The overall goal of this procedure is to fabricate and investigate metal lanthanum-aluminate-strontium type heterostructures using a combination of pulsed-laser deposition magnetron sputtering x-ray photoemission spectroscopy and magneto transport experiments. This method can help answer key questions about the interplay between electrostatic and chemical phenomenon of the quasi two-dimensional electron gas formed in these types of systems. The main advantages of this technique are it's simplicity and versatility, which together allow high-quality layer-by-layer growth of multi-power ultra-thin films.
First, fill a 40 kilohertz ultra sonic cleaner with water and heat the bath to 60 degrees Celsius. Add to a borosilicate glass beaker sufficient acetone to fill the beaker to at least 20%of it's maximum volume. Place in the beaker an out-of-the-box mix-terminated single-side polished 001 oriented STO single crystal substrate.
Sonicate the substrate in acetone for three minutes. Dry the substrate under a stream of nitrogen gas from a blow gun. Repeat the sonication in isopropanol and de-ionized water in sequence.
Mount the clean substrate in a polyvinylidene fluoride dipper. Fill another borosilicate glass beaker with flowing de-ionized water. Then, fill about 20%of a similarly sized polytetrafluoroethylene beaker with hydrofluoric acid buffered with ammonium fluoride.
Use the dipper to immerse the substrate in the buffered oxide etch for precisely 30 seconds. Immediately transfer the etched substrate to the beaker of flowing de-ionized water. Gently agitate the substrate in the flowing de-ionized water for two minutes.
Dry the substrate under a stream of nitrogen gas and anneal the etched substrate in an oxygen environment at 1000 degrees Celsius. Next, gently mechanically polish a single-crystal one-inch LAO target with 180-grit aluminum oxide sandpaper lubricated with isopropanol. Once the surface appears homogenous dry the target under a flow of nitrogen gas.
Mount the polished LAO target in a carousel that allows target rotation. Insert the carousel into the load-lock chamber, evacuate the chamber, and transfer the carousel to a pulsed-laser deposition chamber. Once the pulsed-laser deposition chamber pressure reaches 10 to the 9th millibars, supply oxygen to the chamber to achieve an oxygen partial pressure of 2x10 to the 4th millibars then place an excimer laser energy meter in the krypton fluoride laser beam path between the second converging lens and the quartz window.
Set the laser to an arbitrary frequency and read the energy on the meter. Select the target rotation, remove the energy meter, and pre-ablate the target at three or four hertz for 20, 000 pulses in order to clean it's surface. Prior to sample preparation, use atomic force microscopy to confirm the termination, morphology, and cleanness of the terminated STO substrate surface.
Use the conductive silver paste to fix the terminated STO substrate to a sample holder with the terminated surface face up, ensuring that the substrate is in the center of the holder. Heat the sample holder on a hot plate at 100 degrees Celsius for 10 minutes to evaporate solvent and solidify the paste for optimal thermal conduction. Then, allow the sample holder to cool to room temperature.
Insert the sample holder into the load lock chamber and transfer the holder to the x-ray photoelectron spectroscopy chamber. Align the sample with the surface normal parallel to the electron analyzer axis. Move the x-ray gun as close as possible to the sample without allowing mechanical contact between the gun and the sample holder.
After ensuring that the chamber is under ultra-high vacuum, turn on the x-ray gun. Acquire a survey spectrum between 12, 000 and 0 electron volts, with a selected step of 0.2 electron volts, a dwell time of 0.2 seconds, a pass energy between 30 and 60 electron volts, and the smallest spot size possible. Acquire a spectra of peaks of interest for later elemental analysis.
After initial XPS analysis of the sample, transfer the sample holder to the PLD chamber and position the holder so the substrate faces the LAO target. Supply oxygen to the chamber to achieve an oxygen partial pressure of 2x10 to the 4th millibars. Ramp the sample holder temperature to 730 degrees Celsius at 25 degrees Celsius per minute, then set the reflection high-energy electron diffraction gun to a source voltage of 30 kilovolts, and a current of about 50 microamps.
With the sample holder, place 63 millimeters away from the target, align the read electron beam with the substrate surface at a grazing angle between one and three degrees, so that the diffraction spots appear on the phosphor screen. Monitor the spot intensities in real-time via a CCD camera and image-analysis software. Once the reading stabilizes, start the laser.
Observe the plume and monitor the read oscillations. Stop the laser at the peak of an oscillation upon achieving the desired thickness. Shut down the read gun, reduce the holder temperature to 500 degrees Celsius, and increase the oxygen partial pressure of the chamber to 0.1 millibars.
Once the temperature and pressure are stable, increase the oxygen partial pressure to 300 millibars and anneal the sample at 500 degrees Celsius for 60 minutes. Analyze the LAO/STO sample with XPS. Transfer the sample holder with the LAO/STO substrate to the sputtering chamber under vacuum.
Position the holder so that the substrate is about 7cm from the cobalt target. Supply pure argon gas to the sputtering chamber. With the target shutter closed, ramp the current to about 100 milliamps to ignite the plasma.
Reduce the current to 80 milliamps and decrease the argon supply to 5.2 sccm, pre-sputter the cobalt target and then open the shutter and deposit cobalt for 25 seconds. Close the shutter to end deposition. For transport experiments, deposit a capping layer of three nanometers of aluminum to prevent oxidation of the underlying metal layer upon exposure to air.
Otherwise, analyze the sample with XPS after cobalt deposition. Remove the sample holder from the system and transfer the sample to a transport measurement holder. On the sample stage of an ultrasonic wedge bonding machine, use aluminum or gold wires that are 0.025 millimeters in diameter to wire-bond the four corners of the sample to one channel of the transport measurement holder in hull geometry.
Ensure that the wires stick well to the sample's surface. Then, wire bond the four corners to a second channel in vanderpow geometry. Check the contacts with a multimeter.
Mount the holder in a transport setup and measure the magneto resistance and hull effect at various temperatures. Magneto transport measurements showed strikingly varied behavior depending on the metal layer deposited. LAO/STO samples capped with reactive metals showed s-shaped hull traces, indicating that an inter-facial quaside two-dimensional electron system had formed.
Similar two-unit cell LAO/STO samples capped with noble metals, showed linear hull traces with resistance changes of only a few tens of milliohms over 4 teslas. Consistent with an insulating interface at which only the metal layer was detected. XPS of a tantalum-capped one unit cell LAO/STO sample showed tantalum oxide peaks and partial to total suppression of the metallic tantalum feature, suggesting that oxygen tended to diffuse toward the metal from the perovskite surface.
Titanium atoms hosted some of the electrons released to the lattice by the formation of oxygen vacancies at the perovskite surface, changing the titanium valence state from 4 to 3, and forming a quaside 2D electron system. The titanium 3+to titanium 4+intensity ratio was about 20%Changing the electron take-off angle from 0 to 50 degrees had a minimal effect on the titanium 3+intensity, indicating that the quaside 2D electron system extended beyond the maximal XPS probing depth of five nanometers. Once mastered, the full procedure for one sample can be done in about one or two days.
While attempting to perform PLD growth, pay special attention to the quality of both the substrates and the targets used. Perform extra analysis if the samples grown do not show the expected properties. Due to it's versatility, this procedure allows the highly-controlled creation and analysis of virtually any type of oxide heterostructures, as well as the possibility to explore a multitude of new interfacial and surface physical effects.
