Figure 4. Stacked structure of the patterned different material inside the base material green sheet.
as is shown in Figure 5. This sample was supposed for the single layer of the multilayer inductor. The base material was designed for the ferrite magnetic material and the patterned material was nonmagnetic material. The internal conductor was supposed to be screen printed on the nonmagnetic layer and then the minor magnetic loop between the internal conductors was expected to be suppressed.
3.2. Preparation of Powders and Slurries
The weight ratio of the glass powder and the alumina powder in the glass alumina composite material was 63 and 37. The glass powder was composed of SrO-B2O3- Al2O3-SiO2 and the particle size was about 1 μm. The particle size of the alumina powder was 0.2 μm. The ferrite powder was synthesized by calcination process. The start materials were NiO, ZnO, CuO and Fe2O3 and the molecular ratio was 8.8-32-10-49.2. The ferrite powder was crashed to 0.3 μm by ball mill method.
Mixing the glass powder, the alumina powder, organic solvents and some additives by the ball mill method for 3 hours, poly vinyl butyral (PVB) was added. The ferrite powder was mixed with organic solvents and some additives for 24 hours. After adding PVB binder, the mixing time was 24 hours.
It is often that the glass alumina composite slurry contains phosphoric ester as a dispersant agent. However, the phosphoric ester is known as a sintering inhibitor of the ferrite. Therefore, the slurry contained the non phosphoric ester dispersant.
Different from usual ceramic slurry, filling and patterning capability is required in this method. In order to satisfying these requirements, the rheologic properties of both low viscosity and high thixotropy were required. For this reason, the slurry was adjusted to the appropriate property, varying the content of the binder and organic solvent. Moreover, the density of the ferrite is higher than that of the glass alumina composite. Therefore, the weight ratio of the ferrite powder contents was made larger than that of the glass alumina powder in order to adjust similar volume contents. The adjusted slurry composition is shown in Table 1. The organic solvent was mixture of toluene, xylene and isopropyl alcohol. Also additives such as dispersant and plasticizer were used. In the case of the ferrite, two kinds of specified slurry were arranged for each the patterning and the blank layer.
3.3. Patterning Procedure
First step, the photo resist film was exposed and developed for forming the sacrifice pattern that was corresponding to the through pattern. The thickness of the photo resist film used for the specimen was 90 μm and the film was attached on the Poly Ethylene Terephthalate (PET) carrier film.
The strength of the ultraviolet light used for the exposure was adjusted to 160 mW/cm2 at the gross exposure energy. The exposed film was developed in the developing solution of 1 wt% sodium carbonate. Soaked in the developing solution for 5 minutes, it was rinsed in distilled water at room temperature. After that, the photo resist film was soaked again in the developing solution for 1 minute and then rinsed in distilled water to remove residual material. Then the designed sacrifice pattern was achieved after dry process at room temperature.
The ceramic slurry was coated on the PET film where the sacrifice pattern had been formed. The coating method was doctor blade. At this process, the gap between the blade and the surface of the sacrifice pattern was adjusted to zero. Therefore, the slurry was filled surrounding the sacrifice pattern. The speed of the blade was 10 mm/sec.
Figure 5. Pattern with “J” figure.
Table 1. Composition ratios of slurry.
After the obtained sheet was dried, the sacrifice pattern was dissolved. The dissolving solution was tetra methyl ammonium hydroxide (TMAH) and the solution temperature was 45 degrees Celsius. Through these process the green sheet with the designed through pattern was achieved.
The thickness of the photo resist film used for masking patterned ceramic was 35 μm. The exposure condition of the mask pattern was same condition as that of the sacrifice pattern. The exposed resist film was soaked in the same developing solution for 3 minutes and then rinsed in distilled water at room temperature. After that, the resist film was soaked in the developing solution for 1 minute and then rinsed in distilled water to remove residual material. The alignment of the formed resist film and the patterned ceramic was performed with an equipment of the X-Y-θ stage. The heat press was used for bonding of the mask film and the patterned ceramic. The laminating temperature was 70 degrees Celsius, and the pressure was 20 MPa. The press time was 2 minutes.
The doctor blade method was used for the filling the slurry of different material into the through pattern around the patterned ceramic. The same filling condition as the base ceramic was used in this process. After drying of the green sheet, the resist film for the mask was dissolved in TMAH. The dissolving condition was same as that of the through pattern. Then, the green sheet that was formed the pattern of different ceramic material inside was achieved.
3.4. Lamination and Heat Press
The green sheet that was patterned different ceramic material was dried for 3 hours at 80 degrees Celsius in a dry oven. Then the sheet was removed from the PET film. The removed green sheet was diced into 25.4 mm square, and also the blank sheet of the same material was diced same way. These sheets were stacked in the designed order, and then these sheets were laminated by the heat press with a 1 inch die. The lamination temperature was 70 degrees Celsius, the pressure was 20 MPa. The press time was 2 minutes. After the pressure bonding process, the specimen was diced in 5 mm square.
Each specimen was evaluated by microscopic observation by optical microscope and Scanning Electron Microscope (SEM: Hitachi S-4500). The viscosity was measured by rheology meter (BROOKFIELD DV-II+ Pro).
4. Result and Discussion
4.1. Viscosity of the Slurry
The viscosity of the each ceramic slurry used in this study is shown in Figure 6. The each slurry was adjusted
Figure 6. Viscosity of the slurries.
to be thixotropic. In the present method, the viscosity of the slurry was important because the each slurry is filled surrounding the sacrifice pattern and the through pattern. Therefore, the viscosity of the slurry for the pattern was adjusted to be lower than that the conventional green sheet. Particularly, the high thixotropy of the slurry for the ferrite pattern was achieved. Moreover, the ferrite slurry for the blank layer was adjusted to be low viscosity to obtain the same dry shrinkage as the slurry for the patterning.
4.2. Evaluation and Discussion of Patterning
The optical microscopic image of the obtained green sheets is shown in Figures 7 and 8, which are corresponding to the top view and cross section, respectively. The white and black areas are the glass alumina and ferrite, respectively. It is observed that the designed pattern was formed inside the single sheet and the border was clear in Figure 7. Although, the border of each material is not very clear in Figure 8, the cross section might receive damages during cutting preparation of the sample. The thickness of each material is almost same in Figure 8. Therefore, it is found that the adjustment of the slurry compositional ratio and the thickness of the resist film were appropriate, and the sufficient filling and the homogenous drying shrinkage were also achieved.
In Figure 8, the round bump shape is observed around edge part of the ferrite pattern. The reason is thought to be surface tension of the slurry. The filled slurry shrank during the drying process, and then, peeling off the resist film. The edge part became the round bump shape by the surface tension. After removing the resist film, the bump shape remained. The schematic illustration of the mechanism is shown in Figure 9.
Moreover, it is important that the thickness of the resist film for the mask to achieve the flat different patterns with same thickness. For example, in the case of using the thinner resist film of 15 μm, the difference of the
Figure 7. Top view of the green sheet with the pattern of the different material; white part is glass alumina and black part is ferrite. (a) Base material was glass alumina and the patterned material was ferrite; (b) Base material was ferrite and the patterned material was glass alumina.
Figure 8. Cross sectional view of the green sheet with the pattern of the different material; light part is glass alumina and dark part is ferrite. (a) Base material was glass alumina and the patterned material was the ferrite; (b) Base material was ferrite and the patterned material was glass alumina.
thickness between glass alumina sheet and ferrite pattern was 20 μm. The cross sectional image of the 15 μm case is shown in Figure 10. This phenomenon is occurred by the shortage of the filled slurry amount because of the thin resist film. The schematic illustration of the mechanism of the different thickness is shown in the Figure 11.
The optical microscopic image of the “J” figure is shown in Figure 12. The line width of the pattern was 444 μm. This pattern was non magnetic part. Inside the pattern, the conductor pattern should be formed. About 200 μm of the conductor width, which is standard in the width of conductive wire in the multilayer inductor, is suitable.
Figure 9. Schematic illustration of the mechanism of the round bump shape.
Figure 10. Cross sectional view of the green sheet with the pattern of the different material using the 15 mm resist film mask.
Figure 11. The schematic illustration of the mechanism of the different thickness.
Figure 12. Top view of the green sheet with the pattern for the inductor; white part is glass alumina and black part is ferrite.
The patterning process of the ferrite and the insulation material for the multilayer ceramic device was proposed. First, the green sheet with the through pattern was formed by making use of the sacrifice pattern of the photo resist film. Then the photo resist film that was patterned in the same position of the remained ceramic area was attached for masking the ceramic. The different ceramic material was filled only the through pattern through the mask. After drying process, the resist film for the mask was dissolved and removed. Finally the green sheet with the different material pattern inside the sheet was achieved.
In this study, the ferrite as magnetic material and the glass-alumina composite materials as the insulation material were patterned inside the green sheet each other. By adjusting slurry composition and film thickness, measure structural damage was not observed in the green sheet and the laminated structure.
The fabrication of the specimen was supported by Research Center for Micro Functional Devices of Nihon University. This work was supported by KAKENHI (22560254). The authors appreciate the support.