Intramolecular Charge Transfer

Ever since James Clerk Maxwell wrote down his famous equations, it has been known that an accelerating charge generates an electromagnetic (EM) field. Heinrich Hertz first verified this experimentally in 1887. The generation of an EM field from an accelerating charge is the basis behind light sources as specialized as free electron lasers and synchrotrons, or as mundane as communication transmitters. This is in fact the basis behind generation of terahertz (THz) pulses by photoconductive antennas.

The exact nature of the resulting EM field produced depends upon the details of the generation event. How long did the acceleration last? How large was it? In what direction did the charge accelerate? During intramolecular electron transfer, an electron is transferred from a donor to an acceptor site. Because the movement of charge generates an electromagnetic field, the most direct measurement of charge transfer is measurement of the field generated. In effect, the molecule broadcasts its dynamics to the world - we just need to listen in.

Time-domain THz technology can measure electric field transients directly in the time-domain with sub-picosecond resolution. We therefore use this technology to probe intramolecular charge transfer events. From these measurements we obtain two important results. First, the polarity of the emitted field determines the direction of charge transfer unambiguously. Second, the shape of the field encodes the dynamics of the charge transfer - a slower transfer rate produces a broader temporal pulse.

We have extensively studied the field emitted by intramolecular charge transfer in Betaine-30 (also known as Reichart's dye), and compared it to that of DMANS and biased GaAs. The molecular structures are shown to the right. Biased GaAs is used as a photoconductive antenna to generate THz radiation and serves as a reference for these studies. Photoexcitation of GaAs promotes electrons from the valance band into the conduction band. Once in the conduction band, the applied field accelerates the electrons. The acceleration is in the direction of the positive electrode, and the emitted field is therefore defined as positive.

The ground state dipole moment in Betaine-30 is 15 Debye and in the excited state it is -6 D. We partially orient the molecules in a static field, and upon photoexcitation charge transfer takes place in the opposite direction to the applied field. Thus, the emitted field has the opposite polarity of that generated by photoexciting GaAs. On the other hand, the excited state dipole in DMANS is in the same direction as its ground state dipole, but is much larger (31 D compared to 7 D). The charge transfer is therefore in the direction of the applied field, and its polarity is the same as that of biased GaAs.

In order to measure the emitted field from such an event, we must coherently excite many molecules, and all the fields emitted must constructively add up to give a large field which is measured. Furthermore, the molecules must have some degree of orientation, otherwise the emitted field from one molecule would cancel that of another molecule. We satisfy these two constraints by orienting the molecules in a static electric field and initiating the charge transfer with a ultrashort laser pulse (~100fs). A static field is not required if other methods exist to orient the molecules.

We model the generation of EM radiation using a finite-difference time domain (FDTD) pulse propagation method. The effects of generating and concurrently propagating the EM transient through a dispersive medium are accurately described. A non-linear least squares routine is used to fit the model to the observed data. The model also takes into account propagation through the cuvette walls and the detector crystal, both of which can distort the shape of the measured signal. We obtain the time-dependent polarization, from which we determine the electron transfer rates.

Future work will be to extend this method to systems that are more difficult to study by traditional means. The method described here is very general since it is the motion of the charge that generates the signal and does not rely on secondary effects.