City University of Hong Kong

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Electronic Engineering  >
EE - Doctor of Philosophy  >

Please use this identifier to cite or link to this item:

Title: Behavior and reliability of radiation-processed, single-mode lensed-fiber affixing joints in pump laser packaging
Other Titles: Ying yong yu beng pu ji guang qi jing guang re chu li jia gong gu ding dan mo tou jing guang xian jie he dian de xing neng biao xian he ke kao xing yan jiu
Authors: Tan, Chee Wei (陳志偉)
Department: Dept. of Electronic Engineering
Degree: Doctor of Philosophy
Issue Date: 2006
Publisher: City University of Hong Kong
Subjects: Lasers
Microelectronic packaging
Solder and soldering
Notes: CityU Call Number: TK7870.15.T36 2006
Includes bibliographical references (leaves 165-175)
Thesis (Ph.D.)--City University of Hong Kong, 2006
xviii, 179 leaves : ill. (some col.) ; 30 cm.
Type: Thesis
Abstract: For the packaging of a pumped laser in a 14-pins butterfly package, i.e. a 980nm pump laser, the most crucial assembly step is the fiber-to-laser diode coupling and fixing. Relatively, the use of laser welding as the fixing method offers several advantages, strong joint strength, short process time and less contamination. The first part of this study reports on laser welds formed by various physical schedules corresponding to various incident beam energies (IBE); laser welding process characteristics, and the behavior of laser-welded joints between pure nickel (Ni200) weld clips and KovarTM base metal under various mechanical loadings, i.e. shear and fatigue in shear were characterized. The mechanical stability of fiber-solder-ferrule (FSF) joints under temperature cyclic loading is discussed. Four different thicknesses of solder filler of FSF joints were examined. In the second part of this thesis, the effectiveness of a pulsed laser beam and soft beam energy as a heat source for an optimum solder joint that fixes a lensed-fiber permanently on a Ni/Au-plated substrate are also reported. The penetration depth and fusion zone of laser spot welds was found to be a complicated function of laser pulse energy, power intensity, and beam diameter. The weld strength was found to be dependent on the overlapping area between the two joining materials which strongly depends on the charge voltage, pulse width, input power, and size of the focal spot to the rate of energy input to the workpieces. The pulse width has a dominant effect on the weld width while the charge voltage dominates the depth of penetration and thus increases the absorbed power density. Surface roughness, Ra, has an influence on the fraction of energy absorbed, A, and therefore, affects the penetration depth. From the results of a simple mathematical model, a higher peak temperature is achieved when the lateral distance is set closer to the heat source, and it reaches a peak temperature in a shorter time. When the lateral distance decreases, the peak temperature increases and the thermal gradient decreases. The heat-affected zone (HAZ) that was produced by the selected incident beam energy (IBE) range is very thin. A larger grain size was obtained in the center of the weld pool that has a lower thermal gradient and a smaller grain size was observed near the cooled wall that has a higher thermal gradient, which is the base metal in this case. A larger number of grains were obtained for weld pools that were produced using a lower IBE. In the fusion zone, a cellular structure and/or an elongated cellular structure were observed due to the effect of the solidification rate on the growth mechanism. A lower temperature gradient will result more often in a cellular structure while a columnar structure is concentrated in regions with a higher temperature gradient, near to the base metal. A Nd:YAG laser source of wavelength 1064nm was used to weld 1.0x1.0x0.2mm Ni200 pieces onto a KovarTM substrate by a single pass spot weld. These samples were then subjected to shear tests. The maximum shear force required to break a joint ranged between 90-95N, 70-75N, and less than 45N when welded by 380V, 360V and 340V, respectively. The shear strength very much depends on the spot weld diameter and beam penetration depth. In general, a higher pulse width at the same energy increases the joint diameter, reduces penetration, i.e., produces conduction welding and thus reduces the shear strength. The condition of dimples observed in the fracture surface after shear tests provides an important indication on the properties of the weld joints. Three different surface conditions of KovarTM-based substrates, i.e., bare KovarTM, Au plated and Ni plated have been employed to form laser-welded joints with pure Ni weld clips. Shear fatigue cyclic tests were performed on four weld spots that joined Ni200 saddle shape weld clips onto the KovarTM base metal. The effects of various welding conditions, i.e., pulse width and charge voltage on the fatigue life of the laser-welded joints were evaluated. It was found that the fatigue life mostly depends on the cross sectional area of the weld joint for the same surface treatment, and that depends on the pulse width and charge voltage. In general, Ni plating and Au plating drastically lower the fatigue life. The fatigue life for the weld formed using the selected welding condition ranges from 1225 to 1497 cycles for a bare KovarTM substrate while it ranges from 24 to 235 cycles for metallized surfaces. The surface condition of a fracture surface after a shear fatigue test provides a direct indication on the crack propagation rate. The fracture surfaces of these samples were observed to be smoother and showed less plastic deformation, which suggested that these joints were brittle. In addition, a shallow sunken fracture mode with more Ni-particles on the fracture surface was observed at the fracture surface of the plated substrates suggesting that these joints had fractured underneath the Ni-plated layer. Important mechanical characteristics of these joints, i.e., constants A’ & K, logN50% and µN were estimated. These values are useful for modeling the fatigue characteristics of laser-welded joints and for the prediction of fatigue life. In this study, four different thicknesses of solder filler of FSF joints were examined. By using a finite element method (FEM), their equivalent creep strains of the eutectic lead-tin solder were compared. The joints were subjected to 5 cycles of a temperature cycling test. The applied temperature profile consisted of a cycle of 15 minutes exposure to a high temperature 150°C, and then 15 minutes exposure to a low temperature of -65°C and the ramping time was about 60 seconds. It was found that the thicker solder filler would be subjected to a larger equivalent creep strain than the thinner solder filler. The discussion and argument surrounded the vertical shift (Y-axis) because it is relatively more sensitive to temperature and has more effects to coupling loss. Modeling and experimental results showed that 0.5 mm is the best inner diameter of a ferrule to provide the lowest displacement and thus the lowest power lost under temperature cycles. PbSn solder was laser-soldered onto solder pads coated with a typical Ni/Au metallization. The results show that the morphology of the microstructure and intermetallic formation is strongly influenced by the laser input power. Overall, well-spread solder joints were obtained at higher laser power inputs, while poor wetting was generally observed at lower power levels with very irregular joint strengths. With sufficient laser input energy, Au dissolved into the molten solder completely, as needle-like AuSn4 intermetallic particles. This exposed the underneath Ni-metallization which then formed a Ni3Sn4 intermetallic layer with Sn that provided the actual joint integrity. In aged samples, an AuSn4 intermetallic layer re-deposited at the joint interface and formed Au0.5Ni0.5-Sn4 with Ni from the metallization underneath. The Au0.5Ni0.5-Sn4 intermetallic layer grew thicker with the aging time. Higher temperatures caused Au0.5Ni0.5-Sn4 intermetallic spalling which moved into the bulk solder, and appeared as a mixture of globular and rod structures. After 1000 hours of aging, Ni3Sn4 intermetallic was also detected in the bulk solder together with Au0.5Ni0.5-Sn4 intermetallic. The microstructure of bulk solders slowly changed from a degenerate structure into a lamellar-like structure, and continuous Pb-rich colonies were observed at the joint interface. In the meantime, changes in the microstructure and growth of intermetallic layers that were due to the thermal history significantly modified the shear strength of these solder joints. Solders, i.e., Pb37Sn, Au20Sn and Sn3.5Ag0.5Cu (SAC) [wt%] were evaluated for a fluxless application using a soft beam heating system. The microstructures of the solder joints have been examined using SEM, in order to understand the response of these solder materials to the focussed white light. Obviously, the exposure time has a greater effect on the soldering temperature before reaching the peak temperature which is determined by the power. A power setting of 40W can reach approximately 340oC; 30W can reach about 310oC while 25W can easily reach 260oC. In general, a soldering temperature higher than the melting temperature is required to form well wetted solder joints for fluxless applications. However, too high an input thermal energy may result in premature aging for the cases of Pb37Sn and SAC, and lateral cracks for the case of Au20Sn. The thermal cracks and voids observed in Au20Sn solder joints were attributed to the fact that the soft beam heating profile does not suit the AuSn preform. Out of these 3 solder types, SAC demonstrated just the right response to the soft beam, i.e, good wetting, a fine and homogeneous structure, and no cracks or other visible failures.
Online Catalog Link:
Appears in Collections:EE - Doctor of Philosophy

Files in This Item:

File Description SizeFormat
fulltext.html157 BHTMLView/Open
abstract.html157 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer