All-electronic Nanosecond-resolved Scanning Tunneling Microscopy: Facilitating the Investigation of Single Dopant Charge Dynamics

The overall goal of electronic time-resolved scanning tunneling microscopy is to study dynamic processes occurring on the atomic scale. In these experiments, the charged dynamics of dopants in silicon are studied. These methods can be used to investigate interesting phenomena in the field of nanotechnology and physics such as the rate at which dopants in semiconductors are supplied with electrons or how tunneling currents can be used to drive the magnetic resonance of individual atoms.

The main advantage of these techniques is that fast dynamics can be studied without the need to integrate optics into the tunneling junction. These techniques are not limited to low temperatures or ultra-high vacuum scanning tunneling microscopes. To begin, cool the scanning tunneling microscope capable of ultra-high vacuum to cryogenic temperatures.

Ensure that the microscope's tip is equipped with high frequency wiring capable of about 500 megahertz. Then, connect an arbitrary function generator with at least two channels to the tip. Prepare the cycles of voltage pulse pairs used for pump-probe experiments by configuring the arbitrary function generator so that the pump and probe voltage pulses are generated independently and summed before being fed into the tip.

Next, apply the DC bias voltage used for imaging in conventional spectroscopy to the sample. Then, connect two radio frequency switches to the arbitrary function generator's output channels. Configure the switches so that the tip will be grounded during scanning, tunneling, imaging, and conventional spectroscopy.

Collect the tunneling current for all measurements through a preamplifier connected to the sample. Using a silicon carbide scriber, scratch the back of a silicon wafer. Then, use a pair of glass microscope slides to gently snap the sample off the wafer.

Make sure to use ceramic tweezers to handle silicon crystals. Affix the sample to a scanning, tunneling microscope holder. Once secured in the holder, introduce it to the ultra-high vacuum chamber.

Then, degas the sample by resistively heating it to 600 degrees Celsius and holding it at that temperature for at least six hours in the ultra-high vacuum. Once the sample has been cooled, use the same chamber to degas a tungsten filament by resistively heating the filament to 1, 800 degrees Celsius and waiting for the system to recover to base pressure. Then, remove the oxide from the sample surface by flashing the sample to 900 degrees Celsius and holding it at that temperature for 10 seconds before cooling it back to room temperature.

Progressively flash the sample to higher temperatures while attempting to reach a final flash of 1, 250 degrees Celsius. Let the sample cool to room temperature after each flash. To ensure the sample remains clean, abort flashes where the pressure rises above nine times 10 to the negative 10 Torr.

Then, leak hydrogen gas into the chamber at a pressure of one times 10 to the negative six Torr and heat the tungsten filament to 1, 800 degrees Celsius. This will crack the gas into atomic hydrogen. Hold the sample in these conditions for two minutes before flashing the sample back to 1, 250 degrees Celsius.

After five seconds at 1, 250 degrees Celsius, cool it back down to 330 degrees Celsius. Hold the temperature at 330 degrees Celsius for one minute and then simultaneously close the hydrogen leak valve and turn off the tungsten filament. Let the sample cool to room temperature.

Finally, approach the tip to sample by engaging the current feedback controller with a sample bias of negative 1.8 volts, the current setpoint of 50 picoamperes, and using the control software's auto approach function. Take area scans of the surface on the order of 50 by 50 nanometers to verify the surface's quality. There should be large terraces separated by single atom steps in a defect rate of less than 1%The dimer rows of the silicon one zero zero two by one reconstruction should be clearly resolved.

Dangling bonds appear as bright protrusions in field state images. Look for an area on the sample surface free from large surface defects by taking large area scans that are 50 nanometers by 50 nanometers. To characterize the electronic ringing, position the tip over a hydrogen atom on the surface.

Because each surface silicon atom is terminated with a hydrogen atom, this is equivalent to positioning the tip over the dimer rows on the surface. Turn off the current feedback controller and then set the DC voltage to minus 1.0 volts, then pump voltage to minus 0.5 volts, the probe voltage to minus 0.5 volts, the width of the pump and probe pulses to 200 nanoseconds, and the rise and fall time of the pulses to 2.5 nanoseconds. Then, send a series of trains of pump and probe pulses where the relative delay of the pump and probe is swept from minus 900 nanoseconds to 900 nanoseconds.

Ringing is manifested through smaller amplitude oscillations in the tunneling current on both sides of the origin. By increasing the rise times on the pulses from 2.5 nanoseconds to 25 nanoseconds, a strong suppression of the ringing is observed. Eliminate ringing by increasing the rise and fall times of the pulses to obtain the most accurate spectroscopic results.

However, be aware that increasing the width of the pulse will decrease the resolution. Position the tip over a silicon dangling bond which appear as bright protrusions at negative tip sample biases. Turn off the current feedback controller and send a train composed of only the probe pulse used as a reference signal with a repetition rate of 25 kilohertz.

Over a series of pulse trains, sweep the bias of the probe pulse over a range of 500 millivolts from the DC bias of negative 1.8 volts. Each pulse should have a different bias. This step is equivalent to standard IV spectroscopy but with reduced duty cycle.

Ensure that the width of the pulses are long enough to get a signal to noise ratio of at least 10. Send a train composed of pump pulses set at a fixed bias so that the DC voltage added to the pump voltage is greater than the threshold voltage with a repetition rate of 25 kilohertz. Next, send a train composed of the pump pulses with the probe pulses followed by a delay of 10 nanoseconds.

Compare the probe only in the pump and probe signals by plotting them in the same graph. Any hysteresis in little signals is an indication of dynamics that can be probed with time-resolved STM techniques. To measure relaxation dynamics, position the tip of the microscope over a silicon dangling bond and turn off the current feedback controller.

Then, send a train composed of pump pulses set at a fixed bias such that the DC voltage added to the pump voltage is greater than the threshold voltage with a repetition rate of 25 kilohertz. Next, send another train of pump and probe pulses. Ensure that the probe pulses have an amplitude smaller than the pumps and comparable to the range at which hysteresis occurs.

Finally, sweep the delay between the pump and probe pulse up to several tens of microseconds. For determination of excitation dynamics, again send a train composed of pump pulses so that DC voltage added to the pump voltage is greater than the threshold voltage with a repetition rate of 25 kilohertz. Sweep the duration of the pump pulse from several nanoseconds to several hundred nanoseconds.

Then, send a train of pump and probe pulses. The probe pulses should have an amplitude smaller than the pumps and comparable to the range at which hysteresis occurs. Using conventional scanning tunneling microscopy, silicon dangling bonds occur as bright protrusions in field state images negative sample bias.

In some cases, they display a sharp change in the conductance at minus two volts. For this specific dangling bond, images taken at minus 2.1 volts, minus two volts, and minus 1.8 volts also show this. In these time-resolved measurements, a pump-pulse transiently brings the system above the voltage threshold and immediately after a probe-pulse, interrogates the conductance of the transient state.

To maximize the visibility of the hysteresis, the duration of the probe-pulse should be shorter than the relaxation rate of the high conducted state. When measuring dopant relaxation dynamics, the pulse sequences vary in delay between the pump and probe. A single exponential decay is extracted which settles to a fixed offset with time.

For excitation times, the pulse sequence pump voltage width is changed. By sweeping the width of the pump, the average rate at which the dopant is ionized is mapped. It is important that you characterize any electronic ringing in your microscope setup.

Ringing can result in signal artifacts and must be eliminated. Also, be aware that eliminating ringing by extending pulse length results in a loss of time resolution and will need to be optimized. While we have demonstrated time-resolved scanning tunneling microscopy to study dopants in silicon, these techniques can be applied to many other physical systems.

The main advantage of these techniques is that no optical systems need to be integrated with your microscope. Don't forget that working with cryogens and high voltages can be dangerous. Always ensure a safe working environment and follow proper safety protocol when setting up and using your microscope.