GB/T 4937.22-2018Semiconductor devices -- Mechanical and climatic test methods -- Part 22: Bond strength (English PDF)

1.1 Test Description

Seven test methods are specified, each of which has different purposes, as follows.

--- Method A and Method B. Test the bonding strength inside the device by directly applying a tensile force to the inner leads;

--- Method C. for bonding outside the device, applying pull-off stress between the lead or the lead and the wiring board or substrate;

--- Method D. for internal bonding, applying shear stress between the chip and the substrate or to a similar surface bonding structure;

--- Method E and Method F. for external bonding, applying a push-on stress or a pull-up stress between the chip and the substrate;

--- Method G. Used to test the strength of wire bond shear resistance.

1.2 Test device (applicable to all methods)

The test device shall include equipment that applies specified stresses at the bond points, leads or terminals as required by the specified test method. The device

Calibrated measurements and indications can be provided for stress applied in failure (in N). The device measures force within 100mN (package

Including 100mN) should have an accuracy of +/-2.5mN, the force between 100mN and 500mN should have an accuracy of +/-5mN, and the measurement is super

A force of over 500 mN should have an accuracy of +/-2.5% of the indicated value.

2 Method A. Lead Tension (Single Bond Point) and Method B. Lead Tension (Double Bond Point) (see Appendix A)

2.1 Scope

The method is suitable for use in a semiconductor device package with internal leads bonded by soldering, thermocompression bonding, ultrasonic welding and other related technologies

Lead-to-die bonding, lead-to-substrate bonding, or inner lead-lead bonding.

2.2 General description of the test

2.2.1 Method A procedure

The leads connecting the chips or the substrate are cut so that the tensile test can be performed at both ends. In the case of a short lead, close to a certain end

The lead is broken to perform a tensile test at the other end. For pin-type bonding, use appropriate equipment to secure the leads, then lead or clamp the leads

The device of the wire applies a pulling force that does not exceed an angle of 5° with the surface of the chip or substrate.

2.2.2 Method B Procedure

A hook is inserted under the lead connecting the chip or the substrate and the lead end, and the hook is applied to the hook after the device is fixed. Try to be on the lead

The center applies a pulling force, and the angle between the pulling force and the normal line of the chip or the substrate surface or the perpendicular to the line connecting the two bonding points is not more than 5°.

2.2.3 Force applied (applicable to Method A and Method B)

Gradually apply a pulling force to the lead or bond point to crack [a in 2.2.4), or to achieve b) in the specified minimum tensile force 2.2.4.

2.2.4 Failure criteria

The failure criteria are as follows.

a) Record the value of the pull force at the time of cracking of the lead or bond point and compare it with the values given in Table 2 to determine if it is acceptable. in case

There is no corresponding lead diameter in Table 2, and the curve in Figure 3 should be used to determine the limit value of the bond pull. This curve only applies

The bonding tension is perpendicular to the chip.

b) Gradually apply tension to the specified minimum. If the lead and bond points are not cracked, the bond is considered acceptable.

Note. Method B corrects the pull value according to the information provided in Appendix A.

2.2.5 Failure category

When specified, the cracking of the leads or bonding points is classified as follows.

a) the lead is broken at the necking point (the position where the section is reduced due to the bonding process);

b) the lead is broken at the non-necked point;

c) the bonding on the chip (the interface between the lead and the metallization layer) fails;

d) the bonding of the substrate, the package stud or the non-chip position (the interface between the lead and the metallization layer) fails;

e) the metallization layer is lifted from the chip;

f) the metallization layer is lifted from the substrate or package terminal;

g) chip rupture;

h) The base is ruptured.

Note. Method B is not recommended for measuring the absolute value of bond strength (see Appendix A). However, this method can be used as a bond quality monitoring in the production process.

3.1 Scope

This method is typically used for external bonding of device packages.

3.2 Procedure

The lead or the lead and the device package are fixed, and a pulling force is applied at an angle between the lead or the lead and the wiring board or the substrate. If

There is no other rule, the angle is 90°.

3.3 Applied force

Gradually apply tension to the lead (or lead) or bond point crack (see 3.4.1), or to achieve the specified minimum pull (see 3.4.2).

3.4 Failure criteria

3.4.1 Record the tensile value at the point of cracking of the bond point and compare it with the values given in Table 2 to determine if it is acceptable. If not in Table 2

For the corresponding lead diameter, the curve in Figure 3 should be used to determine the limit of the bond pull. This curve only applies to bond pull perpendicular to

The case of a wiring board or a pedestal. The tension is gradually applied to the specified minimum value, and only the bonding point itself is first invalidated, and the test result is effective. only

Cracking of the bond point itself can be considered as invalidation.

3.4.2 Gradually apply tension to the specified minimum. If the lead (or the lead) and the bonding point are not cracked, the bond is considered

qualified.

3.5 Failure category

When specified, the cracking of the leads or terminals or bonding points should be classified as follows.

a) the lead or the lead end is cracked in the deformation zone (welding affected zone);

b) the lead or the lead end is cracked in the non-bonding process affected zone;

c) failure at the bonding interface (welding in the solder or between the lead or the lead and the wiring board or the substrate conductive layer bonded thereto)

interface);

d) the conductive layer is lifted from the wiring board or the substrate;

e) The inside of the wiring board or substrate is broken.

4.1 Scope

This method is generally used to verify the internal bonding between the semiconductor chip of the surface bonding structure and the substrate. Can also be used to inspect substrates and installation

Bonding between the intermediate carrier or sub-substrate of the chip.

4.2 Procedure

Use a suitable tool or file to contact the chip (or carrier) at a location just above the main substrate, perpendicular to the chip (or

Stress is applied to one side of the body and parallel to the direction of the main substrate, and the bonding failure is caused by the shear stress.

4.3 Applied force

Gradually apply tension to the bond point to crack (see 4.4.1), or to achieve the specified minimum stress (see 4.4.2).

4.4 Failure criteria

4.4.1 Record the stress values at the time of bond failure to determine if it is acceptable. The stress should be no less than 50mN times the number of bonding points. Stress applied

When the bonding point itself first fails, the test result is valid. Only the crack of the bond point itself can be considered as invalid.

4.4.2 Alternatively, increase the stress to 50mN times the number of bonding points. If the bond point and the substrate or chip are not cracked,

The bond can be considered acceptable.

4.5 Failure Category

When specified, failures should be classified as follows.

a) failure of the bonding material or bonding base when applicable;

b) the chip (or carrier) or substrate is broken (ie, a portion of the chip or substrate immediately below the bonding point moves);

c) The metallization layer is lifted (ie, the metallization layer or bonding pedestal is separated from the chip/carrier or substrate).

5.1 Scope

This method is used for devices with beam leads.

Method E is typically used for process control, for semiconductor wafers bonded to a specific substrate, and therefore not suitable for production batch or inspection

Random sampling in the inspection lot.

Method F is typically used for beam lead samples bonded to a ceramic substrate or other suitable substrate.

5.2 Method E procedure

A metallized substrate with holes is used. The hole should be close to the center of the substrate and large enough to provide clearance for the push-off tool, but not so large

Bonding area. The push-off tool should be large enough to minimize device breakage during the test, but not as large as the beam fixed in the bond area

Lead contact.

Firmly secure the substrate and insert the push-open tool through the small hole. The push-off tool should not cause significant impact when it comes into contact with the device (less than

0.25mm/min). The lower side of the bonding device is applied at a constant rate to a value specified in 5.5 or a failure occurs.

5.3 Method F Procedure

The calibrated pull-out device shall include a pull-off tool (eg, a nichrome wire thermoelectric ring) that is attached to the beam lead chip

A hard setting adhesive material (for example, a heat-sensitive polyvinyl acetate resin) is connected. Make sure that the adhesive material does not spread under the beam or the bottom of the chip

unit. The base is securely held in the pull-out fixture and the pull-out tool should maintain a fixed mechanical connection with the adhesive material. Within 5° of the device normal

Gradually increase the stress, at least to the value specified in 5.5 or the upper surface of the chip is 2.5 mm away from the substrate.

5.4 Failure criteria for Method E and Method F

The failure criteria are as follows.

a) the semiconductor chip is broken;

b) the beam lead is lifted from the bonding point;

c) the beam lead breaks at the bond;

d) the beam lead is broken at the edge of the semiconductor chip;

e) the beam lead is broken between the bonding point and the edge of the semiconductor chip;

f) the bonding point is lifted from the substrate;

g) The metallization layer is lifted (the metallization layer is separated from the chip or bonding region).

5.5 Applied Force (Applicable to Method E and Method F)

For a nominal beam width of non-deformed (before bonding), a force of 500 mN is applied per mm (straight line). Bond strength should be divided by breaking stress

The nominal beam width of the front is determined.

6.1 Scope

Method G was used to test the strength of wire bond shear resistance. It is recommended to add this method outside of Method A and Method B. It can provide

More information about the robustness of metallurgical bonding, relative to methods A and B, is more concerned with bonding itself than with exposure

Bonding is not directly related to failure (such as breakage of the lead at the root, neck, and span).

The method provides a standard procedure for detecting a series of spherical bond shear strengths bonded by thermoforming or thermosonic bonding techniques.

6.2 Overview

Bonding shearing uses a chisel-like tool to cut or push a spherical or wedge-shaped bond point away from the bond area (see Figure 1). Record score

The force of departure is used as the bonding shear force. If the bonding shear force of the gold ball bonding is related to the diameter of the bonding ball, the gold ball bonding point and the key may be indicated.

The metallurgical bonding quality between the metallization layers of the junction is shown in Figure 2 and Table 1. Aluminum wedge bond bonding shear force, with the metal of the carrier

The wire tensile strength comparison can indicate the integrity of the weld between the aluminum wire and the metallization layer of the bond zone.

Spherical bonding includes a larger spherical surface or a nail head portion of the lead (hot-welding or first-time bonding operation in thermocompression or ultrasonic welding or both)

In the process, the base bonding zone and the spherical bonding bonding zone interface or the welding surface are obtained.

These test methods are spherical bonds of small diameter leads (from 18µm to 76µm), which are commonly used in integrated circuits and hybrid microelectronics.

In the component. Only when the height and diameter of the ball are large enough and there is sufficient distance from the adjacent influential structure, allowing the periphery of the bonded ball to be removed

The area so that the shear test tool is placed between adjacent bond points.

This method is destructive. Suitable for process development or process control or quality assurance with a suitable sampling plan.

6.3 Terms and definitions

6.3.1

Cut mode definition (see Figure 1) definitionofshearmodes

6.3.1.1

Mode 1---bonding out of mode1-ballift

The entire ball bond point is detached from the bond area leaving only an impression on the bond area. A small amount of intermetallic compound residue on the bonding area

Or no residue.

6.3.1.2

Mode 2 - Bonding tear/ball bond tear/wedge bond shear mode2-bondshear/balshear/wedgeshear

The bonded spheres are separated from the intermetallic compound. An intermetallic compound of a bonding sphere and a portion of gold remain on the metallization layer of the bonding region.

6.3.1.3

Mode 3 - Bonding area is out of mode3-bondpadlift

The bonding zone metallization layer is separated from the substrate. Some bonding zone metallization layers and intermetallic compounds are attached to the bonding balls.

6.3.1.4

Mode 4---Sag (broken) mode4-cratering(chip-out)

The bonding area is lifted up and a part of the substrate is taken up. The bonding area metallization layer and a portion of the substrate are attached to the bonding balls.

6.3.1.5

Mode 5a---instrument/operator error (cutting tool too high) mode5a-instrument/operatorerror(sheartooltoo

High)

The sample is not placed properly, the shear height is too high or the instrument is malfunctioning. The cutting tool only removes the top portion of the spherical or wedge bond point.

6.3.1.6

Mode 5b---instrument/operator error (cutting tool too low) mode5b-instrument/operatorerror(sheartooltoo

Low)

The sample is not placed properly, the shear height is too low, or the instrument is malfunctioning. The cutting tool is in contact with the metallization layer or the protective coating, resulting in an ineffective

Cut value.

a) Mode 1 ball lift b) Mode 2 bond shear - gold/aluminum

Figure 1 Bonding shear mode

c) Mode 3 Bonding area lifting d) Mode 4 Depression

e) Mode 5a Cutting tool is too high f) Mode 5b Cutting tool is too low

Figure 1 (continued)

6.3.2

Cutting tool or cutting arm sheartoolorarm

A tungsten carbide file or equivalent tool with a certain angle at the bottom and back ensures shear movement. The cutting tool should be at least flat

The shearing surface, the sharp cutting edge, and the shearing surface of the bonding diameter or bonding length of 1.5 to 2 times the width.

The cutting tool is designed to ensure that no wave motion or drag movement occurs during the test. Tools should be clean, no gaps or other shadows

The defect of the shear test.

6.3.3

Thermosonic gold (Au) spherical bond thermosonicgold (Au) balbond

Gold ball bonding by a thermosonic wire bonding process.

6.3.4

Ultrasonic aluminum (Al) wedge bonding ultrasonicaluminium (Al) wedgebond

Aluminum wedge bonding or welding by ultrasonic wire bonding process.

6.4 Equipment and materials

6.4.1 Equipment capabilities

The equipment used in the lead ball bond shear test shall have the following capabilities.

a) Spherical bond shear test equipment is capable of accurately positioning the shear arm (+/-2.54µm) above the substrate, equipped with a zoom microscope (most

Small magnification 70X).

b) The designed shearing tool shall ensure that no wave motion or drag movement occurs during the test. Tools should be clean, no gaps or

He affects the defects of the shear test.

c) During the test, the clamp shall be fixed horizontally and perpendicular to the shearing tool and shall have sufficient clamping force to ensure the specimen

Safety.

d) The spherical bond shear test apparatus is capable of accurately positioning the shear arm (+/-2.54 µm) above the substrate. It is at the top of the bonding surface

The distance should ensure that the cutting tool does not touch the surface of the chip and is less than the top of the bonding surface to the center of the spherical or wedge bond point

The distance of the line.

e) Optical microscope/measurement equipment capable of measuring the bond diameter (with an error of +/-2.5 µm).

6.4.2 Calibration

Before the bond shear test, make sure that the equipment has been calibrated according to the manufacturer's instructions and is within the calibration validity period. If equipment

The placement position changes and needs to be recalibrated.

6.5 procedure

6.5.1 For specific devices, the sample size shall be the minimum specified by the SPC control. Standard sample size for process development, unless otherwise specified

At least 50 bonded balls should be randomly selected.

6.5.2 The test is only applicable to unmolded samples, and the ball height (>=10.16µm) and the ball spacing should be large enough to be completely and unobstructed.

Apply spherical bond shear.

6.5.3 Install the test equipment according to the manufacturer's instructions, check the cutting tool, and replace if damaged.

6.5.4 Place the test sample horizontally in the fixture, perpendicular to the shear tool.

6.5.5 Place the cutting tool close to the bonding point (distance less than 2 bonded ball diameters) perpendicular to the surface of the device and less than the key

Half of the height of the joint is shown in Figure 1 mode 1 to mode 4. The bottom of the cutting tool should be kept below the centerline of the bond point, chip bonding

Above the metallization layer.

6.5.6 Start the test button and move the cutting tool to the bond point until the bond ball is sheared off or separated from the substrate.

6.5.7 Check the shear zone, record the shear mode (as defined in 6.3.1) and the value of the bond ball shear force in grams.

6.5.8 Repeat the test for each test sample in the batch (6.5.3~6.5.7).

6.5.9 Calculate and record the following values for each sample.

a) shear force average (gf);

b) minimum value (gf);

c) maximum value (gf);

d) Variance.

6.5.10 Data for invalid tests should not be included in determining the average shear force. Including improper placement of the cutting arm (too high, too low or not touching)

Bonding balls) and other tests that affect the shear test, such as mode 5a and mode 5b.

6.5.11 Calculate the average value of the bond shear strength per test lot using the known (or measured) bond ball diameter for each test sample.

Minimum value (average stress in grams/average size of known bonded spheres).

6.6 Acceptable test limits

6.6.1 Acceptable test limits for spherical bonds

If the single minimum and a...