This paper presents design and fabrication of ring cavity semiconductor lasers with simple fabrication processes and good potential for integration. A 1.55-£gm symmetric quantum well InGaAsP epi-layer wafer is used to fabricate the lasers. The fabrication processes involve a bi-level deep etching to reduce the bending losses. Two geometric types of ring cavity semiconductor lasers have been investigated. For the type 1 ring cavity in the form of race tracks, two different designs are presented. One has a single ring resonator (SRR) design and the other has a coupled double ring resonators (DRR) design. The resonator of the type 2 ring cavity is formed between a cleaved facet and a loop mirror. Both a single ring resonator (SRR) design and a double ring resonator (DRR) design are presented for this type of cavity also. The maximum saturation output light powers of 0.479 and 0.409 mW are observed in room temperature L-I measurements for type 1 and type 2 ring cavity semiconductor lasers respectively. The spontaneous emission spectra of the type 1 ring cavity semiconductor lasers show a red-shift phenomenon under increasing drive currents. The type 1 ring cavity semiconductor lasers with ring resonators of 100 and 200 £gm radii have also been found to exhibit an interesting wavelength clamping phenomenon of the output light.

This paper presents design and fabrication of ring cavity semiconductor lasers and optical filters with simple fabrication process. A 1.55

In this thesis, we use InP based multiple quantum well epi-wafer with modulation doping in the active layer to design the semiconductor optical amplifier and ring laser for the optical communication at 1.55£gm wavelength. We also finish the mask design and fabrication of theses two devices. Besides, we have established an optical measurement system, including the L-I measurement, the optical spectrum measurement and the far field measurement, to test the device parameters. In the device process, we use the new method, called the Multi-Step Undercutting, to precisely control the undercut in the wet etching process. With this technique, we can get a smooth and vertical sidewall for our devices. For the semiconductor optical amplifier, we design two different types, one is the Fabry-Perot Amplifier and the other is the Traveling Wave Amplifier. We use the Multi-Step Undercutting process in the fabrication of these two devices. The main parameters for semiconductor optical amplifier are the change of the output power versus the input current, the spontaneous emission spectrum and the photocurrent spectrum. For the ring laser, we combine the concept of Loop Mirror and Asymmetric Mach-Zehnder Interferometer to obtain the laser with good side mode suppression for a single wavelength light source.

In this thesis, we focus on the investigation of semiconductor optical amplifier and ring laser. We use InP based multiple quantum well epi-wafer with modulation doping in the active layer to design the semiconductor optical amplifier and ring laser for the optical communication at 1.55£gm wavelength. We combine the concept of Loop Mirror and Asymmetric Mach-Zehnder Interferometer to obtain the laser with good side mode suppression for a single wavelength light source. For the semiconductor optical amplifier, we design two different types, one is the Fabry-Perot Amplifier and the other is the Traveling Wave Amplifier. Furthermore, We use the Multi-Step Undercutting process in the fabrication of these two devices. We have established an optical measurement system, including the L-I measurement, the optical spectrum measurement and the far field measurement, to test the device parameters. After annealing, these devices with two different serial number exhibited the contact resistances of 9£[ and 16£[, respectively. Under CW operation, these FPA exhibited the threshold current of 62mA and 70mA at 20¢J, respectively. The stimulated emission wavelength was at 1531nm and 1522nm, respectively.

Precise measurements of both absolute frequencies and small frequency differences of atomic energy levels have played an important role in the development of physics. For example, high precision measurements of absolute frequencies of the 2S½ → 2P ½ transition (D1 line) of alkali atoms form an important link in the measurement of the fine structure constant, α. Similarly, precise interferometric measurements of the local gravitational acceleration (g) rely on the knowledge of the absolute frequencies of the 2S½ → 2P 3/2 transition (D2line) in alkali atoms. Difference frequency measurements of hyperfine structure and isotope shifts of atomic energy levels provide valuable information about the structure of the nucleus, which in turn helps in fine tuning the atomic wave functions used in theoretical calculations. The work reported in this thesis starts with the development and refinement of high precision measurement of absolute frequencies using a ring-cavity resonator. The measurement technique is relatively simple and cost-effective, but the accuracy is comparable to that achieved with the frequency comb technique (10¯11) when the accuracy is limited by the natural linewidth of the transition being measured. The technique combines the advantages of using tunable diode lasers to access atomic transitions with the fact that the absolute frequency of the D2 line in87Rb is known with an accuracy of 6 kHz. A frequency-stabilized diode laser locked to this line is used as a frequency reference, along with a ring-cavity resonator whose length is locked to the reference laser. For a given cavity length, an unknown laser locked to an atomic transition has a small frequency offset from the nearest cavity resonance. We use an acousto-optic modulator (AOM) to compensate for this frequency offset. The measured offset is combined with the cavity mode number to obtain a precise value for the frequency of The unknown laser. We have used this technique for absolute frequency measurements Of the D lines in133Cs and 6,7Li, and the 398.8nm line in Yb. We have also developed a technique to measure the ‘difference frequency’ of atomic energy levels using a single diode laser and an AOM. In this technique, the laser is first locked to a given hyperfine transition. The laser frequency is then shifted using the AOM to another hyperfine transition and the AOM frequency is locked to this difference. Thus the AOM frequency directly gives a measurement of the hyperfine interval. Applying this AOM technique we have measured the hyperfine interval of the D1 lines of all alkali atoms with high precision. We have further developed a technique of coheren-tcontrol spectroscopy (CCS) using co-propagating control and probe beam that is useful for highresolution spectroscopy. In this technique, the probe beam is locked to a transition and its absorption signal is monitored while the control beam is scanned through neighbouring transition. As the control comes into resonance with another transition, the probe absorption is reduced and the signal shows a Doppler free dip. This technique allows us to resolve transitions that are otherwise swamped by crossover resonances in conventional saturated absorption spectroscopy (SAS). We have applied this technique to measure hyperfine intervals in the D2 line of several alkali atoms. Thus, we were able to do high-precision measurements of both absolute and difference frequency of atomic transitions. The precision of the absolute frequency measurement is finally limited by the accuracy of 6 kHz with which the reference frequency is known. The nearby two photon transition in Rb, i.e. the 5S1/2→5D3/2 transition at 778 nm, is known with an accuracy of 1 kHz. In future, we hope to improve the accuracy of our technique using this transition as the reference. This thesis is organized as follows: In Chapter1,we give a brief introduction to our work.. We review the importance of frequency measurements and precision spectroscopy, followed by a comparison of the frequency comb and our ring cavity technique. In Chapter2, we describe measurements of the absolute frequency of the D lines of 133Cs using the ring cavity. We give a detailed discussion of the technique, the Possible sources of errors, and ways to check for the errors. The measurement of the absolute frequency of the D lines of Cs allows a direct comparison to frequency comb measurements, and thus acts as a good check on our technique. In Chapter 3, we describe the absolute frequency and isotope shift measurements in the 398.8 nm line in Yb. We probed this line by frequency doubling the output of a tunable Ti:Sapphire laser. We obtained< 60 kHz precision in our measurements and were able to resolve several discrepancies in previous measurements on this line. In Chapter 4, we describe the measurement of hyperfine structure in the D1 lines of alkali atoms. We used conventional saturated-absorption spectroscopy in a vapor cell to probe different hyperfine transitions and then used our AOM technique to measure the hyperfine interval with high precision. In Chapter 5 we discuss our measurements of hyperfine structure in the D2 lines of several alkali atoms. In the case of 23Na and 39K, the closely-spaced hyperfine transitions are not completely resolved in conventional saturatedabsorption spectroscopy due to the presence of cross over resonances. We have used coherent control spectroscopy to obtain crossover-free spectra and then measured the hyperfine intervals using an AOM. This technique was also used for high resolution spectroscopy in the D2 line of 133Cs. Finally, we describe our measurements of hyperfine structure in the D2 line of Rb using normal saturated absorption spectroscopy. Chapter 6, describes the relative and absolute frequency measurements in the D lines of6,7 Li at 670nm. High-precision measurements in lithium are of special interest because theoretical calculations of atomic properties in this simple three electron system are fairly advanced. Lithium spectroscopy poses an experimental challenge and we describe our efforts in doing highresolution spectroscopy on this system. Chapter 7 describes the hyperfine spectroscopy on the1P 1 state of 173Yb. Measurement of hyperfine structure in 173Yb has a problem because two of the hyperfine transitions overlap with the transition in 172Yb. In our earlier work (described in chapter 4), we had solved this problem by using multipeak fitting to the partially resolved spectrum. Here, we directly resolve the hyperfine transitions by using transverse laser cooling to selectively deflect the 173Yb isotope. In Chapter 8 , we give a broad conclusion to the work reported in this thesis and suggest future avenues of research to continue the work commenced here.