高解像度フォノンアシスト準共鳴蛍光分光

The overall goal of Phonon-assisted Quasi-resonance Florescence Spectroscopy is to achieve laser linewidth limited resolution of optical spectra of atom-like semiconductor structures with a standard photoluminescent Spectroscopy setup. This method facilitates study in semiconductor optical Spectroscopy related to zero dimensional systems, such as quantum dots, nitrogen-vacancy centers in diamond, and defects in silicon carbide. The main advantage of this technique is that it has the ability to obtain high-resolution optical spectra of energy features less than 10 micro-electron volts, while retaining polarization information.

The implications of this technique extend the power of optical Spectroscopy due to it's ability to obtain high-resolution optical spectra. Individuals new to the method will struggle with finding the signal. The greatest hurdle is using a single spectrometer and extinction filter to eliminate stray laser light.

Start with the material or device that will be used in the experiment. This device has quantum dot molecules mounted onto a chip. An overview of the quantum dot molecules is provided in this schematic.

There are two indium arsenide quantum dot layers that are embedded between the gallium arsenide layers. Each quantum dot can be optically addressed. The quantum dot molecules are in the intrinsic region of a Schottky diode, allowing application of an electric field.

The first step is to mount the device onto the cryostat cold finger for cool down. Before mounting the chip on the cold finger, apply indium foil for the chip to rest on. Next, put the chip in position on top of the foil.

Use screws and washers at each end of the device to firmly tighten it to the cold finger and ensure good thermal contact. Continue by attaching the devices electrode wires to the contacts in the cryostat that connect to a source meter. Now mount the cryostat on an XYZ Translation Stage and make the sample accessible to optical probes.

In this case, the translation stage is already in the experiment's optical path. Evacuate the cryostat before cooling it down to the desired temperature. Once the sample has cooled, begin to work with the optical elements.

For photoluminescence, have a long working distance objective and collimation lens in place to capture light from the sample. From there, use mirrors to direct the light onto a single-stage spectrometer. Employ a nitrogen cooled CCD camera to detect the spectra.

In this experiment, we use the last stage of a triple spectrometer. For alignment purposes, mount a white light source near the sample. Illuminate the sample with the source before continuing.

Employ a separate camera to capture the image of the sample in the spectrometer. On the bench, work to properly align the collimating and focusing lenses to sharpen the image of the sample. Conclude alignment efforts only when a clean, focused image appears in the camera.

After removing the light source, prepare a laser to excite the sample. Use a Tunable Diode Laser set to an energy above the ground state transitions. Direct the laser onto the sample at an oblique angle to reduce detecting scattered light.

Employ a lens to focus the beam spot to the smallest size possible on the region captured by the camera. An overview of the setup at this point is given in this schematic. Note the oblique angle of the laser light onto the sample for noise reduction.

Use the laser to excite a higher, non-resonant energy, and run the spectrum acquisition software in focus mode to observe the signal. Use the sample Housing Translation Stage to scan the sample through the laser spot. When the CCD captures the discreet lines of the ground state transitions, stop the scan.

Concentrate on one quantum dot molecule, and optimize the signal by fine-tuning the laser beam's position. The next step is to generate a Bias map using lap control software. The software first applies a Bias potential across the sample electrodes with a source meter.

It then records the spectrum associated with that Bias value. The map is generated by incrementing the Bias potential at regular intervals, recording the energy spectrum at each Bias value, and using the data to produce this representation. Given a Bias map, identify the transition that will be excited.

Note the transition Energy and the Bias range of interest. In this figure, the transition is stepped through by scanning the temperature of the sample. A detection signal will be present when the laser is resonant with the transition.

Use the chosen transition to determine edge filters for the optical setup. The transition energy justifies a 960 nanometer short pass filter for the excitation branch. A longitudinal optical phonon emission energy for the semiconductor alloy justifies a 960 nanometer long pass filter for the detection branch.

The type of filter in the setup can make later steps easier. Interference cut-off filters are ideal because they can be tuned by adjusting the angle. At the computer, set the laser excitation energy and the minus one longitudinal optical phonon center frequency.

Start the CCD, collecting spectra in continuous mode. The signal may be hidden by scatter at this point. In order to maximize the signal, return to the bench.

There, tune the short pass filter by adjusting it's angle to have the proper wavelength cut-off. Monitor the signal for feedback to determine the best angle. In the lab control software, set the Laser energy to scan a range centered on the transition energy, and set the range of Bias voltages to scan over.

This Bias map was created by setting the laser energy, varying the Bias over it's range to collect a spectrum, then repeating this over the entire range of laser energies. By collecting data on the background concurrently, the two can be used during post-processing to remove spurious signals and create an improved image. As seen in these schematics, the experimental setup for the Phonon-assisted Spectroscopy method is almost the same as for standard Spectroscopy.

The only difference is the presence of edge filters on both the detection and excitation beam paths in the Phonon-assisted method. To compare this technique with others, consider the resolution of a neutral exciton. This data is from a single spectrometer with non-resident excitation around 918 nanometers.

The spectral resolution is about 26 micro-electron volts per pixel, and the electron-hole exchange splitting cannot be resolved. When the same spectral region is studied with a spectrometer in triple additive mode, the resolution is about 10 micro-electron volts. The electron-hole splitting is beginning to be resolved.

Using the Quasi-resonant Phonon-assisted Spectroscopy method in the same spectral region, the resolution is laser limited, giving better than one micro-electron volt resolution, and the two peaks are well resolved. The red curve is the result of a double-Lorentzian fit. It suggests an isotropic electron-hole exchange splitting of about 23.3 micro-electron volts.

Once mastered, this technique can be done in the matter of a couple of hours. While attempting this procedure, it's important to remember that after obtaining photoluminescence signal, the key to a good spectrum is being able to separate the laser from the photoluminescence. Following this procedure, other additions can be made to the setup, such as adding polarization analyzers to both the excitation and detection path.

After it's development, this technique paved the way for researchers in the field of Semiconductor Physics to further explore zero dimensional systems, such as quantum dots. After watching this video, you should have a very good understanding of how to use this Phonon-assisted Quasi-resonance Fluorescence Spectroscopy method in order to obtain very high-resolution optical spectra of atom-like solid state systems. Don't forget that working with lasers can be extremely hazardous, and precautions such as proper safety gear should be worn at all times when performing these experiments.