Flip chip is a method of attaching Integrated Circuits (IC) to substrate boards
which involves a series of controlled collapse chip connection (C4) high lead
solder bumps on the IC which form metallurgical interconnections with the eutectic
metal bond sites on the board substrate. The die with the integrated circuit
board is flipped downward in order to make sure the contact between the C4 bumps
on the chip and the eutectic solder bumps on the substrates are well aligned.
Plasma treatment becomes a useful method for surface modification and deposition
of various materials in recent years which has also created a greater impact
in IC packaging applications. Plasma-enhanced contamination removal is used
to prepare surfaces prior to eutectic die attach and wire bonding (Getty,
2002). The plasma treatment can effectively remove a very thin layer of
contaminants from the substrate surface. For this reason, plasma treatment is
widely exploited to improve the interfacial adhesion and wire bondability (Djennas
et al., 1993; Lee et al., 1999; Hsieh
et al., 1999). Plasma-enhanced surface preparation prior to the underfill
process has proven to increase underfill wicking speed, fillet height, uniformity
as well as to improve underfill adhesion for flip chip devices. Plasma treatment
alters surface of the substrate for better adhesion in mold and encapsulation
processes (Zhao and Getty, 2005). Moreover, plasma-enhanced
contamination removal and surface activation processes improve the reliability
and yield as well as enhance the manufacturability of advanced technology products.
Plasma is partly ionized gas, consisting of ions, electrons, free radicals,
neutral species and photons. It is also an electrically neutral mixture of physically
and chemically active gas phase species (Zhao et al.,
2004). Physical interaction is performed by ionized atoms and molecular
species throughout sputtering, physical bombardment of the ionized atom and
molecular species change the topography of the surface by increasing the surface
roughness at the molecular level, thereby improving the interface adhesion (Egitto
and Matienzo, 1994; Getty, 2002; Zhao
et al., 2004). Chemical reaction is done by radical species. Free
radicals decrease the activated energy in a chemical reaction resulting material
removal. Plasma treatment process is conducted by activating the vacuum pump
which is attached to the chamber, is capable in removing the volatile materials
or chemical reaction grown products from the substrate surface.
of various plasma processes towards surface enhancements
Plasma mode can be categorized into two types which are direct and downstream
plasma. Direct plasma mode is a process where ion bombardment of the plasma
occurs directly onto the surface of substrate under glow discharge zone which
has a faster and more effective plasma process. Meanwhile, direct plasma mode
is not applicable for certain devices which are sensitive to ion bombardment
and ultraviolet (UV) light. For the devices that have the limitation mentioned
above can be solved by using downstream mode. In the downstream mode, Ion Free
Plasma (IFP) is used. Most of the ion bombardment and UV is filtered before
it reaches to the substrate surface. It is a mild plasma process in which the
substrate is normally placed outside the plasma glow discharge zone in the downstream
of the gas flow. The ITP plasma can be employed to improve the pull strengths
for certain epoxy strength test.
Performance of plasma in the surface modification process in semiconductor
packaging can be classified into 4 categories, which are contamination removal,
surface activation, etch and lastly cross linking (Getty, 2002).
Surface modification is sensitive to time and environmental exposure where the
surface may lose its plasma-induced physical and chemical properties (Getty,
2002). Surface modification enhances the fluid and adhesion. Adhesion between
underfill material and die passivation depends on the surface properties. Surface
contaminants reduce the wetting action and interfere with the flow of underfill
material under the chip. When the substrate and die surfaces are clean and activated,
they hold a higher surface energy, increased wicking speed, improved fillet
height, uniformity and good adhesion (Takyi et al.,
In this study, two types of commercially used flip chip PBGA packages are investigated and experimental results are demonstrated. Table 1 shows effect of various plasma processes towards surface enhancements. The main objective of this research is to study the effect of the plasma cleaning process in enhancing the surface wettability as well as package robustness.
MATERIALS AND METHODS
In this study, two different commercially available substrates named as package A and B are used to investigate the effect of plasma treatment and prebake process in enhancing the wettability of flip chip PBGA substrate. Substrates A and B are using two different types of solder masks namely SR7200G and AUS 703. Both substrates have the same size of 33x33 mm with 1023 I/O pins and 0.18 mm bump pitch. Meanwhile, the substrates have different die size of 10.9x11.6 mm and 12.3x8.3 mm, respectively. Figure 1 shows the process flow of fresh substrates A and B that promotes better wettability to the substrates as found from the contact angle measurement shown in Table 3.
At first, VCA Optima is used to conduct the contact angle measurement. Figure
2a and b show the contact angle measurement by using VCA
Optima before and after the liquid dispensing. A syringe of 100 μL, which
is attached to the VCA Optima, is initially filled up with deionized (DI) water.
One drop of 0.3 μL DI water is set for each dispensing. Contact angle of
a group of six substrates per boat are obtained at the beginning. The droplets
are then dispensed at the four edges of the die cage to ensure that the dispensing
area is always near to the flux dispensing area located at the center of the
The process then proceeds to prebake process in the Blue M oven at temperature
of 125±5°C, which is in nitrogen ambient of 120-180 Standard Cubic
Feet per Hour (SCFH). The duration of prebake time is between 10 to 96 h. The
main purpose of the prebake is to remove unwanted moisture that appears on the
flip chip PBGA substrate surface. In this experiment, bake time of 10 h has
been found sufficient since the prebake process has not shown a significant
result in enhancing the wettability in this case. The third contact angle measurement
takes place when the substrates are sent for the plasma treatment. The batch
plasma treatment system is equipped with oxygen and argon plasma where, gases
were supplied at the rate of 250 and 1500 mL min-1, respectively
with a power supply of 800 Watts and pressure of 9.992 mbar.
flow of substrate cleaning and evaluation
angle measurement by VCA Optima (a) before and (b) after the liquid dispensing
The plasma treatment is conducted to six substrates per boat per loading.
The clean flux is water soluble, halide free and non-corrosive to under bump
metallization with viscosity of 60 cst which is in deep amber to yellow color.
The clean flux consists of 48-58% wt. of proprietary alcohols, 24-28 wt.% of
isopropyl alcohol and 18-24 wt.% of proprietary activators. The substrates are
then forwarded to reflow process after the clean flux is dispensed on the center
die cage area. The peak temperature reflow profile should be in the range of
235-255°C to ensure the complete interaction between the clean flux and
activator. Moreover, both preheat time and temperature are essentially important
to ensure that flux is active with the required rising slope rate which fulfills
the specification shown in Table 2. Figure 3
shows the reflow oven profile that matches to the s pecification in Table
2. Various color lines in Fig. 3 have shown the repeatability
of the reflow profile. Next, the contact angle measurement is done after the
clean flux is sent back from cleaning. The substrates are then staged for two
days under nitrogen ambient to observe the changes of the contact angle. The
contact angle did not show any significant change at this point of process flow.
Thereafter, substrates are divided into two groups where one group is sent for
plasma treatment followed by prebake process, whereas the other group of substrates
is at first subjected for prebake process followed by the plasma treatment.
temperature profile of reflow process
specification for the interaction of clean flux
angle measurement results at different conditions
Further study has also been conducted to observe the changes between both processes
which contribute to the better wettability on the substrate surface. The whole
process is repeated for the substrate package B.
RESULTS AND DISCUSSION
Contact angle measurement results conducted at different stages are shown in
Table 3. It has been found that the prebake process of 10
h is sufficient since, the prebake has not shown significant impact on reducing
the contact angle that indirectly improved the wetting of the substrate. Prebake
process only removes the surface moisture, whereas plasma cleaning process improves
the surface wetting. Upon plasma treatment, substrate surface becomes cleaner
and wettability improves as well. Therefore, the substrates that were subjected
for the plasma treatment at later time have shown better wettability as the
contact angle obtained is smaller as can be found in Table 3.
Both substrates A and B do not show significant changes in contact angle measurement although these have used different solder masks. There is no visual difference between plasma treated and untreated substrates in microscopy as well. However, the difference of contact angle measurement can be observed as shown in Table 3 when DI water is dispensed on the four corners of the die cage.
There has been approximately 70% decrease in contact angle which means significant improvement as a result of plasma treatment. The active plasma will dissociate and react with the substrate surface by creating different chemically functional groups on the surface with the aid of different gases such as oxygen, hydrogen, nitrogen and ammonia. Better adhesion on a larger area can be created by implying different functional groups which modify the chemical activity of the surface and create stronger chemical bonds with the bulk material.
The physical process involves the positive ions which employs ablation process
dislodges the contamination by bombarding the ions onto the substrate surface.
The contamination on the surface has been removed by both physical and chemical
energy at micron level. The substrate surface becomes rough by the ion bombardment
at atomic scale which creates better wettability of flux reflow at later time.
The chemical process widely employs the reduction and oxidation chemistry through
the gas phase radicals. Plasma chemistry has the capability to etch the surface
selectively in the presence of other materials. The dissociation of carbon tetrafluoride
(CF4) and oxygen in the appropriate concentrations produces highly reactive
oxy, oxyfluoro and fluoro radicals that rapidly break carbon-carbon bonds within
numerous materials (Getty, 2002).
Plasma induces cross linking in removing the atomic species from the surface
and generates reactive surface radials. The cross linking improves the adhesion
of metal layers to the plasma treated polymer laminate (Getty,
2002). The plasma treatment process has been optimized by varying the time,
power, pressure and gas flow rate to substrate (Noh et
al., 2007). Different composition of plasma source gas has its own process
applications. Argon plasma treatment induces the cross-linking between substrate
and polymer and it has the ability to remove the ablation contamination (Liu
et al., 2005). Whilst for the oxygen treatment is able to remove
the chemical contamination on the substrate and induces surface activation by
decreasing the surface energy of the polymers which can etch away the organic
contamination on the oxidation process and indirectly improves the adhesion
properties (Alberici et al., 2004).
The plasma treated substrates demonstrate the better wettability. Gibbs equation
defines the wettability as any spontaneous change occurs at an interface such
as oxidation of metals must lower the interfacial energy. The driving force
for wetting can be expressed as Fwet = γcosθ. A reduction
in the organic and oxide contamination levels leads to an increase in the wetting
force, Fwet. Plasma cleaning or flux application removes some of
the surface contaminants leading to a lower contact angle and thus an increase
in the surface energy of the substrate happens (Takyi et
al., 1998). Introduction of plasma treatment has the capability of enhancing
the surface wetting of substrate from hydrophobic to hydrophilic. This is attributed
to the oxide layer removal and plasma etching on the solder mask for both substrates.
When the substrate surface becomes cleaner, obviously better wettability will
be achieved. Although, these substrates have different die size and solder mask
material, no significant difference was found in contact angle measurement.
Plasma surface activation prior to die-attach provided better contact, improved
heat transfer with minimal voiding. The purpose of the mold or encapsulant material
for semiconductor applications is to provide adequate mechanical strength, adhesion
to various package components, good corrosion and chemical resistance, matched
CTE to the materials it interfaces, high thermal conductivity and high moisture
resistance in the temperature range used (Getty, 2002).
The prebake process does not have any constructive influence to the wettability, however plasma treatment shows significant impact in removing the contamination on the substrate surface to contribute in a better wettability. Both packages give positive results although vary in specification. As found in this study, the prebake process is recommended to be conducted in the beginning since prebake at the later time does not have any positive influence to wettability. Better wettability can be achieved when the plasma treatment is conducted at the end of each process.
The authors would like to extend gratitude to Freescale Semiconductor Malaysia for providing various supports throughout the study.