iNEMI project evaluates BFR-free PCB materials
effective Df**
were then dipped in 500˚F (260˚C)
molten Sn/Pb solder contained in a Waage
hF laminate resin content 1ghz Df 5ghz Df 10ghz Df 20ghz Df
melting pot. Individual samples failed if
rich/Poor
blistering larger than 0.25" occurred. The
Material A Rich, 66% 0.0231 0.0252 0.0269 0.0284 results show that at the eight-hour expo-
(control) Poor, 55% 0.0209 0.0231 0.0245 0.0259
sure, the laminates made from Materials
B, F, G, H and J prepregs were passing.
Material B Rich, 70% 0.0176 0.019 0.019 0.019
Materials C, D, E, I and K performed less
Poor, 53% 0.0147 0.0163 0.017 0.0175
favorably by showing excessive blistering.
Material C Rich, 73% 0.0167 0.0175 0.0183 0.019
Dielectric constant and loss tangent as-
Poor, 53% 0.0161 0.017 0.017 0.0175 sessment using short pulse propagation
The ranges of dielectric constant (Dk or Er)
Material D Rich, 73% 0.0141 0.0156 0.0165 0.0179
and effective loss tangent (Df) are shown in
Poor, 53% 0.0131 0.0147 0.0158 0.0167
Tables 1-4. Both resin rich and resin poor
Material E Rich, 73% 0.0161 0.0174 0.018 0.0192
structures were assessed. The laminate
resin contents varied from 51–73%. Dif-
Poor, 53% 0.0168 0.018 0.0181 0.0181
ferences between this test data and other
Material F Rich, 73% 0.0193 0.0201 0.0202 0.0207
data in this report may be explained by
the specific resin content of the structures
Poor, 53% 0.0154 0.0162 0.0164 0.0174
tested, as well as differences driven by
Material G Rich, 73% 0.0196 0.0207 0.0207 0.0207 measurement technique. The structures
Poor, 53% 0.0158 0.017 0.017 0.0182
tested here are representative of those used
in actual product.
Material H Rich, 73% 0.02 0.021 0.021 0.021
Even though RTF Cu foil was request-
Poor, 53% 0.0153 0.0166 0.0175 0.019
ed to be used on all parts (which typically
results in IPC VLP-like Cu foil roughness
Material I Rich, 70% 0.0203 0.0216 0.0211 0.0217
parameters), there were a wide variety of
Poor, 51% 0.0198 0.0215 0.0215 0.0215
Cu foil profiles observed within the parts.
Material J Rich, 70% 0.0227 0.0247 0.0256 0.0264 • 1oz Cu foil: Peak to valley
Poor, 51% 0.0206 0.0225 0.0231 0.0237
distance: 0.15-0.306 mils.
Control material: 0.122 mils.
Material K Rich, 67% 0.0209 0.0213 0.0203 0.02
• 1/2oz Cu foil: Peak to valley
Poor, 53% 0.0155 0.0165 0.0165 0.0165
distance: 0.125-0.34 mils.
Control material: 0.059 mils.
** Effective Df: includes Cu skin effects
In general, the larger the effective surface
Table 2. Effective loss tangent prior to bake/reflow.
area of the signal trace, the greater the
impact on signal performance. The effec-
tive loss tangent reported in this analysis
to impart V0 rating. The above Tg tion. Obviously, the lower the moisture
is a lumping of the laminate losses and the
out of plane expansion ranged between level present in the material, the better.
effects due to Cu foil roughness as received
250-350 ppm/˚C. The glass transition A major concern regarding bromine-free
by the laminate vendor or imparted by
temperatures were determined to be at 140- materials is that alternative flame retar-
the adhesion promotion treatment at the
200˚C as measured by the DSC middle dants used (organo-phosphorus, nitrogen
fabricator.
point method, with the majority being in containing polymers, and inorganics) can
The Cu resistivity was determined
the range of 150˚C. be hygroscopic and may increase moisture
through resistance measurement of the
The thermal degradation temperature levels in the resins. Moisture up-take at the
long and short traces, calculating the
(5% weight loss) for all laminates was 24-hour RT exposure and one-hour PCT
effective resistance per unit length. That
above 330˚C. T260 values with Cu on the showed saturation levels between 0.1%
information was used with the knowledge
outside ranged from 18 to >120 minutes, and 0.35%. The moisture absorption levels
of the trace cross section obtained during
with Materials B and J being in the low were relatively low. The one-hour PCT
multiple physical sections along each trace
end of this range. T-300 values with Cu exposure pointed out weakness in some of
to calculate the effective Cu resistivity.
on the outside ranged from 0 to >120 the samples, which showed levels as high as
minutes, with the majority in the 0-26 1.4%. Others, such as Materials B, E and J, • 1 oz: 1.81-1.99 µΩ-cm
minute range. showed moisture absorption levels compa- • 1/2 oz: 1.83-2.07 µΩ-cm
Along with thermal stability, mois- rable to the phenolic brominated material.
With this information, the overall losses
ture also plays a key role in determining The extent of hydrothermal stability of the
can be accurately modeled and compared
whether a material is capable of withstand- halogen-free laminates was evaluated by ex-
in a consistent fashion, so the performance
ing lead-free processing or assembly condi- posing the one-sided copper clad laminate
of the materials can be compared without
tions. At elevated temperatures, the vapor samples in a pressure cooker vessel (ten 4”
manufacturing tolerances or differing de-
pressure of trapped moisture absorbed x 4" x 0.035") for 30 minutes, 60 minutes,
sign points affecting the data. This analysis
in the resin can lead to rapid delamina- four hours and eight hours. The samples
was not completed at the time this paper
12 – Global SMT & Packaging – March 2009
www.globalsmt.net
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62