READ ME File For 'Stephen Richardson PhD Data' Dataset DOI: 10.5258/SOTON/D3787 ReadMe Author: Stephen Constantine Richardson, University of Southampton, ORCID ID 0009-0008-2346-0372 This dataset supports the thesis entitled "Membrane Quantum Well Lasers and Tantalum Pentoxide Optical Waveguides" AWARDED BY: University of Southampton DATE OF AWARD: 2025 Date of data collection: 11/11/2020 - 26/09/2024 Information about geographic location of data collection: University of Southampton, Southampton, UK, SO17 1BJ & National Physical Laboratory, Teddington, UK, TW11 0LW Related projects/Funders: University of Southampton/ Mayflower Studentship -------------------- DATA & FILE OVERVIEW -------------------- This dataset contains: Folders for each figure and table with exception of a few (please see notes). Wherever required, MATLAB scripts and COMSOL models are included. If data was derived from another source, list source: 1. Woods, Jonathan RC, et al. "Coherent waveguide laser arrays in semiconductor quantum well membranes." Optics Express 30.18 (2022): 32174-32188. 2. Woods, Jonathan RC, et al. "Supercontinuum generation in tantalum pentoxide waveguides for pump wavelengths in the 900 nm to 1500 nm spectral region." Optics Express 28.21 (2020): 32173-32184. 3. Coen, Stéphane, et al. "Modeling of octave-spanning Kerr frequency combs using a generalized mean-field Lugiato–Lefever model." Optics letters 38.1 (2012): 37-39. 4. Daykin, Jake. Supercontinuum and frequency comb generation in tantalum pentoxide waveguides. Diss. University of Southampton, 2024. 5. Jung, Hojoong, et al. "Tantala Kerr nonlinear integrated photonics." Optica 8.6 (2021): 811-817. 6. Tropper, A. C., et al. "Vertical-external-cavity semiconductor lasers." Journal of Physics D: Applied Physics 37.9 (2004): R75. 7. Kahle, Hermann, et al. "Semiconductor membrane external-cavity surface-emitting laser (MECSEL)." Optica 3.12 (2016): 1506-1512. 8. Dudley, John M., Goëry Genty, and Stéphane Coen. "Supercontinuum generation in photonic crystal fiber." Reviews of modern physics 78.4 (2006): 1135-1184. -------------------------- METHODOLOGICAL INFORMATION -------------------------- 1. Data presented in Figures 2.4, 2.5, 2.6, 2.9, 2.13 and 2.15 were obtained by capturing real and reciprocal space images using CCD cameras. These images were calibrated (pixel conversion) by using an item of known physical dimensions (real space images) and the laser wavelength (reciprocal space images). Regarding Figure 2.15, the pump spot was changed by varying the position of the optical fibre collimation lens. Figure 2.15 was obtained from Dr Jake Daykin (University of Southampton). 2. Data presented in Figures 2.7 and 2.8 were obtained by averaging the interference pattern of a reciprocal space image and applying a Fourier transform to obtain laser cavity length and fitting parameters. 3. The results presented in Figure 2.14 were obtained as follows. The optical fibre collimation lens was translated thus changing the pump spot - FWHM on top of the membrane gain chip. In addition, the pump power was required to change to attain laser threshold conditions. The FWHM information was obtained by scanning the pump spot from the real space images. 4. The laser emission spectrum shown in Figure 2.16 was obtained by using an OSA. 5. Regarding Figure 2.17, an OSA was used to capture the laser emission spectrum for different TEC cooler temperatures and for constant pump power. Figure 2.18 is a detailed depiction of a laser mode transition by using the data from Figure 2.17. Figures 2.20 and 2.21 are also a result from the data shown in Figure 2.17. 6. Figure 2.19 is similar to Figure 2.17, however the temperature is constant and the pump power varies. 7. A COMSOL model was developed to simulate guided modes inside a membrane quantum well laser on a silicon carbide or silicon dioxide substrate (Figures 2.22 and 2.23). Figure 2.24 is a result of a wavelength parametric sweep of the COMSOL model, which produces effective refractive indices for each wavelength. 8. The etch rate of GaAs and the membrane quantum well laser (Figure 2.26) were produced by measuring time and etch depth with a stylus profiler in a cleanroom facility. Etching was achieved with an acid solution. A SEM image was taken (Figure 2.30) to evaluate the device's sidewall quality. 9. Figure 2.28 was obtained the same way as in point 1 for different pump powers. The photoluminescence spectra corresponding to the two different pumping conditions were obtained with an OSA. 10. Figure 3.3 was obtained by recording the laser drive current and measuring the laser output optical power with a power meter. Then a linear fit was implemented after laser threshold to obtain the slope efficiency. 11. Figure 3.5, 3.6 and 3.7 was obtained the same way as in point 1. 12. Figure 3.8 was obtained the same way as in presented in point 1 with the addition of including a reciprocal space simulation and a line scan of the membrane's end facets. 13. Figure 3.9 was produced by capturing real and reciprocal space images (point 1) for different laser stripe separations and an OSA for the laser emission spectra. Pump laser stripe separation change was possible by using the digital micromirror device (DMD) connected to the PC. A MS PowerPoint slide contained two rectangular shapes that shifted manually and was projected on the DMD monitor thus varying the pump laser shape. 14. Figure 3.10 was produced by capturing real and reciprocal space images and performing line scans of the membrane's end facets and interference patterns. As a pump laser stripe approaches the other one, the changes in real and reciprocal space images were captured. 15. Figures 3.6 - 3.8 and 3.10 were produced from Dr Nicholas Klokkou (University of Southampton). 16. Figures 3.11 - 3.14 were obtained the same way as in point 1 and 13. Figures 3.12 - 3.14 were obtained from Dr Nicholas Klokkou (University of Southampton). 17. Figure 3.15 was produced by the same means described in point 1 and 4. A knife edge was used to conceal the pump laser stripes to observe how that affects laser array coherence. 18. Various COMSOL models were developed to simulate light propagation in a medium characterised by optical gain resembling the membrane quantum well lasers and to create Figures 3.16 - 3.23. Line scans of electric field intensities were taken (Figure 3.17, 3.18, 3.19 and 3.22) to show evolution of intensity with respect to optical gain parameter or mesh size. 19. Figure 4.3 simulations were produced from Dr Peter Horak (University of Southampton) by using MATLAB. 20. Figure 4.4 was provided by Dr Jake Daykin (University of Southampton) by using MATLAB. 21. Figures 4.5, 4.6 and 4.8 were created by using the MATLAB script developed by Dr Peter Horak. The spectral outputs utilise the dispersion curve (Figure 4.4) and other physical input parameters. Figure 4.7 is similar and was reproduced from Dr Jake Daykin's PhD. 22. Figure 4.9 was produced from a COMSOL model simulating a micro resonator and investigating the guide modes. From a wavelength parametric sweep, the effective and group refractive indices and effective areas are retrieved and used to calculate the dispersion curve (Figure 4.4) and then the spectral broadening (if it occurs). 23. Figure 4.13 was obtained by a microscope. 24. Figure 4.14 was obtained by two different SEMs. One was obtained by Dr Oliver Trojak (University of Southampton). 25. Figures 4.15 and 4.16 were also obtained by a SEM. 26. Figure 4.18 depicts the pump laser calibration, which is similar to point 10. 27. Figures 4.19a and b are top view images obtained by a CCD camera. 28. Figures 4.20, 4.22 and 4.23 were obtained with an OSA. Figure 4.22 shows the spectral difference before and after the laser beam passes through the optical waveguide. 29. Figures 4.21 and 4.24 are obtained by capturing the transmission spectra. A sum of five sinusoids was fit to those waveforms in order to extract the waveguide optical losses assuming that the linear optical waveguide is a Fabry - Perot interferometer. Table 4.1 presents the calculated optical losses. Laser coupling is achieved by butt coupling. 30. Figures 4.25 - 4.29 were obtained from Dr Jonathan Silver (National Physical Laboratory). The setup used allowed a laser beam to pass through optical waveguides which included ring and racetrack resonators and also to perform a laser wavelength sweep revealing resonance pump wavelengths in the transmission spectra. Tables 4.2 and 4.3 present the waveguide geometries, resonant wavelengths, quality factors and coupling coefficients. Laser coupling is achieved by butt coupling. 31. Figure 4.30 and Table 4.4 depict data similar to point 30. The wafer is similar to the ones involving the previous points of Chapter 4, however it is processed differently. Laser coupling is achieved by coupling from an optical fibre to a diffraction grating etched on the sample. This data was provided by Dr Oliver Trojak (University of Southampton) as a comparison to the data of point 30. 32. Figure A.2 was obtained as in point 1. Figure A.3 presents the evolution of laser profiles across a laser line profile after a laser spot passes through a cylindrical lens and varying the distance between two plano - convex lenses. Figure A.4 depicts the optimal laser line profile and therefore optimal lens separation. 33. Figures B.1 and B.2 are obtained by a laser spot passing through an axicon lens and a CCD camera captures the newly formed laser shape (ring shape) at different locations after passing through the lens. Figure B.4 was obtained by manually measuring the outer radius of the ring by placing the CCD at different locations away from the axicon lens. Figure B.6 is similar to Figure B.4 only it involves the FWHM of the pump laser spot. Figures B.7 and B.8 involve a pair of axicons with their tips facing each other. The ring radii were measured by placing the CCD camera at different positions after the second axicon lens. The separation between axicons is fixed. 34. Figures C.1 and C.2 were captured by using a microscope to monitor the cleaning process. 35. MATLAB and COMSOL were used to analyse data and simulate. 36. All experiments were conducted in the dark. -------------------------- DATA-SPECIFIC INFORMATION -------------------------- 1. All necessary MATLAB scripts are included to read the related data. Comments and descriptions are included in the algorithms. 2. Any spreadsheet that contains data from CCD images contains many rows and columns with no labels. The values refer to intensity (a. u.) captured by each pixel. These spreadsheets can be found in the following folders: 'Figure 2.4 to 2.8', 'Figure 2.14', 'Figure 2.16-2.18, 2.20 & 2.21', 'Figure 2.19', 'Figure A.3', 'Figure A.4', and 'Figure B.6'. 3. Some scripts and images are denoted with a 'r' or 'k'. The former involves real space images whereas the latter involves reciprocal space images. 4. Data for Figure 3.3 can be found in the MATLAB script. 5. Some images are named after the experimental conditions. For example, the .tif image found in the Figure 3.5 folder, is named '4200mA', which refers to the drive current of the pump diode laser. 6. The .txt files in Folder 'Figure 3.17' refer to the line scans for different gain values. The .txt files in Folder 'Figure 3.18' refer to the line scans for different mesh sizes. The .txt files in Folder 'Figure 3.19' refer to the line scans for different gain values. 7. The Folder 'Figure 2.4 to 2.8' includes an image named '488mA 15C'. 488mA is the pump laser drive current and 15C is the TEC cooler temperature. 8. The 'TST' spreadsheet in the folder 'Figure 2.16-2.18, 2.20 & 2.21' refers to 'Temperature Spectral Tuning'. In folder 'Figure 2.19' the name 'PST' refers to 'Power Spectral Tuning' and '14deg' is 14 degrees Celsius. 9. All OSA spreadsheets start with the letter 'W'. 10. Folder 'Figure 4.4', 725x1650 refers to the waveguide cross-section dimensions, 100R is the radius of resonator, 100 um, and 0PE means no (0 um) partial etch. Similar to the folders 'Figure 4.5 to 4.6' & 'Figure 4.7 to 4.8'. 11. The effective area in 'Tantala COMSOL Etch Model Results' (Folder 'Figure 4.9') is measured in 'm^2'. 12. The 'scope' spreadsheets found in the folders 'Figure 4.21 and Table 4.1' and 'Figure 4.22 to 4.24' consist of four (4) columns. Column A is the x-axis of the oscilloscope, which is time (s). Column B, C and D are vertical axes measured in V. B is channel 1, C is channel 3, and D is channel 4. Channel 1 is the piezo feedback, channel 3 is the data measured after the waveguide, and channel 4 is data measured before the waveguide. -------------- Notes: 1. Figure 2.9, 2.10, 2.11, and 2.12 are described in great detail in Woods, Jonathan RC, et al. "Coherent waveguide laser arrays in semiconductor quantum well membranes." Optics Express 30.18 (2022): 32174-32188. 2. Data for Figures 3.6 - 3.8 and 3.12 - 3.14 were provided from Dr Nicholas Klokkou (University of Southampton).