Steven E. Wheeler

Postdoc: Houk Group

Previous Research (Schaefer group, CCQC)

Links are provided to relevant publications. A complete list of my publications is available here.

In an ongoing effort to produce definitive thermochemistry for key combustion intermediates, we have been computing high-accuracy enthalpies of formation and isomerization energies of key intermediates in soot formation. The intial targets were the disputed soot formation intermediates i- and n-C₄H₃ and C₄H₅. More recent work focused on the propargyl radical and other C₃H₃ isomers. More recent work involved the prediction of accurate thermochemistry for the HOSO radical, a key intermediate in NO_x chemistry during the combustion of sulfur containing fuels (ie: fossil fuels).

Second order Møller-Plesset perturbation theory (MP2) is the most commonly used electron correlation method. However, such approaches are applied under the implicit assumption that the perturbation series is convergent, or at the very least that the series is sufficiently well-behaved to yield good results at 2^nd order. The former assumption has been known to be largely false for some time now: Møller-Plesset perturbation series are rarely convergent for molecular systems with large basis sets. We have been exploring the underlying causes of oscillating and diverging closed-shell MP series. Additionally, by modifying the determinant-based full CI program DETCI, we have computed very high-order ZAPT energies (up to ZAPT1000 in one case), to compare ZAPTn energy series to other open-shell perturbation theories (RMP, ROMP, OPT1, OPT2, UMP, etc.). We have also provided the first comparison of ZAPT4-predicted bond lengths and harmonic vibrational frequencies to analogous ROMP4 and RMP4 predicted values and experimental results.

We have provided benchmark structures and atomization energies of small lithium clusters (Li_n) and hydrogenated lithium clusters (Li_nH) using coupled cluster theory with large basis sets. We have also predicted accurate ionization potentials of small Li_n and Li_nH clusters, revealing a substantial error in the accepted experimental IP for Li₄.

More recently, I completed work in collaboration with Michael Duncan (UGA) involving tunable VUV-photoionization measurements on small carbon clusters (C₃ through C₁₅). The experimental IPs were compared with predicted IPs derived using the focal point approach, providing conclusive evidence regarding the structures of C₈-C₁₀ in a molecular beam.

With Paul Schleyer, I have developed a "Symmetry Adapted Stochastic Search" algorithm (SASS), which utilizes framework groups to execute a simple stochastic search of complex quantum mechanical energy surfaces. In contrast to previously described stochastic search algorithms, SASS exploits point group symmetry in the generation of candidate structures, greatly increasing the efficieny with which the 3N-6 dimensional configuration space can be sampled and accelerating geometry optimizations of candidate structures.

I had the opportunity to work on many other chemical application projects at CCQC, including work on:

• Vinyl radical and fluorinated vinyl radical
• Electron affinities of radicals derived from DNA bases and base pairs:
  • Cytosine
  • Guanine-cytosine
  • Adenine-thymine
  • Thymine-water complexes
• Torsional energy surface of oxalyl chloride
• Pentacyanocyclopentadienyl radical and anion
• Protonated carbonyl sulfide (HOCS⁺ and HSCO⁺)
• Renner-Teller bending frequencies of A ²Π OCS⁺