|
| |
| Anti-CAF “Conductive
Anodic Filament” CCL Material |
Eric LIN 林坤正
Nan Ya plastics corporation, Electronics material Div.
Technical Department of Copper Clad Laminate |
|
| |
| Introduction |
Conductive Anodic Filament
(CAF) has been increasingly concerns on printed circuit boards in
the last few years. The major factors are current trend toward smaller
size, lighter weights, thinner and higher performances that the density
of PCB shall be designed with more closely spacing conductors, much
smaller pitches and single-ply dielectrics such as mobile telephones,
notebook computer and camcorder. In the other hand, some electrical
applications must be exposed in extreme environment. For example,
the safe necessity of automotive like engine control and brake control
system. There cannot occur electrical shorts phenomenon, so the insulation
resistance of material becomes very importance. These factors are
pushing the demand for anti-CAF materials.
CAF is an electrochemical corrosion process
that causes electrical shorts, when copper metal is dissolved at the
anode and migrates to the cathode through the resin matrix/ glass
fibers interface. CAF can occur between any two oppositely adjacent
conductors when the conductors are in contact with the glass fibers.
There are four ways that include hole-to-hole, line-to-line, hole-to-line
and layer- to-layer. (See figure1) |
|
 |
| |
| |
| |
| Mechanism of CAF |
CAF failure has two major step
.The first step is degradation at resin/glass fibers interface that
result in path formation. This may be poor wetting during treating
or lamination, chemical hydrolysis of the coupling agent and impure
raw materials. The second step is electrochemical corrosion reaction
that result in loss of insulation resistance. After degradation and
path formation with a continuous moisture medium between conductors,
an applied voltage produces an electrochemical process that copper
ion dissolved from anode and migrated to cathode through the resin/glass
fibers interface. The insulation resistance will be drop and final
to occur CAF phenomenon.
The reaction process as follows.
| 1. |
Cu
® Cu2+ +2e- |
(Copper dissolved at anode) |
| |
H2O ® H++OH- |
|
| |
2H+2e-
® H2 |
|
| 2. |
Cu2++2OH-®Cu (OH) 2 |
(Copper move from anode to cathode) |
| |
Cu (OH) 2 ®CuO+H2O |
|
| 3. |
CuO+H2O®Cu (OH) 2®Cu2++2OH- |
(Copper deposited at cathode) |
| |
Cu2++2e-®Cu |
|
|
|
| |
| |
| |
| Influence Factor of CAF |
CAF is an electrolytic
corrosion process. The essential elements of electrolytic corrosion
are combination moisture, potential difference and poor material quality.
(See figur2) If we remove any element of triangle and the electrolytic
corrosion stops.
|
|
 |
The formation of CAF should
be dependent on operating condition, test pattern, laminate quality
and manufacturing process. All these factors are discussed as follows:
The operating condition includes temperature, humidity, and voltage.
Electrochemical corrosion will be accelerated by critical operating
condition. The reaction process of CAF has explained that moisture
is an electrochemical corrosion medium because mechanism of CAF requires
moisture to operate. Most laminate materials will absorb moisture
by surface absorption and diffusion in the interior when exposed in
high temperature and high humidity environment. Higher temperature
and humidity will accelerate the speed of moisture absorption that
can result in quicker degradation and path formation. The difference
moisture absorption of resin and glass fibers also can lead to interface
stress. Resin will swell due to absorption of moisture. The swelling
stress can lead to de-bonding at resin/glass fibers interface. Voltage
bias is a driving force for CAF formation. Higher voltage can lead
to quicker failure.
The test patterns have four configurations that include hole-to-hole,
line-to-line, hole-to-line and layer- to-layer. Experiment shows that
test pattern with smaller spaces between adjacent conductors are easier
to occur CAF. The hole-to-hole failure is the most common CAF failure.
(See figure 3) This reason is that through hole has higher area of
glass fibers contact with copper metal. There can detect copper on
the resin/glass fibers interface by EDS analysis.(see figure 4) In
the others , hole-to-line or line-to-line failure are not absolute
relationship with CAF. Because of the probability of glass fibers
contact with copper metal are very small. |
|
 |
Laminate qualities have several
items to prevent CAF formation. The major factors have glass cloth,
resin, impregnation and lamination. Glass cloth must be make proper
treatment. Silane is a bridge between glass fibers and resin, so silane
type and treatment are a significant process. Chemical hydrolysis
of silane can lead to de-bonding due to absorption of moisture. In
addition to finish, a special process will let glass fibers splay
more uniform and puffy to help resin complete wetting each glass fiber.
To insure the bonding of glass fibers/resin is absolute. Resin must
be pure which mean low content of ion impurities. Because of ion impurities
can occur corrosion phenomenon when matrix absorb moisture. There
will accelerate copper sale production. Different resin type also
has different CAF performance. Much paper show when resin has higher
glass transition temperature and low moisture content that will has
excellent anti-CAF performance. For example, Bismaleimide triazine
(BT). Cyanate esters (CE) and FR5 can be higher anti-CAF properties
than FR4 laminates. Path formation of CAF may result from incomplete
bonding between resin and glass fibers interface. To enhance wetting
ability in impregnation process that will increase bonding strength
at the resin/fiber interface. Foreign materials are conductive medium
that can decrease insulation resistance, so cleaning environment is
also important in impregnation and lay up process. The voids must
be removed completely during lamination. Because of voids will promote
the accumulation of moisture in laminate.
The PCB processes include line manufacture, lamination, drilling and
surface coating that can result in CAF failure. In lamination process
include temperate, pressure must be correspond with performance of
prepreg to insure adhesive strength of resin/core and none any lamination
voids in PCB. These defects can result in delamination under environmental
stresses, which provide CAF path. The roughness of hole must be minimized
that means straight and smooth. If drilling qualities are poor that
include delaminations, voids and cracks on the wall of hole. Chemicals
can be trapped inside the voids and cracks during PTH process. Moisture
may also be transferred by capillary action a long cracks of glass
fibers / resin interface during high temperature and humidity environment.
|
|
| |
| |
| |
| Test Specification |
| There are not any public specifications
for CAF test. The test condition, pattern, method and judgment standard
are dependent on requirements of PCB markers. Generally, the test
condition include that humidity set from 85%RH to 95%RH, temperature
set from 60℃ to 90℃, applied voltage set from 10VDC to 100VDC, test
voltage set from 10VDC to 500VDC and total time are from 240hrs to
1000hrs. Test pattern include that structure is from 2 layers to 10
layers, hole to hole distance is from 0.35mm to 1.2mm, hole to line
distance is from 0.4mm to 0.85mm and line to line distance is from
0.100mm to 0.8mm. The test methods usually have two ways. First way,
sample can be measured in chamber by automatic resistance machine
every time that call the on line test. The machine can automatic stop
when occur CAF migration (see figure 5). Second way, sample must be
removed from the chamber to measure resistance when pass at regular
intervals that call the off line test. (see figure6). The judgment
standard is insulation resistance that must be large 106
ohms in the on line test and large 108
ohms in the off line test. |
|
 |
| |
 |
| |
 |
| |
| |
| |
| Promotion and Evaluation
System |
There have been
promoted from raw materials, manufacturing processes and qualities
of laminate to develop anti-CAF FR4.
There are three major directions.
1. To prevent pathway formation of CAF.
2. To reduce disruption of epoxy resin and minimize water absorption.
3. To enhance bonding strength between the interface of epoxy resin
and glass fibers interface.
In order to evaluate CAF performance of new products, there have
design various test pattern and test condition to evaluate anti-CAF
FR4 materials. The details are described in tableⅠ. Figure 9,11
show evaluation results regarding standard FR-4 materials. Figure
10,12 show evaluation results regarding Anti-CAF FR-4 materials.
Standard FR-4 materials failed at 1250 –1750hours in test1 and failed
at 120-168hours in test2. Anti-CAF materials have been passed each
test that keep high insulation resistance.
| Pattern |
|
Test1 |
Test2 |
| Layers |
Double side |
8 layers |
| Board thickness |
1.6mm |
1.6mm |
| Hole diameter |
0.7mm |
0.35mm |
| Hole wall to Hole wall distance |
0.4mm, 0.5mm
0.6mm |
0.35mm, 0.45mm |
| Test condition |
Temperature |
85℃ |
85℃ |
| Relative humidity |
85%RH |
85%RH |
| Applied voltage |
50VDC |
50VDC |
| Measure voltage |
500VDC |
100VDC |
| Measure frequency |
Samples are removed from the
chamber and measurement IR at every 250 hours (off line test) |
Samples are measured IR in chamber
at every 24 hours (on line test) |
| Total time |
2000hours |
240hours |
| Judgment standard |
Bellow 108ohms
NG |
Bellow 106ohms
NG |
| TableⅠ: Various test pattern and condition |
|
 |
| |
 |
| |
 |
| |
 |
| |
| |
| |
| Conclusions |
| 1. |
How to reduce the degree of degradation at the
resin/glass fibers interface, it is a key point to promote CAF
performance. |
| 2. |
CAF failure is also highly dependent on the moisture
content that relationship with resin type and hardener. |
| 3. |
Besides quality of laminate, the PCB manufacturing
process will can contribution to CAF performance. |
| 4. |
Although there are many different test pattern
and test condition for CAF test, the purpose is evaluation CAF
performance. Generally, there will be dependent on requirement
of PCB makers. |
|
|
| |
| |
| |
|
References
|
| 1. |
Balu Rudra and Michael Pecht “Assessing Time-to-Failure
Due to Conduction Filament Formation in Multi-Layer Organic
Laminates” |
| 2. |
Balu Rudra and David Jennings “Failure Mechanism
Models for Conductive Filament Formation” |
| 3. |
J.P.Mitchell and T.L.Weisher “Conductive Anodic
Filament Growth in Printed Circuit Materials” |
| 4. |
Anand A. Shukla, Terrance J. Dishongh, Michael Pecht and David
Jenning ”Hollow Fibers in Woven Laminates” |
| 5. |
Simeno J. Krumbein “Electrolytic Models for Metallic
Electro migration Failure Mechanisms” |
|
|
| |