JEDEC
STANDARD
M
ic
ro
n
Wire Bond Shear Test Method
JESD22-B116B
(Revision of JESD22-B116A, August 2009)
APRIL 2017
JEDEC SOLID STATE TECHNOLOGY ASSOCIATION
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
NOTICE
JEDEC standards and publications contain material that has been prepared, reviewed, and
approved through the JEDEC Board of Directors level and subsequently reviewed and approved
by the JEDEC legal counsel.
JEDEC standards and publications are designed to serve the public interest through eliminating
misunderstandings between manufacturers and purchasers, facilitating interchangeability and
improvement of products, and assisting the purchaser in selecting and obtaining with minimum
delay the proper product for use by those other than JEDEC members, whether the standard is to
be used either domestically or internationally.
JEDEC standards and publications are adopted without regard to whether or not their adoption
may involve patents or articles, materials, or processes. By such action JEDEC does not assume
any liability to any patent owner, nor does it assume any obligation whatever to parties adopting
the JEDEC standards or publications.
n
The information included in JEDEC standards and publications represents a sound approach to
product specification and application, principally from the solid state device manufacturer
viewpoint. Within the JEDEC organization there are procedures whereby a JEDEC standard or
publication may be further processed and ultimately become an ANSI standard.
ic
ro
No claims to be in conformance with this standard may be made unless all requirements stated in
the standard are met.
M
Inquiries, comments, and suggestions relative to the content of this JEDEC standard or
publication should be addressed to JEDEC at the address below, or refer to www.jedec.org under
Standards and Documents for alternative contact information.
Published by
©JEDEC Solid State Technology Association 2017
3103 North 10th Street
Suite 240 South
Arlington, VA 22201-2107
This document may be downloaded free of charge; however JEDEC retains the
copyright on this material. By downloading this file the individual agrees not to
charge for or resell the resulting material.
PRICE: Contact JEDEC
Printed in the U.S.A.
All rights reserved
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
PLEASE!
DON’T VIOLATE
THE
LAW!
This document is copyrighted by JEDEC and may not be
reproduced without permission.
n
For information, contact:
M
ic
ro
JEDEC Solid State Technology Association
3103 North 10th Street
Suite 240 South
Arlington, VA 22201-2107
or refer to www.jedec.org under Standards-Documents/Copyright Information.
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
n
ro
ic
M
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 1
TEST METHOD B116B
WIRE BOND SHEAR TEST
(From JEDEC Board Ballot JCB-17-10, formulated under the cognizance of the JC-14.1 Subcommittee
on Reliability Test Methods for Packaged Devices.)
1
Scope
This test provides a means for determining the strength of a ball bond to a die or package bonding surface,
and may be performed on pre-encapsulation or post-encapsulation devices. This measure of bond strength
is extremely important in determining two features:
1) the integrity of the metallurgical bond which has been formed, and
2) the quality of ball bonds to die or package bonding surfaces.
This test method covers thermosonic (ball) bonds made with small diameter wire from 15 µm to 76 µm
(0.6 mil to 3.0 mil).
ro
n
This test method can only be used when the bonds are large enough to allow for proper contact with the
shear test chisel and when there are no adjacent interfering structures that would hinder the movement of the
chisel. For consistent shear results the ball height must be at least 4.0 µm (0.16 mils) for ball bonds, which
is the current state of the art for bond shear test equipment at the time of this revision.
M
ic
This test method can also be used on ball bonds that have had their wire removed and on to which a 2nd bond
wire (typically a stitch bond) is placed. This may be known as “stitch on ball” and “reverse bonding”. See
Annex A for additional information.
The wire bond shear test is destructive. It is appropriate for use in process development, process control,
and/or quality assurance.
This test method may be used on ultrasonic (wedge) bonds, however its use has not been shown to be a
consistent indicator of bond integrity. See Annex B for information on performing shear testing on wedge
bonds.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 2
2
Terms and definitions
For the purposes of this standard, the following terms and definitions apply:
2.1
ball bond: The adhesion or welding of a small diameter wire, typically gold or copper, to a
bonding surface metallization, usually an aluminum alloy, using a thermosonic wire bond process.
NOTE 1 The ball bond includes the enlarged spherical, or nail-head, portion of the wire (provided by the flame-off
and first bonding operation), the underlying bonding surface and the ball bond-bonding surface metallurgical weld
interface.
NOTE 2 Gold wire implies a gold alloy in which the gold content is likely 99% or greater. Copper wire implies a
copper alloy of similarly high copper content and also includes copper wire with a very thin coating of palladium.
NOTE 3 At the time of this revision, other wire materials and wire coatings are being evaluated, but there is not
enough information collected to confirm that the fail modes listed in this test method are valid for any of the new wire
types.
2.2
bonding surface: Either 1) the die pad metallization or 2) the package surface metallization to
which the wire is ball bonded.
ro
n
2.
bond shear: A process in which an instrument uses a chisel-shaped tool to shear or push a ball
bond off the bonding surface (see Figure 1).
M
ic
NOTE The force required to cause this separation is recorded and is referred to as the bond shear force. The bond
shear force of a ball bond, when correlated to the diameter of the ball bond, is an indicator of the quality of the
metallurgical bond between the ball bond and the bonding surface metallization.
Figure 1 — Bond shear set-up for bond on die bonding pad
(Similar setup for bonds on other bonding surfaces, such as package substrate/leadframe)
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 3
2
Terms and definitions (cont’d)
2.4
shear tool; shear arm: A chisel (made of tungsten carbide or an equivalent material with similar
mechanical properties) with specific angles on the bottom and back of the tool to ensure a shearing action.
2.5
stitch bond: The second bond during the ball (thermosonic) bonding process, in which the wire is
typically bonded to the package bonding surface.
NOTE 1 A stitch bond may also be referred to as a crescent bond.
NOTE 2 For some unique constructions (e.g., “stitch on ball”), the second bond may be formed on top of another ball
bond, from which the wire has been removed.
2.6
wedge bond: The adhesion or weld of a thin wire, typically aluminum, copper, or gold to a die pad
metallization or the package bonding surface, usually a plated leadframe post or finger, using an ultrasonic
wire bonding process.
NOTE See Annex B for information on performing shear testing on wedge bonds.
Apparatus and material
n
3
Inspection equipment
ic
3.1
ro
The apparatus and materials required for bond shear shall be as follows:
M
An optical microscope system or scanning electron microscope providing a minimum of 70X magnification.
A higher magnification may be necessary for 15 µm (0.6 mil) diameter wire.
3.2
Measurement equipment
An optical microscope/measurement system capable of measuring the bond diameter to within ± 2.54 µm
(0.10 mil).
3.3
Workholder
Fixture used to hold the part being tested parallel to the shearing plane and perpendicular to the shear tool.
The fixture shall also eliminate part movement during bond shear testing. If using a caliper controlled
workholder, place the holder so that the shear motion is against the positive stop of the caliper. This is to
ensure that the recoil movement of the caliper controlled workholder does not influence the bond shear test.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 4
3.4
Bond shear equipment
The bond shear equipment must be capable of repeatable, precision placement of the shearing tool with
respect to the ball height and the bonding surface. The specified distance (h) above the topmost part of the
bonding surface (e.g., passivation layer on IC, solder mask on organic substrate) shall ensure the shear tool
does not contact the bonding surface (e.g., top passivation or polyimide layer, soldermask) and shall be less
than the distance from the topmost part of the bonding surface to the center line (CL) of the ball bond (see
Figure 2). See Annex C for guidance when the passivation, or other structures on the die surface and
excessive Al splash prevent the shear tool from contacting the ball below the center line.
The ball height ‘Center
Line’ is ½ the distance
from the top of the ball
to the top of the
bonding surface
C
L
ro
n
Passivation
layer
h
Bond shear chisel tool setup
M
3.5
ic
Figure 2 — Proper height placement of shear tool with respect to ball center line
When choosing the proper chisel for the bond being sheared items to consider include but are not limited to:
flat shear face, sharp shearing edge, shearing width of a minimum of 1.2X the bond diameter, and bond
length. The sample and chisel face should be clean and free of chips or other defects that will interfere with
the shearing test.
Bonds should also be examined to determine if adjacent interfering structures are far enough away to allow
suitable placement and clearance (above the bonding surface and between adjacent bonds) for the shear test
tool.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 5
4
Procedure
4.1
Calibration
Before performing the bond shear test, it must be determined that the equipment has been calibrated in
accordance with manufacturer's specifications and is presently in calibration. Recalibration is required if the
equipment is moved to another location.
4.2
Visual examination of bonds to be tested after decapsulation
n
In addition to being a manufacturing process monitor, this test method can also be used to assess bond
strength of encapsulated devices after soldering operations or after reliability stress testing. To do this, the
encapsulation material needs to be removed in a manner that does not significantly degrade the wire, the
bond, the bonding interface, or the bonding surface. Shear force values are often lower for bonds that have
been decapsulated, and therefore cannot be compared to values for similar, unencapsulated bonds. If the
decapsulation process is well controlled and repeatable, which is the case for gold wire, then this test
method can be used for lot to lot comparison; however, it may be hard to consistently control the
decapsulation process for copper wires to ensure the accuracy of the results. For Cu wires, the effectiveness
of etch has been seen to vary due to the encapsulation material and the level of reliability stress testing
performed on the samples. See Annex D for additional information regarding the decapsulation process of
devices with Cu wire bonds.
4.2.1
M
ic
ro
Bonds must also be examined to determine that enough encapsulation material has been removed to allow
suitable placement and clearance (above the bonding surface and between adjacent bonds) for the shear test
tool.
Bond pad examination and acceptability criteria for both Al and Cu bond pad metallization
If performing bond shear testing on a device which has been opened using wet chemical and/or dry etch
techniques, the bond pads shall be examined to ensure there is no absence of metallization on the bonding
surface area due to chemical etching, and wire bonds are attached to the bonding surface. Those bonds on
Al or Cu bond pads with significant chemical attack or absence of metallization shall not be used for ball
shear testing. The shear results for any damaged bonds found during post shear inspection may also be
excluded. It is possible that wire bonds on bonding surfaces without degradation from chemical attack may
not be attached to the bonding surface due to other causes (e.g., package stress). These wire bonds are
considered valid and shall be included in the shear data as a zero (0) shear force value.
4.2.2
Copper bond and Cu wire examination and acceptability criteria
If performing bond shear testing on a part with copper wires, the Cu bond and Cu wire shall be examined
before or after the shear test to ensure there is no significant loss of metal or other damage due to
decapsulation process that might affect the results of the shear test. The shear result can be excluded for a
Cu bond or Cu wire with significant chemical attack or other damage due to the decapsulation process.
Annex D provides additional information to assess what level of damage is acceptable.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 6
4.3
Measurement of the ball bond diameter to determine the ball bond shear failure criteria
Once the bonding surfaces have been examined and before performing bond shear testing, the diameter of
all ball bonds to be tested shall be measured and recorded. The ball is measured at the widest point of the
ball bond. For symmetrical ball bonds (those basically round) only one measurement per bond needs to be
taken.
For asymmetrical bonds, determine the average diameter using both the largest (dlarge) and the smallest
(dsmall) diameter values (see Figure 3). These two ball bond diameter measurements shall be used to
determine the mean, or average, diameter value. The resulting mean, or average, ball bond diameter shall
then be used to establish the failure criteria as defined in bond shear qualification standards.
Diameter
Measured at widest
point of the ball
ASYMMETRICAL
n
SYMMETRICAL
ro
Figure 3 — Ball bond measurement: side view and top view (for symmetrical vs. asymmetrical)
M
ic
To determine whether the sheared bond has passed the acceptability criteria that is stated in JESD47, the
shear force must be divided by the ball bond area. The equation for area is:
A = πr2 = π(d)2/4
Where d is the above measured diameter (or mean diameter for an asymmetrical ball bond) for the bond
being sheared.
To facilitate faster testing a statistically representative ball bond diameter may be used with all of the bonds
sheared within a sample when calculating the shear force per unit area for each bond sheared (in lieu of
using the corresponding ball diameter for each ball sheared to calculate its shear force per unit area value).
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 7
4.4
Performing the bond shear test
The bond shear equipment shall pass all self-diagnostic tests before beginning the test. The bond shear
equipment and test area shall be free of excessive vibration or movement. Examine the shear tool to
verify it is in good condition and is not bent or damaged. Check the shear tool to verify it is in the up
position.
Adjust the workholder to match the part being tested. Secure the part to the workholder. Make sure the
surface of the die is parallel to the shearing plane of the shear tool. It is important that the shear tool does
not contact the surface of the die or adjacent structures during the shearing operation as this will give
incorrect high readings.
Position the part so that the bond to be tested is located adjacent to the shear tool. Lower the shear tool,
or raise the part depending upon shear equipment used, to approximately the height from which the bond
is to be sheared but not contacting the surface. (See Figure 2, distance “h”).
Position the ball bond to be tested so that the shear motion will travel perpendicular to the surface edge.
Position the shear tool within approximately the diameter of one ball of the bond to be shear tested and
shear the bond.
Examination of sheared bonds
ic
4.5
ro
n
Historical data shows that variation in shear speed may have a slight effect on the shear results for gold
ball bonds. Care must be taken to ensure that if test results from different wire bonding lots are to be
compared that any variation in test speed be taken into account.
M
All bonds shall be sheared in a planned/defined sequence so that later visual examination can determine
which shear values should be eliminated because of an improper shear. The bonds shall be examined per
codes in 4.6 using at least 70X magnification to determine if the shear tool skipped over the bond (type 5)
or the tool scraped or plowed into the surface of the die (type 4). Type 4 and type 5 defective shear
conditions are invalid, and shall be eliminated from the shear data (see Figure 4).
Sheared bonds in which a type 3 cratering condition has occurred shall be investigated further to
determine whether the cracking and/or cratering is due to a preexisting condition in the silicon and/or
metallization under the bond pad prior to the bonding operation or was due to the act of bonding.
Cratering resulting from the bonding process shall be considered valid and included in the shear data.
Any bonds with a preexisting condition in the silicon and/or metallization under the bond pad are invalid
for this test method and shall not be included with the shear data. If a preexisting condition in the silicon
and/or metallization under the bond pad is found to cause cratering, it must be addressed.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 8
4.6
Bond shear codes for ball bonds
Wire bond separated from
bonding surface and no
evidence of bond formation
Wire bond separated from bonding
surface. Little or no intermetallic
formed during bonding operation
Slight imprint on
bonding surface
Bonding surface
TYPE 1: Bond Lift
Copper/Aluminum, Copper/Copper,
and Gold/Gold
TYPE 1: Bond Lift
Gold/Aluminum
Layer of bonding surface
metalization remains on
wire bond
Wire bond separated from
bonding surface and no
evidence of bond formation
Bonding surface has
some metal removed,
and visual evidence
of metal shear
M
ic
ro
n
Slight imprint on
leadframe/substrate
bonding surface
TYPE 1: Bond Lift
All metal systems on leadframe or
substrate
Intermetalic formation on wire bond
and bonding surface.
Bonding surface
TYPE 2: Bond Shear – Gold/Aluminum
Variation B – Separation wholly within
intermetalic layer
TYPE 2: Bond Shear – All metal systems
Variation A – Separation within bonding
surface metalization
Separation at interface with
bonding surface and (mandatory)
evidence of shear on bonding
surface.
Bonding surface
TYPE 2: Bond Shear – All metal systems
and surfaces, except Gold/Aluminum
Variation B – Separation at bonding surface
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 9
Major portion of wire bond
Shear occurs at both the material
Interface and within bulk material.
TYPE 2: Bond Shear – All metal systems
and bonding surfaces
Variation C – Separation at material
interface and within bulk material
Ball or Wedge bonding
weld area intact. Visual
evidence of metal shear
TYPE 2: Bond Shear – All metal systems
Variation D – Separation within ball bond
Residual bonding surface
and substrate (bulk) material
attached to wire bond.
Bonding surface lifted taking
portion of substrate (bulk)
material.
M
ic
ro
Ball or Wedge bonding
weld area intact. Visual
evidence of metal shear
n
Major portion of wire bond
TYPE 2: Bond Shear – All metal systems
on leadframe or substrate
Variation D – Separation within ball bond
TYPE 3: Cratering
Minor portion of wire bond
attached to wire.
Arm contacted bonding
surface metallization
instead of wire bond.
Bonding surface separated
from die surface.
TYPE 4: Bonding Surface Contact
Wire bond sheared too
high. Only portion of
wire bond removed
TYPE 5: Shearing Skip
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 10
Metallization on leadframe
or substrate bonding pad
remains with the ball bond
Underlying bonding pad
metallization remains with
the ball bond
Bonding surface metallization
separated from leadframe or
substrate.
Bonding surface metallization
separated from die surface.
TYPE 6: Leadframe or Substrate bond pad
or bonding surface metalization lift
TYPE 6: Bonding Pad Surface Lift
Figure 4 — Bond Shear Codes
4.6.1 Type 1 - bond lift: A separation of the entire wire bond from the bonding surface with only an
imprint being left on the bonding surface (see Figure 5).
ro
n
NOTE A bond lift may require an assessment of the bonder settings and/or cleanliness or integrity of the bonding
surface.
ic
4.6.1.1 Type 1 - bond lift - gold/aluminum: Has the additional requirement of very little evidence of
intermetallic formation, welding, or shearing of the bonding surface metallization.
M
4.6.1.2 Type 1 - bond lift - copper/aluminum, copper/copper, and gold/gold: Has the additional
requirement of no visual evidence of shearing of the bonding surface metallization.
NOTE 1 The copper/aluminum system forms a very thin intermetallic layer that is not generally visible. When the
bond wire and the bonding surface are the same material (e.g., copper/copper, gold/gold) no intermetallic is formed.
NOTE 2 Copper/copper includes the bonding of Cu ball bonds onto bond pad structures that have Cu metallization
with a barrier metal. The barrier metal may also include other thin layers to prevent oxidation of the barrier metal.
Some commonly used structures include Cu-NiPd and Cu-NiPdAu.
4.6.1.3 Type 1 - bond lift from leadframe/substrate: Has the additional requirement of no visual
evidence of disturbance of the bonding surface metallization.
Figure 5 — Imprints on Al pad from lifted bonds with no evidence of shearing (Type 1)
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 11
4.6.2
Type 2 - bond shear
4.6.2.1 Type 2 - bond shear - gold/aluminum
A separation of the wire bond with visual evidence of shearing of the bulk metal where:
a) a thin layer of the bonding surface metallization remains with the sheared wire bond and there is visual
shearing of the bonding surface metallurgy (see Figure 6);
b) the shear occurs wholly within the intermetallic layer with intermetallics remaining on the bonding
surface and with the sheared wire bond (see Figure 7);
c) the shear occurs on multiple planes, partially within the intermetallic layer and within the bulk material;
or
d) the shear occurs above the intermetallic layer and solely within bulk material of the sheared wire bond
(see Figure 9).
4.6.2.2 Type 2 - bond shear - copper/aluminum, copper/copper, and gold/gold
A separation of the wire bond with visual evidence of shearing of the bulk metal where:
n
a) a thin layer of the bonding surface metallization remains with the sheared wire bond and there is visual
shearing of the bonding surface metallurgy (see Figure 6);
ro
b) the shear occurs wholly at the interface with the bonding surface and there must be evidence of sheared
metal on the bonding surface;
M
ic
c) the shear occurs on multiple planes, partially at the interface with the bonding surface and partially
within the bulk material, (see Figure 8); or
d) the shear occurs above the material interface and solely within bulk material of the sheared wire bond
(see Figure 10).
NOTE The copper/aluminum system forms a very thin intermetallic layer that is not generally visible. When the
bond wire and the bonding surface are the same material (e.g., copper/copper, gold/gold) no intermetallic is formed.
For this reason the definition of Variation B for 4.6.2.2 (and 4.6.2.3) does not use the term intermetallic and thus is
different from 4.6.2.1.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 12
4.6.2.3 Type 2 - bond shear on leadframe/substrate
A separation of the wire bond with visual evidence of shearing of the bulk metal where:
a) the shear occurs wholly at the interface with the bonding surface and there must be evidence of sheared
metal on the bonding surface;
b) the shear occurs in both the material interface and within the bulk material, and a portion of each
remains on the bonding surface and with the sheared wire bond (see Figure 7); or
c) the shear occurs above the material interface and solely within bulk material of the sheared wire bond.
Figure 7 — Shear wholly within gold/aluminum
intermetallic layer (Type 2 - Variation B)
M
ic
ro
n
Figure 6 — Shear of aluminum pad
(with copper wire) (Type 2 - Variation A)
Figure 8 — Shear in bulk copper ball bond and
at material interface (Type 2 - Variation C)
Figure 9 — Shear wholly within gold ball bond
(Type 2 - Variation D)
Figure 10 — Shear wholly within Cu ball bond
(Type 2 - Variation D)
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 13
4.6.3
Type 3 – cratering
A condition under the die pad metallization in which the insulating layer (oxide or interlayer dielectric) and
the bulk material (silicon) separate or chip out (see Figure 11).
NOTE 1 Separation interfaces that show pits or depressions in the insulating layer (not extending into the bulk) are
not considered craters.
NOTE 2 Cratering can be caused by several factors including the wire bonding operation, the post-bonding
processing, and even the act of shear testing itself.
n
NOTE 3 Detailed failure analysis is required to determine whether a cratering fail is due to a preexisting condition in
the silicon and/or metallization under the bond pad or due to the bonding process (see Figure 12). Any bonds that are
found to have a preexisting silicon issue are invalid for this test method and their results shall not be included with the
shear data, but the preexisting condition must be addressed.
M
ic
ro
Two optical images of cratering
SEM image of cratering 1
Figure 11 — Bond pad cratering after shear test
Figure 12 — Bond pad cratering (pad and ball view) and validation of crack and
thin Al on another pad
1
R.T.H. Rongen, A. van IJzerloo, C. Cotofana, K.M. Lan, “Cratering response method to study the effect of ultrasonic energy on
Cu-wire”, Microelectronics Reliability 51 (2011) 1865–1868
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 14
4.6.4
Type 4 - arm contacts specimen (bonding surface contact)
The shear tool contacts the bonding surface to produce an invalid shear value (see Figure 13).
NOTE This condition may be due to improper placement of the specimen, a low shear height, or instrument
malfunction. This bond shear type is an invalid result and shall be eliminated from the shear data.
Figure 13 — Images of shear tool contacting the bonding surface (shear tool set too low)
4.6.5
Type 5 - shearing skip
n
The shear tool removes only the topmost portion of the ball (see Figure 14).
M
ic
ro
NOTE This condition may be due to improper placement of the specimen, a high shear height or instrument
malfunction. This bond shear type is an invalid result and shall be eliminated from the shear data.
Figure 14 — Images of shearing skip (shear tool set too high)
4.6.6
Type 6 - bond pad (or bonding surface) lift
A separation between the bonding surface metallization and the underlying substrate or base material with
evidence of bonding surface metallization remaining attached to the ball (see Figure 15).
NOTE This condition if observed on unencapsulated devices may indicate weak adhesion of bond pad, leadframe, or
substrate metallurgy/films as compared to the bond strength, and may need to be addressed in the wire bonding
process, wafer fab back end of line process, leadframe plating line, or substrate plating processes.
Figure 15 — Images of bonding surface lifting
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 15
4.7
Bond shear data
Data shall be maintained for each bond sheared. The data shall identify:
•
•
•
the bond (location, ball diameter, wire material, method of bonding, and material bonded to),
the shear strength, and
the shear code as defined in clause 4.6.
The following are additional data that are recommended to be recorded:
•
•
•
•
5
ball height
device/package
wire bonder
shear tester machine settings
Summary
The following details shall be specified in the applicable procurement documents:
ro
n
a) sample size: number of devices tested and number of ball bonds per device under test, if different
from that stated in JESD47;
JESD47;
minimum accepted process capability data (Ppk), for unencapsulated bonds only;
M
c)
ic
b) minimum accepted shear value per unit area for an individual bond, if different from that stated in
d) acceptability and/or non-acceptability of any fail modes;
e)
sample preconditioning, and/or any stressing prior to shear testing.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 16
Annex A (informative) Performing this test method on “stitch on ball” bonds
Figures A.1 and Figure A.2 show close up views of what it commonly called “stitch on ball” or “reverse
bonding”.
Figure A.1 — Top view of “stitch on ball” bond
Figure A.2 — Side view of “stitch on ball” bond
This test method can be used for “stitch on ball” bonds, also known as “reverse bonding”. Bond shearing
can be performed on three distinct bonds/bond combinations:
ro
n
1) on the ball bond that has had the wire sheared off, but before the second wire has been stitched to it;
2) on the ball for the second bonding operation;
3) on the first ball bond after the stitch bond has been applied.
M
ic
The placing of the stitch bond on the first ball can induce additional thermal and mechanical stress to the
first ball bond; and if there were an issue with a bonding process, performing shear after each of the two
bonding steps would be a useful tool to isolate at which step the issue was occurring.
No evaluation has been done at this time to determine whether the shear modes and force values for a
standard ball bond fully apply to the shearing of “stitch on ball” bonds.
Figure A.3 — Die to die bonding
Figure A.4 — “Reverse” bond, with ball on leadframe
Figures A.3 shows “die to die” bonding with the ball (with sheared wire) is first formed on the right die and
then the second wire has the ball formed on the left die and its stitch is placed on top of the ball on the right
die. Figure A.4 shows “leadframe to die” bonding, which allows for a very low profile package. In this
image a ball (with sheared wire) is first formed on the die bond pad and then the second wire has the ball
formed on the leadframe and its stitch is placed on top of the ball on the die.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 17
Annex B (informative) Performing this test method on ultrasonic wedge bonds
Much work has occurred within the industry to correlate shear results for gold ball bonds and more recently
copper ball bonds across multiple wire diameters and bonding tools to allow the industry to use the shear
test method as a tool to gauge the integrity of thermosonic ball bonds. When this test method was being
revised, the working group did not find a similar amount of data and industry rigor to support the use of this
test method for ultrasonic wedge bonds. It was still very much felt that the test method can still provide
useful, qualitative data on the integrity of ultrasonic wedge bonds. For this reason it was decided to move
those portions of the test method that deal solely with shear testing of wedge bonds into this Annex.
The subclauses in this annex provide additional information that is unique for shear testing of ultrasonic
wedge bonds and are aligned with the clauses of the B116 Test Method.
NOTE The text in B.1 through B.4 may be additional information to a clause or may replace the existing clause as
noted. If this annex skips over a clause in the test method, then there is no unique information that applies only to
wedge bonds for that clause. B.5 does not exist in the test method, it is an item unique only to shearing of ultrasonic
wedge bonds.
B.1
Scope (Additional text to 1):
ro
n
This test method covers ultrasonic (wedge) bonds made with larger diameter wire (minimum of 3 mils) of
the type used in integrated circuits and hybrid microelectronic assemblies.
M
ic
This test method can only be used when the bonds are large enough to allow for proper contact with the
shear test chisel and when there are no adjacent interfering structures that would hinder the movement of the
chisel. For consistent shear results the wire height at the compressed bonded area for wedge bonds should
be roughly ¼ the wire diameter, and at a minimum be at least 32 µm (1.25 mils) in height. This test method
may also apply to gold wedge bonds if the wire is thick enough to allow shearing of the wire from the
bonding surface without smearing the wire, similar to a shearing skip.
B.2
Terms and definitions
B.2.1 bonding surface (Modified definition for 2.2): Either 1) the die pad metallization or 2) the
package surface metallization to which the wire is wedge bonded.
B.2.2 bond shear (Modified definition for 2.3): A process in which an instrument uses a chisel-shaped
tool to shear or push a wedge bond off the bond pad.
NOTE The bond shear force of a wedge bond, when compared to the manufacturer's tensile strength of the wire, is an
indicator of the integrity of the weld between the wire and the bond pad or package surface metallization.
B.2.3 bond footprint (of a wedge bond) (New definition replaces 2.7): The area of the wire that has a
physical bond interface (intermetallic or recrystallized) with respect to the compressed area of the wire.
NOTE The wedge bond includes the compressed (ultrasonically bonded) area of the wire and the underlying bonding
surface. When bonding aluminum wire to an aluminum alloy die bond pad, or bonding copper wire to a copper alloy
leadframe there is no wedge bond-bond pad intermetallic because the two materials are of the same composition, but
the two materials are recrystallized together by the ultrasonic energy of the welding process.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 18
B.3
Apparatus and equipment
B.3.1
Bond shear equipment (Replaces 3.4)
The bond shear equipment must be capable of repeatable, precision placement of the shearing tool with
respect to the wedge height and the bonding surface. The specified distance (h) above the topmost part of
the bonding surface (e.g., passivation layer on IC, solder mask on organic substrate) shall ensure the shear
tool does not contact the bonding surface (e.g., top passivation or polyimide layer, soldermask) and shall be
less than the distance from the topmost part of the bonding surface to the center line (CL) of the wedge bond.
B.4
Procedure
B.4.1
Performing the bond shear test (Replaces 4.4)
Position the wedge bond to be tested so that the shear motion will travel toward the long side of the
wedge bond and is free of any interference (i.e., shear the outside bond first and then shear toward the
sheared wedge bond).
Position the shear tool within approximately one wire diameter of the bond to be shear tested and shear
the bond.
B.4.2
Examination of sheared bonds (Replaces 4.5)
M
ic
ro
n
All wedge bonds to both the die bond pad and the leadframe post shall have a bond footprint inspection
performed. For wires too small for bond shear testing (less than 1.25 mils (31.75 µm) in height at the
compressed bonded area), the wire shall be removed at the bond location using a small, sharp blade. The
removal of the wire shall be sufficient such that the bond interface can be visually inspected and the
metallurgical bond area determined. For larger wires after bond shear testing, all devices shall be
inspected to examine the failure mode and to determine the bond footprint coverage.
B.4.3
Bond shear codes for ball and wedge bonds (Additional text to 4.6)
All of the shear codes also apply to (ultrasonic) wedge bonds.
B.5
Shear failure criteria for aluminum wedge bonds (Replaces 5)
The following failure criteria are not valid for devices that have undergone environmental stress testing,
have been desoldered from circuit boards, or were preconditioned prior to decapsulation (if procurement
or qualification documents required that the samples be preconditioned prior to the performing of this test
method).
The wedge bonds on a device shall be considered acceptable if the minimum shear values are equal to or
greater than the manufacturer's tensile strength of the bond wire.
In addition, the percent of the bond footprint in which bonding should occur shall be no less than 50%. If
it is necessary to control the wire bonding process using SPC for percent coverage, a Cpk value can be
calculated to this limit.
The number of devices and/or number of bonds to be shears as well as any stressing of the samples prior
to shear testing shall be specified in the applicable procurement documents.
NOTE The failure criteria for thermosonic ball bonds has been moved to JESD47 and no longer resides in this test
method.
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 19
Annex C (informative) Performing shear testing when tool cannot reach below bond centerline
As wire diameters decrease, the height of the corresponding ball bonds also decrease, as short as 4µm. With
these shorter ball bonds, some die structures and some extra thick passivation layers (i.e., polyimide wafer
coats) may prohibit the positioning of the shear tool to contact the ball bond below the centerline of the ball
(see Figure C.1). In these cases the act of shearing the bond not only creates a shear force, but also projects
a torque onto the ball/bond interface. As the shear tool is placed higher on the ball, more of the force at the
interface will be due to a torque than pure shear; and thus it will be very hard to obtain correlation from one
test set-up to the next, let alone correlate to shear testing when the tool can be placed below the centerline
and only a shear force is applied. With the change in tool position above the centerline, the fail mode will
also be affected. For these reasons, it is not recommended to attempt to perform quantitative comparisons
between different shear test runs, but the results may be used for qualitative purposes.
ro
n
C
L
Polyimide coating
h
ic
Passivation
layer
Due to addition of a
polyimide coating, the
shear tool is not able to
contact the ball below
the ball ‘Center Line’.
M
Figure C.1 — Passivation preventing proper height placement of shear tool
Figure C.2 is an image of shear testing performed on a ball with the shear tool set above the ball
centerline. The resulting shear would be classified as a Type 5 shear mode, since more than 1/2 of the
ball bond is still remaining on the bond pad; specifically for low ball bond height (< 4 µm) and in case of
coating material on top of passivation (with a thickness of a few µm).
Figure C.2 — Remnant due to shear tool placement above centerline
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 20
Annex C (informative) Performing shear testing when tool cannot reach below bond centerline
(cont’d)
ro
n
Another issue that could prevent the shear tool from making contact below the ball centerline is excessive
Al splash when bonding with Cu wire. Figures C.3a and Figure C.3b show excessive Al splash that would
prevent setting the shear tool at the proper height.
Figure C.3b — Excessive Al splash
M
ic
Figure C.3a — Cross section showing excessive Al splash
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 21
Annex D (informative) Concerns with decapsulation processes for devices with copper wirebonds
Besides being used as a process monitor, bond shear testing is also used to assess the integrity of bonds after
encapsulation and after stress testing. It can be used on samples that have been subjected to solder attach
simulation (preconditioning), to thermal cycling stress, and to any other stress test. Some qualification
standards require bond testing after qualification stresses.
n
With the transition to small diameter copper wire bonds (1.3mils and less), some of the etchants that have
been successfully used to decapsulate packages with gold wire bonds have been shown to attack copper wire
bonds. In some cases, this attack has been so severe that the bonds can no longer be pulled or sheared.
Figure D.1 shows some images of copper wire bonds severely damaged by the etching process.
ro
Figure D.1 — Images of copper ball bonds showing severe damage from etching process
M
ic
Most chemical etchants used for devices with gold wire bonds were designed to minimize the damage to
the aluminum bond pad; however, copper can be easily attacked by some of these etchants. A seen in
Figure D.2, new etching chemistries have been successfully introduced to minimize this issue.
Original etchant
New etchant
Figure D.2 — Comparison images showing degree of Cu attack due to two different etchants
With the original etchant, the ball contact area was still intact, and shear could still be performed, however,
there was a small loss in shear strength. For this product a new etchant was able to minimize the damage to
the copper wire bond.
Several factors can affect the amount of damage the etching process will inflict upon copper wire bonds.
These include:
-
molding compound chemistry
type of copper wire (bare Cu wire versus palladium coated copper wire)
etchant chemistry used (type of chemical – wet, or type of plasma – dry)
sample history (thermal exposure, duration and type of stress testing performed on samples)
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 22
Annex D (informative) Concerns with decapsulation processes for devices with copper wirebonds
(cont’d)
Copper bond wires that show some sign of having been attacked by the etchant (slight reduction of wire
diameter) may still provide acceptable bond shear results. Therefore, at this time the members of the
working group that generated this annex cannot recommend any visual inspection criteria for determining
“good” and “bad” bonds for the purpose of performing shear testing. It is noted that SEM images do tend
to provide a better indication of what may be “bad” than do optical images.
One method that may be used for determining if a wire is over etched is to first perform wire pull. If the
pull value is significantly less than the typical value for wire pull of unencapsulated product, then it is very
likely that the ball bond has also been damaged due to the etching process. If this does occur then this ‘over
etched’ sample should not be used for shear testing.
For those devices that are more susceptible to etchants, a two-step decapsulation process may be used.
Either laser or mechanical milling may be performed initially to remove most of the encapsulant. A
mechanical milling process could be used to remove all encapsulant material to just above the wires. Laser
ablation will not damage the copper wires so it can be used to expose the wires themselves. However, the
laser will damage the die surface, so to prevent damage of the bond pads, some amount of encapsulant must
remain on the top of the die surface.
ro
n
The laser or mechanical milling is followed by a chemical decapsulation procedure to remove the remaining
encapsulating material from around the copper wire spans, the ball or wedge bonds, and the bonding
surfaces. The purpose of the two step process is to limit the exposure time to the chemical etchant as much
as possible, and thus minimize the level of damage to the copper wire and ball bonds during the
decapsulation procedure.
M
ic
Laser ablation of the molding compound is an effective method of removing the encapsulant without
damaging the copper wire. Figure D.3 shows an image of a copper wire stitch bond fully decapsulated using
laser ablation with no visible damage to the stitch bond. However, laser ablation cannot remove material
that is underneath the wire (see Figure D.4), and it will damage the die surface, thus severely affecting the
ability to perform this test method properly.
Figure D.3 — Stitch bond after decapsulation using laser ablation
Figure D.4 — Die and wirebonds decapsulated using laser ablation
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 23
Annex E (informative) Bond contact area – Valid method for comparing shear force
In the past, using the ball bond diameter has been the method to approximate the area over which the shear
force is applied. For gold ball bonds on aluminum bond pads the ratio of ball bond area to bond contact area
was consistent and thus to make the necessary calculations to perform this test method easier and faster, the
minimum acceptable shear value charts were altered to be based on ball bond diameter, which is a much
easier value to obtain than the (metallurgical) bond contact area between the bond pad and the ball bond.
The bonding surface area of copper wire ball bonds on aluminum or copper alloy bond pads is noticeably
smaller than the horizontal cross-sectional area of the ball bond and the ratio varies from one bonding
system and silicon technology to the next. For gold wire ball bonds, the difference in areas is smaller and
quite consistent from one product to another. For copper ball bonds the ratio of contact area to ball size
tends to decrease as the wire diameter increases, especially for wire diameters of 2 mils and greater.
When this test method was being revised to accommodate copper wire bonds, the working group used the
much more difficult to measure bond contact area to generate force per unit area values in order to correlate
shear values across multiple data sets of ball bonds with varying wire diameters and ball bond diameters.
M
ic
ro
n
Figure E.1 shows one method of determining the contact area is to perform cross-sections on multiple ball
bonds in multiple directions to get a representation of the diameter of the contacting surface between the ball
and bond pad. Ideally the diameter of Cu/Al intermetallic (IMC) layer should be used, but even if the
sample is thermally aged to grow the intermetallic layer, it is still difficult to identify the IMC layer in the
cross-section.
Figure E.1 — Sample cross section of a copper wire bond
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 24
Annex E (informative) Bond contact area – Valid method for comparing shear force (cont’d)
To be more accurate in its calculations, the working group determined the bond area by etching away
representative Cu wirebonds and calculating the area of the aluminum-copper intermetallic layer that
remained. The calculation of percent intermetallic (% IMC) is typically obtained from a SEM image using
imaging algorithms to differentiate the Cu/Al IMC from the Al bond pad. Figure E.2 shows the output from
a software tool for calculating % IMC from an SEM image
ro
n
Figure E.2 — Image analysis of pixel distribution within the fitted circle (represents ball). Light
grey distribution represents IMC, in this case coverage is 73%. 2
M
ic
Using SEM images can be time consuming, so if a correlation study is performed, optical images can be
used in lieu of SEM images. Figure E.3 shows the results from such a correlation study.
Figure E.3 — Images of “optical vs. SEM” correlation study
2
Reference: G.M. O’Halloran, Arjan van IJzerloo, Rene Rongen, Frank Zachariasse, “Planar Analysis of CopperAluminium Intermetallics”, Proc. International Symposium for testing and Failure Analysis (ISTFA), San Jose, CA,
Nov 3-7 2013, pp.297-30
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
JEDEC Standards No. 22-B116B
Page 25
Annex F (informative) Differences between JESD22-B116B and JESD22-B116A
This revision was a major rewrite of the whole test method, therefore this table briefly describes the more
significant changes made to this standard, JESD22-B116B, compared to its predecessor, JESD22-B116A
(August 2009).
Clause
Description of change
1
1
1
Expanded scope to include testing of copper wire bonds.
Updated the range of ball diameters upon which this test method if valid.
Added guidance in a new Annex A for performing shear testing for ball bonds formed for
the ‘stitch on ball’ (or reverse) bonding process.
Removed reference to shearing ultrasonic wedge bonds from the body of the test method
and put all references into a new Annex B.
Modified the definition of the term ball bond to include copper ball bonds.
Replace the term “bond pad” with “bonding surface” in multiple locations
Modified/clarified the definitions for the terms shear tool and wedge bonds.
Added the definition for the term stitch bond.
Added guidance that a higher mag. is necessary for viewing 0.6 mil diameter wire.
Modified text and added a new Figure 2 to better define the parameter “h”, the height of
shear tool from bonding surface. Also added new Annex C to provide guidance when
shear tool cannot contact the ball below the ball centerline.
Provided significant guidance on performing shear on bonds that have been encapsulated,
due to issue of etchants can degrade copper wire bonds. Added new Annex D for
decapsulation concerns and separate subclause for inspection criteria of copper wire
bonds.
Removed old clause 4.3, Sample sizes. Removed tables that were in clause 4.4 which
contained minimum shear values for Au bonds on Al metallurgy. This information along
with all of the pass/fail criteria that was in clause 5 have been moved to JESD47. JEDEC
policy is for test methods to only state how to perform the test and all pass/fail criteria to
be located in qualification documents.
Provided additional guidance (text and drawing) for determining ball bond diameter, for
calculating the shear force per unit area, and for using a statistically representative bond
diameter for the whole sample shear.
In the 2nd paragraph clarified instruction for the cratering fail mode created by the
bonding process versus preexisting cratering prior to the bonding process.
The bond shear fail code definitions have been removed from clause 2.5 and moved to
this clause along with the shear code images that had been in clause 4.9. New drawings
and definitions were added for Type 1 and Type 2 to convey variations between Au and
Cu wirebonds, as well as different bonding surfaces. Color images have also been added.
Several optional data types were added to the list of data that is required to be recorded
for each sheared bond.
Expanded this clause. Several items point to JESD47 for requirements for sample size
and pass/fail criteria.
Added five informative annexes.
1
n
2.1
2.2
2.4
2.5
3.1
3.4
M
ic
ro
4.2
4.3
4.3
4.5
4.6
4.7
5
Annexes
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
M
ic
ro
n
JEDEC Standards No. 22-B116B
Page 26
Test Method B116B
Revision of Test Method B116A
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
Standard Improvement Form
JEDEC JEDS22B116B
The purpose of this form is to provide the Technical Committees of JEDEC with input from the industry
regarding usage of the subject standard. Individuals or companies are invited to submit comments to
JEDEC. All comments will be collected and dispersed to the appropriate committee(s).
If you can provide input, please complete this form and return to:
JEDEC
Attn: Publications Department
3103 North 10th Street, Suite 240S
Arlington, VA 22201
Fax: 703.907.7583
1.
I recommend changes to the following:
Requirement, clause number
Test method number
Clause number
n
The referenced clause number has proven to be:
Unclear
Too Rigid
In Error
ro
Other
Recommendations for correction:
3.
Other suggestions for document improvement:
M
ic
2.
Submitted by
Name:
Phone:
Company:
E-mail:
Address:
City/State/Zip:
Date:
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST
n
ro
ic
M
Downloaded by Pika Xiong (pika1piece@hotmail.com) on Feb 24, 2023, 6:36 pm PST