Department of Chemistry, Machine VP
The University of Newcastle
Rovibrational Calculations of Bent Triatomic Molecules
The aim of this project is to carry out a thorough and systematic investigation of the electronic and structural properties of electronic and excited states of the XYO(+,n) and XYO(2+,n) ions (where X=first or second row atoms and Y=first and second row atoms). The initial focus was on the alkali metal vapours. The motivation for their study was due to their predicted importance in such diverse technologies as: arc lamps (Na); flash lamps (Cs, K); high resolution optically pumped lasers (Li2, Na2, K2); continuous wave tunable infrared lasers; solar pumped lasers; discharge pump high energy laser or laser amplifiers; infrared photography (K,Rb); use as coolants (Li, Na, K); tritium breeders (6LI,7Li); measurement of tritium inventory in molten lithium (Li2,LiT); shields for fusion reactors (Li liquid); neutral beam injection charge exchange cells (Na, Cs); alternative negative ion for neutral beam injection.
What are the basic questions addressed?
The specific aim is to calculate potential energy surfaces of ground electronic and excited states of the XYO(+,n) and XYO(2+,n) ions (where X=first or second row atoms and Y=first and second row atoms) in order to determine their rotationally resolved infrared spectrum. Attention also focused on calculating ab initio vibrational-rotational eigenenergies as well as evaluating rovibrational transitional probabilities. However, the initial research centred on the alkali metal vapours. The following basic questions were addressed.
(a) What are the strategies required to accurately solve the `complete' Schrodinger problem? Attempts to solve the `complete' molecular Schrodinger equation are not common place, even with the advent of supercomputers.
(b) What lasing states exist for the alkali metal vapours?
(c) What are their electronic and rovibrational properties?
What are the results to date and the future of the work?
The project was initially funded in 1992. The project is continuing to progressing well with the potential energy surfaces and rovibrational states of a number of alkali metal ions being characterised (as well as the rovibrational states of the water cation). There have been a number of papers published in international journals , conference papers and a successful completion of a number of post-graduate research projects arising from access to the VP2200. Moreover, a monograph has been published which summarizes the work completed to-date.
Work will now focus on the ground electronic and excited states of the XYO(+,n) and XYO(2+,n) ions (where X = C, N, O, F and Y = He or Ne) in order to determine their rotationally resolved infrared spectrum.
What computational techniques are used and why is a supercomputer required?
Solving the electronic Schrodinger eigenvalue problem for electron dense systems is at present an open-ended problem. To that end the GAUSSIAN suite of programmes enable `effective' truncations to be made to force tractable solutions. Using the GAUSSIAN programmes with valence double zeta (+polarisation) basis sets coupled with SDCI or MP4 level of theory enables computation of ab initio discrete potential energy surfaces of spectroscopic quality. For example, the computation time required to produce just one surface with a total of 50 points consumes at least 60 SU. Therefore five discrete electronic surfaces would require c.a. 300 SU.
Rovibrational States of Triatomic Silicon, F. Wang and Ellak I. von Nagy-Felsobuki, Theoretica Chimica Acta, 88 (1994) 131-145. J. Hayward, Honours project, The University of Newcastle, Newcastle (1994).
Ab Initio Calculations of Rovibrational States of Alkali Metal Ions, F. Wang, PhD thesis, The University of Newcastle, Newcastle (1994).