(a) Sketched cross section of NSThM probe: The inner body consists of borosilicate glass, the diameter at the upper end being a few μm. A platinum (Pt) wire is incorporated inside, protruding from the glass by about 0.5–1 μm. The whole assembly is coated with a gold (Au) layer of 50–200 nm thickness. At the diverging point of the Pt wire and the glass, a thermocouple is formed. (b) SEM micrograph of a NSThM probe before any electron bombardment was applied. (c) The same probe as shown in (b), but after being heated up in mode 1 by an electron pulse of 230 ms duration ( ), while the maximum temperature reached was about 1220 K. The glass mantle is deformed, in some areas the Au vanished.
Scheme of electrical setup. The user can specify different parameters of the procedure using a PC. These parameters are submitted to a microcontroller, which in turn is connected to an ADC and two semiconductor switches. The ADC is used to convert analog signals regarding the temperature of the NSThM probe and the probe current from the filament to the probe into digital values and transmits them to the microcontroller. A filament is used as electron source, which, when heated by a current, thermally emits electrons. By applying a voltage between the filament and the probe, the former being on a negative potential with respect to the latter, electrons are accelerated toward the NSThM probe, thus heating it up. The legs of the thermocouple can be shorted for protection against electrostatic discharges. The microcontroller and the PC are galvanically insulated from the rest of the circuit, as during electron pulses all components within the dashed box are on the same electrical potential .
(Color online) Experimental data recorded during an electron bombardment procedure in mode 1. The plots in subfigures (a) and (b) show the evolution of the NSThM probe's temperature T and probe current over time. In subfigure (b), a magnified time scale is used, the state of the electronic switches is plotted additionally. At the beginning of the treatment, was enabled for 230 ms, yielding . During this phase, the short between the legs of the thermocouple was activated, hence the thermovoltage did not rise considerably, yielding a false reading of . After the deactivation of and the short, ceased and the thermovoltage rose, yielding a value of the probe's temperature of more than 1220 K. The probe cooled down to ambient temperature within approximately 2 s. SEM micrographs of the probe before and after the procedure are shown in Figs. 1(b) and 1(c) , respectively.
(Color online) Experimental data of an electron bombardment procedure in mode 2. The plots are organized as in Fig. 3 . In contrast to the data recorded during a mode 1 procedure, in mode 2 the probe's rising temperature T can be observed as shown in subfigure (a). This is due to short intervals of time during which and the short between the legs of the thermocouple are deactivated, allowing the measurement of the thermovoltage and hence the probe's temperature. The pulses can be identified using the lower plot of subfigure (b). The intermediate values of T and stem from relaxation processes after the toggling of and the short between the legs of the thermocouple.
(Color online) STM micrographs taken with the same NSThM probe on Au(111) in constant current mode ( , sample bias 500 mV, set point of tunnel current 500 pA). (a) Before electron bombardment, (b) after treatment in mode 2 during which the probe has been heated to more than 950 K. Less distortions and artifacts occurred in (b) than in subfigure (a). The sample was not sputtered or treated otherwise between the measurements.
Article metrics loading...
Full text loading...