Making Record-efficiency SnS Solar Cells by Thermal Evaporation and Atomic Layer Deposition

The overall goal of this procedure is to establish a reliable platform for developing solar cells based on new materials. This is accomplished by first preparing a clean electrically insulating substrate. The second step is to deposit two layers of sputtered molybdenum for the rear hole, selective electrical contact.

Next deposit the P-type tin sulfide absorber by either atomic layer deposition or thermal evaporation. The the final step is to finish the PN junction by depositing a stack of N type layers comprised of a compound zinc oxy sulfide buffer layer, followed by a transparent conducting oxide and metal fingers deposited to assist charge collection. Ultimately, the resulting solar cells are tested by measuring current voltage curves under simulated solar light and quantum efficiency under monochromatic illumination.

Although this method is demonstrated for the development of tin sulfide solar cells, it can also be used to develop inorganic thin film substrate style solar cells based on other new materials. The deposition of tin sulfide by atomic layer deposition will be demonstrated by Schwan Yang, a grad student in the Gordon Group at Harvard, far Steinman. A postdoc in the photovoltaics research lab at MIT will demonstrate the deposition of 10 sulfide by thermal evaporation.

This video begins after the preparation of the silicon wafer substrates. Use one square inch polished 500 micron thick silicon wafers with a 300 nanometer or thicker thermal oxide. First sputter a 360 nanometer adhesion layer of molybdenum, and then deposit a second 360 nanometer thick conduction layer of maum.

Store the prepared substrates in a nitrogen glove box until it is time for deposition of the tin sulfide to ready the substrates for atomic layer deposition. First, transfer them to a UV ozone cleaner to remove organic particles. Clean the substrates for five minutes.

As the cleaning takes place, ensure the substrate holder for the deposition chamber is nearby. When they are ready, recover the substrates and place them in the sample holder. This sample holder is loaded with substrates and is ready for placement in the deposition chamber.

Insert the holder into the furnace, then pump down the furnace using a roughing vacuum pump. Next, stabilize the furnace temperature at 200 degrees Celsius. Continue by checking on the precursors.

Maintain tin ate one precursor for the deposition of tin sulfide in an oven at 95 degrees Celsius. Arrange gas piping to have pure nitrogen gas to assist the delivery of tin ate vapor. The second precursor is hydrogen sulfide in a gas mixture of 4%Hydrogen sulfide in nitrogen gas proceed with the deposition at a growth rate of 0.33 angstroms per cycle.

During each atomic layer deposition cycle, supply three doses of 10 ate for a total exposure of 1.1 tor second. Then purge the chamber with nitrogen gas. Next, inject the hydrogen sulfide in nitrogen gas for an exposure of 1.5 tor second.

Finally, purge again with nitrogen. An alternative deposition method is thermal evaporation. Ensure the process chamber pressure is two times 10 to the minus seven tor or lower.

Use a sample holder with appropriately sized pockets to hold the substrate securely. As before, use a previously prepared substrate with two layers of molybdenum. Here depicted schematically.

Load the holder with this substrate and small additional CYS substrates for later characterization. Load the sample holder into the thermal evaporator through the load lock. Pump down the load lock, and then transfer the substrates into the growth chamber.

Using the transfer arm when they are in position, ramp the source and substrate heaters to achieve a growth rate of one angstrom per second and substrate temperature of 240 degrees Celsius. Now raise the substrates in order to measure the deposition rate with the quartz crystal monitor. Engage the linear translation arm to move the quartz crystal monitor into the process chamber.

Once it is in place, open the source shutter to begin the measurement. The growth rate is indicated on the front panel of the Crystal Monitor controller. When the measurement is completed, close the shutter and remove the quartz monitor from the chamber.

Then lower the substrates to the growth position and open the source shutter to start the deposition. For the target film thickness of 1000 nanometers, the deposition will take about three hours. When the deposition is complete, close the source shutter.

Repeat the growth rate measurement using the quartz crystal monitor. After the measurement, close the source shutter, turn off the heaters and wait for the substrates to cool before transferring them to the load lock. Vent the lock to air and unload the substrates.

Quickly unmount them from the sample holder and transport them into storage in a nitrogen glove box. After either the atomic layer or thermal evaporation deposition process, the sample will have a tin sulfide layer. All samples then undergo a kneeling in hydrogen sulfide gas followed by surface pacification with a native oxide.

The next step is deposition of N type zinc oxy sulfide and zinc oxide buffer layers by atomic layer deposition. The next step is deposition of the transparent conductor. We use indium tin oxide before deposition of indium tin oxide.

The active areas of the devices must be established. Use metal shadow masks to define the transparent conducting oxide pad size, which sets the active device area. The masks define 11 rectangular devices that are 0.25 square centimeters in size, plus a larger pad in one corner that is used for optical reflectivity measurements in the drawing.

An indication of the size of the indium tin oxide layer to be added for one device is given by the shaded area, mount the devices and masks in a mask aligner. The masks should be sufficiently thick so that they don't flex in the mask aligner. Otherwise, the oxide pad area can become poorly defined.

Take the mask aligner with the mounted devices to the sputtering chamber. Put the mask aligner into position for the deposition step. The next step is to grow a 240 nanometer thick indium tin oxide film.

To do this, heat the substrate to 80 to 90 degrees Celsius and enable substrate rotation. Set the radio frequency sputtering, power, the argon and oxygen gas flow rate and the pressure, and then monitor the deposition schematically. This is the state of the device after indium tin oxide deposition.

The next step is metalization. Use a shadow mask and electron beam evaporation to deposit 500 nanometers of conductive metal, either silver or a nickel aluminum stack. This is a completed sample.

Notice that the sample shown here has already been scratched to expose the molybdenum back contact after fabrication. Move on to device characterization Using a solar simulator calibrated for an air mass coefficient of 1.5. First, use a scalpel blade to scratch away the buffer and tin sulfide layers and expose the molybdenum back contact.

Next place the sample on the rail mounted sample chuck, a vacuum holds the sample in place and the chuck temperature is controlled at 25 degrees Celsius. Now, make electrical contact with each of the devices on the sample and the scratched back contact. This is done here with a custom probe card that contacts all devices simultaneously with copper beryllium.

Double probe tips in four wire mode for routine measurements. Omit the use of a light aperture for more rigorous testing, and aperture should be used to define the active area. The transparent conducting oxide shadow mask can be used for this purpose.

Move the chuck so the devices are in position to collect light and dark current voltage data using an automated test system. Program, the system to probe the individual devices sequentially while opening and closing the light shutter. The testing can be seen in this video, but normally should be conducted behind a black curtain to reduce ambient light.

After the current voltage measurements slide the sample chuck into the external quantum efficiency system, the illuminated current voltage data, the external quantum efficiency data and the optical reflectance measurements provide a baseline characterization of the devices. Here are solar cell performance data for devices on two samples made using thermal evaporation. The samples are distinguished color gray for one, red for the other for each device.

The open circuit voltage, the short circuit current density, the fill factor, and the power conversion efficiency is given. The best devices have about a 4%efficiency. The apparent correlation between the efficiency and the fill factor across both devices is consistent with devices that suffer from shunt or series resistance losses.

The data in red also show a correlation between efficiency and the open circuit voltage as expected for shunt resistance. Losses for comparison. Here is solar cell performance data for devices made using atomic layer deposition.

These devices have better performance than the thermal evaporation devices with the best showing 4.6%efficiency. Various factors may account for this difference. One factor is that devices produced with atomic layer deposition seem to suffer from less shunt resistance.Losses.

Don't forget that working with hydrogen sulfate can be very dangerous. Make sure that the hydrogen sulfate tank is stored in a air ventilation device like a film hood with hydrogen sulfate detectors. We think of this baseline fabrication protocol as a platform to carry out controlled and focused experiments designed to improve the overall solar cell performance and to better understand particular loss mechanisms.

After watching this video, you should have a good understanding of how to make a reproducible substrate style solar cell.