Improving Thermoelectric Properties of Bi₂Te₃ Thin Films By Manganese Co-Sputtering

My research examine the change in the thermoelectric performance of the manganese-doped bismuth telluride thin film deposited through control co-sputtering. Existing methods failed to adequately examine the impact of the manganese doping on the bismuth telluride thin film. So this protocol addresses this gap through systematic investigation.

After placing the prepared and cleaned soda lime glass substrates on the rotating substrate holder, close the chamber door and initiate pump down. Apply gentle pressure around the chamber door at the start to ensure proper sealing and confirm the pressure drop on the gauge. Continue pumping until the chamber pressure reaches less than or equal to five times 10 to the power of negative three torr.

Switch on the cooling system and set the temperature to 288 Kelvin to ensure continuous circulation of recirculated deionized water through both sputter guns prior to plasma ignition. Introduce high purity argon gas and allow it to stabilize for approximately two to five minutes, or until it remains steady at greater than or equal to 3.9 standard cubic centimeters per minute. Set the substrate rotation to 10 RPM in a single fixed direction and maintain continuous rotation throughout pre-sputtering and deposition.

Power on the radiofrequency power supply and the matching network controller. Adjust the matching network settings for both radiofrequency guns using the minimum or maximum function until the load value and the tune value reach 50 watts. Switch the controller mode from manual to auto-matching.

Set the radiofrequency power to 50 watts on the bismuth telluride gun and 10 watts on the manganese gun before plasma ignition. Minimize reflected power before plasma ignition, keeping it at or below five watts. Ignite the plasma for both guns and confirm stable plasma by observing a steady glow maintained for approximately one to two minutes without visible flicker or sudden pressure spikes.

Once stable plasma was achieved at both guns, pre-sputter both targets for 15 minutes to remove surface contamination and native oxides after stable plasma is achieved. Remove the test substrates after pre-sputtering. Before loading the substrates for the deposition run, remove dust from the chamber interior and the sample holder region using a clean rubber blower.

Keep the viewing window protected with a clean Petri dish lid for plasma observation. Inspect the targets to ensure that they are properly seated, centered, and in good condition. Verify that each target is firmly clamped and properly centered within the sputter gun.

Place the cleaned substrates symmetrically near the outer region of the rotating holder at approximately 5 to 10 millimeters from the holder center. Align the longer side of each substrate parallel to the holder edge. Fix the target to substrate distance at 55 millimeters for both sputter guns.

Pump down the chamber and continue evacuation for at least five hours or until the base pressure stabilizes at the target value. Confirm base pressure stability when the pressure drift remains less than or equal to 5%over approximately 10 minutes and the base pressure reaches five times 10 to the power of negative five torr or lower. Stabilize the cooling system at 288 Kelvins after reaching base pressure.

Set the argon flow to four standard cubic centimeters per minute and allow it to stabilize. Maintain substrate rotation at 10 RPM. Confirm the matching network settings for both guns and enable auto-matching on the controller.

Initiate the bismuth telluride gun at 70 watts for easier plasma ignition. Increase the power by five watts every 10 seconds until reaching 100 watts. Start the deposition timer only after the bismuth telluride power readback remains at 100 watts for approximately 30 seconds and plasma stability is confirmed.

Deposit the films for 30 minutes while maintaining stable plasma conditions during the deposition process. Using the same parameters, repeat the deposition steps for manganese at radiofrequency powers of 10 watts and 15 watts using fresh substrates. Initiate shutdown through the system controller by switching off the radiofrequency output first.

Reduce the argon flow to 10 to 11 standard cubic centimeters per minute while keeping the cooling system active until the chamber is vented and the sputter guns return to near ambient temperature. Confirm that the turbomolecular pump has stopped completely and the speed reading is zero before venting the chamber. Vent the chamber slowly using dry nitrogen through the vent valve to allow gradual pressure, equalization, and minimize particle disturbance.

Open the chamber only after the pressure returns to approximately atmospheric pressure at approximately 760 torr and the chamber door can be released without resistance. Wear a mask and gloves during sample removal to reduce contamination and exposure to fine particles. Remove the deposited samples from the chamber.

Transfer the samples into a clean Petri dish. Store the samples in a vacuum storage container and perform characterization within 24 to 72 hours after deposition. After sample removal, rough pump the chamber for 10 to 15 minutes to reach the required pre-TMP pressure, minimizing exposure to ambient air and moisture.

X-ray diffraction patterns confirm that all samples retain the rhombohedral bismuth telluride phase across the manganese power series. The dominant diffraction peak shifted toward higher two theta values with increasing manganese input. The crystallite size generally decreased from 14.23 nanometers at zero watts to 8.35 nanometers at 15 watts, consistent with reduced coherent domain size and increased microstrain at higher manganese power.

Surface field emission scanning electron microscopy images demonstrated clear changes in film microstructure with increasing manganese input. The zero watt control film showed larger polygonal grains with distinct grain boundaries, while the five watt manganese-doped film exhibited a more compact surface with finer grains. Films deposited at 10 watt and 15 watt manganese power appeared denser with fewer intergranular gaps than the zero watt and five watt samples.

Elemental composition analysis confirmed that manganese incorporation increased with increasing manganese target power. All films exhibited a negative Seebeck coefficient, confirming n-type conduction behavior, and the magnitude of the Seebeck coefficient generally increased with temperature for all samples. The highest Seebeck coefficient magnitude was observed for the 15 watt film, reaching minus 162.08 microvolts per Kelvin at 523 Kelvins.

All films exhibited semi-conducting transport behavior where resistivity decreased with increasing temperature. The maximum power factor was achieved for the five watt film, reaching approximately 529.33 microwatts per meter per Kelvin squared at 523 Kelvins. Although the 15 watt film exhibited the highest Seebeck coefficient magnitude, the increased resistivity reduced the power factor to 300.59 microwatts per meter per Kelvin squared.

This protocol allowed the researcher to co-sputter the manganese-doped on the bismuth telluride film, and evaluate how the doping input level change the thermoelectric performance. The key challenge of this protocol is to maintain the stable plasma and also the sputtering condition to incorporate mechanisms reliably and consistently. For the future study can be built by optimizing thermoelectric properties, post-synthesis, and bifurcating a thin film thermoelectric generator for the device level testing.