Decapsulation of Multi-Chip BOAC Devices with Exposed Copper Metallization Using Atmospheric Pressure Microwave Induced Plasma

Abstract

With the introduction of new packaging technologies and the great variety of semiconductor devices, new decapsulation tools are needed to improve failure analysis with a higher success rate, and to improve quality control with a higher confidence level. Conventional downstream microwave plasma etchers use CF 4 or other fluorine containing compounds in the plasma gas that causes unwanted overetching damage to Si 3 N 4 passivation and the Si die, thus limiting its use in IC package decapsulation. The approach of atmospheric pressure O 2-only Microwave Induced Plasma (MIP) successfully solves the fluorine overetching problem. Comparison between MIP, conventional plasma, acid etching based on several challenging decapsulation applications has shown the great advantage of MIP in preserving the original status of the die, wire bonds, and failure sites. One of the challenging failure analysis cases is Bond-Over-Active-Circuit (BOAC) devices with exposed thin copper metallization traces on top of Si 3 N 4 passivation. The BOAC critical die structures present a challenge to both conventional plasma and acid decapsulation. The use of MIP to solve the BOAC device decapsulation problem will be discussed in detail through multiple case studies. It appears that the MIP machine is the only approach to decapsulate BOAC devices without causing any damage to the exposed copper on passivation critical structure, which demonstrates the failure analysis capabilities of the MIP system.

Full text

Decapsulation of Multi-Chip BOAC Devices with Exposed Copper Metallization
Using Atmospheric Pressure Microwave Induced Plasma
Jiaqi Tang
JIACO Instruments B.V., Feldmannweg 17, 2628 CT, Delft, The Netherlands
jiaqi@jiaco-instruments.com
Kristopher D. Staller
Texas Instruments, Tucson Device Analysis Operations Laboratory
5411 E. Williams Blvd, Tucson, Arizona 85711, USA
Kees Beenakker
Delft University of Technology, Mekelweg 4, 2628 CD, Delft, The Netherlands
Abstract
With the introduction of new packaging technologies and the
great variety of semiconductor devices, new decapsulation
tools are needed to improve failure analysis with a higher
success rate, and to improve quality control with a higher
confidence level. Conventional downstream microwave
plasma etchers use CF4 or other fluorine containing
compounds in the plasma gas that causes unwanted
overetching damage to Si3N4 passivation and the Si die, thus
limiting its use in IC package decapsulation. The approach of
atmospheric pressure O2-only Microwave Induced Plasma
(MIP) successfully solves the fluorine overetching problem.
Comparison between MIP, conventional plasma, acid etching
based on several challenging decapsulation applications has
shown the great advantage of MIP in preserving the original
status of the die, wire bonds, and failure sites. One of the
challenging failure analysis cases is Bond-Over-Active-Circuit
(BOAC) devices with exposed thin copper metallization traces
on top of Si3N4 passivation. The BOAC critical die structures
present a challenge to both conventional plasma and acid
decapsulation. The use of MIP to solve the BOAC device
decapsulation problem will be discussed in detail through
multiple case studies. It appears that the MIP machine is the
only approach to decapsulate BOAC devices without causing
any damage to the exposed copper on passivation critical
structure, which demonstrates the failure analysis capabilities
of the MIP system.
I. Introduction
Semiconductor devices are routinely decapsulated for failure
analysis and quality control. During the package decapsulation
process, successful failure analysis relies on the preservation
of the original state of the die, bond wires, bond pads, critical
structures like exposed copper traces on Si3N4 passivation,
and original failure sites like Electrical Overstress (EOS) and
surface contamination. The trend to use copper wire bonding
and multi-chip package in IC packaging presents challenges to
both acid [1, 2] and conventional plasma [3] decapsulation
techniques.
Conventional acid decapsulation induces corrosion damage to
the copper wires and reduces wire mechanical strength. Acid
corrosion also removes the original surface features on the
copper wire and the surface contaminants on the bond pads,
making unambiguous failure analysis difficult. Thermal
stressing during quality tests further cures the epoxy chains in
the mold compound, and results in strongly increased acid
decapsulation time. For devices that went through High
Temperature Storage (HTS), Temperature Cycling (TC),
Highly Accelerated Stress Test (HAST), the copper bond
wires are often found severely damaged after acid
decapsulation.
Conventional plasma decapsulation uses carbontetrafloride
(CF4) or other fluorine containing compounds and oxygen
(O2) as precursors. Atomic fluorine is needed to remove the
silica filler in the mold compound. However, the etching
selectivity is low and fluorine always causes unwanted
overetching damage to the Si3N4 passivation layer and Si die.
Any original surface contamination on the passivation layer
will be removed due to fluorine etching. For multi-chip
packages, the inhomogeneity of the plasma and the die
structures often results in one die been severely overetched
while the other dies are not fully exposed yet. The
maintenance of vacuum system and cleaning of the valves
from silica filler particles bring additional operation
difficulties. Because conventional plasma etchers are
commonly designed for wafer cleaning or etching purposes,
they are not suitable for decapsulation.
Recent work published by our group has demonstrated that an
O2-only atmospheric pressure Microwave Induced Plasma
(MIP) system [4] has great advantage compared to the
conventional techniques mentioned before. The highly
confined plasma jet results in a high flux of oxygen neutral
radicals in the plasma afterglow, which contributes to a high
mold compound etching rate and a high etching selectivity.
The MIP system has demonstrated its unique capability in
numerous applications including decapsulation of thermally
stressed copper wired IC packages [5, 6] and 3D stacked die
package with contamination failure [7].
ISTFA 2015: Conference Proceedings from the
41st International Symposium for Testing and Failure Analysis
November 1–5, 2015, Portland, Oregon, USA
Copyright © 2015 ASM International®
All rights reserved.
asminternational.org
485

In this paper we explore the applicability of the MIP afterglow
decapsulation in complex multi-chip Bond-Over-Active-
Circuit (BOAC) devices [8, 9].
The challenge of decapping BOAC sample is from the
exposed copper metallization trace that is located on top of the
Si3N4 passivation layer (see Fig.1). The BOAC devices tend to
conduct high current and dissipate heat. By putting the copper
traces on top of passivation, heat can be dissipated better.
Similar copper-antenna-on-die [10] or copper-antenna-on-
PBO structures are also used in RF communication chips.
Similar exposed copper trace structures are used in Wafer
Level Chip Scale Package (WLCSP) with polyimides or
polybenzoxazole (PBO) as Redistribution Layer (RDL).
Compared to copper wire bonded ICs, exposing such exposed
copper metallization is even more challenging due to its
delicate structure and its location on the Si3N4 passivation
layer surface [11-13].
Acid decapsulation directly comes into contact with the
exposed copper metallization and causes unwanted corrosion
damage and dissolution into copper sulfate.
Conventional downstream microwave plasma decapsulation
attacks the Si3N4 passivation layer due to the use of fluorine,
and it is not always possible to fully clean the die surfaces
without breaching the passivation and floating the upper
metallization patterns.
Figure 1. Schematic representation of the cross section
structure of: (a) Original BOAC packaging structure. (b) Goal
of decapsulation is to expose the copper metallization and
remove all mold compound between the trench gaps, without
causing any damage to the copper metallization traces and
Si3N4 passivation.
II. The fully-automatic atmospheric-pressure
MIP machine
A fully-automatic atmospheric-pressure O2-only Microwave
Induced Plasma (MIP) decapsulation machine was built to
solve the copper wire package and copper metallization
BOAC device decapsulation tasks (see Fig.2). The machine
consists of a microwave generator, a custom-built TM010 mode
Beenakker type microwave resonant cavity, a gas discharge
tube, two mass flow controllers, a CCD camera, a
programmable XYZ-stage, cleaning unit and drying unit, and a
computer to control all the components. The MIP machine is
able to generate a stable plasma at atmospheric pressure.
As the MIP machine operates at atmospheric pressure and the
mean free path of ions at that pressure is extremely low, only
isotropic high selectivity etching occurs by long lived oxygen
radicals. The prevention of ions and microwave leakage fields
on the IC package sample is crucial to avoid damage to the
device inside the package. It has been shown that
semiconductor devices remain fully functional after their
packages have been decapsulated by the MIP machine [4].
The MIP machine went through several versions to overcome
the 4 major limitations of conventional plasma decapsulation
tools:
 Use of CF4 gas that results in low etching selectivity and
overetching damage to passivation and chip
 Low mold compound etching rate that results in long
processing time
 Difficulty in silica filler removal that results in frequent
manual handling
 Use of vacuum that results into frequent pump oil changes
and rapid pump wear out by the silica particles
The latest version of the MIP has upgraded the system into a
fully-automatic machine. O2-only etching and silica filler
removal are done fully automatically through software, thus
greatly enhancing the usability of the MIP machine for fully
automatic O2-only decapsulation. A list of MIP unique
application range in handling the most challenging
decapsulation and failure analysis cases is provided (Table 1).
Figure 2. Schematic representation of the fully-automatic
atmospheric-pressure O2-only-MIP machine.
486

Table 1. MIP application range in handling the most challenging decapsulation and failure analysis cases
MIP-O2
Conventional CF4+O2 Plasma
Acid/chemical
HAST, HTS, TMCL
stressed Cu, PdCu,
Au wire packages
No damage to bond
wire, bond pad, and
die.[5, 6]
Potential to cause overetch damage
to Si3N4 passivation and undercut.
Potential to cause corrosion damage
to bond wires and reduce wire
strength that leads to difficulty in
quality control.
EOS failures on Si,
GaAs, GaN devices
Clean exposure of
original EOS failure
sites without damage to
Si [6], GaAs [14], GaN
[14] die.
Potential to remove original surface
feature of failure site due to
overetching damage to Si3N4, or
GaAs die material.
Difficult to expose EOS failure that
has a melting structure of
metallization/passivation/mold.
Potential to remove original surface
feature of failure site due to
corrosion damage to Cu and Al
metallization or GaAs die material.
Surface
contamination and
corrosion
Clean exposure of
contamination and
corrosion failure
sites.[7]
Potential to remove/alter original
surface contamination and corrosion
sites due to fluorine-induced
overetching damage.
Potential to remove original surface
contamination and corrosion sites
due to acid corrosion.
Open failure on ball
bonds, intermetallic
Clean exposure of lifted
ball bond, bond pad, and
intermetallic.[6]
Potential to remove/alter original
surface failure feature due to
fluorine.
Potential to remove/alter original
surface intermetallic feature due to
acid corrosion.
Stacked-die 3D IC
package
Layer by layer exposure
of stacked-die without
damage.[7]
Severe overetch damage to stacked-
die due to the 3D structure and the
use of fluorine.
Can decap, difficulty in preserving
the mold compound package frame
in certain cases.
Sensors with
transparent epoxy
and fillers
Clean decapsulation
without damage.[14]
Overetch damage to the die due to
fluorine.
Dynasolve cannot cleanly decap in
certain cases, sometimes swelling
of epoxy takes place.
System in Package
(SiP) with high
integration
Clean localized
decapsulation without
damage.[14]
Severe overetch damage to
components in SiP due to the 3D
structure and the use of fluorine.
Potential to remove/damage
components in SiP due to the
heterogeneous 3D integration.
Localized
decapsulation
Focused MIP can
localize decap without
using mask.[14]
Complex masking is required, often
found not practical. Overetch damage
to the die due to fluorine.
Not practical due to difficulty to
control acid etching location and
process.
BOAC (power or
RF devices) with
exposed copper
metallization traces
on top of Si3N4 layer
or PBO RDL layer
Clean decapsulation
without damage. (results
shown in this paper)
Fluorine attacks the Si3N4 passivation
layer. Difficult to clean the die
surfaces without breaching the
passivation and floating the upper
metallization patterns under the
passivation.
Acid directly comes into contact
with the copper metallization trace
and cause corrosion damage.
Other MIP applications under development
487

III. BOAC samples decapsulation
Case 1. Glob top multi-chip BOAC
(The most complex case)
Case 1 is a BOAC multi-chip BGA package (23 mm by 23
mm) containing two dies covered with glob top encapsulate
and with 1.2 mil gold bond wires. The thickness from the top
of the encapsulation material to the top of the bond wires is
about 100 um. The thickness from the top of the bond wires to
the die is about 100 um.
Conventional acid decapsulation turns the glob top into an
expansive rubber-like compound. Nitric acid etches the copper
metallization on the dice, while sulfuric acid etches the
aluminum bond pads. A mixture of sulfuric and nitric acid rips
off the bond wires due to expansion of mold compound. Acid
also directly comes into contact with the copper metallization
(see Fig.3) and aluminum bond pads (see Fig.4) and causes
unwanted corrosion damage.
Figure 3. Acid etching causes corrosion damage copper metal
structures on the Si3N4 passivation. [12]
Figure 4. Acid etching etches aluminum bond pad metal,
leaving barrier metallization.
Conventional downstream microwave plasma decapsulation
attacks the Si3N4 passivation layer due to the use of fluorine
(see Fig.5), and it is not possible to fully clean the die surfaces
without breaching the passivation and floating the upper
metallization patterns on long exposures (see Fig.6). Without
fluorine conventional plasma cannot remove the silica filler in
the mold compound. Although there is automatic filler
removal function in the plasma etcher, it is often found that
filler removal is insufficient and in practice manual cleaning
of the filler by gas blowing is needed every 15 minutes
between O2/CF4 plasma etching intervals.
In order to preserve the surface of PCB substrate that is
integrated into the IC packages, aluminum tape with a
butylene-rubber-based adhesive was used to cover the areas of
the package that needs to be retained. An opening on the tape
marks the intended decapsulation area. Laser ablation is used
to remove the top layer of mold compound before applying
MIP decapsulation (see Fig.7).
Figure 5. Metal traces sprung due to loss of passivation layer
during long conventional microwave plasma deprocessing
Figure 6. Uniform decapsulation of entire die is difficult with
conventional downstream microwave plasma. The wire bonds
on the left side is still not fully exposed, yet other locations on
the die are already damaged due to fluorine overetching.
488

Figure 7. Case 1 sample after laser. Laser ablation depth is
shallow in order to avoid potential laser damage to the die.
After laser ablation the thickness from the top of the
encapsulation material to the top of the bond wires is 75 um.
Figure 8. Case 1 sample after MIP decapsulation. Package
size: 23mm by 23mm. Gold wire diameter: 1.2 mil.
In order to achieve maximum MIP etching selectivity and to
preserve the original state of Si3N4 passivation layer, Si die,
and fine copper metallization traces, we used an O2-only MIP
decapsulation process. The process contains two steps, first
step is selectively remove epoxy in the mold compound by the
high flux of atomic oxygen neutral radicals in the MIP
afterglow, second step is ultrasonic cleaning of the etched IC
package in de-ionized water to selectively remove the SiO2
filler residues. Typical MIP decapsulation process requires 1
hour to fully decapsulate a 5 mm by 5 mm BGA package
without causing process-induced damage. However the
molding compound used in this type of samples is extremely
difficult to etch and the size of the device is large. It took 10
hours to fully decapsulate this BGA sample with MIP.
Figure 9. Undamaged aluminum bond pad and gold bond
wire after MIP decapsulation
Figure 10. Undamaged aluminum metallization traces under
passivation layer after MIP decapsulation
After MIP decapsulation, the two dice, bond wires, and bond
pads are cleanly exposed (see Fig.8). The fine copper
metallization patterns can be seen at the periphery of the two
dice. Since the entire MIP decapsulation process is an O2-only
etching process, copper metallization traces, gold bond wires,
aluminum bond pads (see Fig.9), Si3N4 passivation layer, Si
die, and aluminum metallization traces beneath passivation
layer (see Fig.10) are not damaged by the decapsulation
process itself.
Optical microscopy images of the copper metallization show
that the copper remains in excellent state, the nickel palladium
plating layer on the copper surface has been removed by MIP
(see Fig.11). The copper metallization pattern is hanging
above the Si3N4 passivation layer, the empty trenches between
the copper metallization traces are originally filled with epoxy
mold compound.
489

Figure 11. Undamaged copper metallization of 15 um high, 9
um wide, and with gap width of 10 um. (focus on the top
copper metallization layer)
Figure 12. Silica filler residue particles between the copper
metallization patterns in a trench gap structure (focus on one
layer below the copper metallization layer).
On areas that are not fully cleaned, silica filler residues are
observed in the trench gap structure between the copper
metallization (see Fig.12). To clean the fillers in the trench, it
is not advisable to use any fluorine in the plasma gas. Fluorine
will cause under-cut beneath the thin copper metallization and
that is the reason why conventional downstream microwave
plasma decappers cause breaching of the passivation and
floating the upper metallization patterns. However,
conventional plasma cannot remove the silica filler residues in
the trench with O2-only etching.
Figure 13. Cleanly exposed copper metallization. (focus on
the top copper metallization layer)
Figure 14. Silica filler particles are cleaned from the copper
metallization patterns in the trench structure (focus on one
layer below the copper metallization layer).
It is found that MIP can fully remove the silica filler residues
in the trench structure between the copper metallization (see
Fig.13 and Fig.14). The focused MIP effluent is capable to
direct the flux of atomic oxygen radicals to reach into the
trench structure and selectively remove the mold compound in
one layer beneath the copper metallization layer. The clean
exposure of copper metallization patterns facilitates further
analysis like PEM and OBIRCH during fault localization. The
preservation of original surface condition on copper
metallization, aluminum bond pads, gold bond wires, Si3N4
passivation, and Si die facilitates true root cause failure
analysis after MIP decapsulation.
490

After resolving the most complex BOAC copper metallization
on Si3N4 passivation case, the same MIP decapsulation
process is applied to different BOAC samples with various die
and package structures. The results are compared between
MIP, conventional downstream microwave plasma
decapsulation tool with O2+CF4 as etching gas, and acid
etching by jet etcher regarding the performance, selectivity,
speed, and the amount of manual handing.
Case 2. Copper Leadframe bonded on die
Case 2 is a package with copper leadframe bonded on die (2.5
mm by 2.5 mm) (see Fig.15).
Conventional acid decapsulation causes corrosion damage on
the copper leads that are bonded on the die surface. Note that
one of the copper lead beams is lost on the lower left lead due
to acid corrosion damage (see Fig.16).
Conventional downstream microwave plasma decapsulation
attacks the Si3N4 passivation layer due to the use of fluorine
(see Fig.17). The passivation layer shows variation in color
indicating variation in oxide thickness. Frequent manual filler
cleaning is needed every 15 minutes of O2/CF4 plasma etching
with an overall processing time of 2.5 hours.
Figure 15. Case 2 sample after laser ablation
Figure 16. Acid etching causes corrosion damage to copper
leads and removal of one lead.
MIP decapsulation cleanly exposes the copper leads, Si3N4
passivation and Si die without damage (see Fig.18 and Fig
19). MIP etching and filler removal are done fully
automatically with an overall processing time of 45 minutes.
Figure 17. The CF4 in conventional downstream microwave
plasma causes unwanted overetching damage to the Si3N4 and
SiO2 on the die.
Figure 18. Case 2 sample after MIP decapsulation
Figure 19. The copper leads, Si3N4 passivation and Si die are
exposed without damage after MIP decapsulation
491

Case 3. BOAC with copper metallization
Case 3 is a BOAC with copper metallization sample (9.5 mm
by 4.5 mm) (see Fig.20).
Conventional acid decapsulation causes corrosion damage on
the copper metallization traces (see Fig.21).
Conventional downstream microwave plasma decapsulation
exposes the BOAC die clean and no difference in Si3N4
passivation thickness is observed in this case (see Fig.22).
Frequent manual filler cleaning is needed every 15 minutes of
O2/CF4 plasma etching with an overall processing time of 6.5
hours.
Figure 20. Case 3 sample after laser ablation
Figure 21. Acid etching causes corrosive attached of the thin
copper traces, note the copper sulfate crystals along the sides.
Figure 22. BOAC structure after conventional downstream
microwave plasma decapsulation.
MIP decapsulation cleanly exposes the BOAC copper
metallization, Si3N4 passivation and Si die without damage
(see Fig.23, Fig.24, and Fig 25). MIP etching and filler
removal are done fully automatically with an overall
processing time of 1.5 hours.
Figure 23. Case 3 sample after MIP decapsulation
Figure 24. Cleanly exposed NiPd plated copper metallization
after MIP decapsulation. (focus on the top copper
metallization layer)
Figure 25. Cleanly exposed die surface after MIP
decapsulation. (focus on one layer below the copper
metallization layer)
492

Case 4. BOAC with copper metallization
Case 4 is a BOAC with copper metallization sample (14 mm
by 6 mm) (see Fig.26).
Conventional acid decapsulation causes corrosion damage on
the copper metallization traces (see Fig.27 and Fig.28).
Conventional downstream microwave plasma decapsulation
attacks the Si3N4 passivation layer due to the use of fluorine
(see Fig.29 and Fig.30). The passivation layer shows variation
in color indicating variation in oxide thickness. Frequent
manual filler cleaning is needed every 15 minutes of O2/CF4
plasma etching with an overall processing time of 6.5 hours.
Figure 26. Case 4 sample after laser ablation
Figure 27. Acid decapsulation causes corrosive attack of the
thinner BOAC metal traces and discoloration of the NiPd
plating over the larger copper traces.
Figure 28. Acid etching causes corrosive attached of the thin
copper traces, note the copper sulfate crystals along the sides.
MIP decapsulation cleanly exposes the BOAC copper
metallization, Si3N4 passivation and Si die without damage
(see Fig.31, Fig.32, and Fig 33). MIP etching and filler
removal are done fully automatically with an overall
processing time of 3 hours.
Figure 29. Case 4 sample after conventional downstream
microwave plasma decapsulation. The variation in die color
indicates variation in passivation layer thickness due to CF4
overetching damage.
Figure 30. BOAC exposed, however the CF4 in conventional
downstream microwave plasma causes unwanted overetching
damage to the Si3N4 and SiO2 on the die.
Figure 31. Case 4 sample after MIP decapsulation
493

Figure 32. Cleanly exposed copper metallization after MIP
decapsulation. (focus on the top copper metallization)
Figure 33. Cleanly exposed die surface after MIP
decapsulation. (focus on one layer below the copper
metallization layer)
Based on the 4 case studies of different BOAC sample
decapsulation, an apple-to-apple comparison is made between
MIP, conventional downstream microwave plasma, and acid
(see Table 2).
Table 2. Comparison between different decapsulation tools
MIP-O2
Downstream
microwave
CF4+O2
plasma
Acid
Preserving
BOAC Cu
Yes
Yes
No
Preserving
Si3N4, Si
Yes
No
Yes
Processing
time
Save 70% time
compared to
Conv. plasma
Hours to days
~10
minutes
Automation
Fully
automatic
Manual filler
removal
IV. Conclusions
Based on the clean exposure of delicate copper metallization
structures on the surface of Si3N4 passivation layer in complex
multi-chip BOAC packages, we conclude that MIP afterglow
decapsulation is a successful approach in addressing failure
analysis and quality control of complex BOAC packages.
Compared to acid decapsulation, MIP has great advantage in
preserving the original copper metallization surface structures.
Compared to commercial downstream microwave plasma,
MIP does not cause any damage to the passivation and die.
The fully automatic feature of the MIP machine greatly
reduces the manual handling time. The high efficiency of the
MIP machine saves 70% processing time compared to
commercial downstream microwave plasma decapsulation
machine.
O2-only etching process, atmospheric operating pressure, and
high power density contribute to the superior MIP
decapsulation performance. The fully automatic feature and
the wide application range in handling the most challenging
decapsulation cases make the MIP machine a unique tool that
greatly enhances failure analysis competence and improves
quality control.
Acknowledgements
The authors would like to thank Delft University of
Technology for access to the cleanroom lab facility, and EKL
colleagues A. van den Bogaard, C. C. G. Visser, and R. P. van
Viersen for their help on experiments.
References
1. S. Murali and N. Srikanth, "Acid decapsulation of epoxy
molded IC packages with copper wire bonds," IEEE
Transactions on Electronics Packaging Manufacturing,
vol. 29, pp. 179-183, 2006.
2. C. P. Liu, Y. F. Liu, C. H. Li, H. C. Cheng, Y. C. Kung,
and J. Y. Lin, "A novel decapsulation technique for failure
analysis of epoxy molded IC packages with Cu wire
bonds," Microelectronics Reliability, vol. 52, pp. 725-734,
2012.
3. J. Thomas, J. Baer, P. Westby, K. Mattson, F. Haring, G.
Strommen, J. Jacobson, S. S. Ahmad, and A. Reinholz, "A
unique application of decapsulation combining laser and
plasma," in 59th Electronic Components and Technology
Conference (ECTC), 2009, pp. 2011-2015.
4. J. Tang, D. Gruber, J. B. J. Schelen, H.-J. Funke, and C. I.
M. Beenakker, "Fast Etching of Molding Compound by an
Ar/O2/CF4 Plasma and Process Improvements for
Semiconductor Package Decapsulation," ECS Journal of
Solid State Science and Technology, vol. 1, pp. P175-
P178, 2012.
494

5. J. Tang, C. H. Chen, S. K. Liang, E. G. J. Reinders, C. T.
A. Revenberg, J. B. J. Schelen, and C. I. M. Beenakker,
"Microwave induced plasma decapsulation of stressed and
delaminated high pin-count copper wire bonded IC
packages," in 63rd Electronic Components and
Technology Conference (ECTC), 2013, pp. 1717-1723.
6. L. Zhang, T. C. Chai, A. D. Trigg, L. C. Wai, K. L. Ho, D.
P. Vallett, and J. Tang, "Reliability Test Failure Analysis
using Advance FA techniques on Cu Wire Bonded
Devices," in to be published at 17th Electronics
Packaging Technology Conference (EPTC), 2015.
7. J. Tang, M. R. Curiel, S. L. Furcone, E. G. J. Reinders, C.
T. A. Revenberg, and C. I. M. Beenakker, "Failure
analysis of complex 3D stacked-die IC packages using
Microwave Induced Plasma afterglow decapsulation," in
65th Electronic Components and Technology Conference
(ECTC), 2015, pp. 845-852.
8. G. Heinen, R. J. Stierman, D. Edwards, and L. Nye, "Wire
Bonds Over Active Circuits," in 44th Electronic
Components and Technology Conference (ECTC), 1994,
pp. 922-928.
9. T. Efland, D. Abbott, V. Arellano, M. Buschbom, W.
Chang, C. Hoffart, L. Hutter, Q. Mai, I. Nishimura, S.
Pendharkar, M. Pierce, C. C. Shen, C. M. Thee, H.
Vanhorn, and C. Williams, "LeadFrameOnChip offers
integrated power bus and bond over active circuit," in
IEEE International Symposium on Power Semiconductor
Devices and ICs (ISPSD), 2001, pp. 65-68.
10. J. Schaper, T. Barbieri, and R. Stone, "Application of
electron microscopy in the development of a decapsulation
methodology for microcircuits with copper on-chip passive
devices," Microscopy and Microanalysis, vol. 9 (SUPPL.
2), pp. 1050-1051, 2003.
11. D. Nguyen, B. Davis, and C. Lewis, "Copper Bond over
Active Circuit (BOAC) and Copper over Anything (COA)
Failure Analysis," in Proc. 29th International Symposium
for Testing and Failure Analysis, 2003, pp. 76 - 81.
12. K. D. Staller, "Low temperature plasma decapsulation of
copper-wire-bonded and exposed copper metallization
devices," in Proc. 36th International Symposium for
Testing and Failure Analysis, 2010, pp. 127-132.
13. J. E. Klein and L. Copeland, "Decapsulation of Copper
Bonded Plastic Encapsulated Integrated Circuits Utilizing
Laser Ablation and Mixed Acid Chemistry," in Proc. 36th
International Symposium for Testing and Failure Analysis,
2010, pp. 133-136.
14. J. Tang and C. I. M. Beenakker, "MIP decapsulation
unpublished work,"
495