Page 239 - Six Sigma Advanced Tools for Black Belts and Master Black Belts

OTE/SPH
OTE/SPH
3:4
Char Count= 0
JWBK119-15
August 31, 2006
224 Development of A Moisture Soak Model For Surface Mounted Devices
diffusion. This moisture inside the plastic package will turn into steam, which expands
rapidly when exposed to the high temperature of the vapor phase reflow infrared sol-
dering. Under certain conditions, the pressure from this expanding moisture can cause
internal delamination, internal cracks and other mechanical damage to the internal
structure of the package. 1−8 Such damage can result in circuit failure, which could
immediately affect the yield or could be aggravated over time, thus affecting device
reliability. This has been a major quality and reliability concern for semiconductor
manufacturers and users.
A proven technique in detecting various package problems is a preconditioning test
9
as specified in JESD22-A113A (1995). This crucial test is usually conducted prior to
other reliability tests in order to simulate the types of processes an integrated circuit
(IC) package is likely to go through before being assembled. For SMDs, the proce-
dure includes electrical and functional tests, external visual inspection, temperature
cycling, baking, moisture soak and reflow. There are various levels of moisture soak
and reflow, which are determined by the classification of the package during its qual-
ification stage. Moisture absorption and desorption tests are seldom carried out alone
except for identifying the classification level of a plastic molded compound. This test
is used for identifying the classification level of plastic SMDs so that they can be prop-
erly packed and stored. This will better protect the package and avoid any subsequent
mechanical damage during reflow or any repair operation.
There are two basic issues in moisture absorption: the rate of absorption and the
saturation level. Plastic mold compounds transport moisture primarily by diffusion.
Galloway and Miles 10 tackled the absorption and desorption of moisture in plastic
molded compound through the measure of moisture diffusivity and solubility as a
function of time. It was established that transportation strictly by diffusion could be
modeled using the standard transient diffusion equation,
2
2
2
∂ C ∂ C ∂ C 1 ∂C
+ + = ,
∂x 2 ∂y 2 ∂z 2 α ∂t
where C is the concentration level, α is the diffusivity, (x, y, z) are spatial coordinates,
and t is time. Using the standard separation technique results in an expression for
the local concentration, which in turn gives an analytical expression for the total
weight gain as a function of time. Saturation concentration, which is a function of
temperature, humidity and material, determines the maximum possible weight gain
per sample volume for a particular ambient condition. This forms the basis for the
functions under consideration in this case study.
Two similar international standards, EIA/JESD22-A112-A 11 and IPC-SM-786A, 12
have been developed to classify SMD packages. Both define six levels of classification.
Each level corresponds to the maximum floor life that a package can be exposed to in
the board assembly environment prior to solder reflow without sustaining package
damage during the reflow operation. Table 15.1 shows the levels defined in JESD22-
9
A113-A for the moisture soak of the preconditioning test.
5
Huang et al. found that delamination and ‘popcorn’ (a term commonly used to
describe the failure phenomenon due to the rapid expansion of water vapour trapped
in plastic packages) of plastic packages were highly dependent on reflow parameters.
The dominant reflow factor affecting delamination is the total heat energy applied
to the package, which is represented by a combination of the total time above 100 C
◦