Experiment
#2: Ongoing
Characterize the
Effectiveness of Arathane 5750 (formerly known as Uralane 5750) Conformal Coat Material in Prohibiting Tin
Whisker Formation and/or Tin Whisker Penetration
|
|
Results To Date
Purpose:
To determine
if Arathane 5750 (formerly known as Uralane 5750) conformal coat material over a bright tin plated
surface provides a sufficient barrier to prevent tin whiskers from penetrating
the protective coating. If this material proves to be successful
at stopping whisker penetration, then its use may be suitable as a workaround
to protect systems that already have pure tin plated components that cannot
be replaced with non-pure tin plated alternates.
Abstract:
Most published literature on tin whiskers agree that brass with
bright tin finish is a combination that promotes whisker formation. For
this experiment several brass substrates (some with copper flash) have been
plated using a "bright" tin finish. To further promote whisker
formation some of the plated samples were scratched using a knife blade.
The samples
were segregated into two test groups. Portions of each sample were
coated with Uralane 5750 conformal coat material. One test
group has been placed in an oven at 50°C because the literature suggests
that whisker growth is greatest at this temperature. The other test
group is being stored at room temperature. All samples will be visually
examined periodically to determine
 | If conformal coat
inhibits whisker formation |
| If whiskers are
capable of growing outward through the conformal coat |
| As a side benefit |
| The incubation
period for whisker formation |
| The rate of whisker
growth |
This experiment
is ongoing. If possible, some substrates with whiskers grown during
this experiment may be used in Experiment #3
Sample Preparation:
The experiment
team utilized the plating services of Alexandria Metal Finishers (Lorton,
VA)**. At the request of the experimenters AMF plated
several brass substrates using "bright" tin processes ("bright" tin is
considered to be susceptible to tin whisker formation). Two separate
plating processes were used:
| 1/2 of the samples
were "bright" tin plated directly over the brass substrate |
| 1/2 of the samples
received a Copper flash over the brass substrate prior to "bright" tin
plating |
The application
of a Copper flash is part of Alexandria's normal plating process.
Since this experiment desires the formation of whiskers, the non-Copper
flash samples were ordered because most of the literature regarding tin
whisker formation suggests that pure tin over brass is a combination that
is highly susceptible to whisker growth.
The samples
were intentionally scratched with the goal of encouraging whisker formation
at the stressed locations.
The samples
were then coated on one half of the substrate with Uralane 5750 to a
nominal thickness of 1 to 2 mils. Several samples (both Cu and
non-Cu flashed) were placed in an oven maintained at 50°C (literature
suggests that 50°C is the optimal temperature for whisker formation).
The remaining samples were maintained at room ambient conditions.
Inspection
Procedure:
Periodic optical
microscope inspections using 40X to 100X will be performed to look for
any whisker growth with special emphasis on inspecting the conformally
coated areas. As necessary, Scanning Electron Microscopy (SEM) will
be used to confirm and document findings. Some whisker sites will
be SEM examined again on a regular basis to determine growth rate.
Results of this experiment will be posted to this web site.
Observations/Findings
The links below
provide an assortment of Scanning Electron Microscope (SEM) photographs
taken to date:
As of November, 2008:
As of 3/1/02:
| The latest inspection of conformal coating showed that more nodules have
formed under the conformal coating (by the density increase of whisker
domes). None of the nodules have penetrated through nominal 2 mils
thickness of the coating after over 3 years of experiment. However, early
observations did note that a whisker did penetrate ~0.25 mils of conformal
coat. |
| Inspection of four specimens that were subjected to conformal coat
removing process showed that the nodules from stripped side are still in
nodule form after 6 months of additional storage (some at room
temperature, some at 50°C) since the removal of conformal coating. |
As of 6/18/01:
| Four specimens that were stored in oven at 50°C for 2.5 years have been
subjected to conformal coat removing process to compare nodule formation
in shape and size with the nodules from non-coated side. The specimens
were soaked in Uralane stripper for hours until all the coating was
removed. Subsequent SEM inspection revealed most of the nodules that were
growing under the conformal coat were irregular shapes. Some of these
nodules are 2 or 3 times bigger in size than nodules found on non-coated
side. |
| Two of the four specimens were returned to the 50°C oven for storage.
The other two specimens were not returned to the oven in favor of keeping
them stored in room ambient conditions because GSFC observations to date
suggest that whisker formation occurs faster at room temperature compared
to 50°C storage. |
As of 3/1/01:
| A whisker is breaking
through the conformal coating (see photo).
At this point, only one whisker has penetrated through the coating. The
thickness where the whisker penetrate through is confirmed to be very thin
(1/10 mil). |
| The pure tin plated
samples without Cu flash stored at room temperature (25 °C)
show a faster whisker growth rate (10x) compared to samples stored at 50
°C. The pure tin with Cu flash sample comparison
has not yet been made. |
As of
09/15/00:
| Still none of these
domes made complete penetration through the conformal coat, but the number
of domes in the conformal coat has increased since the last observation. |
| The whiskers from
the non-conformally coated side are now long enough that they are visible
without the microscope. |
| The interim report
titled "Effect of Conformal Coat on
Tin Whisker Growth [pdf]" is published in the Proceedings of the 37th
IMAPS Nordic Conference, September 2000. |
As of
03/20/00:
| The latest observation
indicates that the domes observed in the conformal coat from the last inspection
become larger and narrower at the top, which suggests that the whisker
tips are beginning to penetrate the conformal coat. To date no
whiskers have been observed piercing through from underneath the conformal
coat. |
| Two random samples
(sn#1T and sn#7C) have been coated with gold (vapor deposition, 200 angstrom)
to reduce electronic charging effect during SEM inspection on coated side.
This approach made the inspection of conformal coated side easier. |
| The longest whisker
observed on the non-conformally coated area is 800 um (31 mil). |
As of
09/15/99:
| Picture sn2T_04_0830.jpg
(new) shows possibility of whisker penetrating through conformal coating.
It is still difficult to predict whether whiskers are breaking through
the conformal coating based on just one picture. However, this picture
shows strong evidence that conformal coating may not be the solution for
preventing whisker growth.
|
As of 07/13/99:
| To date it is still
not known if the observed tin whiskers will be able to penetrate the 1
to 2 mil thick Uralane coating. Domes observed in the conformal coating
are indicative of whisker growth beneath the coating. However, we
have not yet confirmed that whiskers are breaking through the conformal
coating. |
| The longest whisker
observed from the non-conformally coated area is 85.5 um (3.3 mil). |
| The estimated average
rate of whisker growth during 6 months is 0.098 mm/year. |
| The whisker shapes
can be categorized into 3 shapes; pyramid, toothpaste, and triangular.
In addition, several irregular shaped whiskers have been observed.
|
As of 05/10/99:
| To date it is still
not known if the observed tin whiskers will be able to penetrate the 1
to 2 mil thick Uralane coating. |
| The longest whisker
observed from non-conformally coated area is 38 um (1.5 mil) which is the
nominal thickness of the conformal coating. |
| The estimated rate
of whisker growth during 4 months is 0.08 mm/year. |
| Regular re-examination
will be continued. |
As of 03/17/99:
| It has been observed
by optical microscope inspection that there are more tin whisker nodules
growing under the conformal coat than there are growing on the non-conformally
coated side. One hypothesis for this observation is that the uncoated
portion of the sample oxidizes due to exposure to the atmosphere.
The tin oxide that forms may slow down the rate of new whisker formation.
On the conformally coated side of the test samples the conformal coat may
limit oxidation of the tin leaving a surface that could be more susceptible
to whisker initiation. |
| Some of the whiskers
photographed previously were re-examined and photographed to identify changes
in growth. |
| Future examinations
will include repeated observations of selected areas to determine the rate
of new whisker formation. |
As of 02/17/99:
| Initiation of whisker
growth was noted within the first 3 weeks after plating. |
| Observation has
confirmed the initiation of whiskers on areas coated with Uralane 5750. |
| It is not yet known
if the whiskers will penetrate the 1 to 2 mil thick coating.
|
** The GSFC Code
562 experiment team commends Alexandria Metal Finishers for their quick
turnaround in preparing the test samples utilized in Experiment 2 and for
their attentiveness to the specific plating requirements imposed by the
team.
|
|
