Radio Frequency Scanning Tunneling Microscopy
Scanning tunneling microscopy (STM) has shown itself to be one of the most important and ubiquitous tools in nanoscience. This microscopy technique routinely shows atomic lateral resolution and sub-atomic surface morphology sensitivity. However a major limitation of existing systems is milli-second time resolution. This is due to the high tip-to-sample tunnel impedance combined with instrumentation capacitance (cables and amplifiers) which produce audio frequency detection time-scales. This is far from the fundamental bandwidth limit which is given by the tunneling rate of the electrons, typically 100ps to 1nsec for typical currents.
In order to improve this situation, we have improved the electronic detection by employing a standard tool of radio-frequency engineering: impedance transformation with a high frequency resonant circuit. We have imbedded the high impedance STM tip into a LC network which transforms the impedance at the resonant frequency to a much lower value, a value much closer to the characteristic impedance of high-frequency measurement systems. This is the same technique demonstrated by the Yale and Chalmers group to improve the readout of single electron transistors (rf SET).
This work was accomplished by partnering with Prof. Kamil Ekinci at Boston University: Kamil’s group provided the expertise with STM, with my group providing the rf engineering expertise as a result of our experience with the rf SET. Together with Kamil as PI and myself as Co-PI we received an NSF instrumentation grant in 2003 to develop this technique. The experiments were carried-out in Kamil’s laboratory and began to show success in the summer of 2006 with his PhD student Utku Kemiktarak.
We have shown three applications of the rf STM technique. Firstly, we have shown that the detection bandwidth can be as high as a 10 MHz. Using this bandwidth we have shown imaging scan rates of 10 microns/second with sensitivity to resolve atomic steps on a gold surface. Secondly, we have shown that fast thermometry is possible by measuring the high frequency noise generated by the electron tunneling. Thirdly, we have shown that it is possible to detect radio-frequency nanomechanical motion with femto-meter per root Hertz sensitivity. Both the noise thermometry and the motion sensitivity will enable interesting thermal and mechanical motion imaging in future work.
This technique has been patented with BU as the lead institution.
