STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS WITH ULTRASENSITIVE INFRARED LASER TECHNIQUES

Richard J. Saykally Department of Chemistry UC Berkeley Berkeley, CA 94720

Over the last decade, modeling of the chemistry occurring in inter­stellar gas clouds has evolved drfmatically. The early qualitative versions of Solomon and Klemperer and Herbst and Klemperer,2 which first predicted the preeminence of ion-molecular reactions in these cold, diffuse environments, are now supplanted by the modern sophisti­cated co~puter models of Mitchell, Ginzberg and Kuntz,3 Prasad and Huntress,4 Watson,5 and others. Quantitative predictions of molecular abundances are now given for an impressive number of species, including some two dozen of the most important molecular ions. In Table 1, the molecular ions expected to be most abundant in cold (500 K), dense (n~IQ6 cm-3) clouds are listed, along with abundances (relative to Hi) predicted by ~itchell, Ginzberg, and Kuntz,3 appropriate for a cloud density of 10 cm-3.

Generally, the stable, closed-shell "proton adducts" are the most abundant ions in cold, dense clouds, because open-shell species general­ly react rapidly with H2 until these thermodynamically favored products are finally obtained. Actually, the synthesis of these ions is most likely to proceed from the direct proton transfer reaction of the principal reactant ion, H3+' with abundant stable molecules (e.g., CO, H20, CO2 , NH3 , etc.) to yield the respective proton adducts. Until the present, only three such protonated species have been definitively identified in interstellar clouds--HOC+, HNN+ and HCS+. In addition, CH+ (the unstable isomer of HCO+)6 and HC02+ 7 have been tentatively identified. This situation is simply the result of a general lack of high-resolution laboratory spectroscopic information on polyatomic ions. While approximately 45 diatomic ions have now been studied at high resolution in the laboratory,8,9 only 15 polyatomic species lO have thus far been detected by spectroscopic techniques capable of resolving their rotational energy level structures. Definitive identification of molecular transitions observed by microwave, millimeter; or infrared astronomical techniques clearly requires comparison with such high­precision laboratory data for the general cases of nonlinear molecules, even though the linear molecules HCO+, HNN+, and HCS+ were, in fact, originally identified quite reliably from millimeter observations alone. 8 ,9

403

G. H. F. Diercksen et al. (eds.), Molecular Astrophysics, 403-419. © 1985 by D. Reidel Publishing Company.

404 R. J. SAYKALLY

TABLE 1 Molecular Ions Predicted to be Abundant in Cold (SOOK), Dense (n~106 cm-3) Interstellar Clouds

Ion (x) nx/ Log nH

H+ 3

-10.3

CH/ -11. 99

CH + 5 -14.3

o + 2 P -11.4

OH+ P -14.8

H 0+ 2 P -14.2

H 0+ 3 -11. 0

HCO+ * -9.8

CO+ P -14.8

HO + 2 P -13.8

H CO+ 2 P -11. 9

HCO + 2 -14.6

H3CO+ -14.1

NH + 2 -14.8

NH + 3 P -11. 9

NH4+ -l3.4

NO+ -12.7

HNN+ * -11. 2

H CN+ 2 -l3.1

CCN+ -14.6

SH+ P -11. 3

H S+ 2 P -14.8

SO+ P -10.8 S +

2 P -14.8

HCS+ * -14.5

H CS+ 2 P -l3.8

HSiO+ -12.9

P "* paramagnetic

* "* detected by radioastronomy

STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS 405

A number of interesting open-shell polyatomic ions are also pre­dicted by current models to be abundant in dense clouds (n=106); for example, Mitchell, Ginzberg, and Kuntz,3 predict high abundances of 02+ H02+, SH+, H2S+, S2+' NH3+' H2CO+' and HCS+ in such regions. For these cases as well, there exist very little spectroscopic data which can be of use in identifying spectra observed in astronomical sources. Indeed, while several open-shell polyatomic ions have~ in fact, been studied, both in the laboratory and in comet tails,8, principally by optical emission spectroscopy, these open-shell species are likely to be even more difficult to study by high-resolution techniques than closed-shell ions because of their extreme chemical reactivity.

In this paper we discuss the development of two infrared laser techniques--far-infrared laser magnetic resonance and velocity-modula­tion infrared laser spectroscopy--which have demonstrated the capability to study molecular ions with sufficient precision to produce a positive identification of their corresponding astronomical spectra. We antici­pate that in the near future these methods will make it possible to detect, identify, and make detailed studies of spectra of a variety of astrophysically important molecular ions. The extension of astro­nomical measurements to new ionic species is of obvious importance to understanding the chemistry of the interstellar medium, particularly in view of the facts that most of the ion-molecule reaction data used as the input for modern chemical models have rarely been obtained at the low temperatures appropriate for the ISM, and that many of the criti­cally important dissociative recombination and radiative association reactions have not been studied at all.

The first infrared spectrum of a gaseous molecular ion (NO+) was observed in connection with atmospheric nuclear weapons tests conduct­ed in the 1960s. 11 In 1979, Schwartz developed a method for observing broadband IR absorption spectra of ions generated by pulsed radio lysis at atmospheric pressure, and applied it to H30+(H20)n (n=0-6) in 197512 and to NH4+(NH3) (n=0-4) in 1979. 13 • The first laboratory IR spectra of ions to b~ obtained with high resolution were those of HD+, observed1~y the Doppler-tuned fast ion beam technique of Wing and coworkers in 1f~5. Sygsequentlv. his group was able to study transitions in HeH+, D3+' and H2D+ 17 by this method. Carrington has incorp?§ated a 1jecond20aser ,.d.th the fast ion beam experiment to study HD+, HeH+,l CH+, H3+,21 and other ions at internal energies near or even exceeding their lowest dissociation limits. In 1979, K.M. Evenson and 122 reported the application of far-infrared laser magnetic resonance spectroscopy for the m~asurement of rotational transitions in the ground 2IT state of HBr. This work has been cont-in.lIed at Berkeley, and has been extended to the ground states of HCl+,23 HF+,24 OH+,25 and H20+. 26 In 1980, Oka27 opened the door to the investigation of vibrational spectra of molecular ions with tunable infrared lasers with his work on H3+' subsequently extended to HeH+ 28 and Ne~ 29 by Amano and coworkers at Ottawa. Brault and Davis30 observed the vibration-rotation spectrum of ArH+ and KrH+ as well. Van den Huevel and Dymanus 32 pioneered the application of tuna­ble far-infrared laser sources to rotational spectroscopy of ions in 1983 with their study of CO+, HCO+, and HNN+. In 1983, Gudeman,

406 R. J. SAYKALL Y

Begemann, Pfaff, and myself reported the development of a new method for observing molecular ion vibration-rotation spectra, called "velocity-modul.ation laser sp;ctroscopy," and its described application to HCO+,33 HNN ,34 and H30+.3J This approach has ~?w been extended to include NH4+,36,37 H2F+,38,39 D30+ 40 and DNN+ in studies at Berkeley, to NH4+,42 H30+,43, HCNffT,44 and OH+ 45 in studies by Oka's group at Chicago, and to HCO+ 46 studied by Davies at Cambridge. R~7 centlv Amano and coworkers at Ottawa have reported studies of HCO+, H2D+,48 HD2+,48 and DCNH+ 49 with infrared difference frequency spec­tro~cobY of a discharge, and McKellar and coworkers have studied HCO ,5 HNN+, DCO+ and DNN+ 51 with a diode laser. The molecular ions that have been studied with high-resolution infrared methods are listed in Table 2. C.S. Gudeman and myself lO have reviewed most of this work very recently; consequently, only a brief discussion of the results of laser magnetic resonance spectroscopy and infrared laser spectroscopy experiments that are directly relevant to astrophysical studies is given here.

TABLE 2 Molecular Ions Studied by High-Resolution Infrared Spectroscopy

Ion References

HD+ 14,18

HeH+ 5,19,28

NeH+ 29

KrH+ 31

ArH+(ArD+) 31,31,71

HF+ 24

HCl+ 23,55

HBr+ 22,55

CH+ 20

O~ 25,45

HCO+(DCO+) 33,46,47,50

HNN+(DNN+) 34,41,51,64

+( + + +) H3 H2D ,HD2 ,D3 27,16,17,48

H2F 38,39

H 0+ 2 26

HP-(D3O+) 35,40,43

NH4+ 36,37,42,70

HCNH+(DC~) 44,49

STUDIES Of ASTROPHYSICALLY IMPORTANT MOLECULAR IONS 407

Characteristics of Infrared Spectroscopy

Unlike in the microwave and millimeter regime, where the spectral resolution is typically limited by collisional line broadening, infra­red spectra are usually Doppler-limited. For typical ~ight molecules, this corresponds to a linewid~? (hwhm) which is -lxIO- of the tran­sition frequency. At 3000 em ,this is roughly 100 MHz, whereas at 30 cm-1 (in the far infrared) it is 1 MHz. Considering that one can generally measure a line center -10 times more accurately than the fractional linewidth, this implies a conservative Doppler-limited measurement precision ranging from -10 MHz in the IR to -100 kHz in the FIR. Typical measurement precision for the microwave or millimeter region is 10 kHz. The loss of precision in the IR is partially offset by the fact that often 50 to 100 rotational lines can be measured for a given molecule. When a small number of parameters are required to fit the spectra, as in the cases of HCO+, HNN+, or HCNH+, knowledge of the astrophysically interesting lowest transition frequency essentially increases like the standard deviation of a mean--i.e., as the square root of the number of measurements. Hence, one can, in principle, re­gain an order of magnitude in precision from infrared studies, yielding -1 MHz measurement precision at 3000 cm-I • Moreover, sub-Doppler experiments can sometimes be done to increase this precision even fur­ther. In view of these facts, studies of molecular ions at frequencies ranging from the infrared to the FIR can yield predictions of either vibrational or rotational transitions of molecular ions with sufficient precision to afford a definitive identification of astronomical rota­tional and vibrational spectra--particularly if several rotational transitions are observed. A disadvantage of infrared spectroscopy is that the small closed-shell quadrupole hyperfine splittings, often measured in microwave studies, are not normally observable; hence this extremely useful diagnostic tool is lost. As we shall see, infrared spectroscopy often has greater capability for studying a given poly­atomic ion than does microwave spectroscopy, particularly because of the difficulty involved with searching suitably large frequency regions by microwave techniques. In any case, microw;ve measurements are highly complementary to infrared vibration-rotation spectroscopy, and clearly both should be employed whenever possible.

Our approach has been to study the very reactive open-shell ions by the most sensitive absorption method known--far infrared laser magnetic resonance (LMR)--and to study the more stable closed-shell species by velocity-modulation spectroscopy with a narrowband tunable infrared laser. These two approaches are contrasted in the following sections.

Laser Magnetic Resonance Rotational Spectroscopy

The LMR experiment, initially developed by K.M. Evenson and his coworkers, has been described in detail several times;52,53 its appli­cation to molecular ions generated in discharge plasmas has been dis­cussed as well,54,23,24 hence no further experimental details will be given here. The two principal advantages of LMR are its extremely

408 R. J. SAYKALLY

high sensitivity (-106 ions/cm3 for HCI+) and its high resolution (-0.1 MHz). These features make LMR a powerful tool for studying spec­tra of open-shell ions, which usually react rapidly with even trace amounts of water or other hydrogen-containing molecules to yield more stable ions, and hence generally occur in very low concentrations (-108 cm-3) in laboratory plasmas.

We have given a preliminary account 23 of our study of the ground state of HCI+, diagrammed in Figure 1, by LMR. In a subsequent and more detailed stud¥S we have measured numerous hyperfine-Zeeman transi­tions and analyzed them with both the effective Hamiltonian of Brown56 and with the Hund's Case C formalism of Veseth. 57 A typical LMR spectrum

Figure 1.

N :x: I-

w

10

8

6

4

2

o

HCI+

A = -19.289 THz B = 0.29342 THz

r---+

2.9 GHz

94.61Lm

J =!! 2

J = ~ ~~ __ + 2

I. 7 GHz

840 MHz + 7 ""--___ J=2

340 MHz_ 5 J =­+ 2

207.61Lm

90MHz+ J = ~ (x200) - 2

n. = 3/2

_ _ -C::55::G::H:::Z - 5 + J= 2

_ ""T--. ... 1=9=G=H=Z I + J =2 (x2)

9J =0

19.00 THz

n. = 1/2

Energy level diagram of the x2rr ground state of HCI+.

STUDlFS OF ASTROPHYSICALL Y IMPORTANT MOLECULAR IONS 409

HCI+ iS3~hown in Figure 2, in which the hyperfine splitting from the I=3/2 CI nucleus is revealed, as well as the larger lamda-doublet split§}ng. A similar spectrum, exhibiting the same hyperfine effects for H CI+ is shown ~~ Figure 3. A total of 80 lines were measured for H35CI+ and for H CI+, and analyzed with a 10-parameter fit for each isotope to give an average deviation of -2 MHz. Unfortunately,

H3~CI+ x2n .n. 3/2 .. 0

J. 5/2-7/2 MJ '-3/2--1/2

147.9 I'm CH3NH2

1.027-1.127 T

Figure 2. LMR rotational spectra of H35CI+. a) 2 mTorr HCI in 1 Torr He. b) 6 mTorr HCI in 1 Torr He. c) 20 mTorr Hel in 1 Torr He. The small pair of triplets is due to NH.

Figure 3.

H37CI+ x2rr .n = 3/2 v = 0

J = 7/2 - 9/2

116 I'm 13CH30H

'--_---''--__ '--_---ll MJ = 712 -5/2

'--__ .l..... __ -'-_---ll M J = 5/2 - 312

0.730-0.830 T

LMR rotational spectra of H37C1+.

410 R. J. SAYKALL Y

parameter correlation is a problem in such complex analyses as these, impairing our ~bility to make accurate determinations of all eigenva­lues of the X IT state (in particular, our poor determination of the d hyperfine parameter results in a relatively large uncertaintly for the n=~ eigenvalues). Nevertheless, it is possible to predict both the lambda-doubling transitions as well as the pure rotational transi­tions for the astronomically relevant lower spin state n=3/2 with ±2 MHz uncertainty (at the 10 level). These predictions will be pre-sented in a separate paper.55 4 .

In Figure 4 we present a LMR spectrum measured 2 for the jsovalent species HF+, in which the very small proton hyperfine splitting is res~lved; unfortunately we have not been able to accomplish this for Hel so far, but Figure 4 is indicative of the capability of LMR to make such high precision measurements in favorable cases. In FiguZ6 5 is shown a recent LMR detection of the HZO+ ion by my group. While

a detailed analysis has not yet been carr led out for the three rota­tional transitions observed thus far, we certainly expect to obtain a precise determination of the pure rotational frequencies for this im­portant ion shortly.

Figure 4.

10.0

--+

2 X n3/2

MJ = -3/2--1/2 J = 3/2-5/2

M (F)- +1 I 2

i

11.0 12.0

kG

+ LMR rotational spectra of HF .

+--

M (F).-1. I 2

i

13.0

While LMR clearly possesses the sensitivity and resolution to produce a great deal of valuable information about molecular ions, it possesses one major liability; the major problem now encountered when attempting to detect LMR spectra of ions generated in discharge plasmas is simply the obfuscation of weak ion transitions by the much stronger and more numerous lines of neutral species also formed in these environments. It is just this difficulty which must be tran­scended if LMR is to become a truly general tool for mole·cular spectro­scopy.

STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS

2.4 3.4 4.5

",0· XIS"y·O

301- 3 11

J. 7/2 -7/2

5.4

B (kGouIl)

3041'm DCOOD 985.8897 GHI

6.4 7.4

Figure 5. LMR rotational spectra of H20+. A singlet hyperfine pattern is observed for this rotational transition.

411

The same difficulty -- that of distinguishing the spectra of ions from those of more abundant neutrals--also limited vibration-rotation studies of ions to the cases of very simple species which could be generated in chemically ¥ncomp~icated environments. Until nearly 1983, only HD+, HeH+, HJ ' H2D , and Ne~, HrH+, and ArH+ had been studied by high-resolut10n infrared techniques. lO On Christmas Eve of 1982, Gudeman, Begemann, and Pfaff performed the first velocity- + 3 modulation detection of a molecular ion spectrum (the vI band of HCO ).3

412 R. 1. SA YKALL Y

+ + + + + + . Subsequently, HNN , DNN , H10 , D30 , NH4 ' and HZF were studled at Berkeley by this method,34-41 in which the acceleration of an ion to drift velocities near 300 m/sec. in a discharge plasma and the result­ing Doppler shift in its transition frequencies is exploited to switch vibration-rotation transitions into and out of resonance with a color­center laser beam in an AC discharge cell, thereby permitting ions to be detected with very high sensitivity in such plasmas without inter­ference from neutral molecules through the use of simple phase sensi­tive detection techniques. The vel~4!ty-modulation method was also used by ~ka and his collaborators4Z in recent studies of NH4+' H30+, and HCNH. Subsequent to the detection of these species by velocity modu*ation, Amano and coworkers 47 - 49 have reported spectra of HCO+, DCNH , and HZD+ by the multipassed direct a~sorption method with a differ­ence frequency laser first developed bi Oka27 in his original study of H3+' and McKellar and coworkers 50 ,5 have reported the ~easurement of lower frequency heavy atom stretching vibrations of HCO , DCC+, HN~, and DNN+ with multipassed diode laser spectroscopy of a hollow cathode discharge. In these latter two methods, velocity modulation discrimination against neutral was not employed, but the high frequen-cy discharge amplitude modulation afforded some degree of discrimination against the longer-lived neutrals generated in the plasmas, which might otherwise obscure the ion spectra. The information with direct rele­vance to astrophysics obtained in these very recent studies is summa­rized in the following paragraphs.

Astrophysical Implications

A. HCl+ The cosmic elemental abundance58 of chlorine is -5xl0-6 , that of

HZ, about the same as that of phosphorous. The ionization potential of Cl is lZ.7 ev, 0.9 ev less than that of hydrogen, implying that a substantial fraction of interstellar chlorine may be ~resent in ionized form. Neutral HCl has been detected by Blake+et al. S in the ISM. HCl+ can be produced by the fast reaction HZ CI+ ~ HCI++H. Hence HCI+ may be expected in regions with low density and high exci­tation. A search for the lambda-doubling transitions in HCI+, analogous to those already detected for OH and CH, must be made at RF frequencies; this constitutes a difficult endeavor, requiring a very large antenna. The lowest rotational transition (J=3/Z ~ 5/Z) occurs near Z07.6~m, making a search at FIR frequencies possible with modern FIR techniques. Both are viable Bossibilities, given the large predicted dipole moment of H~l+ (1.8D).6 Because HCI+ will react rapidly with HZ to give HZCl , much of the ionized chlorine is likely to be present as the triatomic ion in denser sources. Although this species has yet to be detected by spectroscopic methods, the recent study of t~g VI and v3 vibrations of the isovalent ion H2~ made at Berkeley38, indicates that the infrared spectrum of HZCI+ will very likely be observed in the near future, providing predictions of rotational transitions which could be detectable by microwave or sub-millimeter astronomy.

STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS 413

B. HCO+(DC~+) The HCO ion has been studied extensively in the millimeter region

by the R.C. Woods group at Wisconsin. 61 In fact, HCO+ is the first polyatomic ion to yield a microwave equilibrium structure. Gudeman and Woods62 have observed the J=O+: rotational transition in all three singly excited vibrational modes, thus providing the capability to study vibrationally excited ions in interstellar space. In the case of HCO+, the recent vibrational spectroscopy studies provide a precise determination of the V1(CH and CD) and V3(CO) stretching fre­quencies, but do not improve on the microwave determinations of rota­tioanl transition frequencies.

C. HNW (Dmrt) Curiously, the microwave studies63 made for ground state HNW and

DNN+ have not been extended to excited vibrational states. Hence, the Berkeley studies of the VI fundamentals and bend-excited hot bands of both species44 ,41,64 and the Ottawa studies51 of the V3(N-N) vibra­tions, now produce reliable predictions for the astrophysically impor­tant J=O+l rotational transiti~n in all singly excited fundamentals, and in the doubly excited (1 1 0) mode. Moreover, the P,-doubling fre­quencies in the v 2=1 modes of both ions are accurately predicted to in the region near 30 GHz.41,64

D. 44 The Oka group's recent study of protonated hydrogen cyanide,

followed by Amano's work49 on DCNH+, provide good predictions of the astrophysically accessible rotational transitions for these important ions. The difficulty with detecting these sEecies is that their dipole moments are very small--approximately 0.3 D. 5-67 This situation is improved somewhat for the HCND 6 in which the center of mass is shifted away from the center of change, 7 increasing ~he dipole moment some­what. Because of this small polarity of HCNH , searches made for its laboratory microwave spectrum have not yet been successful. 68

E. H30+ While the extremely important hydronium ion escaped detection by

high-resolution spectroscopy until 1983, measurements on three of its four allowed normal vibrations have now been carried out by velocity­modulation spectroscopy. The VI and v3 bands (symmetric and antisymmetric stretches respectively) have been observed for H30+ and D 0+ at Berkeley,40,69 while the V2 (nondegenerate bend) mode has ~een studied at Chicago. 43 Experiments on the asymmetric deutrated forms are also in progress at Berkeley.69

Astronomical detection of H30+, or one of its isomeric forms, could be accomplished by far-infrared observations of its rotation-inversion transitions, by microwave or FIR observation of rotational transitions in the asymmetric isomers, or by infrared detection of vibrational tran­sitions, most likely the V~ band, near 1000 em-I.

F. NH + Th~ v3 mode (triply degenerate asymmetric stretch) of ammonium

414 R. J. SAYKALLY

was detected initially by both my group36 and by Oka's group.42 An extensive study of this complicated spectrum has been carried out by Schaefer, Robiette, and myself,37 leading to accurate predictions of the forbidden transitions, necessary to determine the structure of this molecule. Croften and Oka70 have recently reported the observation of these forbidden transitions.

Detection of NH4+ in the ISM would probably best be accomplished by FIR observation of the rotation-inversion spectrum of the symmetric top isotopomer NH3D+, which will be quite strongly allowed by the shift of the center-of-mass away from the the center of charge.

G. OH+ The ground state of the hydroxylion is X3 L-. It has been detected

only very recently by high-reso2~t~~n insZsuments, in discharges of He and H20,25 and in Hand 02.' Oka has reported observation of its vibrational funaamental by velocity modulation spectroscopy, and Gruebele, Muller and myself 2J have measured three different rota­tional transitions with six FIR laser lines by laser magnetic resonance. A typical LMR rotational spectrum measured at Berkeley is shown in Figure 6. Unfortunately, a detailed analysis of these LMR spectra is still in progress, but clearly these measurements of rotational spectra of O~ will provide precise predictions of the FIR transitions from the lowest rotational state, as well as of the fine-structure and hyper­fine splitting patterns, which will serve to unambiguously identify astrophysical spectra of this species.

Figure 6.

f2

OH­.1,". y-o .-1-2 J·2-,

14

1.1160" ..

IS f6

A typical LMR spectrum of OH+, observed in a H2/02 plasma.

STUDIES OF ASTROPHYSICALL Y IMPORTANT MOLECULAR IONS

Acknowledgements The LMR effort at Berkeley is supported by the Director, Office

of Basic Research, U.S. Department of Energy, with equipment provided by the National Science Foundation. The infrared velocity modulation spectroscopy work is supported by the National Science Foundation (CHE 84-02861) and the Petroleum Research Fund. Postdoctoral support was provided by the California Space Institute. I thank all of my students and postdocs at Berkeley who did the work described in this paper.

415

416 R. J. SAYKALL Y

REFERENCES

1. P.M. Solomon and W. Klemperer, Astrophys. J. 178, 389 (1972).

2. E. Herbst and W. Klemperer, Astrophys. J. 185, 505 (1973).

3. G.F. Mitchell, J.L. Ginsberg, and P.J. Kuntz, Astrophys. J. Suppl. 38, 39 (1978).

4. S.S. Prasad and W.T. Huntress, Jr., Astrophys. J. Suppl. ~, 1 (1980).

5. W.D. Watson, Astrophys. J. 188, 35 (1974).; Astrophys. J. 189, 22 (1974).

6. R.C. Woods, Private Communication.

7. P. Thaddeus. M. Guelin, and R.A. Linke, Astrophys. J. 246, L41 (1981); J.L. Destombes, C. Demuynck, and M. Bogey, preprint.

8. R.J. Saykally and R.C. Woods, Ann. Rev. Phys. Chem. 11, 403 (1981).

9. T.A. Miller and V. Bondybey, editors, Molecular Ions: Structure and Chemistry, North-Holland, Amsterdam (1983).

10. C.S. Gudeman and R.J. Saykally, Ann. Rev. Phys. Chem. 12, 387 (1984).

11. Billingsley, Chem. Phys. Lett. 11, 160 (1973).

12. H.A. Schwarz, J. Chem. Phys. ~, 5525 (1977).

13. H.A. Schwarz, J. Chem. Phys. ]2, 284 (1980).

14. W.H. Wing, G.A. Ruff, W.E. Lamb, and J.J. Spezeski, Phys. Rev. Lett. 36, 1488 (1976).

15. D.E. Tolliver, G.A. Kyrala, and W.H. Wing, Phys. Rev. Lett. 43, 1719 (1979).

16. J.T. Shy, J.W. Farley, W.E. Lamb, and W.H. Wing, Phys. Rev. Lett. 45, 535 (1980).

17. J.T. Shy, J.W. Farley, and \-l.H. Wing, Phys. Rev. ! 24,1146 (1981).

18. A. Carrington and J. Buttenshaw, Mol. Phys. 44, 267 (1981); A. Carrington, J. Buttenshaw, and R. Kennedy~Mol. Phys. 48 775 (1983); J. Mol. Structure 80, 47 (1982).

STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS

19. A. Carrington, J. Buttenshaw, R. Kennedy, and T.P. Softley, Mol. Phys. 44, 1233 (1981); A. Carrington, R. Kennedy, T. Softley, P. Fournier, and E.G. Richard, Chern. Phys. ~, 25 (1983).

20. A. Carrington, J. Buttenshaw, and R. Kennedy, Mol. Phys. ~, 747 (1982) .

417

21. A. Carrington, J. Buttenshaw, and R. Kennedy, Mol. Phys. 45. 753 (1982); A. Carrington and R. Kennedy, in Ions and Light, ed. M. Bowers (1982); A. Carrington, Proc. Roy. Soc. London A367, 433 (1979); A. Carrington and R. Kennedy, J. Chern. Phys. 81, 91 (1984).

22. R.J. Saykally and K.M. Evenson, Phys. Rev. Lett. 43, 515 (1979).

23. K.G. Lubic, D. Ray, and R.J. Saykally, Mol. Phys. 46, 217 (1982).

24. D.C. Hovde, E. Schafer, S.E. Strahan, C.A. Ferrari, D. Ray, K.G. Lubic, and R.J. Saykally, Mol. Phys. ~, 245 (1984).

25. M. Grueble, R.P. Mueller, and R.J. Saykally, in preparation.

26. S. Strahan, R.P. Mueller, and R.J. Saykally, in preparation.

27. T. Oka, Phys. Rev. Lett. ~, 53 (1980).

28. p. Bernath and T. Arnano, Phys. Rev. Lett. 48, 20 (1982).

29. M. Wong, P. Bernath, and T. Arnano, J. Chern. Phys. 1l, 693 (1982).

30. J.W. Brault and S.P. Davis, Physica Scripta~, 268 (1982).

31. J.W.C. Johns, J. Mol. Spec. 106, 124 (1984).

32. F.C. Van den Heuvel and A. Dyrnanus, Chern. Phys. Lett. 21, 219 (1982).

33. C.S. Gudernan, M.H. Begernann, J. Pfaff, and R.J. Saykally, Phys. Rev. Lett. 50, 727 (1983).

34. C.S. Gudernan, M.H. Begernann, J. Pfaff, and R.J. Saykally, J. Chern. Phys. ~, 5837 (1983).

35. M.H. Begernann, C.S. Gudernan, J. Pfaff, and R.J. Saykally, Phys. Rev. Lett. ~, 554 (1983).

36. E. Schafer, M.H. Begernann, C.S. Gudernan, and R.J. Saykally, J. Chern. Phys. ~, 3159 (1983).

37; E. Schafer, R.J. Saykally, and A.G. Robiette, J. Chern. Phys. 80, 3969 (1984).

418 R. J. SAYKALLY

38. E. Schafer and R.J. Saykally, J. Chem. Phys. 80, 2973 (1984).

39. E. Schafer and R.J. Saykally, J. Chem. Phys., in press.

40. D.J. Nesbitt, H. Petek, C.S. Gudeman, R.J. Saykally, and C.B. Moore, in preparation.

41. D.J. Nesbitt, H. Petek, C.S. Gudeman, R.J. Saykally, and C.B. Moore, J. Chem. Phys., in preparation.

42. M.W. Crofton and T. Oka, J. Chem. Phys. 11, 3157 (1983).

43. N.N. Haese and T. Oka, J. Chem. Phys. 80, 572 (1984).

44. R.S. Altman, M.W. Crofton, and T. Oka, J. Chem. Phys. 80, 3911 (1984).

45. T. Oka, 39th Symposium on Molecular Spectroscopy, Columbus, Ohio; June, 1984.

46. P.B. Davies and W.J. Rothwell, J. Chem. Phys., in press.

47. T. Amano, J. Chem. Phys. 11, 3595 (1983).

48. T. Amano and J.K.G. Watson, 39th Symposium on Molecular Spectro­scopy, Columbus, Ohio; June, 1984, paper TE-3. K.G. Lubic and T. Amano, same conference, paper TE-13.

49. T. Amano, 39th Symposium on Molecular Spectroscopy, Columbus, Ohio; June, 1984, paper TE-7.

50. S.C. Foster, A.R.W. McKellar, and T.J. Sears, J. Chem. Phys. ~, 579 (1984).

51. S.C. Foster and A.R.W. McKellar, J. Chem. Phys., in press.

52. K.M. Evenson, R.J. Saykally, D.A. Jennings, R.F. Curl, Jr., and J.M. Brown, "Chemical and Biochemical Applications of Lasers, Vol. V," C.B. Moore, Editor (Academic Press, New York, 1980).

53. K.M. Evenson, Faraday Discussions Chem. Soc. 1l, 7 (1981).

54. R.J. Saykally, K.G. Lubic, and K.M. Evenson, in Molecular Ions: Geometric and Electronic Structures, J. Berkowitz and K. Groeneveld, eds. (Plenum Inc., 1980, New York).

55. K.G. Lubic, D. Ray, L. Veseth. and R.J. Saykally, J. Chem. Phys., submitted.

56. J.M. Brown and J.K.G. Watson, J. Mol. Spec. 65, 65 (1977); J.M. Brown, E.A. Colbourn, J.K.G. Watson, and F.D~Wayne, J. Mol. Spec.

STUDIES OF ASTROPHYSICALLY IMPORTANT MOLECULAR IONS 419

J..!!.., 294 (1979).

57. L. Veseth, J. Mol. Spec. ~, 51 (±976); 63, 180 (197~).

58. V. Trimble, Rev. Mod. Phys. ~, 877 (1975).

59. G. Blake, private communication.

60. S. Green, private communication.

61. R.C. Woods, T.A. Dixon, R.J. Saykally, and P.G. Szanto, Phys. Rev. Lett. 35, 1269 (1975); R.C. Woods, R.J. Saykally, T.G. Ander­son, T.A. Dixon, and P.G. Szanto, J. Chern. Phys. ~, 4256 (1981).

62. C.S. Gudeman and R.C. Woods, in preparation.

63. R.J. Saykally, T.A. Dixon, T.G. Anderson, P.G. Szanto, and R.C. Woods, Astrophys. J. 205, L101 (1976); P.G. Szanto, T.G. Anderson, R.J. Saykally, N.D. Piltch, T.A. Dixon, and R.C. Woods, J. Chern. Phys. ~, 4261 (1981).

64. J. Owrutsky, C.S. Gudeman, C.C. Martner, M.H. Begemann, N.H. Rosen­baum, L. Tack, and R.J. Saykally, in preparation.

65. N.N. Haese and R.C. Woods, Chern. Phys. Lett. &l, 396 (1979).

66. T.T. Lee and H.F. Schaefer, preprint.

67. P. Bottschwina, private communication.

68. R.C. Woods, private communication.

69. M.H. Begemann and R.J. Saykally, in preparation.

70. M. Crofton and T. Oka, private communication.

71. N.N. Haese, F.S. Pan, and T. Oka, Phys. Rev. Lett. 50, 1575 (1983).