Loading…
Glass multilayer bonding for high density interconnect substrates
The aim of this research was the investigation of bonding borosilicate glass sheets, its trade mark CMZ, 100μm thickness, to create multilayer substrates capable of supporting high-density electrical interconnections. CMZ glass was chosen as it has a coefficient of thermal expansion that is close to...
Saved in:
| Main Author: | |
|---|---|
| Format: | Default Thesis |
| Published: |
2010
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/6463 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1818176061873061888 |
|---|---|
| author | Fuad Mohamed Hassan Khoshnaw |
| author_facet | Fuad Mohamed Hassan Khoshnaw |
| author_sort | Fuad Mohamed Hassan Khoshnaw (7202303) |
| collection | Figshare |
| description | The aim of this research was the investigation of bonding borosilicate glass sheets, its trade mark CMZ, 100μm thickness, to create multilayer substrates capable of supporting high-density electrical interconnections. CMZ glass was chosen as it has a coefficient of thermal expansion that is close to that of silicon, thereby minimising thermal stresses in assemblies generated by manufacturing processes or service conditions. Two different methods of bonding the glass were used in this study; pressure assisted low temperature bonding (PALTB), and water glass bonding, using Sodium Trisilicate (Na2Si3O7) solution. These two bonding methods have already been applied in electronics manufacturing applications, such as silicon wafer bonding and multichip modules (MCMs). However, glass-to-glass bonding is a relatively new subject and this study is an attempt to standardise bonding processes. Additionally, the concept of using glass as a multilayer substrate provides a foundation for further exploration by other investigators. Initial tests that were carried out before standardising the procedures for these two methods showed that a two-stage bonding process provided optimum results. A preliminary stage commenced by placing the cleaned (using Decon 90 solution) samples in a vacuum oven for 15 minutes, then heating at 100oC for 1hr. The permanent stage was then achieved by heating the samples in a conventional oven at temperatures from 200 to 400oC, for different periods. At this stage, the main difference between the two methods was the application of pressure (1-2MPa) during heating of the PALTB samples. To evaluate the quality of the bonds, qualitative tests such as visual, optical microscope and dye penetrant were used. In addition, to estimate the strength and the rigidity of the interlayer bonds, two quantitative tests, comprising of deflection under cyclic stresses and crack opening were used. Thermal cycling and humidity tests were also used to assess resistance of the bonds to environmental effects. The results showed that heating to 100oC was insufficient to enhance the bonds, as occasionally a sudden increase of deflection was observed indicating slippage/delamination. These bonds were enhanced during the permanent bonding stage by heating to 300oC in PALTB, under a pressure of 1-2MPa. The crack-opening test showed that the delamination distances of the bonds in the permanent stage were lower than that for preliminary bonding in both bonding methods. The delamination distances from the crack opening tests were used to calculate the strain energy release rate (GIC) and fracture toughness (KIC) values of the interlayers. The results showed that the KIC values of the permanent PALTB and water glass interlayers were higher than 1MPa.m0.5, while the KIC value of the CMZ glass, determined by linear elastic fracture mechanics, was around 0.8MPa.m0.5. The optical observations revealed that the prepared bonded sheets did not delaminate or break after thermal cycling and humidity tests. |
| format | Default Thesis |
| id | rr-article-9533309 |
| institution | Loughborough University |
| publishDate | 2010 |
| record_format | Figshare |
| spelling | rr-article-95333092010-01-01T00:00:00Z Glass multilayer bonding for high density interconnect substrates Fuad Mohamed Hassan Khoshnaw (7202303) Mechanical engineering not elsewhere classified untagged Mechanical Engineering not elsewhere classified The aim of this research was the investigation of bonding borosilicate glass sheets, its trade mark CMZ, 100μm thickness, to create multilayer substrates capable of supporting high-density electrical interconnections. CMZ glass was chosen as it has a coefficient of thermal expansion that is close to that of silicon, thereby minimising thermal stresses in assemblies generated by manufacturing processes or service conditions. Two different methods of bonding the glass were used in this study; pressure assisted low temperature bonding (PALTB), and water glass bonding, using Sodium Trisilicate (Na2Si3O7) solution. These two bonding methods have already been applied in electronics manufacturing applications, such as silicon wafer bonding and multichip modules (MCMs). However, glass-to-glass bonding is a relatively new subject and this study is an attempt to standardise bonding processes. Additionally, the concept of using glass as a multilayer substrate provides a foundation for further exploration by other investigators. Initial tests that were carried out before standardising the procedures for these two methods showed that a two-stage bonding process provided optimum results. A preliminary stage commenced by placing the cleaned (using Decon 90 solution) samples in a vacuum oven for 15 minutes, then heating at 100oC for 1hr. The permanent stage was then achieved by heating the samples in a conventional oven at temperatures from 200 to 400oC, for different periods. At this stage, the main difference between the two methods was the application of pressure (1-2MPa) during heating of the PALTB samples. To evaluate the quality of the bonds, qualitative tests such as visual, optical microscope and dye penetrant were used. In addition, to estimate the strength and the rigidity of the interlayer bonds, two quantitative tests, comprising of deflection under cyclic stresses and crack opening were used. Thermal cycling and humidity tests were also used to assess resistance of the bonds to environmental effects. The results showed that heating to 100oC was insufficient to enhance the bonds, as occasionally a sudden increase of deflection was observed indicating slippage/delamination. These bonds were enhanced during the permanent bonding stage by heating to 300oC in PALTB, under a pressure of 1-2MPa. The crack-opening test showed that the delamination distances of the bonds in the permanent stage were lower than that for preliminary bonding in both bonding methods. The delamination distances from the crack opening tests were used to calculate the strain energy release rate (GIC) and fracture toughness (KIC) values of the interlayers. The results showed that the KIC values of the permanent PALTB and water glass interlayers were higher than 1MPa.m0.5, while the KIC value of the CMZ glass, determined by linear elastic fracture mechanics, was around 0.8MPa.m0.5. The optical observations revealed that the prepared bonded sheets did not delaminate or break after thermal cycling and humidity tests. 2010-01-01T00:00:00Z Text Thesis 2134/6463 https://figshare.com/articles/thesis/Glass_multilayer_bonding_for_high_density_interconnect_substrates/9533309 CC BY-NC-ND 4.0 |
| spellingShingle | Mechanical engineering not elsewhere classified untagged Mechanical Engineering not elsewhere classified Fuad Mohamed Hassan Khoshnaw Glass multilayer bonding for high density interconnect substrates |
| title | Glass multilayer bonding for high density interconnect substrates |
| title_full | Glass multilayer bonding for high density interconnect substrates |
| title_fullStr | Glass multilayer bonding for high density interconnect substrates |
| title_full_unstemmed | Glass multilayer bonding for high density interconnect substrates |
| title_short | Glass multilayer bonding for high density interconnect substrates |
| title_sort | glass multilayer bonding for high density interconnect substrates |
| topic | Mechanical engineering not elsewhere classified untagged Mechanical Engineering not elsewhere classified |
| url | https://hdl.handle.net/2134/6463 |