My Ph.D research was to study the properties and characteristics of semiconductor materials, specifically surface features and corresponding deposition growth sites in Silicon and GaAs. I employed ab initio DFT methods, highly realistic potentials and specialized electron correlation treatments to also examine band structures of novel materials. The models that we use to predict the transport properties of new materials suffers from under-predicting the experimentally determined energy gap. Some work I did was aimed at attempting to correct this problem, by treating the electronic interactions more realistically. (These models are highly simplified because computing time can increase with the cube of the number of particles you want to simulate.) The model is called the "Coulomb hole" and it tries to model the way electrons repel each other. It's based on the fact that one single electron repels all the other electrons in the material, but if you pick a different single electron, the same thing is happening. So the charge can't go too far -- in fact, it goes away by some characteristic distance (the radius of the "hole") and then piles up on the other side. Imagine digging a hole and throwing the dirt out. You'd have a hole next to a big pile of dirt, whose volume must be the same as that of the hole. (This is how I incorporated charge conservation into the model.) If you're interested (or incredibly bored), see more in the first paper listed below.

    I also did some beginning investigation into the conductive properties of spin-dependent half-metallic transition metal compounds. This field is now combining with electronics to be known as "spintronics". How can something be "half metallic" you ask? As with many scientific words, the word "metal" actually means something very specific and precise. It means a material whose Fermi level (the energy that divides conduction electrons from valence electrons) is located inside a partially filled band. Thus, there is NO energy gap separating the valence bands and the conduction bands. Therefore, many electrons can be thermally excited to higher energy levels and becoming conducting with great ease. In the case of a "half metal", the conducting states and valence states depend on the spin of the electrons. Electrons have only two spin possibilities: up or down. Suppose the spin-up electrons have a Fermi level that lies within a partially filled spin-up band, while the spin-down electrons have a Fermi level that lies within a gap separating the filled (valence) states from the unfilled (conduction) states. Then you'd have a material that conducts easily, but only with spin-up electrons. Thus, this is a "half metal" because only half (or some fraction) of the electrons are able to act like a metal and conduct heat or electricity easily.

    These materials will become highly useful in the future, because we can use them to create spin polarized currents, or as spin polarized current switches. All that needs to happen for us to create a spin polarized current is to pass electrons of both spin types through a half metal. This can prevent the spin-down electrons from passing, but allow the spin-up electrons to speed through. Then, by varying the applied voltage on the half metal, we can push the Fermi levels up and down, which means we can pass the spin-down electrons while blocking the spin-up electrons. Thus, if a spin-up current is being used, we've just shut it off. Pretty neat!!


    • M.D. Watson, C.Y. Fong, Solid State Communications, 124, 12, pp. 457-461, 2002. Full text (pdf).
    • C.Y. Fong, M.D. Watson, L.H. Yang, S. Ciraci, Modelling and Simulation in Material Science and Engineering, 10, 5, pp. R61-R77, 2002.
    • C.D. Consorte, C.Y. Fong, M.D. Watson, L.H. Yang, S. Ciraci, Materials Science & Engineering B, 96, pp. 141-144, 2002.
    • C.D. Consorte, C.Y. Fong, M.D. Watson, L.H. Yang, S. Ciraci, Physical Review B, 63, pp. 041301R, 2001. Full text (pdf).
    • Ph.D Dissertation Title: Investigations of the Reconstruction and Growth on the Si (100) Surface, and Studies of an Interelectronic Correlation Function (2001).


    Studies of air quality have shown that the gas-phase pollution cycles are well-understood. Currently, the major focus is on aerosol particulate matter (PM). PM can range from a collection of a few molecules to 10 microns in diameter (for comparison, human hair has a diameter of around 60 microns). During inhalation, the lung has problems not due to more mass of PM, but rather more number concentrations of PM. By far the highest number concentration of PM is located in the sub-micron size range. PM is emitted from (for example) hot diesel exhaust, and these small particles can form nucleation sites for condensation growth to larger size ranges. I modeled regional emissions and investigated the growth of nuclei and establishment of secondary PM, and how different sources in the region contribute to this secondary aerosol PM in the atmosphere.


Charge density plot of bulk Silicon



Graph of the Coulomb hole function



Charge density plot of layered GaAs.



A half-metal schematic band structure.


This photograph was taken by a former student during her trip to Japan. We had just discussed the principles of reflection and she saw this!

Question: why do we see the underneath of the bridge?







©2013 M. D. Watson

Homepage quick returns: