The overall goal of this procedure is to fabricate lateral quantum dots capable of reaching the single electron regime. This is accomplished by first choosing a proper substrate and etching a mesa to delimit a region of the sample where the two dimensional electron gas will be present. The second step is to fabricate omic contacts using photolithographic and rapid thermal anil processes.
Next, the leads and gates defining the dot are fabricated by electron beam lithography. The final step is to fabricate the bonding pads and the larger leads using photolithography. Ultimately, stability diagrams are measured at low temperature to show that the few electron regime can be reached.
The main advantage of this device over existing spin qubits, such as vertical quantum dots, is that the properties of the lateral quantum dots are tuneable. Though this protocol is used to fabricate lateral quantum dots on gallium maronite aluminum, gallium maronite, hetero structures, it can also be easily adapted to fabricate devices on other substrates such as silicon. Silicon germanium.
Start with a substrate of modulation, doped, gallium arsenide aluminum, gallium arsenide, hetero structure from which traces of resist and organic compounds have been removed. This sample a square with sides slightly longer than one centimeter will hold 20 identical fabricated devices. Clean the sample twice in a low power sonic bath with acetone for five minutes.
Then clean it in the sonic bath with isopropyl alcohol for an additional five minutes. Blow dry the sample with nitrogen from a compressed nitrogen gun. Next, bake the sample in an oven at 125 degrees Celsius for at least 15 minutes to dehydrate the surface after dehydration, let the sample cool to room temperature, then spin with Shipley S 1818 photo resist at 3, 500 RPM for 30 seconds.
Then bake the sample on a hot plate at 115 degrees Celsius for 60 seconds. During the spinning of the resist, the sample has formed an edge bead on its contours. A photo lithographic process is used to remove this edge bead and allow better contact between the mask and the sample surface.
During the development, only the exposed resist on the edge of the sample is removed, leaving a perfectly flat surface of resist in the center. Proceed by using a photo lithic graphic mask aligner to position an edge bead mask composed of a one centimeter by one centimeter chrome square on a glass plate. Expose the resist on the outer edge of the sample to UV light for 10 seconds.
In this setup, the UV light has a wavelength of 436 nanometers and a power of 15 milliwatts per squared centimeter. As equipment varies, the exposure should be adjusted so as to obtain well-defined features, develop the exposed resist on the sample by immersing the sample in MF three 19 developer for two minutes and 10 seconds while agitating it slowly rinse the sample in water for 15 seconds and blow it dry with compressed nitrogen. With the edge bead removed, continue to use the photo lithographic mask aligner now in conjunction with the mesa mask that will create the region for the two dimensional electron gas.
Expose the sample to UV light for seven seconds. Again, develop the exposed resist by immersing it in developer for two minutes, 10 seconds. Rinse it in water for 15 seconds and blow it dry with compressed nitrogen.
The remaining resist on the sample is now in the shape of the mesa. Remove all traces of resist in the previously exposed area by placing the sample in a plasma Asher with oxygen plasma for one minute at 75 watts. Next, prepare an etching solution of water, hydrogen peroxide and sulfuric acid in a ratio of 55 to one to five.
Let the solution rest for 20 minutes before proceeding. It is necessary to etch the surface of the sample past the layer of silicon dots almost all the way down to the two dimensional electron gas to ensure that there is no longer a two dimensional gas in the etched region. To do this, immerse the sample in the acid solution for five to 10 seconds, and then immediately rinse it in water for 30 seconds.
To stop the reaction, measure the etch depth with a profilometer. After each immersion perform several five to ten second immersions and depth measurements until the desired etch depth is reached. For this substrate, a total of 34 seconds of immersion leads to the 90 to 100 nanometer etch depth.
Strip the resist by immersing the sample in 1165 remover at 65 degrees Celsius for three hours. Rinse the sample in acetone for five minutes, then in isopropyl alcohol for an additional five minutes. Here is a device at this point in the protocol to add omic contacts to the sample with a mesa spin coat, a layer of LOR five A, then S 1813, and remove the edge bead.
Next, use the photo lithographic mask aligner with the omic contact mask to align the patterns for the omic contacts on the etched mesa exposed with UV light for six seconds. Develop, rinse and dry the sample. Then use oxygen plasma to remove any remaining resist from the exposed area.
Prepare a solution of sulfuric acid and water in a one to five ratio. Remove the gallium arsenide native oxide by immersing the sample in the solution for 30 seconds. Next, rinse in water for 30 seconds and then blow dry with compressed nitrogen as soon as possible.
Place the sample in an e-beam evaporator deposit. 25 nanometers of nickel, 56 nanometers of germanium and 80 nanometers of gold. The advantage of using a double layer of resist becomes clear at this step when the exposed sample is placed in the developer.
The bottom layer of LOR five A is dissolved faster than the top layer of S 1813 creating an undercut when the metal is deposited over the entire surface of the sample. This profile in the resist prevents walls of metal from being formed at the edges of the metal shapes. The undercut also allows the metal deposited on top of the resist to be removed easily.
To remove the unwanted metal, dissolve the resist by immersing the sample in remover at 65 degrees Celsius for three hours. To help lift unwanted metal, use a pipette to gently spray the surface of the sample with warm remover. Rinse the sample in acetone, then isopropyl alcohol for five minutes each diffuse the deposited metal down to the two dimensional electron gas with a rapid thermal annealing process in forming gas during the rapid thermal annealing process, the gold that has been deposited on the surface of the sample diffuses into the gallium arsenide aluminum gallium arsenide substrate through a certain number of channels.
These metallic channels will ensure the electrical continuity between the two dimensional electron gas and contact pads that will be deposited on the surface of the sample in later steps. A device after the annealing stage is shown here. Fabrication of the titanium gold shock key leads begins with preparing the sample so that it has a layer of PMMA high molecular weight resist over a layer of PMMA low molecular weight resist.
The leads are using a scanning electron microscope equipped to do electron beam lithography. Place the sample in the scanning electron microscope, adjust the microscope parameters such as the focus and the stigmatism, and measure the beam current with a Faraday cup. Have a previously prepared CAD file containing the pattern for the leads that will be exposed by the microscope.
Align the beam with the omic contacts and set the magnification to 300 times the scanning electron microscope is programmed to expose the desired areas by exposing an array of dots within them. An area dose of 43 micro kula per squared centimeter is chosen to obtain sharp features of the desired dimensions. Prepare a solution of nine parts, isopropyl alcohol to one part water at a temperature of 20 degrees Celsius.
Develop the resist by immersing the sample in the solution for 30 seconds. Rinse in water for 30 seconds and blow dry with compressed nitrogen. To remove any trace of resist from the exposed patterns, place the sample in an oxygen plasma for four seconds.
At 50 watts, remove the gallium arsenide aluminum gallium arsenide native oxide by immersing the sample in sulfuric acid solution. Next, move the sample to an ebeam evaporator and deposit 10 nanometers of titanium as well as 20 nanometers of gold. Then the sample is placed in a remover at 65 degrees Celsius to remove the remaining resist and excess metal.
A device at this stage is shown here. Aluminum shot key leads and gates that define the dot are fabricated using the same process used for the fabrication of the titanium gold shot key leads. Once these are in place, do not use the ultrasonic bath for cleaning the sample since it can damage the final aluminum gates.
To fabricate the bonding pads, prepare the sample with a layer of LOR five A resist followed by S 1813.Resist. Then remove the edge bead using a photo lithographic mask aligner with the shot. Key leads mask.
Align the patterns for the bonding pads with the previously fabricated structures exposed for six seconds. After developing the exposed patterns, place the sample in an e-beam evaporator and deposit 30 nanometers of titanium and 350 nanometers of gold. After excess metal is removed and the sample is rinsed, a device should be as depicted here.
The final step of the protocol consists of dicing the sample to separate the individual devices. Spinco S 1818 resist onto the surface of the sample at low speed. This thick layer of resist will protect the fabricated structures during dicing.
Place a layer of dicing tape on the top and bottom faces of the sample. Then place the sample in the ER once the sample is diced, place it in acetone for a few minutes to remove the dicing tape and protective resist, rinse the individual devices in isopropyl alcohol and gently blow dry with compressed nitrogen. Here the 20 final devices obtained are shown next to an initial substrate of gallium arsenide.
The sample is now ready to be mounted and bonded onto a sample holder. Once this is done, place the sample in a helium cryostat for low temperature characterization. A simple test of gait integrity can be conducted by applying a voltage bias between two omic contacts and measuring the current driven through the sample as a function of voltage applied on a pair of gates.Shown.
Here is an example of such measurements taken at 1.4 kelvin with a voltage bias of 500 microvolts between source and drain. The meaning of the gait labels is shown on the right depletion of the 200 nanometer wide leads is seen at about minus 300 millivolts. Depletion of the 60 nanometer gates is completed at about minus 750 millivolts.
The plateaus seen at lower voltages correspond to the quantization of conductance. Typical values for pinch off points between pairs of gates, range between minus 500 millivolts and minus two volts. In this sample, the gates are likely to be geometrically similar since the top center with top left and top center with top right pairs pinch off the current at approximately equal voltages.
The same goes for the top center with bottom left and top center with bottom right pairs. However, the fact that the current still flows between the top center and bottom center gates for high voltages indicates that the bottom center gate is either broken or missing. Usually about 50%of the tested devices have fully functional gates and OMI contacts.
Once a fully functioning device has been found, measure a stability diagram to confirm that the zero electron regime can be reached. In this diagram, the left and right numbers in parentheses give the numbers of electrons in the left and right dots respectively for different values of voltages on the bottom left and bottom right gates. The green lines are used to deli regions with a constant number of electrons.
The color scale represents the current flow through the. at different gate voltages. Following this procedure, microm magnets can be added on top of the device to enable ultra fast manipulation of the electron spin using microwave pulses After its development.
This technique paved the way for researchers in the field of experimental quantum computing to explore the interaction between multiple electron and double quantum dots.
