International Workshop on Atomic Interactions in Laser Fields - Abstracts


Laser Spectroscopic Studies of Metal atom--Hydrocarbon Interactions

P.D. Kleiber, J. Chen, T.H. Wong, Y.C. Cheng, D.A. Olsgaard, and J. Holmes

Department of Physics and Astronomy, University of Iowa, Iowa City, IA 52242

 

Abstract

Photodissociation spectroscopy serves to mimic a bimolecular "half-collision", and provides an especially fruitful experimental approach to the study of excited state molecular interactions and dynamics. We have applied this approach to study the chemical interactions of metal ions with small alkane and alkene hydrocarbons. Our studies give insight into the orbital interactions and nonadiabatic effects that dominate the chemical dynamics and show that there are fundamental similarities in the factors that control the activation of H-H and C-H σ-bonds, and C-C π-bonds, despite the greater complexity and reduced symmetry of the metal-hydrocarbon systems.

Introduction

Our research focuses on studies of excited state molecular dynamics, including both chemically reactive and nonreactive (energy transfer) processes, using state-resolved "half-collision" methods [1-12]. Our work may be divided into two related project areas:

  1. studies of excited state metal atom chemical dynamics by scattering state spectroscopy [1-6];
  2. photodissociation spectroscopy of weakly bound bimolecular complexes [7-12].

Far-wing absorption spectroscopy offers a direct probe of the continuum or "scattering states" of a transient bimolecular collision complex [1]. These techniques can be used to selectively excite quasimolecular electronic states of well-defined symmetry, corresponding to a specific electronic orbital alignment of the reagents within the transient reaction complex. This entrance channel selectivity allows novel insight into excited state chemical dynamics. Continuum absorption spectra are sensitive to the excited state potential energy surfaces and can be used, for example, to measure local activation barriers or identify and characterize saddle point regions in the reactive entrance channel. Final state-resolved measurements of the action spectra are sensitive to the dynamical evolution in the excited state, from the point of Franck-Condon excitation in the entrance channel into the distribution of product channels, and can give unique insight into molecular orbital interactions and alignment dependence, nuclear motion dynamics, and nonadiabatic effects on the final state branching and energy partitioning. We have used scattering state spectroscopy to investigate energy transfer in metal atom-rare gas collisions [2], and the activation of H-H [3,4] and C-H [5] σ-bonds in metal atom-H2 and -CH4 reactive collisions.

An alternative (and complementary) approach to the study of excited state molecular dynamics is through the photolysis of a weakly bound bimolecular precursor complex [7-12]. Photodissociation serves to mimic a bimolecular "half-collision", and provides an especially fruitful experimental approach to the study of molecular interactions and dynamics. Weakly bound bimolecular complexes are formed in a supersonic molecular beam source, mass-selected and studied by laser photodissociation spectroscopy in an angular reflectron time-of-flight-mass spectrometer. The structure of the stable precursor can usually be determined through a combination of bound state spectroscopy and electronic structure calculation. Following absorption, the excited state half-collision begins from a well-defined geometry and electronic orbital alignment, and with a restricted range of collision energies and relative angular momenta. Spectral measurement of the photodissociative product yield, translational energy partitioning, and vector anisotropy, give information about the bonding and structure of the complex, intermediate state lifetime, and the dominant reactive and nonreactive quenching pathways including insight into the molecular orbital interactions, nuclear motion dynamics and electronic nonadiabatic interactions that couple adiabatic Born-Oppenheimer potential energy surfaces. In support ofthis experimental work we have also carried out extensive ab initio electronic structure calculations using the Gaussian '94 and GAMESS platforms for several small metal ion-molecule complexes [7-12].

We have used this approach to investigate metal ion interactions with small molecules: H2, and small alkane and alkene hydrocarbons, and it is this work we would like to emphasize in this presentation [7-12]. Our studies of metal atom -H2, -CH4, and -C2H4 dynamics show that there are fundamental similarities in the orbital interactions and dynamical factors that control the activation of H-H and C-H σ-bonds, and C-C π-bonds, despite the greater complexity and reduced symmetry of the metal-hydrocarbon systems.

Results and Discussion

(i) Metal ion-CH4 chemistry [9,10]

Following earlier work on the photodissociation spectroscopy of Mg+(D2) [8], we have investigated the chemical dynamics ofthe weakly bound bimolecular complexes M+(CH4) (M = Mg and Ca) [9,10]. The process can be schematically represented

  M+(CH4) + hν → [M+*(pπ or pσ)CH4] → M+(p)* + CH4 (1a)
or → M+(nl) + CH4+ (1b)
  → MH+ + CH3 (1c)
  → MCH3+ + H (1d)

For Mg+(CH4), a broad continuum absorption spectrum is consistent with a large geometry change on excitation and fast dissociation [9]. We observe both nonreactive (Mg+), and reactive fragmentation products (MgH+ and MgCH3+). C-H bond attack follows on the attractive "Π-like" (2E) surfaces. The dominant reaction product is the methyl fragment, MgCH3+, showing that insertive geometries control the ion-molecule reaction. Measurements of the nonreactive (Mg+) product translational energy release show that Mg+ kicks impulsively off the highly extended H-CH3 bond, so that (even when the insertion reaction does not occur) the nascent CH4 product is left highly vibrationally and rotationally excited [9].

The work was supported by ab initio electronic structure calculations of the ground and low lying excited state potential energy surfaces [9,13]. Absorption is assigned to the metal-centered transition (12E ← 12A1) in C3v geometry, followed by a geometrical relaxation of the complex to states of 2B1 and 2B2 symmetry in η2 coordination. Our ab initio calculations find a region of avoided surface crossing between the lowest two surfaces of A' symmetry allowing a path for efficient nonadiabatic relaxation to the ground state surface through a highly stretched H-Mg+-CH3 insertive transition state. Calculations show the transition state is evenly centered with the Mg+-H and the Mg+-C distances close to their equilibrium values in isolated MgH+ and MgCH3+, respectively. The close approach in the Mg+-CH4 complex may be facilitated by the lack of Mg+(s)-CH(σ) repulsion and the possibility for efficient donation of electron density from the occupied bonding CH(σ) orbital into the unoccupied s-orbital of the metal ion.

We have also completed studies of the structure of Ca+(CH4) using photodissociation spectroscopy [10]. These results offer an intriguing comparison with the Mg+(CH4) work. In contrast to the Mg+ - case, the Ca+(CH4) photodissociation spectrum shows evidence for complex rovibrational structure. No reactive quenching product is observed. The action spectrum exhibits a long progression in the Ca+-CH4 intermolecular stretch, with rotational substructure, and evidence for weaker intermolecular bending vibrations in the excited 2E state. Spectroscopic constants and supporting ab initio electronic structure calculations show CaCH4+ to be weakly bound in a C3v geometry, in both ground and electronically excited states. Calculated spectroscopic constants are in good agreement with the experimental observations [10].

The simplest explanation for the dramatic differences observed in the Mg+- and Ca+-CH4 interactions is that the larger size of the Ca+ orbitals hinder close approach to the C-H bond in the complex, by Pauli repulsion. This constraint on the approach, and the larger pπ-orbital size limits the overlap between the Ca+(4pπ) orbital and the C-H bond-centered CH(σ*) LUMO. Poor MO overlap results in weak chemical interaction and minimal stretching of the exposed C-H bond, which, in turn, ensures that the ground state surface in this configuration will not closely approach the excited state surface. Thus, the nonadiabatic transition rate at the surface "crossing" will be smaller and the excited state lifetime longer.

(ii) Metal ion-C2H4 chemistry [11,12]

We have carried out extensive studies of the UV/vis spectroscopy and photochemistry of π-bonded complexes M+(C2H4) (M = Mg, Ca & Al) [11,12]. Metal - π-bond interactions are important in many biological and organometallic chemical processes.In the case of Mg+(C2H4) the spectrum is complicated because metal-centered bands, ethylene-centered bands, and metal-ethylene charge transfer (CT) bands are all present [11]. Assignment of the 5 observed spectral bands is facilitated by electronic structure calculations that show Mg+(C2H4) to be weakly bound in the ground state in π-bonding geometry with Mg+ lying above the C=C bond of ethylene [10,14]. Three of the bands, 12B2, 12B1, and 22A1 ← 12A1, are assigned to the metal-centered Mg+(3p-3s) atomic transition. One of the remaining bands is assigned as 22B1 ← 12A1, a ligand-centered transition correlating with the a3B1u ← X1Ag forbidden band of C2H4. This observation shows that the cluster environment can significantly modify the radiative cross-section. The final band, 32A1 ← 12A1, is assigned to a metal-ligand CT transition.

The dramatic red shift in the 12B2 ← 12A1 band from the asymptotic resonance energy is indicative of the very strong chemical interaction in this state. In this orbital symmetry the Mg+ p-orbitals have favorable MO overlap with the π*-LUMO of C2H4, allowing for efficient donation of electron density into this antibonding orbital, and resulting in a weakening and stretching of the C=C bond. Consistent with this idea, the 12B2 ← 12A1 band is a broad continuum, indicative of a large geometry change and fast predissociation in the upper state [11]. A nonadiabatic dissociation mechanism involving C-C π-bond activation by Mg+(3p) is suggested by our ab-initio calculations which find a conical intersection surface crossing between the 12B2(A') and the 12A1(A') surfaces at short Mg-C2H4 distances when the C-C bond is allowed to stretch [11]. Reactive path calculations show that the coupling region should be accessible with no significant activation barrier. This is evidence that a bond stretch mechanism may also be important for the activation of C-C π-bonds [15].

In contrast, the 12B1 ← 12A1 band shows pronounced vibrational structure with a long progression in the Mg+-CH4 intermolecular stretch, and weaker progressions assigned to combination bands built on the intermolecular out-of-plane wag, and a CH2-CH2 wag [11]. Measurement of the fragment kinetic energy release determines the bond energies for the ground (D0"(Mg+-C2H4) = 0.7 eV) and 12B1 excited state, (D0'(Mg+-C2H4) = 1.8 eV). Spectroscopic constants and bond energies are in very agreement with our ab-initio predictions [11,14].

Currently we are carrying out similar studies of the photodissociation of Ca+(C2H4). The work is still in progres s and (as of this writing) our results show a similar absorption band structure, but with a much smaller red shift for the continuum 12B2 ← 12A1 band. We interpret this difference as resulting from the lesser MO overlap in the Ca+(4p) state. Because of the larger size of the 4s-orbital in Ca+, the ground state equilibrium geometry shows a longer metal-hydrocarbon bond length. This coupled with the larger size of the 4p-orbitals limits the MO overlap and hence the chemical shift of the 2B2 symmetry state at the Franck-Condon geometry.

We have also extended these techniques to Al+(C2H4) [12], allowing a very interesting comparison with work on Mg+(C2H4) and Ca+(C2H4) [11,12]. Due to the high energy of the Al+ resonance excited state, absorption is dominated by the metal-ethylene CT process. The differences in valence electronic structure for Al+(3s2) and Mg+(3s) have a profound effect on the spectroscopy and dynamics. Again, the experimental work is supported by ab initio electronic structure calculations of the ground and low-lying excited states of the complex [11,12].

Electronic structure calculations show Al+ to be weakly bonded to ethylene in π-bonding geometry (similar to Mg+C2H4 although with much weaker bonding (D0 = 0.4 eV)) [12,16]. Based on these results the three observed molecular absorption bands are assigned to CT transitions that adiabatically correlate with the product channel Al(3s23p) + (C2H4)+. The distinct bands correspond to the differing alignments of the neutral ground state Al(3p) orbital with respect to the C-C bond of the ethylene ion. Note that photodissociation here represents a half-collision analog to the ground state CT reaction,

Al(3s23p) + (C2H4)+ → Al+(3s2) + (C2H4),

and the experiment probes the interaction of ground state neutral Al(3s23p) with C2H4+ [12].

The low energy 11B2 ← 11A1 band is a continuum, consistent with ab initio results that suggest rapid nonadiabatic dissociation through a conical intersection or narrow avoided surface crossing in A' symmetry can be facilitated by a stretch in the C-C bond of ethylene (as found previously in the Mg+(C2H4) case) [11,12]. These results suggest that hard chemistry (involving C-C or C-H bond breaking) may follow photoinduced CT excitation in some cases.

In contrast, the 11B1 ← 11A1 band shows pronounced vibrational structure, consistent with a long-lived, weakly bound excited state [12]. Spectroscopic analysis yields vibrational mode frequencies for the 11B1 excited state: observed modes have been assigned to the intermolecu1ar Al--C2H4 stretch, the Al--C2H4 out-of-plane wag, and two intramolecular ethylene modes. Multiple quanta excitation of the high frequency intramolecular vibrational modes of ethylene result from appreciable geometry change of C2H4 on ionization [12]

References

  1. P.D. Kleiber, in Chemical Dynamics and Kinetics of Small Radicals, edited by K. Liu and A. Wagner, (World Scientific, Singapore, 1996), p. 573.
  2. K.C. Lin, P.D. Kleiber, J.X. Wang, W.C. Stwalley, and S.R. Leone; S. Ananthamurthy, P.D. Kleiber, W.C. Stwalley, and K.C. Lin, J. Chem. Phys. 90, 7605 (1989).
  3. S. Bililign, P.D. Kleiber, W.R. Kearney, and K.M. Sando, J. Chem. Phys. 96, 218 (1992).
  4. P.D. Kleiber, A.M. Lyyra, K.M. Sando, S.P. Heneghan, and W.C. Stwalley, Phys. Rev. Lett. 54, 2003 (1985); P.D. Kleiber, A.M. Lyyra, K.M. Sando, V. Zafiropulos, and W.C. Stwalley, J. Chem. Phys. 85, 5493, (1986).
  5. T.H. Wong, K.H. Yang, and P.D. Kleiber, J. Chem. Phys. 110, 6743 (1999).
  6. T.H. Wong, C. Freel, and P.D. Kleiber, J. Chem. Phys. 108, 5723 (1998); T.H. Wong and P.D. Kleiber, J. Chem. Phys. 102, 6476 (1995).
  7. P.D. Kleiber and J. Chen, Int'l Rev. Phys. Chem. 17, 1 (1998).
  8. L.N. Ding, M.A. Young, P.D. Kleiber, W.C. Stwalley, and A.M. Lyyra, J. Phys. Chem. 97, 2181 (1993).
  9. Y.C. Cheng, J.Chen, L.N. Ding, P.D. Kleiber, and D.K. Liu, J. Chem. Phys. 104, 6452 (1996).
  10. J. Chen, Y.C. Cheng, P.D. Kleiber, J. Chem. Phys. 106, 3884 (1997).
  11. J. Chen, T.H. Wong, and P.D. Kleiber, Chem. Phys. Lett. 279, 185 (1997); J. Chen, T.H. Wong, K. Montgomery, Y.C. Cheng, and P.D. Kleiber, J. Chem. Phys. 108, 2285 ( 1998).
  12. J. Chen, T.H. Wong, Y.C. Cheng, and P.D. Kleiber, J. Chem. Phys. [in press]
  13. C.W. Bauschlicher, Jr., and M. Sodupe, Chem. Phys. Lett. 214, 489 (1993).
  14. M. Sodupe and C.W. Baushlicher, Chem. Phys. Lett. 185, 163 (1994).
  15. I.V. Hertel, in Dynamics of the Excited State, edited by K.P. Lawley (Wiley, New York, 1982), p. 475; P. Halvick, and D.G. Truhlar, J. Chem. Phys. 96, 2895 (1992); and references therein.
  16. D. Stoeckigt, J. Schwarz, and H. Schwarz, J. Phys. Chem. 100, 8786 (1996).