Back to Table of Contents
44481-G6
Electron Transport and Photoexcitation at Interfaces
Daniel S. Kosov, University of Maryland
The last decade has witnessed a remarkable miniaturization of conventional microelectronic devices. If this trend is to continue, elements of microelectronic circuits will soon shrink to the size of single molecules. One of the major goals in nanotechnology is the construction, measurement and modeling of electronic circuits in which molecular systems act as conducting elements. The grant from ACS PRF has supported our research on development of theoretical and computational methods for first principles simulations of electron transport characteristics of molecular-scale devices.
We studied computationally the electron transport properties of dithiocarboxylate terminated molecular junctions. Transport properties were computed self-consistently within density functional theory and nonequilibrium Green's functions formalism. On our calculations we established a microscopic origin of the experimentally observed current amplification by dithiocarboxylate anchoring groups. We predicted a new microscopic mechanism of rectification based on the electronic structure of asymmetrical anchoring groups. We demonstrated that the peaks in the transmission spectra of 4'-thiolato-biphenyl-4-dithiocarboxylate junction respond differently to the applied voltage. Depending upon the origin of a transmission resonance in the orbital interaction picture, its energy can be shifted along with the chemical potential of the electrode to which the molecule is more strongly or more weakly coupled.
We have been developing the computational methodology to model the typical experiments in molecular electronics where the molecular junctions are formed by repeatedly crashing an Au STM tip into and pulling it out of the Au surface in a solution comprised of the molecules intended to form junction. The preliminary results were described in our recent paper. To model these experimental conditions we developed and implemented into the SIESTA program the computational method which calculates molecular conductance ”on-the-fly”. Specifically, we explained by our calculation why amine-terminated molecules show well-behaved conductance in the scanning tunneling microscope breakjunction experimental measurements. We performed density functional theory based electron transport calculations to explain the nature of this phenomenon. We found that amines can be adsorbed only on the apex Au atom, while the thiolate group can be attached equally well to undercoordinated and clean Au surfaces. Our calculations showed that only one adsorption geometry is sterically and energetically possible for the amine anchored junction whereas three different adsorption geometries with very distinct transport properties are almost equally probable for the thiolate-anchored junction. We calculated the conductance as a function of the junction stretching when the molecules are pulled by the scanning tunneling microscope tip from the Au electrode. Our calculations demonstrated that the stretching of the thiolate-anchored junction during its formation is accompanied by significant electrode geometry distortion. The amine-anchored junctions exhibit very different behavior—the electrode remains intact when the scanning tunneling microscope tip stretches the junction.
With the support of the ACS PRF grant we initiated research project on the use of molecular rotation as a probe of complex environments (nonequilibrium condensed phase, nano-cavities, porous materials). We have developed a general approach for the calculation of the single molecule polarization correlation function C(t), which delivers a correlation of the emission dichroisms at time 0 and t. The key dynamic quantities of our analysis are the even-rank orientational correlation functions, the weighted sum of which yields C(t). We demonstrated that the use of nonorthogonal schemes for the detection of the single molecule polarization responses makes it possible to manipulate the weighting coefficients in the expansion of C(t). Thus valuable information about the orientational correlation functions of the rank higher than the second can be extracted from the single molecule polarization correlation function.
It has recently been shown that relaxation of the rotational energy of hot nonequilibrium photofragments (i) slows down significantly with the increase of their initial rotational temperature and (ii) differs dramatically from the relaxation of the equilibrium rotational energy correlation function, manifesting thereby the breakdown of the linear response. We demonstated that the cause of this phenomenon is the angular momentum dependence of rotational friction. We developed the generalized Fokker–Planck equation and extended J-diffussion model with angular momentum dependent rotational friction. The calculated rotational correlation functions correspond well to their counterparts obtained via molecular dynamics simulations in a broad range of initial nonequilibrium conditions. It is suggested that the angular momentum dependence of friction should be taken into account while describing rotational relaxation far from equilibrium. We proposed theoretical model, which describes chemical reaction in the condensed phase far from the equilibrium. To this end we study the influence of the velocity dependence of friction on the escape rate of a Brownian particle from the deep potential. We demonstrated that the effects due to the velocity-dependent friction may be of considerable importance in determining the rate of escape of an under- and moderately damped Brownian particle from a deep potential well, while they are of minor importance for an overdamped particle.
Back to top