Content uploaded by Bernhard Wessling
Author content
All content in this area was uploaded by Bernhard Wessling on Jul 22, 2017
Content may be subject to copyright.
PRIVILEGED AND CONFIDENTIAL MATERIALS
ORMECON CSN interactions
with solder masks
2013-06-26 … 28
Dr. B. Wessling
Projects team:
Holger Merkle, Stefan Benn, Florian
Kloth, Wu Jie, Wu XiaoYu, Andy Xu,
Ju Wei, Zhang Da Chen
PRIVILEGED AND CONFIDENTIAL MATERIALS
22
Preface
●ORMECON CSN is an established ImSn process globally
used since 1998 (first customers). Originally, it was
developed, manufactured and sold by Ormecon GmbH, in
Sept 2008, Enthone acquired Ormecon, since then
Enthone manufactures and sells this process without any
discontinuity.
●Automotive electronics suppliers have started to use
ORMECON CSN in early 2000s. With the change to lead-
free assembly, the requirements for surface finish
performance have increased, also for ImSn, ORMECON
CSN showed to perfectly meet these increased
requirements.
●The process parameters have been fine-tuned and are
continuously strictly monitored and updated.
PRIVILEGED AND CONFIDENTIAL MATERIALS
33
Introduction / Background
●In spite of ORMECON CSN's perfect performance, end customers (like
Continental) every once in a while observe assembly failures which we
analyzed in every case in order to find the root cause.
●Failures don't occur due to incompatibility between solder mask types
and CSN chemicals/process parameters: our customers are either using
solder masks which had been checked for compatibility (procedure)(e.g.
pass / fail / effect of ok s/m) or by mass production experience, such
solder mask type was found to be sufficiently compatible. The following
solder mask related issues are critical for improving assembly results:
–optimal Cu surface preparation before solder mask application (to avoid
undercut)
–optimal curing degree (to avoid excess ionic contamination)
–development and curing steps to be run so that no solder mask residues
remain on the Cu pads prior to entering ORMECON CSN process.
●Solder mask residues on Cu pads and in holes are responsible for
mouse bite and for a big portion of dewetting cases. This has been
concluded from 100s of analyses over the past 15 years.
PRIVILEGED AND CONFIDENTIAL MATERIALS
44
Questions to Solder Mask Suppliers
We therefore have the following questions to the solder mask suppliers:
1) there is broad evidence that solder mask residues are more or less
often, to a more or less critical degree left on the Cu pads after
developing / curing: what is the root cause for solder mask residues?
2) under which conditions, by using which process parameters can our
customers assure that NO solder mask residues are left on the Cu
pads? How to analyze the developer chemicals to decide to dump?
3) how can solder mask residues be detected? Do you have any suitable
method?
4) what training can customers get for improving solder mask edges?
5) how can our customers MEASURE the degree of curing and make sure
it is WELL cured, and not under- or over-cured?
In the following, we will present some evidence focussing on solder mask
residues.
PRIVILEGED AND CONFIDENTIAL MATERIALS
55
Spu tte r Pa ram ete rSa m pl e Pa ram ete r
Sa m p l e :
Com m en ts : Bi +; ;
Sn/ Cu
p os it i v e
Orig i n:
Po la ri t y :
Are a:
Ene rgy :
15 0. 0x 15 0. 0 µ m²
PI:
Curre n t :
Cs
3 ke V
44 .0 0 n A
Fi l e : SN11 0 . tf d
An al ys i s Para me te r
Are a :
En e rgy :
2 9. 3x 29 .3 µ m²
PI:
Curre n t:
Bi1
2 5 ke V
1 .0 0 p A
Time / s
100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
0
10
1
10
2
10
3
10
4
10
Intensity
Su bs ta nc e M as s Col or
Na 2 2. 99
Al 2 6. 98
Si 2 7. 98
K 3 8. 96
63Cu 6 2. 93
120S n 1 19 .91
Ba 1 37 .90
CuCs 1 95 .83
Sn Cs 25 2.8 0
Spu tte r Pa ram e te rSam pl e Para me ter
Sa m p l e :
Com m en ts : Bi +; ;
Sn/ Cu
p os it i v e
Orig i n:
Po la ri t y :
Are a:
Ene rg y :
15 0. 0x 15 0. 0 µ m²
PI:
Curre n t:
Cs
3 ke V
44 .0 0 n A
Fi l e : SN11 0 . tf d
An al ys i s Para me te r
Are a :
En e rgy :
29 .3x 2 9.3 µm ²
PI:
Curre n t:
Bi1
25 k eV
1. 00 pA
Depth
200 400 600 800 1000 1200 1400 1600 1800
0
10
1
10
2
10
3
10
4
10
Intensity
Su bs ta nc e M as s Col or
Na 2 2. 99
Al 2 6. 98
Si 2 7. 98
K 3 8. 96
63Cu 6 2. 93
120S n 1 19 .91
Ba 1 37 .90
CuCs 1 95 .83
Sn Cs 25 2.8 0
Positive ions depth profile
For positive depth profile, CuCs and SnCs represent Cu and Sn metal, while 63Cu and 120Sn
represent oxide.
TOF-SIMS
Solder mask residues in Sn layer
PRIVILEGED AND CONFIDENTIAL MATERIALS
66
Results (positive ions)
● For a board having seen 2 reflow, the
concentration profile for Sn and Cu is normal
● The Sn- and the Cu-oxides have a low
concentration
● Characteristic elements which do not belong
to the ORMECON CSN formulation found here
are:
● Si, Ba and Br
● Na, K are typical surface contaminations
resulting from environment + sample handling
TOF-SIMS
Solder mask residues in Sn layer
PRIVILEGED AND CONFIDENTIAL MATERIALS
77
Spu tte r Pa ram ete rSam p le Para me ter
Sam pl e :
Com m ent s: Bi+ ; ;
Sn/ Cu
po s it iv e
Orig in :
Pol a ri ty :
Area :
Ene rg y :
15 0. 0x 150 .0 µm ²
PI:
Curre n t :
Cs
3 ke V
44 .0 0 n A
Fi l e : SN_1 2 0 . tf d
Ana ly s is Para me te r
Area :
Ene rg y :
28 .3 x2 8.3 µm ²
PI:
Curre n t :
Bi1
25 k eV
1.00 pA
Time / s
10 0 2 0 0 30 0 4 0 0 50 0 6 0 0 7 0 0 8 0 0 90 0 1 0 0 0 11 0 0 1 2 0 0 13 0 0
0
10
1
10
2
10
3
10
4
10
5
10
Intensity
Sub s tan ce M a ss Col or
C 1 2. 00
O 1 5.9 9
S 31 .97
63Cu 6 2. 94
79Br 78. 93
PO3 7 8. 97
120 Sn 11 9.9 0
Ba 1 37 .90
120 SnO2 151.90
Spu tte r Pa ram ete rSam p le Para me ter
Sam pl e :
Com m ent s: Bi+ ; ;
Sn/ Cu
po s i ti v e
Orig in :
Pol a ri t y :
Are a:
Ene rg y :
15 0. 0x 15 0.0 µm ²
PI:
Curre n t:
Cs
3 ke V
44 .0 0 n A
Fi l e : SN_12 0 . tf d
Ana ly s is Para m ete r
Are a:
Ene rg y :
28 .3 x2 8.3 µm ²
PI:
Curre n t :
Bi1
25 k eV
1. 00 pA
Depth
20 0 40 0 60 0 8 0 0 100 0 12 0 0 14 0 0 16 0 0 18 0 0
0
10
1
10
2
10
3
10
4
10
Intensity
Sub s tan ce Ma ss Col or
C 12 .00
O 15 .99
S 3 1. 97
63Cu 62 .94
79Br 78 .93
PO3 7 8. 97
120 Sn 11 9. 90
Ba 13 7.9 0
120 SnO2 1 51. 90
Negative ions depth profile
For negative depth profile, Sn represent true metal, SnO2 represent oxide.
TOF-SIMS
Solder mask residues in Sn layer
PRIVILEGED AND CONFIDENTIAL MATERIALS
88
Results (negative ions)
● Again, Sn oxides are very low in
concentration
● Characteristic elements which do not
belong to the CSN formulation are:
●Br, Ba
TOF-SIMS
Solder mask residues in Sn layer
PRIVILEGED AND CONFIDENTIAL MATERIALS
99
Elements of the solder mask
In a former EDX study, Enthone had
analyzed the composition of the solder
mask, cf following table
Spektrum C O Al Si Co Ba Summe
Spektrum 1 29.23 25.73 1.26 1.09 2.10 40.59 100.00
Br
Solder mask residues in Sn layer
PRIVILEGED AND CONFIDENTIAL MATERIALS
10
10
Contamination caused dewetting
Contamination caused dewetting
This was the first example where we on purpose caused dewetting
by solder mask residue.
ok sample (1): Cu almost clean,
perfect wetting.
2nd reflow sample 2B: 100% dewetting
PRIVILEGED AND CONFIDENTIAL MATERIALS
11
11
Identification & Different Forms of
Solder Mask Residues
Identification & Different Forms of
Solder Mask Residues
PRIVILEGED AND CONFIDENTIAL MATERIALS
12
12
EDX measurements
EDX measurements
Element C O Cl
Si
Cu Sum
weight percent w% 11,42 3,46 0,58
0,88
83,65 100,0
PRIVILEGED AND CONFIDENTIAL MATERIALS
13
13
Element C O
Si
Cu Sum
weight percent w% 9,51 1,81
0,34
88,35 100,0
EDX measurements
EDX measurements
PRIVILEGED AND CONFIDENTIAL MATERIALS
14
14
Element C O Cl
Si
Al
S
K Cu Sum
weight percent w% 50,3 23,31 2,35
0,38
0,38
2,11
0,76 20,4 100,0
EDX measurements
EDX measurements
Br
PRIVILEGED AND CONFIDENTIAL MATERIALS
15
15
Spektrum C O Mg Al Si K Ba Summe
Spektrum 1 45.90 26.95 0.42 0.59 1.38 0.12 24.63 100.00
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
Br
PRIVILEGED AND CONFIDENTIAL MATERIALS
16
16
Solder mask
residues!
Sample -1B
showed small s/m
residue dots, no
dewetting in 2nd
reflow – next page.
Spektrum C N O Mg Al Cl K Cu Ba Summe
Spektrum 1 31.19 1.14 6.73 0.04 0.15 1.05 0.82 58.47 0.41 100.00
Spektrum 2 50.87 4.11 5.25 0.02 0.20 0.05 0.02 39.48 100.00
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
Br
PRIVILEGED AND CONFIDENTIAL MATERIALS
17
17
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
Solder mask
residues!
Sample -2B (75° / 55‘)
These pads showed
dewetting in 2nd
reflow – next page.
Spectrum C O Mg Al Cl Cu Ba Summe
Spectrum 1 12.07 3.90 1.39 82.64 100.00
Spectrum 2 19.62 3.92 0.03 0.12 20.11 55.03 1.17 100.00
Br
PRIVILEGED AND CONFIDENTIAL MATERIALS
18
18
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
PRIVILEGED AND CONFIDENTIAL MATERIALS
19
19
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
PRIVILEGED AND CONFIDENTIAL MATERIALS
20
20
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
PRIVILEGED AND CONFIDENTIAL MATERIALS
21
21
Various forms of Solder Mask Residues
Various forms of Solder Mask Residues
PRIVILEGED AND CONFIDENTIAL MATERIALS
22
22
Overview: Possible negative
influences from
Solder Mask Process Defects
1) suboptimal Cu surface before solder mask
application: lift-off & undercutting
2) suboptimal curing: contamination of CSN bath
and increased ionic contamination
3) over-curing: cleaning of Cu pads before CSN
process ineffective
4) solder mask residues: mouse bite (cf following
pages, can not share more because no NDA in
place) & dewetting (cf previous pages)
PRIVILEGED AND CONFIDENTIAL MATERIALS
23
23
Mouse Bite Cross Section
(example for holes)
PRIVILEGED AND CONFIDENTIAL MATERIALS
24
24
Mouse bite at lift-off
PRIVILEGED AND CONFIDENTIAL MATERIALS
25
25
Solder mask edge – before CSN process!
PRIVILEGED AND CONFIDENTIAL MATERIALS
26
26
This not only shows lift off of the solder mask, but some thin solder mask layer
close to the edge (arrow), as can be seen in SEM / EDX (later pages)
Solder mask edge – before CSN process!
PRIVILEGED AND CONFIDENTIAL MATERIALS
27
27
Solder mask lift off in SEM
Edges not sharp,
but very roughly
developed, lift off
can clearly be seen;
this SEM is made
on raw board before
CSN process!
PRIVILEGED AND CONFIDENTIAL MATERIALS
28
28
Solder mask lift off + thin s/m layer
This black „shadow“
is not a shadow,
but is solder mask
residue, continuous
layer starting under
the lift off of the
solder mask
PRIVILEGED AND CONFIDENTIAL MATERIALS
29
29
EDX showing thin solder mask
layer under lift-off (before CSN!)
In this direction, the Si There is a thin solder
content decreases mask residue layer!
PRIVILEGED AND CONFIDENTIAL MATERIALS
30
30
Solder Mask and Mouse Bite
●Mouse bite is mainly generated at solder mask edges.
●If the edges are not 90° and clean / sharp, they are
lifted off („undercut“ without any action having cut
under it, hence lifted off already from the beginning)
●Under the lift off extending onto the Cu surface, there
can be a continuous thin layer of solder mask residue.
This is causes a continuous mouse bite „ditch“.
●Because there is no NDA in place, we can not share
more evidence.
PRIVILEGED AND CONFIDENTIAL MATERIALS
31
31
High Risk Mouse Bite Areas
mbmbmb
Pads with a very thin
SM layer at the
interface to copper
have a high risk of
mouse bite.
Holes filled with SM
residues have a high
risk of mouse bite in
the hole.
mb
mb
mb
Tented Hole
Via hole wall
Pads with
Cu / solder mask
interface
Cu Solder mask
PRIVILEGED AND CONFIDENTIAL MATERIALS
32
32
mbmbmb
Low Risk Mouse Bite Areas
Pads with a strong
and clear SM edge
at the border zone
have a low risk of
mouse bite.
Holes which are free
from SM have a
small risk of mouse
bite on top of the
PCB
SM opening ViaPads with
Cu / solder mask
interface
Cu Solder mask
mb
mb
PRIVILEGED AND CONFIDENTIAL MATERIALS
33
33
No Risk Mouse Bite Areas
Fully plugged holes
do not have a risk of
mouse bite at all.
Plugged hole
PRIVILEGED AND CONFIDENTIAL MATERIALS
34
34
Questions to Solder Mask Suppliers
Finally, I want to repeat the questions with which I started:
1) there is broad evidence that solder mask residues are more or less
often, to a more or less critical degree left on the Cu pads after
developing / curing: what is the root cause for solder mask residues?
2) under which conditions, by using which process parameters can our
customers assure that NO solder mask residues are left on the Cu
pads? How to analyze the developer chemicals to decide to dump?
3) how can solder mask residues be detected? Do you have any suitable
method?
4) what training can customers get for improving solder mask edges?
5) how can our customers MEASURE the degree of curing and make
sure it is WELL cured, and not under- or over-cured?
Any new solder mask should be tested for compatibility.
PRIVILEGED AND CONFIDENTIAL MATERIALS
35
35
Disclaimer
The information contained herein is based on data
considered accurate and is offered at no charge.
No warranty is expressed or implied regarding the
accuracy of this data. Liability is expressly
disclaimed for any loss or injury arising out of this
information or use of any materials designated.