JPS5937045A - Precision machining - Google Patents

Abstract

PURPOSE:To efficiently perform precise thinning machining and precise symmetrical degree machining of silicone wafers, etc., by interposing a freezing layer between a supporting surface and a raw material and freezingly depositing the raw material onto the supporting surface and machining the material in this state. CONSTITUTION:A finishing table 5 consists of a vertically movable rotary board 4 whose surface 3 faces a supporting surface 2 (table surface) of a rotary table 1. On the supporting surface 2 is placed a water-containing sheet 4a, on which silicone wafers 7 are placed and pressed against the surface 2. Further, the surface 2 is supported in water in a container and, in this state, wafers 7 are held at a certain interval apart from the surface 2, the water is frozen to form a frozen layer, the raw material 7 is freezingly deposited onto the surface 2, and in this state, the raw material 7 is machined. In this case, the frozen layer is very thin and able to maintain the bottom surface of the raw material 7 parallel to the surface 2. Thus, the precise thinning machining and precise symmetrical degree machining of raw materials can be effectively performed.

Description

[Detailed description of the invention] This relates to a precision processing method characterized by freezing and adhering a raw material to a supporting surface and processing the raw material in that state, which is suitable for precision thinning processing of silicon wafers, etc. Its purpose is to efficiently carry out precision machining.

To explain the embodiment of the present invention shown in the drawings, a rotary base plate L is provided on the machine frame, and a rotary plate 4 having a surface 3 facing the support surface 2 (base surface) of the rotary base plate 1 is provided so as to be adjustable up and down. In the finishing machine 5, which is made up of paper, cloth,
Textiles, not #! & A water-containing sheet (4a) made of cloth, sponge, etc. is placed on the same sheet (4aJ2+Akira), and a wafer 7 (thin semiconductor board) is placed thereon and pressed against the support surface 2 (Fig. 4).
. In addition, the support surface 2 is attached to the container l6 as shown in FIGS. 5 to 11.
In this state, the wafer 7 is supported in water 4b in the container 16 at a constant distance from the support surface 2, and the water 4b is frozen with 1- to form a freezing layer, and this is removed from the container 16. l￥ The support shaft l7 of the L support surface 2 is supported by the bearing of the rotary table l. As shown in FIGS. 12 to 21, water grooves or water holes 18 are provided on the support surface 2, and the water that rises slightly due to surface tension from the water grooves or water holes 18 is suppressed by the wafer 7 to form a water film layer. (4~,
7 Å or more). In this way, the temperature of the silicon wafer 7 mounting section, that is, the finishing section 6 is kept at -20°C or below (
The silicon wafer 7 (thin semiconductor board) can be supported on the support surface 2 of the rotary table 1. The freezing layer can be formed not only with water but also with other ingredients, such as methyl alcohol, liquid nitrogen, liquid oxygen, etc. In place of the silicon wafer 7, materials such as glass thin pieces (prints or positive glass thin plates for printing wiring on the wafer 7), GGG (gadolinium potassium φ garnet), ferrite, crystal and ceramics, gallium arsenide and other semiconductors and metals can be used. It may be.

Then, as shown in FIG. 1, the rotary plate 1 and the rotary disk 4 are moved relative to each other in the direction of the arrow with appropriate pressure, and the upper surface of the wafer 7 is polished by both abrasion and grinding using an abrasive. The thickness of the wafer 7 can be formed to be 300 μl or less. Thereafter, the ice is melted by electromagnetic waves or other means, the wafer 7 is separated from the support surface 2, the wafer 7 is inverted, the wafer 7 is supported on all sides 2 in the same manner as described above, and the wafer 7 is Polish the top surface of /\
Both the front and back sides of 17 can be finished into parallel surfaces with the same parallel accuracy as the support surface 2 and the facing surface 3. The support surface 2 is a flat surface as shown in FIGS. 1 to 21, but may be a spherical or curved surface as shown in FIGS. 22 and 23. In addition, what is shown by 8 in the figure is a rotary suction chuck, 9 is a machinable material, and 2a
10 is a spherical support surface, 10 is a crystal oscillation element, and 11 is a spherical support surface.
12 is a rotary wind chuck, 13 is a crystal oscillation element, 14 is a machinable material supporting the crystal oscillation element 13, 15 is a curve generator for machining a spherical surface, and 16 in FIG. 4 is a dust particle.

In this case, the conventional ENO-7 is adhered to the supporting surface 2 using an adhesive. However, since the adhesive is flexible, it is difficult to hold the lower surface of the wafer 7 and the support surface 2 parallel to each other, and when the adhesive is spread, it reaches the limit of its molecular structure and loses its adhesive strength.
There was a defect.

The present invention has been made in view of the above-mentioned defects, and as described above, a freezing layer is interposed between the supporting surface and the raw material, the raw material is frozen and adhered to the supporting surface, and the raw material is frozen and adhered to the supporting surface in that state. Since this is a precision processing method characterized by processing, the supporting surface and the raw material are in close contact with each other through a frozen layer of water and other components, and the frozen layer is extremely thin (4 to 7 people in the case of water) and the supporting surface 2 and the raw material 7. The raw material 7 can be held parallel to the bottom surface of the raw material 7.
This has the effect of improving the symmetry accuracy and parallelism accuracy of the material 7, and significantly reducing the thickness of the raw material 7.

[Brief explanation of the drawing]

Figure 1 is a front view showing the processing state according to the present invention, Figure 2 is a plan view of the supporting surface of the raw material 1, Figures 3 and 4 are front views of the raw material supporting state, and Figure 5 is the supporting surface. 6, 7, 8, 9, 10, and 11 are other embodiments of FIG. A front view of the surface and the wafer in a frozen state in the container, FIG. 12 is a plan view of another embodiment of the support surface,
13 is a front view of FIG. 12, FIG. 14 is a plan view of another embodiment of the support surface, FIG. 15 is a front view of FIG. 14, and FIG. 16 is a front view of FIG.
17 is a plan view of another embodiment of the support surface, FIG. 17 is a front view of FIG. 16, FIG. 18 is a plan view of another embodiment of the support surface, and FIG. 19 is an enlarged front view of FIG. 18. , FIG. 20 is a plan view of another embodiment of the support surface, FIG. 21 is an enlarged front view of FIG. 20,
FIGS. 22 and 23 are side views of the curved surface attached state, respectively. 2. Support surface, 7. Raw material. Patent applicant Yoshiaki Nagaura

Claims (1)

[Claims] 1) A precision processing method characterized by interposing a freezing layer between the supporting surface and the raw material, freezing and adhering the raw material to the supporting surface, and processing the raw material in this state. 2) The precision machining method according to claim 1, wherein the supporting surface is flat. 3) The precision machining method according to claim 1, wherein the supporting surface is a spherical surface. 4) The precision machining method according to claim 1, wherein the support surface is a rotary table surface. 5) The precision processing method according to claim 1, wherein the raw material is a thin semiconductor plate (urchin/\-). 6) The precision processing method according to claim 1, wherein the raw material is silicone urethane. 7) The precision processing method according to claim 1, wherein the raw material is crow. 8) The precision processing method according to claim 1, wherein the raw material material is GGG. 9) Particularly, the raw material is ferrite, i1 Claim 1
Precision machining method described in section. 10) The precision processing method according to claim 1 sn, wherein the raw material is crystal. +1) The precision processing method according to claim 1, wherein the raw material is ceramic. 12) The precision processing method according to claim 1, wherein the raw material is potassium arsenic. 13) The precision processing method according to claim 1, wherein the frozen layer is formed of a water-containing layer. +4) The precision machining method according to claim 1, which is a frozen layer or a frozen water layer. 15) The precision processing method according to claim 1, wherein the freezing layer is liquid nitrogen. 16) The precision machining method according to claim 1, wherein the frozen layer is liquid oxygen.