JPS5974640A - Bonder for wire - Google Patents

Abstract

PURPOSE:To prevent the mixing of air into a necessary gas by the curtain action of an inert-gas flow, to form a ball of the high degree of vacuum and to improve bondability by forming a section on the capillary intruding side of a shield cylinder in double structure and flowing an inert gas to the inside. CONSTITUTION:A bonding head 11 is loaded on an XY table 10, one end of a bonding arm 12 is supported to the head 11 in a vertically vibratory shape, and a capillary 13 is fixed at the nose of the arm 12. An approximately L-shaped electrode 19 is arranged where adjacent to the capillary 13, and the discharge section 20 of a lower side is vibrated as a nose. The electrode 19 is formed to a hollow-pipe shape, a tube 21 communicated with a gas source is connected at an end section on the axial support 18 side, the electrode 19 is supplied with the necessary gas. A nozzle 23 is formed at the nose of the discharge section 20, a shield inner cylinder 24 is set up tightly so as to cover the discharge section 20 of the electrode 19, and the cylinder 24 is covered with a shield outer cylinder 25. The inert gas is flowed through the inner cylinder 24, and the mixing of air into the necessary gas is prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a wire bonder with good bondability.

One of the manufacturing processes for semiconductor devices is a so-called wire bonding process in which a semiconductor element pellet fixed to a lead frame is electrically connected to the lead frame. and,
Wires used for this wire bonding include Au wire and A4. lj! However, in recent years, the low-priced A
T-line is used. Furthermore, even when At@'5 is used, the so-called ultrasonic bonding method is constrained in the wire bonding direction, which makes the bonding work troublesome and the structure of the wire bonder complicated. head bonding method) has come to be adopted.

That is, this wire bonding method was previously proposed by the present applicant, and as shown in FIG. 4, and the electrode 4
The tip of the At wire is covered with a shield tube 5 integrally provided in the shield tube 5, and the inside of the shield tube 5 is maintained in a predetermined gas atmosphere (for example, H2 gas), and in this state, a high temperature is created between the At wire 3 and the electrode 4. A ball is formed at the tip of the At wire by applying a voltage and generating a discharge.

Needless to say, after the ball is formed, the At wire is connected to the pellet or lead frame by thermocompression bonding in the same manner as in the case of the Au wire.

By the way, according to the inventor's experiments, it has been found that the better the sphericity of the ball formed at the tip of the At wire 3, the better the reliability of the connection strength of wire bonding, etc.
It has also been found that in order to form a ball with good sphericity, it is important to keep the N2 gas atmosphere at the tip of the At wire constant during discharge.

However, in the conventional device as described above, a notch hole is formed in a part of the shield tube 5, and the capillary 2 is inserted into the shield tube through the notch hole. Outside air (02, etc.) easily enters the shield cylinder through the gap between the shield cylinder, notch, and hole (there is a problem that it is difficult to maintain a constant N2 gas atmosphere inside the shield cylinder, that is, at the tip of the At wire). be.

In particular, since the shield cylinder is constructed integrally with the electrode 4 and is configured to be moved from a retracted position on the side of the capillary to the capillary position when forming a ball, the shield cylinder is moved to the capillary position and stopped. When this happens, a so-called air entrainment phenomenon occurs, causing air and nitrogen to enter the cylinder.
This causes a turbulent flow of mixture with the two gases, and the stability of the 142 gas atmosphere for forming the ball with high sphericity described above becomes extremely poor.

Therefore, an object of the present invention is to
It is an object of the present invention to provide a wire bonder capable of keeping the gas atmosphere at the tip of the wire in a stable state, thereby forming a ball with high sphericity and completely improving bondability.

In order to achieve such an object, the present invention has a double structure in the part of the shield cylinder on the capillary entry side, and allows the required gas to flow back inside and inert gas to flow outside.
This curtain effect of the inert gas flow is intended to prevent air from entering the required gas.

Hereinafter, the present invention will be explained with reference to illustrated embodiments.

FIG. 2 shows a wire bonder according to an embodiment of the present invention.
A bonding arm 12 is pivoted at one end to a bonding head 11 mounted on a table 10 so as to be able to swing up and down.
Further, a capillary 13 is fixed to the tip of this arm 12. A wire wound around a spool (not shown), that is, an AL wire 14 is inserted into the capillary 13, and the tip of the At wire is made to protrude below the capillary 13. A bonding stage 15 is arranged below the capillary 13,
On this stage 15, a semiconductor structure 16 to which wire bonding is to be performed is moved along a line.

On the other hand, an abbreviated-shaped electrode 19 whose upper end is pivotally supported 18 by a supporter 17 is disposed near the capillary 13, and the entire structure can be swung in a direction perpendicular to the plane of the paper with the discharge section 20 on the lower side as the tip. That's what I do. This electrode 19 is the third
The entire structure is shaped like a hollow pipe, as shown in detail in the figure.
A tube 21 communicating with a gas source (not shown) is connected to the end on the shaft support 18 side, and a necessary gas (N2 gas in this example) is supplied into the electrode. Further, a plurality of gas holes are formed on the upper side of the discharge portion 20 of the electrode 19 to blow out the N2 gas supplied inside. Furthermore, the tip of the discharge section 20 is slightly drawn to form a nozzle 23. The electrode 19 is arranged in a shield inner cylinder 24 so as to cover the discharge part 20.
A shield outer cylinder 25 having an oval-shaped cross section is further fixed to cover the shield inner cylinder 24. At this time, as shown in FIG. 4, the shield inner cylinder 24
Although the shield outer cylinder 25 and the shield outer cylinder 25 are in close contact with each other on the lower side, a gap 26 is defined between them from both sides to the upper side. In addition, holes 27 and 28 for the capillary 13 to enter are completely formed on the upper surfaces of the shield inner cylinder 24 and outer cylinder 25. Note that the void 2
6 is opened at the distal end position, while an inert gas (N2 gas in this example) is supplied through the entire cheep 29 connected to the shield outer cylinder 25 at the proximal end side.

According to the above configuration, the capillary 13 is moved downward after the electrode 19 is swung to one side as shown in the figure, and the capillary is inserted into the hole 28.2 of the shield outer cylinder 25 and the shield inner cylinder 24.
7 and insert the tip of the HT wire 14 into the shield inner cylinder 24.
If a high voltage is applied between the At wire 14 and the electrode 19, a discharge will occur between the discharge portion 20 and the tip of the A4 wire 14, and the energy will form a ball at the tip of the At wire. At this time, if N2 gas or other reducing gas is supplied into the electrode 19, the N2 gas will flow through the gas hole 2 of the discharge section 20.
2 is also ejected into the shield inner cylinder 24 and inside the shield inner cylinder,
In other words, the sphericity of the ball is improved by keeping the atmosphere at the tip of the At wire in an N2 gas atmosphere.

At the same time, when N2 gas is supplied into the shield outer cylinder 25 through the tube 29'f, the N2 gas is forced to flow through the gap 26 between the shield outer cylinder 25 and the inner cylinder 24. A so-called curtain action is generated to isolate the shield inner cylinder 24 and the shield outer cylinder 25. Therefore, the N2 gas inside the shield inner cylinder 24 flows into the hole 27.
The air outside the shield outer cylinder 25 is prevented from passing completely through the holes 28 and 27 and flowing into the shield inner cylinder 24.

This maintains a stable N2 gas atmosphere inside the shield inner cylinder 24 and improves the sphericity of the ball.

In addition, even immediately after the electrode 19 is swung, the curtain action of the N2 gas prevents air from entering the shield inner cylinder, so turbulence due to air intrusion does not occur. As a result, it becomes possible to stably maintain the atmosphere inside the shield cylinder even with high-speed rocking of the electrode, which is effective for high-speed wire bonding.

Here, the N2 gas supplied into the electrode 19 is also discharged from the nozzles 2 and 3 at the tip, but in order to prevent the environment from deteriorating due to the dissipation of the N2 gas into the outside air, the N2 gas is discharged from the nozzle 23.
It is designed to burn two gases. Of course, nozzle 23
It is also possible to connect an exhaust tube to the tube and lead it out to a predetermined processing device.

After the ball is formed, the electrode 19f swings upward to evacuate the discharge portion from below the capillary 13, and then the bonding arm 12 is actuated to attach the ball to the semiconductor structure 1.
Needless to say, the wire bonding is completed by crimping the wire bonding on the wire 6.

As described above, according to the wire bonder of the present invention, the shield tube has a double structure, and the required gas is passed inside and the inert gas is passed outside, and the curtain action of this inert gas allows the required gas to be contained. Since the mixture of air is prevented, the gas atmosphere at the tip of the At wire can be maintained stably, which improves the sphericity of the ball formed on the At wire and improves bondability. It has the effect of being able to do it.

[Brief explanation of drawings]

Fig. 1 is a cross-sectional view of a part of a conventional wire bonder, Fig. 2 is an overall configuration diagram of the wire bonder of the present invention, Fig. 3 is an enlarged sectional view of main parts, and Fig. 4 is an enlarged cross-section taken along the line AA in Fig. 3. It is a diagram. 11... Bonding head, 12... Bonding arm, 13... Capillary, 14... At wire,
16... Semiconductor structure, 19... Electrode, 20... Discharge part, 22-... Gas hole, 24... Shield inner cylinder, 25
...7-old outer cylinder, 26... air gap, 27.28...
hole. Agent Patent Attorney Susuki 1) Toshiyuki τ〉-2, No.
1 figure

Claims (1)

[Claims] 1 A4 wire is used as the bonding wire, and a discharge is generated between the tip of this A4 wire and the discharge part of the electrode.
This wire bonder forms a ball for thermocompression at the tip of the T wire, and has a shield tube attached to the electrode so as to cover the tip of the At wire when discharge occurs, and this shield tube is attached to the tip of the At wire. It has a double structure with a shield inner cylinder that covers the shield and an outer shield cylinder provided outside the shield, and while the required gas is passed through the shield inner cylinder, an inert gas is passed between the two cylinders. A wire bonder characterized in that the shield inner cylinder is isolated from the outside air by the curtain action of the inert gas.