Hostname: page-component-74656f9884-8r8nv Total loading time: 0 Render date: 2026-11-07T09:50:24.256Z Has data issue: false hasContentIssue false
  • English
  • Français

H3+: the initiator of interstellar chemistry

Published online by Cambridge University Press:  07 August 2008

Mats Larsson
Mats Larsson
Affiliation:
Department of Physics, AlbaNova University Center, Stockholm University, SE-10691 Stockholm, Sweden e-mail: mats.larsson@physto.se
Rights & Permissions [Opens in a new window]

Abstract

Second only to H2, protonated molecular hydrogen, H3+, is the most abundantly produced interstellar molecule. Owing to its high reactivity and acidity, it plays the pivotal role in initiating interstellar chemical reactions, something which also reduces its steady-state concentration. Interstellar H3+ is not only destroyed in chemical reactions but also in dissociative recombination with electrons. The rate constant and mechanism of recombination have long been controversial, but great advances have been made during recent years, with the important consequence that the cosmic ray ionization rate in diffuse clouds is now believed to be higher by an order of magnitude than previously assumed.

Keywords

dissociative recombinationinterstellar chemistrycosmic rayionization rate

Information

Type
Research Article
Information
International Journal of Astrobiology , Volume 7 , Issue 3-4 , October 2008 , pp. 237 - 241
Copyright
Copyright © 2008 Cambridge University Press

Introduction

A prime focus of 21st century astrophysics is the study of the molecular universe at far-infrared, millimetre and submillimetre wavelengths, which will provide information about the formation processes of stars and galaxies. The space mission Herschel Space Observatory and the ground-based Atacama Large Millimeter Array are examples of instruments that will be taken into operation between 2008 and 2012 to address these and other questions. During the past 50 years the picture of the interstellar medium has been transformed from something that was thought to be primarily atomic, with a few diatomic molecules present (CO, CN, CH+), to something which is known to be primarily molecular. When Watson (Reference Watson1973) and Herbst & Klemperer (Reference Herbst and Klemperer1973) proposed that interstellar molecules are synthesized in ion–molecule reactions, it was also clear that H3+ must play a pivotal role in these schemes. Although H3+ at that time had been known as a physically stable molecule for more than half a century (it was first discovered in Thomson's (Reference Thomson1911) mass spectrometer), its electromagnetic spectrum was unknown. It was the skilful and persistent work by Oka and collaborators that led to the discovery of the infrared spectrum of H3+ (Oka Reference Oka1980) and then its presence in dark (Geballe & Oka Reference Geballe and Oka1996) and diffuse (McCall et al. Reference McCall, Geballe, Hinkle and Oka1998) clouds. Oka (Reference Oka2006) has reviewed these and subsequent discoveries of H3+ in the Central Molecular Zone near the Galactic centre.

Interstellar H3+ is produced by cosmic ray ionization of H2 followed by

(1)
{\rm H}_{\setnum{2}}^{ \plus } \plus {\rm H}_{\setnum{2}} \to {\rm H}_{\setnum{3}}^{ \plus } \plus {\rm H}

This reaction is extremely efficient and leads to H3+ being the dominant ion in hydrogen plasma, rather than H2+ or H+. Its important role in interstellar chemistry is a combination of this and its low proton affinity (4.4 eV), which results in the proton hop reaction:

(2)
{\rm H}_{\setnum{3}}^{ \plus } \plus {\rm X} \to {\rm HX}^{ \plus } \plus {\rm H}_{\setnum{2}}

where X is an atom or molecule. Reactions (1) and (2) initiate a chain of reactions that produce interstellar molecules. Figure 1 shows such a network of ion–molecule reaction chemistry, with dissociative recombination as the terminating step leading to the formation of stable, neutral molecules such as water and ethanol. Dissociative recombination:

(3)
{\rm H}_{\setnum{3}}^{ \plus } \plus e \mathop {\to} \limits_{k_{{\rm e}} } {\rm H}_{\setnum{2}} \plus {\rm H}\;{\rm or}\;3{\rm H}

is the additional destruction channel to reaction (2). In order to model interstellar chemistry, the rate of reaction (3) is required, as well as the product branching ratios.

Fig. 1.

Network of ion-molecule chemistry (reproduced with permission from McCall (Reference McCall2001)).

Dissociative recombination of H3+: a brief history

When the destruction of H3+ by thermal-energy electrons was measured in a pulsed-discharge (afterglow) experiment by Leu et al. (Reference Leu, Biondi and Johnsen1973) to proceed with a rate constant of k e=2.3×10−7 cm3 s−1, this could hardly have surprised anyone. After all, this was essentially the rate constant with which the atmospheric ions NO+, N2+ and O2+ recombined. About 10 years later, based on experiments using a flowing afterglow/Langmuir probe (FALP) technique, the Birmingham group published the surprising result that the rate constant was less than 2×10−8 cm3 s−1 (Adams et al. Reference Adams, Smith and Alge1984). This study was followed by reports in conference proceedings in which it was claimed that the rate constant could not be larger than 10−11 cm3 s−1 (e.g. Adams & Smith Reference Adams, Smith, Vardya and Tarafdar1987). There are several surprising aspects of this very small rate constant. The result was never published in a regular journal in which one could legitimately have asked for experimental details; the rate constant was much smaller than what one reasonably could expect to measure with a flowing afterglow apparatus; it had a considerable impact on the community of interstellar chemists (e.g. Van Dishoeck & Black Reference Van Dishoeck and Black1986), and the possibility of observing H3+ in diffuse clouds was discussed (where electron recombination is the main destruction process of H3+).

The notion of a non-recombining H3+ ion was so strong at the time that Amano (Reference Amano1988, Reference Amano1990) had some difficulties in convincing the community that his measurement of a rate constant of 1.8×10−7 cm3 s−1 was to be taken seriously. Amano measured spectroscopically the disappearance of H3+ in a pulsed afterglow instead of measuring, as in the FALP technique, the change in electron density along the flow tube.

The ion storage ring studies of H3+ (Larsson et al. Reference Larsson1993; Sundström et al. Reference Sundström1994) giving a rate constant of 1.15×10−7 cm3 s−1 started to tip the scale towards a higher rate, but the situation at the time of the Royal Society Discussion Meeting on H3+ (Herbst et al. Reference Herbst, Miller, Oka and Watson2000) was confusing. Larsson (Reference Larsson2000) has reviewed this in detail, and the reader is referred to this article for many more references than given here. One year later, at the symposium entitled ‘Dissociative Recombination of Molecular Ions with Electrons’ (Guberman Reference Guberman2003), it was not inconceivable that the recombination rate constant rate constant for H3+ at interstellar conditions (i.e. T e<100 K) could be as small as 10−8 cm3 s−1. What had happened, and what was the way forward to come to grips with this elusive rate constant?

From confusion to clarity: almost

The mechanism for the dissociative recombination of H3+ remained a problem for a long time. The lowest vibrational level of ground state H3+ is remote from the doubly excited resonant state of H3 which can drive recombination of vibrationally excited levels in H3+. Schneider et al. (Reference Schneider, Suzor-Weiner and Orel2000) investigated the effect of indirect recombination through the manifold of H3 Rydberg states converging to ground state H3+ and found it to be substantial, but nevertheless unable to give a rate constant larger than 10−9 cm3 s−1. Without the indirect mechanism, the rate was lower by a factor of a hundred.

Around the year 2000 a new stationary afterglow experiment (Glosik et al. Reference Glosik, Plašil, Poterya, Kudrna and Tichý2000) produced results in good agreement with the theoretical result. There was accumulating evidence that experiments at ion storage rings had been successful in determining the rate constant for H3+ in the zeroth vibrational level, but with considerable rotational excitations, and there were indications that the rotational excitations affected the rate constant (Jensen et al. Reference Jensen, Pedersen, Safvan, Seiersen, Urbain and Andersen2001; Kreckel et al. Reference Kreckel, Krohn and Lammich2002; Larsson et al. Reference Larsson and Guberman2003).

Greene's group in Boulder found that a hitherto overlooked mechanism, Jahn–Teller distortion of the H3+ ion as a result of the incoming low-energy electron, increased the recombination rate (Kokoouline et al. Reference Kokoouline, Greene and Esry2001), but only by about a factor of ten as compared with the result of Schneider et al. (Reference Schneider, Suzor-Weiner and Orel2000). Greene concluded in his talk at Guberman's meeting in 2001:

“The rate and cross-section of the dissociative recombination of H3+ are found to be in the range of the microwave afterglow results [i.e. about 10−8 cm3 s−1] and about an order of magnitude smaller than the storage ring experiments.” (Greene et al. Reference Greene, Kokoouline, Esry and Guberman2003).

The question was thus unavoidable: perhaps the rate constant was increased by a factor of 10 by rotational excitations?

Oka (Reference Oka and Guberman2003), being well aware that the observation of H3+ in diffuse clouds (McCall et al. Reference McCall, Geballe, Hinkle and Oka1998) favoured a low recombination rate constant, insisted that:

“Finally a word of caution. This astronomical mystery should not be taken to constrain the value of k e in any definitive way. The assumed canonical astrophysical parameters may be wrong. It is even possible that clouds containing H3+ are indeed very large. What we need is the TRUE value of k e, and not a value which explains the problem quickly. To this end scientists in different disciplines need to work together. Physicists, chemists and astronomers of the world, unite!” (Oka Reference Oka and Guberman2003).

The past five years have seen remarkable progress towards an understanding of dissociative recombination of H3+. In a collaborative effort between groups in Stockholm, Berkeley and Urbana-Champaign, a supersonic expansion discharge ion source was built and tested, and was found to be able to produce rotationally cold H3+. This source was connected to the ion storage ring CRYRING in Stockholm and used as injector. McCall et al. (Reference McCall2003, Reference McCall2004) found a decrease of the rate constant as a result of the rotational cooling of H3+ to about 40 K, but only to 6.8×10−8 cm3 s−1 from the rotationally ‘hot’ value of 1.15×10−7 cm3 s−1.

Using a different procedure of rotationally cooling H3+, namely with a cryogenically cold ion trap, Kreckel et al. (Reference Kreckel2005) used the test storage ring in Heidelberg with the trap as injector and obtained very good agreement with the results from CRYRING.

Kokoouline & Greene (Reference Kokoouline and Greene2003a,Reference Kokoouline and Greeneb) discovered a factor of ten error in their preliminary calculations (Kokoouline et al. Reference Kokoouline, Greene and Esry2001; Greene et al. Reference Greene, Kokoouline, Esry and Guberman2003), and managed to overcome a number of technical hurdles in order to calculate a fully quantum mechanical cross-section. The rate constant derived from their cross-section was in excellent agreement with the storage ring results. Figure 2 shows the storage ring and theoretical results. The agreement between the theoretical and experimental cross-sections has been improved in a recent theoretical work (Fonseca dos Santos et al. Reference Fonseca dos Santos, Kokoouline and Greene2007), but is still not perfect. It is not clear why the theoretical cross-section is more highly structured than the experimental one.

Fig. 2.

Comparison of results for dissociative recombination of H3+. The interconnected black dots display the results from TSR (Kreckel et al. Reference Kreckel2005), the interconnected grey dots display the results from CRYRING (McCall et al. Reference McCall2003), and the continuous grey line shows the theoretical results of Kokoouline & Greene (Reference Kokoouline and Greene2003a,Reference Kokoouline and Greeneb). The small difference between the experimental data of CRYRING and TSR is due to the colder electron beam used in the TSR experiment (reproduced with permission from Kreckel et al. (Reference Kreckel2005)).

Thus, the situation concerning the recombination rate constant for H3+ changed radically during the span of a few years. Now nobody seriously questions that the rate constant at an electron temperature of 300 K and a rotational temperature of 30–50 K is about 7×10−8 cm3 s−1. This has two consequences: it requires a revision of a canonical astrophysical parameter, the cosmic ray ionization rate, and an explanation of the conflicting results derived from plasma experiments.

McCall et al. (Reference McCall2003) explained the observed column density of H3+ in the diffuse cloud towards ζ Persei by an enhanced cosmic ray ionization rate, from the dark cloud canonical value of 3×10−17 to 1.2×10−15 s−1. The latter value, which is the ionization rate of molecular hydrogen, corresponds to a cosmic ray ionization rate per hydrogen atom of 5.2×10−16 s−1. In a more comprehensive analysis, this rate was reduced to 3.2×10−16 s−1 (Indriolo et al. Reference Indriolo, Geballe, Oka and McCall2007), still an order of magnitude larger than the canonical value. Indriolo et al. also investigated H3+ in other diffuse clouds and found an average value of 2×10−16 s−1. The need for an upward revision of the cosmic ray ionization rate based on the detection of large abundances of H3+ in diffuse clouds and the recombination rate constant derived from storage rings and theory has also been pointed out by Dalgarno (Reference Dalgarno2006).

The last piece of the H3+ puzzle is an understanding of the various afterglow experiments, of which only a few have been briefly discussed here. A comprehensive review of all plasma experiments up to the end of 2006 can be found in a very recently published research monograph by Larsson & Orel (Reference Larsson and Orel2008). Glosik's group in Prague, in collaboration with Greene and Kokoouline (Glosik et al. Reference Glosik2008), have made an ambitious attempt to explain the result of all plasma experiments. If they are correct, plasma experiments which yield very low rates can be explained by formation of long-lived H3 Rydberg states acting as storage reservoir for H3+ ions.

Acknowledgement

I would like to thank J. Glosik for providing a preprint of his most recent plasma work.

References

Adams, N.G. & Smith, D. (1987). Recent advances in the studies of reaction rates relevant to interstellar chemistry. In Astrochemistry, IAU Symp. 120, eds Vardya, M.S. & Tarafdar, S.P., pp. 1–18. Reidel, Dordrecht.Google Scholar
Adams, N.G., Smith, D. & Alge, E. (1984). Measurements of dissociative recombination rate coefficients of H3+, HCO+, N2H+, and CH5+ at 95 K and 300 K using the FALP apparatus. J. Chem. Phys. 81, 1778–1784.CrossRefGoogle Scholar
Amano, T. (1988). Is the dissociative recombination of H3+ really slow? A new spectroscopic measurement of the rate constant. Astrophys. J. Lett. 329, L121–L124.Google Scholar
Amano, T. (1990). The dissociative recombination rate coefficient of H3+, HN2+, and HCO+. J. Chem. Phys. 92, 6492–6501.Google Scholar
Dalgarno, A. (2006). The galactic cosmic ray ionization rate. Proc. Natl. Acad. Sci. USA 103, 12269–12273.CrossRefGoogle ScholarPubMed
Fonseca dos Santos, S., Kokoouline, V. & Greene, C.H. (2007). Dissociative recombination of H3+ in the ground and excited vibrational states. J. Chem. Phys. 127, 124309.Google Scholar
Geballe, T.R. & Oka, T. (1996). Detection of H3+ in interstellar space. Nature 384, 334–335.CrossRefGoogle ScholarPubMed
Glosik, J., Plašil, R., Poterya, V., Kudrna, P. & Tichý, M. (2000). The recombination of H3+ ions with electrons: dependence of partial pressure of H2. Chem. Phys. Lett. 331, 209–214.CrossRefGoogle Scholar
Glosik, J. et al. (2008). Recombination of H3+ ions in the afterglow of a He-Ar-H2 plasma. Phys. Rev. Lett. (unpublished).Google Scholar
Greene, C.H., Kokoouline, V. & Esry, B.D. (2003). Importance of Jahn–Teller coupling in the dissociative recombination of H3+ by low energy electrons. In Dissociative Recombination of Molecular Ions with Electrons, ed. Guberman, S.L., pp. 221–233. Kluwer Academic/Plenum, New York.Google Scholar
Guberman, S.L. (ed.) (2003). Dissociative Recombination of Molecular Ions with Electrons. Kluwer Academic/Plenum, New York.Google Scholar
Herbst, E. & Klemperer, W. (1973). The formation and depletion of molecules in dense interstellar clouds. Astrophys. J. 185, 505–533.Google Scholar
Herbst, E., Miller, S., Oka, T. & Watson, J.K.G.(eds.) (2000). Astronomy, physics and chemistry of H3+. Phil. Trans. R. Soc. Lond. A 358, 2359–2559.Google Scholar
Indriolo, N., Geballe, T.R., Oka, T. & McCall, B.J. (2007). H3+ in diffuse interstellar clouds: a tracer for the cosmic-ray ionization rate. Astrophys. J. 671, 1736–1747.Google Scholar
Jensen, M.J., Pedersen, H.B., Safvan, C.P., Seiersen, K., Urbain, X. & Andersen, L.H. (2001). Dissociative recombination and excitation of H3+. Phys. Rev. A 63, 052701-1–5.Google Scholar
Kokoouline, V. & Greene, C.H. (2003a). Theory of dissociative recombination of D 3h triatomic ions applied to H3+. Phys. Rev. Lett. 90, 13201-1–4.Google Scholar
Kokoouline, V. & Greene, C.H. (2003b). Unified theoretical treatment of dissociative recombination of D 3h triatomic ions: Applications to H3+ and D3+. Phys. Rev. A 68, 012703-1–23.CrossRefGoogle Scholar
Kokoouline, V., Greene, C.H. & Esry, B.D. (2001). Mechanism for the destruction of H3+ ions by electron impact. Nature 412, 891–894.CrossRefGoogle ScholarPubMed
Kreckel, H., Krohn, S., Lammich, L. et al. (2002). Vibrational and rotational cooling of H3+. Phys. Rev. A 66, 052509-1–11.Google Scholar
Kreckel, H. et al. (2005). High resolution dissociative recombination of cold H3+ and first evidence for nuclear spin effects. Phys. Rev. Lett. 95, 263201-1–4.Google Scholar
Larsson, M. (2000). Experimental studies of the dissociative recombination of H3+, Phil. Trans. R. Soc. Lond. A 358, 2433–2444.Google Scholar
Larsson, M. & Orel, A.E. (2008). Dissociative Recombination of Molecular Ions. Cambridge University Press, Cambridge.Google Scholar
Larsson, M. et al. (1993). Direct high-energy neutral-channel dissociative recombination of cold H3+ in an ion storage ring. Phys. Rev. Lett. 70, 430–433.Google Scholar
Larsson, M. et al. (2003). Studies of dissociative recombination in CRYRING. In Dissociative Recombination of Molecular Ions with Electrons, ed. Guberman, S.L., pp. 87–94. Kluwer Academic/Plenum, New YorkGoogle Scholar
Leu, M. T., Biondi, M. A. & Johnsen, R. (1973). Measurements of recombination of electrons with H3+ and H5+, Phys. Rev. A 8, 413–419.Google Scholar
McCall, B.J. (2001). Spectroscopy of H3+ in laboratory and astrophysical plasmas. PhD Thesis, The University of Chicago.Google Scholar
McCall, B.J., Geballe, T.R., Hinkle, K.H. & Oka, T. (1998). Detection of H3+ in the diffuse interstellar medium toward Cygnus OB2 no. 12. Science 279, 1910–1913.Google Scholar
McCall, B.J. et al. (2003). An enhanced cosmic-ray flux towards ζ Persei inferred from a laboratory study of the H3+−e− recombination rate. Nature 422, 500–502.Google Scholar
McCall, B.J. et al. (2004). Dissociative recombination of rotationally cold H3+. Phys. Rev. A 70, 052716-1–13.CrossRefGoogle Scholar
Oka, T. (1980). Observation of the infrared spectrum of H3+. Phys. Rev. Lett. 45, 531–534.Google Scholar
Oka, T. (2003). Help!!! Theory for H3+ recombination badly needed. In Dissociative Recombination of Mole-cular Ions with Electrons, ed. Guberman, S.L., pp. 209–220. Kluwer Academic/Plenum, New York.Google Scholar
Oka, T. (2006). Interstellar H3+. Proc. Natl. Acad. Sci. USA 103, 12235–12242.CrossRefGoogle ScholarPubMed
Schneider, I.F., Suzor-Weiner, A. & Orel, A.E. (2000). Channel mixing effects in the dissociative recombination of H3+ with slow electrons. Phys. Rev. Lett. 85, 3785–3788.Google Scholar
Sundström, G. et al. (1994). Destruction rate of H3+ by low energy electrons measured in a storage-ring experiment. Science 263, 785–787.Google Scholar
Thomson, J.J. (1911). Rays of positive electricity. Phil. Mag. 21, 225–249.CrossRefGoogle Scholar
Van Dishoeck, E.F. & Black, J.H. (1986). Comprehensive models of diffuse interstellar clouds: Physical conditions and molecular abundances. Astrophys. J. Suppl. Ser. 62, 109–145.Google Scholar
Watson, W.D. (1973). Rate of formation of interstellar molecules by ion-molecule reactions. Astrophys. J. 183, L17–L20.Google Scholar
Figure 0

Fig. 1. Network of ion-molecule chemistry (reproduced with permission from McCall (2001)).

Figure 1

Fig. 2. Comparison of results for dissociative recombination of H3+. The interconnected black dots display the results from TSR (Kreckel et al.2005), the interconnected grey dots display the results from CRYRING (McCall et al.2003), and the continuous grey line shows the theoretical results of Kokoouline & Greene (2003a,b). The small difference between the experimental data of CRYRING and TSR is due to the colder electron beam used in the TSR experiment (reproduced with permission from Kreckel et al. (2005)).