample (top) and simulation (underneath). Calculated spectrum (middle) for parameters indicated in the figure and corresponding rotational lines (bottom) are shown (see text).
The simulation calculation shown in Fig. 2 is based on a twophoton resonance absorption in a diatomic approximation [19,21,23,28]. This is based on the assumption that the electron transition is largely localized in either of the two HF units of the dimer for excitation of electrons in highest occupied molecular orbitals, corresponding to non-bonding, nZ2, p-orbital fluorine atom electrons. Rotational constants for the lowest vibrational state (v1Z0, v2Z0), sublevels KZ0 of the ground electronic state of (HF)2, were used [3]. The quantum numbers v1 and v2 correspond essentially to the free hydrogen stretch (n1) and the bonded hydrogen stretch (n2) of the dimer, respectively [3]. The total electronic angular momentum (spinCorbital angular momentum; U) needs to be set equal to zero and unaltered (DUZ0) for the overall transition. Thus, the major observed rotational structure could be interpreted as being due to Q rotational lines with transition strengths determined mostly by the zero rank component of the transition tensor [23,28] and weighted by a Boltzmann population distribution corresponding to a rotational temperature close to 24 K. Parameters used and derived from the simulation are listed in Fig. 2. The rotational parameter for the excited state (B Z 0.2416 cmK1) is close to that of the ground state (BZ 0.21666 cmK1). The Bs are related to B and C rotational constants corresponding to the two principal axes b and c as
experimental spectrum is stronger than any other observed features nearby on the wavenumber scale suggesting that the transition is Franck–Condon factor favoured. We therefore conclude that the excited state, most probably, is a ground vibrational state (v1 Z 0, v2 Z 0) of a Rydberg state. Furthermore, for the following reasons, most probably the Rydberg state correlates with the (H*(nZ2, p1) and F(2p5) atomic species. The ground state dimer correlates with the ground atomic species (H(nZ1) and F(2p5)) with a dissociation energy close to that of the monomer (47,340 cmK1)14, since the bond dissociation for the HF dimer is only about 1038 cmK1 [27]. The dissociation energy of the Rydberg state, for which E(v1 Z 0, v2 Z 0)Z86,616.3 cmK1, to form H*(nZ2) and F(2p5) would be close to that of the ground state (i.e. about 43,000 cmK1) whereas the dissociation energy to form H(nZ1)CF*(nZ3) would be much larger (about 63,220 cmK1 for F*(3s1) formation and about 76,720 cmK1 or larger for F*(3p)).
B Z ðB C CÞ=2 for a prolate near-symmetric rotor. In a near-symmetric top approximation, BZC, and B Z Z2 =ð2hcIt Þ Hence, the moments of inertia vertically to the top axis for the two states are ItZ2.06!10K46 kg m2 for the ground state and It Z1.84!10K46 kg m2 for the excited state, indicating that the geometries of the two states are comparable. The
Fig. 3. Schematic figure showing electronic-, vibrational- and rotationalresonance transitions in the (2C1)REMPI, 86,616.3 cmK1 system of (HF)2.
30
A´. Kvaran et al. / Journal of Molecular Structure 790 (2006) 27–30
Furthermore, conservation of electronic angular momentum by two-photon excitation of an electron from a non-bonding 2pF orbital to a Rydberg state, as observed, requires a transition to the non-bonding p orbital (i.e. H(2p1)). Fig. 3 summarizes the nature of two-photon resonance excitation observed in the HF dimer as discussed above. For rotational temperature near 24 K, close to 10% population might be expected in sublevels KZ1 which are about 36.5 cmK1 higher in energy than the KZ0 levels [29]. Therefore, we cannot exclude some contribution from KZ1 transitions. However, alterations of the jet cooling by use of different backing pressures and gas mixture ratios did not show significant structure changes in the spectrum, which makes us believe that the structure is mostly due to KZ0, DKZ0 transitions.
4. Conclusion REMPI-TOF spectra recorded for jet-cooled HF(g)–argon mixtures in the laser wavenumber region 43,306–43,313 cmK1 (two-photon excitation region 86,612–86,626 cmK1) are interpreted as being due to two-photon resonance excitation in the HF dimer followed by one-photon ionisation, forming HFC or H2FC ions. Its partly resolved rotational structure was analysed by approximation simulation calculations to allow assignment and characterization of the electronically excited state involved. The excited state is assigned as a Rydberg state, with zero total electronic angular momentum, lowest vibrational state (v1 Z 0, v2 Z 0), which correlates with H*(2p1)CFCHF. Rotational parameters (B* and D*) as well as band origin (n~0 ) values are determined. Acknowledgements The financial support of the University Research Fund, University of Iceland and the Icelandic Science Foundation is gratefully acknowledged.
References [1] A. Pine, B. Howard, J. Chem. Phys. 84 (1986) 590–596. [2] T. Dyke, B. Howard, W. Klemperer, J. Chem. Phys. 56 (1972) 2442– 2454. [3] A. Pine, W. Lafferty, J. Chem. Phys. 78 (1983) 2154–2162. [4] H. Sun, R. Watts, U. Buck, J. Chem. Phys. 96 (1992) 1810–1821. [5] H. Chang, W. Klemperer, J. Chem. Phys. 100 (1994) 1–14. [6] D. Anderson, S. Davis, D. Nesbitt, J. Chem. Phys. 105 (1996) 4488–4503. [7] H. Chang, W. Klemperer, J. Chem. Phys. 98 (1993) 9266–9278. [8] M. Quack, M. Suhm, J. Chem. Phys. 95 (1991) 28–59. [9] G. Vissers, G. Groenenboom, A.V.D. Avoird, J. Chem. Phys. 119 (2003) 277–285. [10] G. Vissers, G. Groenenboom, A.V.D. Avoird, J. Chem. Phys. 119 (2003) 286–292. [11] M. Quack, J. Stohner, M. Suhm, J. Mol. Struct. 599 (2001) 381–425. [12] G. Chaban, R. Gerber, Spectrochim. Acta 58 (2002) 887–898. [13] T. Blake, S. Sharpe, S. Xantheas, J. Chem. Phys. 113 (2000) 707–718. [14] K.P. Hu¨ber, G. Herzberg, Constants of Diatomic Molecules, Van Nostrand-Reinhold, New York, 1979. [15] F. Carnovale, R. Tseng, C. Brion, J. Phys. B At. Mol. Phys. 14 (1981) 4771–4785. [16] J. Nee, M. Suto, L. Lee, J. Phys. B At. Mol. Phys. 18 (1985) L293–L294. [17] C. Brion, F. Carnovale, J. Phys. B At. Mol. Phys. 16 (1983) 499–504. ´ . Kvaran, H. Wang, A ´ . Logado´ttir,, in: Recent Research Development in [18] A Physical Chemistry, vol. 2, Transworld Research Network, Kerala, India, 1998, pp. 233–244. ´ . Kvaran, A ´ . Logado´ttir, H. Wang, J. Chem. Phys. 109 (1998) [19] A 5856–5867. ´ . Kvaran, B.G. Waage, H. Wang, J. Chem. Phys. 113 (2000) [20] A 1755–1761. ´ . Kvaran, H. Wang, A ´ . Logado´ttir, J. Chem. Phys. 112 (2000) [21] A 10811–10820. ´ . Kvaran, J. Mol. Struct. 563–564 (2001) 235–239. [22] H. Wang, A ´ . Kvaran, H. Wang, B.G. Waage, Can. J. Phys. 79 (2001) 197–210. [23] A ´ . Kvaran, H. Wang, Mol. Phys. 100 (2002) 3513–3519. [24] A ´ . Kvaran, H. Wang, J. Mol. Spectrosc. 228 (2004) 143–151. [25] A [26] D.S. Green, G.A. Bickel, S.C. Wallace, J. Mol. Spectrosc. 150 (1991) 388–469. ´ . Kvaran, To be published. [27] A. Bodi, K. Matthı´asson, H. Wang, A [28] R.G. Bray, R.M. Hochstrasser, Mol. Phys. 31 (1976) 1199–1211. [29] Z. Yu, E. Hammam, W. Klemperer, J. Chem. Phys. 122 (2005) 194318,1–194318,8.