-------
APPENDKA
6.14 Process Waste Disposal—Immersion Silver
Complete the table below for the immersion silver process. Consider both the volume treated from bath
replacements along with any loses due to bailout. Provide a RCRA waste code only if applicable for each baft
type-
Bath Type
Pre-Cleaner
Mkroetch
Pre-Conditioner
Immersion Silver
Other (specify):
Annual Volume
Treated or
Disposed3
.
-
* Annual Volume Treated or
Disposed - Enter the yearly amount
of the specific bath treated or
disposed. Be sure to consider the
volume treated from both bath
change outs aria bailout before
entering the total.
Method of
Treatment or
Disposal"
RCRA Waste
Code
(if applicable)
B Methods of Treatment or Disposal-
IP] - Precipitation pretreatment on-site
PSTJ - pH neutralization pretreatment on-
site
[S] - Disposed directly to sewer with no
treatment
[D] - Drummed for off-site treatment or
PRN] - Recycled on-site
{RFJ - Recycled off-site
[O] - Other (specify)
Container
Type
Container Type - indicate
the type of container used
for disposal of bath wastes
[OHJ- Open-head drum
[CH]- Closed-head drum
[T]- Chemical tote
[O]- Other (specify)
A-48
-------
fi
1.1
J"§
-^J3fe
O ® ^y-N
"*^ j*8
rO '15'
">**•
fill
j
A
s
i ;
; ~- -=---
- -:--
:- -
.
•"* :- "— "
- -~.
o
f
1
-
„ "
*— r
..
I
•2
1
a
^* 2
w &
a «
S «
S
I
fc
it
§
i s*
I n
w »»•«
i
-
11 11s
o'^'O o-S W
a
a
(A
•5
5
1
CO
5
« 5
s
I
i t-
W
I
4>
I. t
a £ «
lasiaaga
•9
/1 i
s
-'i
• ^
if
-« S
4*M O
.1,1
C5-
•J5
§ 1
» N
•s
s
-i
«
I
*5
I
I I
M
I
S
&
I
a
I
3
?
§
o
e
•3
3
A-49
-------
APPENDIX A
7.2
General Data—Immersion Tin
Number of days immersion tin fine
is in operation:
days/yr
Number of hours per day the
immersion tin line is in operation:
hrs/day
Estimated scrap rale (% of
defective product) for immersion
tin process:
Total of PWB surface square feet
processed by immersion tin line per
year:
sstfyr
7.3 Process Area Employees—Immersion Tin
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the immersion tin line, and for what length of time. Consider only workers who have
regularly scheduled responsibilities that require them to be physically within the process room. Specify "other"
entry. Enter "N/A" in any category that is not applicable.
Type of Surface finish
Area Worker
Number of Employees
in Surface Finish
Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
Line Operators
Lab Technicians
Maintenance Workers
Vhrs
Wastewater Treatment Operators
his
Supervisory Personnel
trs
Other (specify):
7.4 Physical Settings—Immersion Tin
Size of the room containing the sq.ft. Height of room:
surface finish process:
Are the overall process areas/rooms Yes Ifo - Air flow rate:
ventilated (circle one)?
Do you, have local vents (circle one)? Yes TSo i Local vent air flow rate:
Overall surface finishing process line dimensions
Length (ft.): Width (ft): Height (ft.):
7.5 Rack Dimensions-Immersion Tin •
Average number of panels per rack: Average space between panels in s:h;;;il!:^j;i::l;;:;:!:!m..
; rack:
Average size of panel in rack: Length (in.); - Width (in,):
Do you purposely slow the withdraw rate of your panels from process baths
... to reduce drag-out? (Circle one)
A-50
-------
APPENDIX A
7.6 Rinse Bath Water Usage-Immersion Tin
Consult the process schematic in sectional to obtain the process step numbers associated with each of the water rinse
baths present in your immersion tin process. Enter, in the table below, the process step number along with the flow
control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a cascade,
you need only report the daily wafer flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal/day
Process Step
Number1
Blow Control"
Daily Water
Flow Rate'
Cascade Water
Process Steps'1
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank
" Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath. .
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank
Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Flow Control Methods Kev
[C] - Conductivity Meter
P]-pH Meter
[VJ - Operator Control Valve
[R]-FlowRestricter
[N] -None (continuousflow)
[O]-Other (explain)
7.7 Filter Replacement—Immersion Tin
Not Applicable
Bath(s) filtered
(enter process step # from flow diagram in 2.1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E]-Eye Protection [G]-Gloves
[L]-Lab coat/Sleeved garment [A]-Apron [N]-None
[R]-Respiratory Protection [B]-Boots
[Z]-A1I except Respiratory Protection
A-51
-------
APPENDIX A
7.8
Rack or Conveyor Cleaning—Immersion Tin
Not Applicable D
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
'.
min, .
Rack Cleaning Method:
[C]-Chemical bath on SF process line
Pi-Chemical bath on another line
m-Teinporaiy chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical nosing or soaking
[S]-Manual scrubbing with chemical
[MJ-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Personal Protective Equipment::
[E]-Eye Protection [G]-Gloves
UJ-Lab coat/Sleeved gannent [A]-Apion
[R]-Respiiatory Protection [B]-Boots
[O]-Continuous Cleaning [N]-None
[Z]-A11 except Respiratory Protection
7.9 Chemical Bath Sampling—Immersion Tin
Bath Type
Example:
Cleaner
Microetch
Prcdip
Immersion Tin
Other (specify):
" Tvne of Sampling
[A]-Automated
[M]-Manual
[N]-None
bFreauencv: Enter 1
time elapsed or numt
ft processed between
Clearly specify units
sq.ft)
Type of
Sampling1
A
Frequency*
3 per day
»
!
Duration of
Sampling0
Smin
--
-
Protective
Equipment"1
E,G,A
Method of
Sampling"
P
c Duration of Sampling: Enter the c Method of Sampling:
average time required to .manually [D]-Drain or spigot
take a sample from the tank. [P]-Pipette
[L]-Ladle
d Protective Equipment; Consult [O]-Qther (specify)
he average the key for the above table and
>er of panel sq. enter the letters for all protective
t samples. equipment used by the person
(e.g., hours, performing the chemical sampling.
A-52
-------
S
iiyajjij:
••3
-
2
e-a»
'2e
§ §
«2 T3
•S ™
r
2 3
II
l|
§
•-S
1
em
2
"S
o
&'
§s
'So
I
1
I
P
s» o>
S %
£1
mers
Time
HIS
••e;;
ill!
1
"3 2
.S |
es S
§ o
sz: >•
l
51
l
li^ffgl
is;! !!!E
I
•S J,
es
ca
11
3£m
®!Pj
a
"«
o
1
ce
I
•ta
«
I
OS
Id
O ets
a -°
w «
IB.
"^ o
e «
•
A-53
-------
APPENDIX A
SOS
S- *S
silt
i-I CJ « -5!
%
GO
^"
>> ^pl^
•— • w3
II
s §
•a ^
s
S3 -3
11
u .a
"*" "3
if
§ ?s
«
J5
u
s
•o
o
.£•
2
1
JO
1
: .
~<
1 Cleaner
•
•
'"
•
cs
CO
•*
rt
I Microetch
-
CM
co
"*
1-1
.s-
•3
CM
co
rl-
^
a
H
I Immersion
cs
CO
•*
"-1
§
1
•«rt
o
CM
co"
•*
— '
CM
co
"*
op
"o
gc
•M
IM
0
e
0
5
^
i
u
o
1
d
•^
f
o
s
_cg
"o
o
•s
_OT
"3
1
C8
«
o
^§
St!
•s
(U
p
f
o>
13
§
-5
«3
A-54
-------
APPENDIX A
A-55
-------
APPENDIX A
A-56
-------
7.14 Process Waste Disposal—Immersion Tin
Bath Type
Cleaner
Microetch
Predip
Immersion Tin
Other (specify):
Annual Volume
Treated or
Disposed3
-•
-
-
3 Annual Volume Treated or
Disposed - Enter the yearly amount
of the specific bath treated or
disposed. Be sure to consider the
volume treated from both bath
change outs and bailout before
entering the totaL
Method of
Treatment or
Disposal *
RCRA Waste
Code
(if applicable)
"
-
B Methods of Treatment or Disposal-
IP] - Precipitation pretreatment on-site
[N| - pH neutralization pretreatment on-
site
[S] - Disposed directly to sewer with no
treatment
[D] - Drummed for off-site treatment or
[RNj - Recycled on-site
[RF] - Recycled off-site
IP] - Other (specify)
Container
Type
Container Type -
indicate the type of
container used for
disposal of bath wastes
[OH]- Open-head drum
[CH]- Closed-head drum
[TJ- Chemical tote
[OJ- Other (specify)
A-57
-------
APPENDIX A
Facility Background Information
Design
for the
Environment
Printed Wiring Board Project
Performance Demonstration Questionnaire
Please complete this questionnaire, make a copy for
your records, and send the original to;
Ellen Moore
Abt Associates
55 Wheeler St
Cambridge, MA 02138
Fax:(617)349-2660
Note: The completed questionnaire must be returned PRIOR TO the
scheduled site visit.
FACILITY AND CONTACT INFORMATION
Facility Identification:
Company
Name:
Site Name:
Street Address:
City:
State: ; Zip:
Contact Identification: Enter the names of the persons who can be contacted regarding this survey.
Name:
Title:
Phone:
Fax:
E-Mail:
A-58
-------
APPENDIX A
Section 1. Facility Characterization
Estimate manufacturing data for the previous 12 month period or other convenient time period of 12
consecutive months (e.g., FY96). Only consider the portion of the facility dedicated to PWB manufacturing
when entering employee and facility size data,
1.1 General Information
Size of portion of facility used for
manufacturing PWBs.
Size of portion of facility used for
surface finishing.
Sq.Ft.
Sq. Ft
Number of days Surface Finish line is
in operation:
days/yr
1.2 Process Type
Estimate the percentage of PWBs manufactured at your facility using the following methods for surface
finishing (SF). Specify "other" entry.
Type of PWB process
HASL
OSP-Thick
OSP-Thin
Immersion Tin
Immersion Silver
Percent of total
%
%
%
%
%
Type of PWB process
Electroless Palladium
Electroless Nickel/Immersion Gold
Other:
Other:
TOTAL
Percent of Total
%
%
%
%
100%
1.3 General Process Line Data
Process Data
Hours
Number of hours the Surface Finishing line is in operation per day:
1.4 Process Area Employees
Complete the following table by indicating the number of employees of each type that perform work duties in
the same process room as the Surface Finishing line and for what length of time. Report the number of
hours per employee. Consider only workers who have regularly scheduled responsibilities physically within
the process room. Specify "other" entry. Enter "N/A" hi any category not applicable.
Type of Process
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other:
Other:
Number of Employees
in Process Area
.
'
_^
Average Hours per Week per
Employee in Process Area
" , .- Hrs.
Hrs.
Hrs.
Hies.
Hrs.
Hrs.
Hrs.
A-59
-------
APPENDIX A
1.5 Wastewater Discharge and Sludge Data
Wastewater discharge type (check one)
Direct
^i^;
Indirect
Annual weight (quantity in pounds) of sludge generated:
Is sludge dewatered prior to disposal?
% water content prior to dewatering:
% water content after dewatering:
Zero
~
i
f
i
Section 2. Process Description: Immersion Tin
2.1 Process Schematic
Fill in the following table by identifying what type of surface finishing process (e.g., HASL) your facility
uses. Then, using the proper key at the bottom of the page, identify which letter corresponds with the first
step in your process and write that letter in the first box (see example). Continue using the key to fill in
boxes for each step in your process until your entire surface finishing process is represented If your process
is not represented by a key below, complete the chart by writing in the name of each process step in your
particular surface finishing line.
A-60
-------
APPEISDKA
A A
<*• t
A
T T T
-9
f-i
or?
CO
to
A-6I
-------
APPENDIX A
22 Rinse Bath Water Usage
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the
water rinse baths present. Enter, in the table below, the process step number along with the flow control
method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a cascade,
you need only report the daily water flow rate of one bath in the cascade.
Amount of water used by the surface finishing line when operating:
Process Step Number*
Example: 8
Flow Control6
'- R ••' "'-"••-
E - 1 •"-'-:''.-"•-
Daily Water Flow
Rate'
2y400 gaL/day
i gaL/day
•'•-'•.'..• '••••',• ;•• •':-' •":' ; 'gal/day-'
gal/day
-•:.• gaL/day
; •'.' :gal/.day
''".'•• : gaL/day
gaL/day
gaL/day
" Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
* Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate -Enter the average daily flow rate for the specific
water rinse tank.
* Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
^ . f '<. gatfday
Cascade Water Process Steps*
-,--.'•-,. • - • s->6: •:•• •- •-••.
Flow Control Methods Key
[C] - Conductivity Meter
[P] - pH Meter
[V] - Operator Control Valve
[R] - Flow Restricter
[N] - None (continuous flow)
[O] - Other (explain)
23 Process Parameters
Size of the room containing the process
Are the overall process areas (not tank vent) ventilated? (Circle one)
Air flow rate:
Do you have local vents?
Local vent air flow rate:
sq.ft.
v - Height of room
No
* cuJt.min.
-No
cu. FL/min.
Type of process automation for surface finishing line: (circle one)
Automated non-conveyorized Automated conveyorized Manually controlled hoise
Manual (no automation) Other, specify: .
A-62
-------
2.4 Physical, Process, and Operating Conditions
Complete the table below by entering the data requested for each specific type of chemical bath listed. If tw<
tanks of the same type are used within the process, list the data for a single tank only.
BATH
Acid cleaner
Microetch
Acid predip
Immersion tin
Other (specify)
LENGTH (inches)
'in.
- in.
' ; ' :" * >
'S- in.
in.
- in.
-£4-
in.
WIDTH (inches)
* ' ibu.
>~ in..
** ^
• _ in..
Hll-
in..
'ii,
in..
in..
NOMINAL VOLUME
< " gal.
gal-
^ ~gal.
gal.
gal
gal.
gal.
' gaL
A-63
-------
APPENDIX A
i
a
t>
1
Ol
cr>
^t*
CIS
19 ^
al Ouan
weight in pounds
A-64
-------
APPENDIX A
A-65
-------
APPENDIX A
•S3 i
8
ra
B
CM
es
R spec
A-66
-------
APPENDIX A
Observer Data Sheet
Observer Data Sheet
DfE PWB Performance Demonstrations
Facility name and location:
Surface finishing process type and name:
Pate: Contact Name:
Installation date:
Test Panel Rum
Overall Surface Finishing process line dimensions
.Length (ft.): Width (ft.):
Height (ft.):
Average number of panels per rack:
Average space between panels in rack:
Average size of panel in rack Length(in):
Width (in.):
At what % of capacity is the line currently running?
At what % of capacity is the line typically running?
What is the overall throughout?
How is it calculated:
surface sq.ft./year
Estimated yield for surface finishing line:
Number of thermal cycles the finished board can withstand:
Note any unusual storage conditions or oxidation.
Load system with layer 4 facing up or toward the operator.
White running the test panels, verify each process step and complete the table on the next page.
Test Board
1.
2. .
Test Panel Serial Numbers
Serial #
Test Board
3,
4.
Serial #
Test Board
5.
6.
Serial #
A-67
-------
APPENDIX A
. ,•-, • , ' *V * ^ *•("*,*)*?*
, Test Panel Run _ x „,
Bath Name
(from schematic)
-I.'' :• ; ..:•...;
2. . • • •• . ..;••;•.
3. ". ' . ; ;,.,-.
4. -'/ ...•"."'•'
5- - '"'•'••••.:•'
6.
T- ; --. •'.''"••••
8. '• .'.- -..
9. .. •'-"'••:':;
10.
.11.'- ; •-;/-•;,;
12.
13.. '•-•••..- '.::••:
14.
Equipment"
Bath
Temp
Immersion
Time
Drip
Time
-
Overall System Time: .
a List Number, type of
Agitation: Vapor Control: Filter Type: Heater Control: Water Rinses:
[PA] -Panel agitation [BC]- Bath Cover (BF]-Bag [TH] - Thermostat [CN] - Continuous
[CPJ-CirculatfoaPump QFE}- Fully Enclosed [CF] - Cartridge [TM]- Timer [DP] - Continuous During Process
[AS] -Air Sparge [VO]- Vent to Outside P>R]- Programmed [PP]- Partial During Process
tVC] - Vent to Control
A-68
-------
APPENDIXA
Verification of Part A (mark any changes on working copy of Part A):
Ventilation:
Verify the type of ventilation as recorded in the Questionnaire: a
Tank Volumes:
Verify the length, width, and volume of each .tank, as recorded in the Questionnaire: n
Water use:
Verify water use data, for each tank:
Daily water flow rate verified n
Cascade process steps verified D
Pollution Prevention:
Have you used any other pollution prevention techniques on the surface finishing line (e.g., covered tanks
to reduce evaporation, measures to reduce dragout, changes to conserve water, etc.)?
If yes, describe and quantify results (note: if results have not been quantified, please provide an.
estimate):
If your throughput changed during the time new pollution prevention techniques were
implemented, estimate how much (if any) of the pollution reductions are due to the throughput
changes:
A-69
-------
APPENDIX A
-Filter Replacement ' " ^r / /'„ "*" ' > ,. < f>' (
Bath(s) filtered (enter process step #)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key: , , t «• ,1 , r «, ' ,',r ""^ii
pS3-EycProtcctiott [Gl -Gloves [Zf- Alt except Respiratory Protection
UJ-Labcoal/SIccved garment JA}- Apron INJ-None ( ?
fRJ-ReqarsatoryPiotectioii. , JB] -Soots -. % x , > *-• ^ vij
Equipment Maintenance
Estimate the maintenance Tequirements (excluding filter changes aadbatia. changes) of the surface
jBnisningprocess equipment for both outsidb services calls (maintenanceTjy vendor or service company)
and in-house maintenance (byfacflily personnel)- r '"
Describe the typical maintenance activities associated with the surface finishing process line (e.g., motor
repair/replacement, conveyor repairs, valve leaks, etc.)
Average time spent per week:
Average downtime:
If there a recurring maintenance problem?
If yes, describe:
A-70
-------
APPENDIX A
? -{
Rack or Conveyor Cleaning Not Applicable "a
j »
Frequency of rack or conveyor cleaning:
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Number of personnel involved:
Personal protective equipment (see key);
Average time required to clean:
Rack Cleaning Method:
[C] - Chemical bam on SF process line
[D] - Chemical bath oa another line
PI - Temporary chemical bam
[S] - Manual scrubbing with chemical
[M] - Non-chemical cleaning
[N]-None
[O] - Continuous cleaning
Conveyor Cleaning Method:
[C] - Chemical rinsing or soaking
[S] - Manual scrubbing with chemical
fMJ -Non-chemical cleaning
[N]-None
[O] - Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection
[L] - Labcoat/Steeved garment
|R] -Respiratory Protection
[O] - Continuous Cleaning
[Z] - All except Respiratory Protection
[G]-Gloves
[A] - Apron
[B] -Boots
|N]-None
Chemical Bath Sampling . -
Bath Type
Cleaner
Microetch
Flux
Solder
Post Clean
Other (specify)
Other (specify)
Type of
Sampling8
Frequency1*
Duration of
Sampling0
Protective
Equipment*
-
•Type of Sampling , 'Dnraflon of Sampling: 'Method of Obtaining Sampi
£Aj- Automate* - _Entotheaveragetimeforinanuany {DJ-Rrafct or spigot
fMJ- Manual takmga san^teftomtnetanfc rpl-Proette
M-N008 . _ ' £L]-lSdte
k , £OJ- Other (specify)
"Frequency: ' *Pr«tectfreEqn^ineilfc
Entertheavesagetimeelapsed Consuttthekeyforfee-abovetable
or number of panel sq.ft. processed " and enter the letters for all protective
betweeusan^les. Clearly specify "- eqn^njentwombythepaspnperfomring
units (e.g., hours, sq. &., -ete) the chemical sampling.
Method of
Sampling6
es:
A-71
-------
APPENDIX A
1 *-«
3
J
II
. 53 *0
-c o ^ -o
jg 1 a =§'•!'!
1 g -111 i
o o S s o <
A-72
-------
APPENDIX A
Comparative Evaluation: "; "" ' \ * ?, ' -% ",
If th& facility has switched from ^previous system to the current system, complete this page.
Product Quality:
What, if any, changes were noticed in the quality of the boards produced? (Yield change?)
Installation:
How long was the debug period when this system was installed?
What were the types of problems encountered:
Manufacturing Process Changes: How did you change your upstream or downstream processes when
this system was installed (e.g., did you have to make changes in your solder mask)?
Waste Treatment:
Have any of your waste treatment methods or volumes changed due to the installation of this system (not
associated with volume changes due to throughput changes)?
If yes, describe the change(s) and attach quantitative information, if available:
Process Safety:
Have any additional OSHA-related procedures or issues arisen as a result of changing to the present
system (e.g., machinery lock-outs while cleaning, etc)? If so, describe:
Customer Acceptance:
Have customers accepted the new process? Why or why not:
Other:
Describe any other issues that have arisen as a result of the new process.
A-73
-------
APPENDIX A
Supplier Data Sheet
DfE Printed Wiring Board Project
Alternative Technologies for Surface Finishing
Manufacturer/Supplier Product Data Sheet
Manufacturer Name:
Address:
Contact
Phone: .
Fax:
How many alternative making holes conductive product lines will you submit for testing?_
Please complete a Data Sheet for each product line you wish to submit for testing. In addition,
if you have not already done so, please submit the material safety data sheets (MSDS), product
literature, and the standard manufacturer instructions for each product line submitted.
Product Line Name:
Category:"_
" Categories of Product Lines:
A.HASL
B. Immersion Tin
C. Immersion Palladium
D. Electroless Nickel/Immersion- Gold
E. Nickel/PalladiunVImmersion Gold
F. OSP-(Thin)
G. OSP-(Thick)
For the product line listed above, please identify one or two facilities that are currently using the
product line at which you would like your product demonstrated. Also, identify the location of the
site (city, state) and whether the site is 1) a customer production site, 2) a customer test site, or 3)
your own supplier testing site.
Facility 1 Name and Location:
Type of Site: .
Facility Contact:
May we contact the facility at this time (yes or no):
Facility 2 Name and Location: '
Type of Site:
Facility Contact:
May we contact the facility at this time (yes or no):
Phone:
A-74
-------
APPENDIX. A
»
ta
I
1
4>
i
•+•*
£
I
§
I
1
-s
•3
g*
1
a
o
1
£
w
l|
" O.
S
* s
II
i
I
5
er
a
o
D D a a
60
S S D«
o D n
n a D D
1
1 1 1 a y
S 8 S.J f
anno D
•g <<-i
•3
o a a a
-
000 a
n D a a
0 §.
"
11
•a *"
Hi..
a o o o
T3
' |,
I
g
-O
&
•s
-a
I
S
(X
4-<
O
3
rf>
S
I
*A
S
£
A-75
-------
APPENDIX A _
Special Product Characteristics
1. Etoes the process operate as a vertical process, horizontal process, or either?
2. Average number of thermal excursions the finished board can withstand?
3. Most likely process step preceding the beginning of the surface finish application?
4. Should the application of solder mask occur after the application of the surface finish, or before?
5. Which of the following technologies is the surface finish compatible with?
(Circle all applicable choices.) '
A. SMT
B. Hip Chip
C. BGA
D. Gold Wire Bonding
E. Aluminum Wire Bonding
F. Other, Explain:
6. Please state cycle time of surface finish process line.
7. Please describe any special process equipment recommended (e.g., high pressure rinse, air knife, dryer,
aging equipment, etc.).
Product Line Constraints
1. Please list any substrate incompatibilities (e.g., BT, cyanate ester, Teflon, Kevlar, copper invar copper,
polyethylene, other [specify])
2. Please list compatibilities with solder masks.
3. Are there any special requirements needed for the soldering process (e.g., type of flux, etc.)?
4. Average shelf-life of finished boards?
5. Other general comments about the product line (include any known impacts on other process steps).
A-76
-------
APPENDIX A
Bath Life
Please fill in the following table (for bath listings, please refer back to your process description on page 2).
Bath
1.
2.
3.
4.
5.
6.
7.
8.
Recommended
Treatment/Disposal
Method"
Criteria for Dumping
Bath
(e-g., tone, surface sq ft of
panel processed,
concentration, etc)
Attach and reierence materials, it necessary.
Recommended Bath
Life
(in terms of criteria listed at
left)
• '
A-77
-------
APPENDIX A
Costs:
Chemical Cost
Please provide the cost per gallon (or pound) of chemical for each chemical product required to operate this
alternative surface finishing product line. It is recognized that the cost of chemicals is, in part, dependant on
the amount of chemical purchased (i.e., volume discounts) and may vary accordingly. If cost would decrease,
please write decreased cost in margin along with volume of chemical required for pricing discount.
Bath Name " .
1.
2.
3.
4.
5.
6.
7.
_>v )
Product Name
r ""•" - t
V. -6 f.
A.
B.
C.
A.
B.
C.
A.
B.
C.
A-
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.
Chemical Cost
($/gal»r$/Ib)
A-78
-------
APPENDIX A
Equipment Cost
Do you recommend or suggest any specific equipment manufacturers to customers for obtaining process
equipment to operate this surface finish line? If so, why? Please provide the contact information for
equipment manufacturer below.
Equipment Company # 1
Company Name:
Contact Name: _^
Phone number:
Equipment Type:
Equipment Company # 2
Company Name:
Contact Name:
Phone number:
Equipment Type:
Do either of the companies listed above manufacture equipment specifically designed for your product line?
Which one? _^^
If so, what is special or different about the equipment design? •
A-79
-------
-------
Appendix B
Bath Chemistry Data
-------
-------
Table B-l.
Table B-2.
Table B-3.
Table B-4.
Table B-5.
Table B-6.
Contents
Bath Concentrations for the HASL Technology ................... ; ... B-l
Bath Concentrations for the Electroless Nickel/Immersion Gold Technology . B-l
Bath Concentrations for the Electroless Mckel/Electroless Palladium/Immersion
Gold Technology . . . ...................... . ........
Bath Concentrations for the OSP Technology . . . ...... . B-2
Bath Concentrations for the Immersion Silver Technology ............... B-3
Bath Concentrations for the Immersion Tin Technology . . ............... B-3
-------
-------
APPENDIX B
Table B-l. Bath Concentrations for the HASL Terhnnln^y
Bath
Cleaner
Microetch
Chemicals
Alkylphenolpolyethoxyethanol
Ethylene glycol monobutyl ether
Fluoboric acid
Phosphoric acid
Sulfuric acid
*9 other confidential chemicals
1,4-Butenediol
Copper sulfate pentahydrate
Hydrogen peroxide
Sodium hydroxide
Sulfuric acid
*7 other confidential chemicals
Concentration in Bath (g/1)
18.00
22.90
12.33
61.11
110.40
12.72
45.00
50.73
0.170
103.50
Table B-2. Bath Con entrations for the Electroless Nickel/Immersion Gold TWh™i«™
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemicals
Phosphoric acid
Sulfuric acid
Hydrochloric acid
Alkylphenolpolyethoxyethanol
*Two other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
*Two other confidential chemicals
Hydrochloric acid
*Four other confidential chemicals
Two confidential chemicals
Nickel sulfate
*13 other confidential chemicals
Potassium gold cyanide
*Four other confidential chemicals
Concentration in Bath (g/1)
50.8
138
17.85
18.00
0.170
35.88
45.00
87.40
55.80
37.24
2.999
B-l
-------
APPENDIX B
Table B-3. Bath Concentrations for the Electroless Nickel/Electroless Palladium/Immersion
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Chemical
Phosphoric acid
*2 other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
*1 other confidential chemical
*4 confidential chemicals
*1 confidential chemical
Nickel sulfate .
*10 other confidential chemicals
*4 confidential chemicals
Ethylenediamine
Propionic acid
Maleic acid
*6 other confidential chemicals
Potassium gold cyanide
*4 other confidential chemicals
Concentration in Bath (g/1)
50.80
0.17
35.88
45.00
156.40
58.65
4.45
7.30
2.00
3.00
Table B-4. Bath Concentrations for the OSP Technology
Bath
Cleaner
Microetch
OSP
Chemical
Phosphoric acid
Sulfuric acid
*3 other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
*6 other confidential chemicals
Copper ion
*5 other confidential chemicals
Concentration in
Bath (g/1)
50.80
9.20
0.170
18.165 '
45.00
250.70
50.50
B-2
-------
APPENDIX B
Table B-S. Bath Concentrations for the Immersion SHver Technoloev
Bath
Cleaner
Mcroetch
Predip
Immersion Sflver
Chemical
Phosphoric acid
1,4-Butenediol
Sulfuric acid
Hydrogen peroxide
Sodium hydroxide
*4 other confidential chemicals
Sodium hydroxide
*5 other confidential chemicals
Concentration in Bath (g/I)
122.90
12.72
4.60
113.00
29.36
26.43
Table B-6. Bath Concentrations for the Immersion Tin Technology
Bath
Chemical
Concentration in Bath (g/1)
Cleaner
Ethylene glycol monobutyl ether
Fluoboric acid
Sulfuric acid
Phosphoric acid
*6 other confidential chemicals
22.90
12.33
184.00
30.25
Microetch
Sulfuric acid
*1 other confidential chemical
18.40
Predip
Methane sulfonic acid
Sulfuric acid
*10 other confidential chemicals
337.50
0.0092
Immersion Tin
Sulfuric acid
Urea
1,3-Diethylthiourea
Tin chloride
Methane sulfonic acid
Stannous methane sulfonic acid
*14 other confidential chemicals
92.18
90.00
20.00
13.98
69.17
111.80
B-3
-------
-------
Appendix C
Chemical Properties Data
-------
-------
Contents
1,3-Diethylthiourea . ..... f r
1,4-Butenediol . ....... ........... *..'.'.'.':'.'.'.'.'.'.'.'.'.'.'."" " ........ " " ..... r 3
Acetic Acid ................ . ............ ............ " ....... ' "
Branched Octylphenol, Ethoxylated ......... . . '.'.'.'.'.'.'.'.'.'.'.""•'"• ........... r~8
Ammonium Chloride .................... ....... ............ • " * " r"
Ammonium Hydroxide . . . ...... ................ '.'.'.'.""'"'• .......... ...... C12
Sodium Citrate (citric acid) ...... ..... ........ '.'.'.'.'.'.'.'' ....... ..... r"l4
Cupric Sulfate (copper ion) .................... '.'.'. '.'.['.•'.'.'.'. ........ ....... C~16
Cupric Acetate (copper sulfate pentahydrate) ...... ........ ................. r"18
Ethylenediamine ....... ..... ................. •' ....... ' r~~~
Ethylene Glycol ..... ...... ...... '.'.'.*.'.'.'.'.'.'.'.'."'. ........ ................... r 22
Ethylene Qycol Monobutyl Ether .......... .''.''.".'. ........... " " c"24
Fluoroboric Acid (fluoride) ... ............ ." ........ ' ...... " " "
Hydrochloric Acid . ................... " " .............. " " ".' .......... '
Hydrogen Peroxide ...... ,,
Lead ................... ............... ................... ............ . C-32
Maleic Acid .......... ............................ " ........... "'" ' ^"^
Malic Acid ..... ..... ........ ;-.:................ ....... '. ........... ^37
Methanesulfonic Acid .... .......................... " " '
Nickel Sulfate ........... ...................................... '
Palladium Chloride ...... .... . ...................... ' " .......... ' •
Phosphoric Acid ...... .... ........... '.'.'.'.'.'.'.'.'.'.'.'.'.'. ................... " r 45
Potassium Aurocyanide . . ............. ......... ............ ' ....... " ' •" '„
Potassium Peroxymonosuh%te ...... ...... ....................... r"4Q
Propionic Acid ........ ... ........ ................. ... ^_4y
Silver Mtrate ............. ' ." ".", ..... ........ " .............. " ........ • ' ' r"^
Sodium Hydroxide ................ *'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. ..... ............. ." " c 56
Sodium Hypophosphite and Sodium Hypophosphite Monohydrate ................ r'ss
Stannous Methanesuuonic Acid ................... T:
Sul&ric Acid ...... ........ ............. •""•
Thiourea . ...... '•• ....... " ............ ''•
Tin ............. ..... ........... ....... ....... • ;- ' '
Tin Chloride ..... ...................
References
-------
-------
APPENDIX C
CHEMICAL SUMMARY FOR 1,3-DIETHYLTHIOUREA
This summary is based on information retrieved from a systematic search limited to secondary sources. The. only
exception is summaries of studies ftom unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of 1,3-diethylthiourea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF 1,3-DffiTHYLTfflOUREA
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
105-55-5
N,N-diethylthiourea
CSH,2N2S
CjHsNHCSNHQHs
buff solid
132.32
78 °C
decomposes
4.56 g/L
Ulmg/m3
no data
49 (estimated)
0.57
0.240 mm Hg at 25 °C (estimated)
no data
no data
no data
no data
no data
no data
6.9xlO"8atm mVmole (estimated)
2 (estimated) •
no data
Reference
Lide (1995)
Lide(1995)
Lide (1995)
Lewis (1993)
Lewis (1993)
Lide (1995)
Lide (1995)
Lide (1995)
PHYSPROP (1998)
Ohm (1997)
HSDB (1998)
PHYSPROP (1998)
PHYSPROP (1998)
PHYSPROP (1998)
HSDB (1998)
H. ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, 1,3-diethylthiourea is not expected to adsorb to suspended solids and sediments in water
based upon an estimated Koc of 49 (HSDB, 1998; Swann et al., 1983), determined from a log Kow of 0.57 (Covers et
al., 1986, as cited in PHYSPROP, 1998) and a regression-derived equation (Lyman et al., 1990). Volatilization from
the water column to the atmosphere is not expected to occur (Lyman et al., 1990) based on an estimated Henry's Law
constant of 6.9xlQ-8 atm-m3/mole (PHYSPROP, 1998; SRC, 1998) . Since thiourea, a structurally similar compound,
was found to be stable to hydrolysis and photolysis (Schmidt-Bleek et al., 1982, as cited in HSDB, 1998), 1,3-
diethylthiourea is also expected to be stable to both hydrolysis and photolysis. According to a classification scheme
C-l
-------
APPENDIX C
(Franke et al., 1994), an estimated BCF of 2 (HSDB, 1998; Lyman et al., 1990) suggests that the potential for
bioconcentration in aquatic organisms is low.
B.
Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), 1,3-diethylthiourea, which has an estimated vapor pressure of 0.24 mm Hg at 25 °C (PHYSPROP, 1998; SRC,
1998), should exist solely as a vapor in the ambient atmosphere. The predominant removal process of 1,3-
diethylthiourea from the atmosphere is reaction with photochemically-produced hydroxyl radicals; the half-life for this
reaction in air is estimated to be 4 hours (Atkinson, 1988). 1,3-diethylthiourea, which has a high estimated water
solubility of 4.56 g/L (PHYSPROP, 1998; SRC, 1998), is expected to adsorb onto atmospheric particulate material; the
small amount of 1,3-diethylthiourea deposited onto particulate material may be physically removed by wet and dry
deposition (HSDB, 1998).
C.
Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 49 (HSDB, 1998), determined from a. log
Kow of 0.57 (Covers et al., 1986, as cited in PHYSPROP, 1998) and a regression-derived equation (Lyman et al.,
1990), indicates that 1,3-diethylthiourea is expected to have very high mobility in soil. Volatilization of 1,3-
diethylthiourea from moist soil surfaces is not expected to be important (Lyman et al., 1990) given an estimated Henry's
Law constant of 6.9xlO'$ atm-mVmole (PHYSPROP, 1998). In addition, 1,3-diethylthiourea is not expected to
volatilize from dry soil given its estimated vapor pressure of 0.24 mm Hg (PHYSPROP, 1998; SRC, 1998).
D. Summary
If released to air, an estimated vapor pressure of 0.24 mm Hg at 25 °C indicates that 1,3-diethylthiourea should exist
solely as a vapor in the ambient atmosphere. Gas-phase 1,3-diethylthiourea will be degraded in the atmosphere by
reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 4 hours.
1,3-Diethylthiourea is not expected to adsorb to suspended solids and sediments in water. An estimated BCF of 2
suggests the potential for bioconcentration in aquatic organisms is low. If released to soil; 1,3-diethylthiourea is
expected to have very high mobility based upon an estimated Koc of 49, and, therefore, it has the potential to leach to
groundwater. Volatilization from water and from moist soil surfaces is not expected to be an important fate process
based upon a Henry's Law constant of 6.9x10'8 atm-mVmole. Volatilization from dry soil surfaces is not expected to
occur based upon the vapor pressure of this compound.
C-2
-------
APPENDIX C
SUMMARY FOR 1,4-BUTENEDIOL
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of 1,4-butenediol are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF 1,4-BUTENEDIOL
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point •
Water Solubility
Density
Vapor Density (air = 1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
110-64-5
2-butene-l,4-diol (mixed isomers)
C4H802
HOCH2CH=CHCH2OH
pale, yellow liquid
88.1
4 °C (cis); 25 °C (trans)
235 °C (cis); 135 °C @ 12 mm Hg (trans)
soluble; estimated to be >lx!03 g/1
specific gravity = 1.07 @ 25 °C (liquid)
no data
8.6 (estimated)
-0.81
4.7xlO'3 mm Hg @ 25 °C (extrapolated)
no data
not flammable: flash point >100 °F
263 °F (Cleveland open cup)
no data
no data
no data
1.54x10-'° atm mVmole (estimated)
0.14 (estimated)
no data
Grafjeetal. (1985)
Grafje et al. (1985)
Grafjeetal. (1985)
Grafjeetal. (1985)
Grafjeetal. (1985)
Grafjeetal. (1985)
Howard and Meylan (1997)
Howard and Meylan (1997)
Grafje et al. (1985); SRC (1998)
Weiss (1986)
Lyman et al. (1990)
Hansch et al. (1995)
Grafjeetal. (1985)
Cote (1997)
Flick (1991)
Meylan and Howard (1991)
Boethlingetal.(1994)
H. ENVIRONMENTAL FATE
A.
Aquatic Fate
An estimated Koc of 8.6, determined from a log Kow of-0.81 (Hansch et al., 1995) and a regression-derived equation
(Lyman et al., 1990), indicates that 1,4-butenediol is not expected to adsorb to suspended solids and sediment in water.
Also, an estimated Henry's Law constant of 1.54xiO-'° atm nf/mole at 25 °C (Meylan and Howard, 1991) indicates that
1,4-butenediol is not expected to volatilize from water surfaces (Lyman et al., 1990). Hydrolysis is not expected to be an
important fate process for 1,4-butenediol due to the lack of hydrolyzable functional groups (Lyman et al., 1990). No
data were available in the scientific literature for the biodegradation of 1,4-butenediol in aquatic media under aerobic or.
anaerobic conditions. However, using a structure estimation method (Boethling et al., 1994), aerobic biodegradation is
C-3
-------
APPENDIX C
expected to be rapid (days to weeks). According to a classification scheme (Franke et al., 1994), an estimated BCF of ,
0.14 (Lyman et al., 1990), obtained from the log Kow, suggests the potential for bioconcentration of 1,4-butenediol in
aquatic organisms is low.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), 1,4-butenediol, which has an extrapolated vapor pressure of 4.7xlO'3 mm Hg at 25 °C (Grafje et al., 1985), is
expected to exist solely as a gas in the ambient atmosphere. Gas-phase 1,4-butenediol is degraded in the atmosphere by
reaction with photochemicaUy-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 5-6 .
hours, depending upon the isomer (Meylan and Howard, 1993). The half-life for the reaction of 1,4-butenediol with
ozone in the atmosphere is estimated to be 1-2 hours, depending upon the isomer (Meylan and Howard, 1993). 1,4-
Butenediol is not expected to directly photolyze in the atmosphere due to the lack of absorption in the environmental UV
spectrum greater than 290 nm (Lyman et al., 1990). Because 1,4-butenediol is miscible with water, physical removal
from the atmosphere by wet deposition may occur.
C.
Terrestrial Fate
An estimated Koc of 8.6 (Lyman, 1990), determined from a log Kow of-0.81 (Hansch et al., 1995), indicates that 1,4-
butenediol is expected to have very high mobility in soil (Swann et al., 1983). Volatilization of 1,4-butenediol from
moist soil surfaces is not expected to be important (Lyman et al., 1990) given an estimated Henry's Law constant of
1.54xlO'10 atm mVmole (Meylan and Howard, 1991). In addition, an extrapolated vapor pressure of 4.7xlQ-3 mm Hg
(Grafje et al., 1985) indicates that 1,4-butenediol is not expected to volatilize from dry soil surfaces. No data were
available in the scientific literature for the biodegradation of 1,4-butenediol in soil under aerobic or anaerobic
conditions. However, using a structure estimation method (Boethling et al., 1994), aerobic biodegradation is expected
to be rapid (days to weeks).
D.
Summary
1,4-Butenediol exists as a mixture of the cis and trans isomers that are expected to behave similarly in the environment.
If released to air, an extrapolated vapor pressure of 4.7xlO'3 mm Hg at 25 °C indicates 1,4-butenediol should exist
solely as a gas in the ambient atmosphere. Gas-phase 1,4-butenediol will be degraded in the atmosphere by reaction
with photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 5-6 hours,
depending upon the isomer. The gas phase reactions of 1,4-butenediol with photochemically produced ozone
corresponds to a half-life of 1-2 hours. Physical removal of gas-phase 1,4-butenediol from the atmosphere may also
occur via wet deposition processes based on the miscibility of this compound with water. If released to soil, 1,4-
butenediol is expected to have very high mobility and is not expected to adsorb to soil surfaces. Volatilization from
water and moist soil surfaces is not expected to be an important fate process based upon an estimated Henry's Law
constant of 1.54xlO-1° atm mVmole. In addition, volatilization from dry soil surfaces is not expected to occur based upon
the vapor pressure of 1,4-butenediol. Biodegradation data were not available from the scientific literature; however, a
computer model estimates that aerobic biodegradation in both soil and water may occur within days to weeks. In water,
1,4-butenediol is not expected to bioconcentrate in fish and aquatic organisms based on its estimated BCF of 0.14.
C-4
-------
APPENDIX C
CHEMICAL SUMMARY FOR ACETIC ACED
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of acetic acid are summarized in Table 1.
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility "
Density
Vapor Density (air =1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
•Dissociation Constant
Molecular Diffusivity Constant
Air Diffiisivity Constant
Henry's Law Constant
Fish Bioconcentration Factor
Odor Threshold
Data
64-19-7
ethanoic acid; vinegar acid
C2H402
CH3COOH
clear liquid
60.05
16.7 °C
118 °C
IxlO3 g/1, 25 °C
d™5, 1.049 ' . .
no data
6.5-228
-0.17
15.7 mm Hg @ 25 °C
corrosive, particularly when dilute
flammable
103 °F (39 °C), closed cup
pKa = 4.76
no data
no data
l.OOxlO'' atm mVmole @ 25 °C
<1 (calculated)
no data
• Reference
Howard and Neal (1992)
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavari et al. (1996)
Budavarietal. (1996)
U.S. EPA (1981)
Budavari et al. (1996)
Sansone et al. (1987)
Hanschetal. (1995)
Daubert and Danner (1985)
Weiss (1986)
Budavari et al. (1996)
Budavarietal. (1996)
Serjeant and Dempsey ( 1 979)
Gaffhey et al. (1987)
Lyman et al. (1990)
II. ENVIRONMENTAL FATE
A.
Aquatic Fate
The dominant environmental fate process for acetic acid in water is expected to be biodegradation. A large number of
biological screening studies have determined that acetic acid biodegrades readily under both aerobic (Zahn and Wellens,
1980; Dore et al., 1975; Price et al., 1974; Placak and Ruchhoft, 1947 as cited in HSDB, 1998) and anaerobic (Kameya'
et al., 1995; Mawson et al., 1991; Swindoll et al., 1988 as cited in HSDB, 1998) conditions. Two aqueous adsorption
studies found that acetic acid exists primarily in the water column and not in sediment (Hemphill and Swanson, 1964;
Gordon and Millero, 1985 as cited in HSDB, 1998). In general, organic ions are not expected to volatilize from water to
adsorb to particulate matter in water to the degree that would be predicted for their neutral counterparts. Volatilization
C-5
-------
APPENDIX C
from the water column to the atmosphere is not expected to occur (Lyman et al., 1990 as cited in HSDB, 1998) based on
a Henry's Law constant of 1x10'9 atm-mVmole at pH 7 (Gaffiiey et al., 1987 as cited in HSDB, 1998). According to a
classification scheme (Franke et al., 1994 as cited in HSDB, 1998), an estimated BCF of <1 (Lyman, 1990 as cited in
HSDB, 1998), calculated from a log Kow of-0.17 (Hansch et al., 1995 as cited in HSDB, 1998), suggests that the
potential for bioconcentration in aquatic organisms is low.
B.
Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman, 1988
as cited in HSDB, 1998), acetic acid, which has a vapor pressure of 15.7 mm Hg at 25°C (Daubert and Danner, 1989 as
cited in HSDB, 1998), should exist solely as a gas in the ambient atmosphere. This is consistent with a study in which
over 91% of the total measured acetic acid in an air sample was found to be in the gas phase (Khwaja, 1995 as cited in
HSDB, 1998). Acetic acid has been identified as one of the major sources of free acidity in precipitation from remote
regions of the world (Keene and Galloway, 1984 as cited in HSDB, 1998), indicating that physical removal by wet
deposition is an important fate process (Hartmann et al., 1989 as cited in HSDB, 1998). Another important removal
process of acetic acid from the atmosphere is reaction with photochemically-produced hydroxyl radicals; the half-life for
this reaction in air is estimated to be 22 days (Atkinson, 1989 as cited in HSDB, 1998). Acetic acid has also been
detected adsorbed to atmospheric particulate material as the acetate (Gregory et al., 1986; Khwaja, 1995 as cited in
HSDB, 1998); the small amount of acetic acid associated with particulate material may be physically removed by wet
and dry deposition (Grosjean, 1992).
C.
Terrestrial Fate
The major environmental fate process for acetic acid in soil is expected to be biodegradation. A large number of
biological screening studies have determined that acetic acid biodegrades readily under both aerobic ,Zahn and Wellens,
1980; Dore et al., 1975; Price et al., 1974; Placak and Ruchhoft, 1947 as cited in HSDB, 1998) and anaerobic (Kameya
etal., 1995; Mawson et al., 1991; Swindoll et al., 1988 as cited in HSDB, 1998) conditions. Based on a classification
scheme (Swann et al., 1983 as cited in HSDB, 1998), Koc values of 6.5 to 228 (Sansone et al., 1987 as cited in HSDB,
1998) indicate that acetic acid is expected to have moderate to very high mobility in soil. This is consistent with a study
in which no sorption was reported for three different soils/sediments (Von Oepen et al., 1991 as cited in HSDB, 1998).
Volatilization of acetic acid from moist soil surfaces is not expected to be important (Lyman et al., 1990, as cited in
HSDB, 1998) given a Henry's Law constant of IxlO'9 atm-mVmole (Gaffiiey et al., 1987 as cited in HSDB, 1998) and
because acetic acid will exist predominantly as the acetate at environmental pH's. However, the potential for
volatilization of acetic acid from dry soil surfaces may exist based on it's vapor pressure of 15.7 mm Hg (Daubert and
Danner, 1989 as cited in HSDB, 1998). Volatilization will be attenuated depending upon pH and the amount of acetic
acid dissociated.
D.
Summary
Acetic acid occurs throughout nature as a normal metabolite of both plants and animals. Consequently, acetic acid's fate
in the environment will, in part, be dependent on its participation in natural cycles. With a pKa of 4.76, acetic acid and
its conjugate base will exist in environmental media in varying proportions that are pH dependent; under typical
environmental conditions (pHs of 5 to 9), acetic acid will exist almost entirely in the ionized (dissociated) form. If
released to air, a vapor pressure of 15.7 mm Hg at 25 °C indicates that acetic acid should exist solely as a gas in the
ambient atmosphere. Gas-phase acetic acid will be degraded in the atmosphere by reaction with photochemically-
produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 22 days. Physical removal of vapor-
phase acetic acid from the atmosphere may occur via wet deposition processes based on its miscibility with water. An
estimated BCF of <1 suggests the potential for bioconcentration on aquatic organisms is low. Adsorption studies
indicate that acetic acid is not expected to adsorb to suspended solids and sediments in water. If released to soil, acetic
acid is expected to have very high to moderate mobility based upon measured Koc values ranging from 6.5 to 228 and,
therefore, it has the potential to leach to groundwater. If released to soil in high concentrations, such as those
encountered in a spill, acetic acid may travel through soil and reach groundwater. Volatilization from water and from
moist soil surfaces is not expected to be an important fate process based upon a Henry's Law constant of IxlO'9 atm-
m'/mole. Yet, volatilization from dry soil surfaces may occur based upon the vapor pressure of this compound.
However, volatilization of acetic acid will be pH dependent; if acetic acid is dissociated, very little (about 1%) will be
C-6
-------
APPENDIX C
available for volatilization. Biodegradation is expected to be rapid and may be the dominant fate process in both soil
and water under non-spill conditions; a large number of biological screening studies have determined that acetic acid
biodegrades readily under both aerobic and anaerobic conditions.
C-7
-------
APPENDIX C
. * CHEMICAL SUMMARY FOR BRANCHED OCTYLPHENOL, ETHOXYLATED1
(alkylphenol polyethoxyethanol)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verity information in these databases and secondary sources. The
search identified sources of information for branched octylphenol, ethoxylated.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of branched octylphenol, ethoxylated1 are summarized in Table
1. • ' '
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF BRANCHED
OCTYLPHENOL, ETHOXYLATED'
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Row
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
9036-19-5, 9002-93-1
Triton X-1001, OPIOSP •
CHH220.(C2H40),M
(C8H17)C6H40(CJH,0)1M
Clear viscous liquid
polymer, >4000
7.2°C
271°C
Dispersible, >100 g/L
d", 1.07
>l
No data
No data
<0.001 torr
No data
No data
288°C
No data
No data
No data
No data
No data
No data
Reference
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
• MSDS
"Howard and Neal (1992)
MSDS
MSDS
MSDS
MSDS
MSDS
MSDS
MSDS
•
The properties are given for TritonXlOO (manufacturer Rohm and Haas).
C-8
-------
APPENDIX C
CHEMICAL SUMMARY FOR AMMONIUM CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ammonium chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF AMMONIUM CHLORIDE
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
12125-02-9
Ammonium muriate
C1H4N
colorless cubic crystals
53.492
sublimes at 3 50 °C
no data
approximately 300 g/L '
1.519 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
1.84X10-'2 mm Hg at 25"C (extrapolated)
no data
not flammable
not flammable
dissociates to NH4* and Cl"
no data
no data
no data; expected to be < IxlO"8
no data
odorless
CAS (1998)
Budavari et al. (1996)
Budavarietal. (1996)
Budavari et'al. (1996)
Lide (1995)
Lide (1995)
Lewis (1993)
Estimated
Lide (1995)
^Estimated
Estimated
Daubert and Danner (1992)
Weiss (1986)
Weiss (1986)
Bodek et al. (1988)
Estimated
Weiss (1986)
1 Estimated from a reported solubility of 37 parts in 100 parts water at 20°C (Dean 1985).
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If ammonium chloride is released into water, it is expected to dissociate into ammonium (NH4+) and chloride (Cr) ions
(Bodek et al., 1988). The counter ion associated with the NH4+ will vary depending on the concentration and type of
ions available and the pH in the receiving water. In addition, NH4+ and NH3 (ammonia) are in equilibrium in the
environment and since the pKa of the ammonium ion, NH4*, is 9.26, most ammonia in water is present as the protonated
form rather than as NH3 (Manahan, 1991). Ammonia is, however, present in the equilibrium and will volatilize to the
atmosphere (based upon its Henry's Law constant of 1.6X10'5 atm mVmole [Betterton, 1992 as cited in PHYSPROP,
C-9
-------
APPENDIX C
1998]); the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature (ATSDR, 1990).
In the aquatic environment, ammonium can undergo sequential transformation by the nitrification and denitrification
processes of the nitrogen cycle; within this process, ionic nitrogen compounds are formed (ATSDR, 1990). In addition,
ammonium can be taken up by aquatic plants as a source of nutrition, and the uptake of ammonium by fish has also been
documented (ATSDR, 1990). Adsorption of ammonium to sediment should increase with increasing organic content,
increased metal content, and decreasing pH; however, ammonium can be produced in, and subsequently released from,
sediment (ATSDR, 1990). The dissociation of ammonium chloride into its component ions indicates that ammonium
chloride is not expected to bioconcentrate in aquatic organisms. Ammonium ions may be adsorbed by negatively
charged surfaces of sediment in the water column, however ammonium ions are expected to be replaced by other cations
present in natural waters (Evans, 1989). The chloride ion may complex with heavy metals, thereby increasing their
solubility (Bodek et al., 1988). Adsorption of the chloride ion to suspended solids and sediment in the water column is
.not expected to be.an important fate process.
B.
Atmospheric Fate
If ammonium chloride is released to the atmosphere, this compound's low vapor pressure (Daubert and Danner, 1992)
indicates it will exist as a particulate in the ambient atmosphere. Ammonium chloride is expected to undergo wet
deposition (ATSDR, 1990) in rain, snow, or fog based upon its high water solubility (Dean, 1985). Dry deposition of
ammonium chloride is expected to be an important fate process in the atmosphere (ATSDR, 1990). The rate of dry
deposition will depend on the prevailing winds and particle size (Bodek et al., 1988). In addition, NH4+ and NH3
(ammonia) are in equilibrium. The gas-phase reactions of ammonia with photochemically produced hydroxyl radicals
has been reported to be 1.6xlO"13 cmVmolc-sec, with a calculated half-life of approximately 100 days; this process
contributes approximately 10%-to the removal of atmospheric ammonia (ATSDR, 1990).
C. Terrestrial Fate
If ammonium chloride is released to soil, it is expected to dissociate into its component ions in moist soils. As noted
above, NHL,* and NH3 (ammonia) are in equilibrium in the environment and since the pKa of the ammonium ion, NH/,
is 9.26, most ammonia in water is present as the protonated form rather than as NH3 (Manahan, 1991). The low vapor
pressure and Henry's Law constant expected for an ionic salt indicates that ammonium chloride will not volatilize from
either dry or moist soil surfaces. Nonetheless, ammonia is present in the equilibrium and will volatilize to the
atmosphere (based upon its Henry's Law constant of 1.6X10"5 atm mVmole [Betterton, 1992 as cited in PHYSPROP,
1998]); the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature (ATSDR, 1990).
The mobility of ammonium ions through soil may be attenuated by attraction to negatively charged surfaces of soil
particles, however ammonium ions are expected to be replaced by other cations present in soil (Evans, 1989). In soil,
ammonium will serve as a source of nutrient taken up by plants and other organisms and converted to organic-nitrogen
compounds. Ammonium can be converted to nitrate.by microbial populations through nitrification; the nitrate formed
will either leach through soil or be taken up by plants and other organisms. It has been determined that minerals and dry
soils can rapidly and effectively adsorb ammonia from air. Chloride is extremely mobile in soils (Bodek et al., 1988).
The chloride ion may complex with heavy metals, thereby increasing their solubility (Bodek et al., 1988) and potential
for leaching into groundwater.
D.
Summary
If released into water, ammonium chloride is expected to dissociate into ammonium and chloride ions. The dissociation
of ammonium chloride into its component ions indicates that ammonium chloride is not expected to bioconcentration in
aquatic organisms. Ammonium, however, will be used as a nutrient source by microorganisms and plants, and rapid
uptake is anticipated. Ammonium is in equilibrium with ammonia, but the majority will be in the ammonium form under
most environmental pHs. When present, ammonia's Henry's Law constant indicates that volatilization from water
surfaces may occur. If released to soil, ammonium chloride is expected to dissociate into its component ions in moist
soils and will be used as a nutrient by microorganisms and plants. The dissociation of ammonium chloride into its
component ions in moist soils indicates that volatilization of ammonium from moist soil surfaces is not expected to
occur. The mobility of ammonium ions in soil is expected to be attenuated by cation exchange processes. The low
vapor pressure expected for an ionic salt indicates that ammonium chloride is not expected to volatilize from dry soil
surfaces, however, when ammonia is present in equilibrium, volatilization may occur. If released to the atmosphere,
C-10
-------
APPENDIX C
ammonium chloride's low vapor pressure indicates this compound will exist, as a particulate. Wet and dry deposition
will be the dominant fate processes in the atmosphere. The rate of dry deposition will depend on the prevailing wind
patterns and particle size. Some atmospheric oxidation may occur.
C-ll
-------
APPENDIX C
CHEMICAL SUMMARY FOR AMMONIUM HYDROXIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ammonium hydroxide are summarized hi Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF AMMONIUM HYDROXIDE
Characteristic/Property ' Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Row
Vapor Pressure
Reactivity
Flam inability
Flash Point
Dissociation Constant •
Molecular Difiusivity Constant
Air Diffusrvity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
1336-21-6 tide (1995)
ammonia solution; aqua ammonia; ammonium hydrate Lewis (1993)
HSNO PHYSPROP (1998)
NH4OH Lide(1995)
colorless liquid • Lewis (1993)
35.05 Lide(1995)
no data
no data
soluble in water Sax (1984)
no data
no data
no data; estimated to be < 10 Estimated
no data; estimated to be < 1 . Estimated
no data
incompatible w/ HCI, HNO3, Ag compounds Sax (1984)
not flammable Weiss (1986)
no data; estimated to be > 350 °C Estimated
9.26 (water solution) • Manahan (1991)
no data
no data
no data1
no data
no data
1 In the environment, ammonium ion is expected to predominate in the ammonia-ammonium ion equilibrium; however, this equilibrium is highly
dependent on both pH and temperature (ATSDR, 1990). Ammonia is expected to have a very high Henry's Law constant, while ammonium is
expected to have a negligible Henry's Law constant (SRC, 1998).
II. ENVIRONMENTAL FATE
A.
Aquatic Fate
If released into the water column at low concentrations, ammonia or ammonium hydroxide will volatilize to the
atmosphere; the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature (ATSDR,
1990). Since the pKa of the ammonia is 9.26, most ammonia in most environmental waters is present as the protonated,
NH4*, form rather than as NH3 (Manahan, 1991). In the aquatic environment, ammonia can undergo sequential
transformation by the nitrification and denitrification processes of the nitrogen cycle; within this process, ionic nitrogen
C-12
-------
APPENDIX C
compounds are formed (ATSDR, 1990). In addition, ammonia can be taken up by aquatic plants as a source of
nutrition, and the uptake of ammonia by fish has also been documented (ATSDR, 1990). Adsorption of ammonia to
sediment should increase with increasing organic content, increased metal content, and decreasing pH; however
ammonia can be produced in, and subsequently released from sediment (ATSDR, 1990). Large releases of the '
concentrated base into water, such as may result from a spill, will result in an increase of the pH (ATSDR, 1990).
B. Atmospheric Fate
If ammonia is released to the atmosphere, its vapor pressure indicates it will exist as a vapor in the ambient atmosphere.
If ammonium hydroxide is released to the atmosphere, it is anticipated that the dominant form will be as a particulate,
but during equilibrium between ammonium and ammonia, the ammonia will rapidly leave the particle as a vapor. The
dominant fate process for the removal of ammonia from the atmosphere is the reaction with acid air pollutants to form
ammonium compounds (e.g., ammonium sulfate, ammonium nitrate); these ammonium compounds can then be removed
by wet or dry deposition (ATSDR, 1990). In addition, gas-phase reactions of ammonia with photochemically produced
hydroxyl radicals has been reported to be 1.6xlQ-13 cm3/molc-sec, with a calculated half-life of approximately 100 days;
this process contributes approximately 10% to the removal of atmospheric ammonia (ATSDR, 1990).
C.
Terrestrial Fate
If ammonia or ammonium hydroxide is released to soil, it will serve as a source of nutrient taken up by plants and other
organisms and converted to organic-nitrogen compounds. Ammonia can be converted to nitrate by microbial
populations through nitrification; the nitrate formed will either leach through soil or be taken up by plants and other
organisms. It has been determined that minerals and dry soils can rapidly and effectively adsorb ammonia from air.
Specifically, ammonia may be either bound to soil or undergo volatilization to the atmosphere. (ATSDR, 1990)
D. Summary
Ammonia is a base, and as such, the environmental fate of ammonia is pH and temperature dependent. If released into
water, ammonia and ammonium hydroxide will volatilize to the atmosphere, depending on the pH. At high pHs, where
the equilibrium more favors ammonia, volatilization will become increasingly important At low pHs, volatilization will
be less important. Adsorption of ammonia to sediment and suspended organic material can be important under proper
conditions (i.e., organic matter content, metal content, and pH). In addition, ammonia will be taken up by aquatic
organisms and plants as a source of nutrition. The dominant fate of ammonia in water will be its participation in the
nitrogen cycle. The predominant removal process of ammonia and ammonium hydroxide from the atmosphere is
expected to be wet and dry deposition. To a lesser extent, reactions with photochemically-produced hydroxyl radicals
will occur. If released to soil, ammonia is expected to be taken up by plants and other organisms and converted to
organic-nitrogen compounds. These compounds will either be taken up by plants or other organisms or leach through
•the soil. Volatilization of ammonia from soil surfaces is expected to occur.
CMS
-------
APPENDIX C
CHEMICAL SUMMARY FOR SODIUM CITRATE (citric acid)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium citrate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM CITRATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Kcc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Henrys Law Constant
Molecular Diffusivity Coefficient
Air DiHusivity Coefficient
Fish Bioconccntration Factor
Odor Threshold
Conversion Factors
Data
68-04-2
trisodium citrate; sodium citrate anhydrous; 2-hydroxy-l,2,3-
propanetricarboxylic acid, trisodium salt
QH5Na307
CH2(COONa)C(OHXCOONa)CH2COONa
dihydrate, white crystals, granules, or powder; pentahydrate,
relatively large, colorless crystals or white granules
258.07
150°C(-2H2O)
decomposed at red heat
72 g/100 mL at 25°C (dihydrate)
1.9
no data
no data
no data
no data
0 (nonreactive, NFPA classification);
aqueous solution slightly acid to litmus
1 (slightly combustible, NFPA classification);
no data
no data
no data
no data
no data
no data
no data; odorless
no data
Reference
Lockheed Martin 1991
Budavari et al. 1989
Osol 1980
Budavari et al. 1989
Budavari etal. 1989
Fisher Scientific 1985
Lewis 1993
Weast 1983-1984
Fisher Scientific 1985
Lockheed Martin 1991
Osol, 1980
Lockheed Martin 1991
Lewis 1993
II. ENVIRONMENTAL FATE
A. Environmental Release
Sodium citrate is a solid with a cool, saline taste that is soluble in water (Fisher Scientific 1985). It is used in soft
drinks, frozen desserts, meat products, cheeses, and as a nutrient for cultured buttermilk; in photography; in detergents;
as a sequestrant and buffer; as an anticoagulant, for blood withdrawn from the body; and in the removal of sulfur dioxide
from smelter waste gases (Lewis 1993). Medicinally, sodium citrate is used as expectorant and systemic alkalizer.
CM4
-------
APPENDIX C
Sodium citrate is a chelating agent and has been used to facilitate elimination of lead from the body (Osol 1980).
No data were found on the environmental releases of sodium citrate. The chemical is not listed on U S EPA's Toxics
Release Inventory, requiring certain U.S. industries to report on chemical releases to the environment (TRI93 1995)
The chemical could potentially enter the environment when used for the removal of sulfur dioxide from smelter waste
gases.
B. Transport
No data were found on the environmental transport of sodium citrate in the secondary sources searched. Its water
solubility suggests that the sodium citrate would remain in the water phase.
C. Transformation/Persistence
No data were found on the transformation/persistence of potassium bisulfate in the secondary sources searched.
C-15
-------
APPENDIX C
CHEMICAL SUMMARY FOR CUPRIC SULFATE (copper ion)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of cupric sulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF CUPRIC SULFATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
LogKow
Vapor Pressure
Reactivity
Reactivity
Data
7758-99-8
cupric sulfate pentahydrate; blue Vitriol
CuO4S-5H2O
CuSO«-5H20
large, blue, triclinic crystals; blue powder'
249.68
decomposes @110°C
decomposes to CuO @ 650°C
316g/L@0°C
2,286 g/cm3
no data
no data
no data
no data
reacts with Mg to produce Cu2O, MgSO.,, and H2
reacts with NHiGl producing (NH^SO., and CuCl2; .
Reference
Lide (1995)
Budavari et al. (1996)
ATSDR(1990)
Lide (1995)
Budavari etal. (1996)
Lide (1995)
Lide (1995)
ATSDR(1990)
Weast(1985)
Lide (1995) '
U.S. Air Force (1990)
HSDB(1998)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diflusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
reacts with excess aq. NH3 producing Cu(NH3)22'1' + OH" •
Decomposition products include SO2.
non-flammable . HSDB(1998)
non-flammable. HSDB (1998)
no data
no data
no data
no data
10-100 for copper; 30,000 for copper in oysters ATSDR (1990)
no data
II. ENVIRONMENTAL FATE
A. Aquatic Fate
Copper (Cu) commonly exists in three valence states, Cu° (metal),.Cu+ (cuprous), and Cu2+ (cupric). It can also be.
oxidized to a Cu3* state, but Cu3* ions are rapidly reduced to Cu2* in the environment (ATSDR, 1990). Copper in
solution is present almost exclusively as the Cu2+ valence state (U.S. EPA, 1987). Copper in the Cu2+ valence state
forms compounds and complexes with a variety of organic and inorganic ligands binding to -NH2, -SH, and, to a lesser
C-16
-------
————— APPENDIX C
•. So,u»0^X'^c ^^Z!^^^^^^^;^-
above pH 6 (U.S EPA, 1987; ATSDR, 1990). The association of copper with organic or inorganic ligands also
Zl ? I P5 ?" CaC°3 alkaHnity- M°St °f ^ C°Pper enterinS surface water is in the form of particulate
matter, which settles out, precipitates, or adsorbs to organic matter, hydrous iron and manganese oxides, and clay;
however, the predominating form can change with the amount of rain, pH, content of runoff and the availability of
hgands (ATSDR, 1990). The processes of complexation, adsorption and precipitation fimit the concentration of copper
(Cu ) to very low values m most natural waters (ATSDR, 1990). Calculations of the bioconcentration factor in fish for
copper have ranged from 10 to 100; however, the majority of copper measurements in fish tissues under environmental
conditions have indicated little, ,f any, bioconcentration. Filter feeding shellfish, especially oysters, however, were found
to significantly concentrate copper with bioconcentration factors as high as 30,000 (ATSDR, 1990).
B.
Atmospheric Fate
Most of the copper in the air is in the form of paniculate matter (dust) or is adsorbed to particulate matter Larser
particles (>5 urn) are removed by gravitational settling, smaller particles are removed by other forms of dry and wet
deposition (ATSDR, 1990). Atmospheric copper resulting from combustion is associated with sub-micron particles that
can remain in the troposphere for an estimated 7-30 days and may be carried long distances (ATSDR 1990)
C.
Terrestrial Fate
.Most of the copper deposited in the soil is strongly adsorbed primarily to organic matter, carbonate minerals clay
minerals, and hydrous iron and manganese oxides. Movement through the soil is dependent on the presence'of the'se
substances the pH, and other physical and chemical parameters. The greatest potential for leaching is seen in sandy
sods w,th ow pH (ATSDR, 1990). Laboratory experiments using controlled models and field experiments utilizing
core samples have shown that very little copper moves through the soil. Core samples showed that some movement
occurred as far as the 22.5-25 cm layer of soil, but little, if any, moved below this zone. The evidence indicates that
hazardous amounts of copper should not leach into groundwater from sludge, even from sandy soils (ATSDR, 1990).
D. Summary
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu state, but there are no important industrial Cu3+ chemicals, and Cu3+ ions are rapidly reduced to Cu2+
in the environment. If released to water, copper in solution will be present almost exclusively as the Cu2+ valence state
The predominant form of copper in aqueous solution is dependent on the pH of the solution. Most of the copper
entering surface water is in the form of particulate matter; however, the predominating form can change with the amount
of ram, PH, content of runoff, and the availability of ligands. Copper in the Cu2+ valence state will form compounds and
complexes with a variety of organic and inorganic ligands. Calculations of the bioconcentration factor in fish for copper
have ranged from 10 to 100; however, the majority of copper measurements in fish tissues under environmental
conditions have mdicated little, if any, bioconcentration. .If released to soil, the majority of copper deposited in the soil is
strongly adsorbed. Movement through the soil is dependent on the presence of organic matter, carbonate minerals clay
minerals, hydrous iron and manganese oxides, the pH, and other physical and chemical parameters The °reatest '
potential for leaching 1S seen in sandy soils with low pH. If released into the atmosphere, copper is expected to exist as a
dust particulate or adsorb to particulate matter. Studies have shown that copper can remain in the atmosphere up to 30
days and be carried long distances. . '
C-17
-------
APPENDIX C
CHEMICAL SUMMARY FOR CUPRIC ACETATE (copper sulfate pentahydrate)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of cupric acetate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF COPPER ACETATE
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air » 1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
6046-93-1
copper (n) acetate monohydrate
(CH3CO2)2Cu-H2O
CuCCjHjOj )2-H20
dark, green monoclinic crystals
greenish-blue, fine powder
199.65
115 °C
decomposes at 240 °C
72 g/L cold water; 200 g/L hot water
1.88 g/cm3
no data
no data
no data
no data
stable
not flammable
not flammable
no data
no data
no data
no data
10-100 for copper; 30,000 for copper in oysters
no data
Lide (1995)
tide (1995)
Aldrich (1996)
Lide (1995)
Budavari et al. (1996)
Lewis (1993)
Lide (1995)
Lide (1995)
Lide (1995)
Weast(1985)
Lide (1995)
Weiss (1986)
Weiss (1986)
Weiss (1986)
ATSDR(1990)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu3* state, but Cu3+ ions are rapidly reduced to Cu2* in the environment (ATSDR, 1990). Copper in
solution is present almost exclusively as the Cu2+ valence state (U.S. EPA, 1987). Copper in the Cu2+ valence state
forms compounds and complexes with a variety of organic and inorganic ligands binding to -NH2, -SH, and, to a lesser
extent, -OH groups (ATSDR, 1990). The predominant form of copper in aqueous solution is dependent on the pH of
the solution. $elow pH 6, the cupric ion (Cu2+) predominates; copper complexes with carbonate usually predominate
C-18
-------
APPENDIX C
above pH 6 OU-S. EPA, 1987; ATSDR, 1990). The association of copper with organic or inorganic ligands also
depends on the pH and on the CaCO3 alkalinity. Most of the copper entering surface water is in the form of particulate
matter, which settles out, precipitates, or adsorbs to organic matter, hydrous iron and manganese oxides and clay-
however, the predominating form can change with the amount of rain, pH, content of runoff, and the availability of
ligands (ATSDR, 1990). The processes of complexation, adsorption and precipitation limit the concentration of copper
(Cu ) to very low values in most natural waters (ATSDR, 1990). Calculations of the bioconcentration factor in fish for
copper have ranged from 10 to 100; however, the majority of copper measurements in fish tissues under environmental
conditions have indicated little, if any, bioconcentration. Filter feeding shellfish, especially oysters, however were found
to significantly concentrate copper with bioconcentration factors as high as 30,000 (ATSDR, 1990).
B. Atmospheric Fate
Most of the copper in the air is in the form of particulate matter (dust) or is adsorbed to particulate matter Larger
particles (>5 urn) are removed by gravitational settling, smaller particles are removed by other forms of dry and wet
deposition (ATSDR, 1990). Atmospheric copper resulting from combustion is associated with sub-micron particles that
can remain m the troposphere for an estimated 7-30 days.and may be carried long distances (ATSDR 1990)
C.
Terrestrial Fate
Most of the copper deposited in the soil is strongly adsorbed primarily to organic matter, carbonate minerals clay
minerals, and hydrous iron and manganese oxides. Movement through the soil is dependent on the presence'of these
substances, the PH, and other physical and chemical parameters. The greatest potential for leaching is seen in sandy
soils with low pH (ATSDR, 1990). Laboratory experiments using controlled models and field experiments utilizing
core samples have shown that very little copper moves through the soil. Core samples showed that some movement
occurred as far as the 22.5-25 cm layer of soil, but little, if any, moved below this zone. The evidence indicates that
hazardous amounts of copper should not leach into groundwater from sludge, even from sandy soils (ATSDR, 1990).
D. Summary
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu state, but there are no important industrial Cu3+ chemicals, and Cu3+' ions are rapidly reduced to Cu2+
m the environment. If released to water, copper in solution will be present almost exclusively as the Cu2+ valence state
The predominant form of copper in aqueous solution is dependent on the pH of the solution. Most of the copper
entering surface water is .in the form of particulate matter; however, the predominating form can change with the amount
of ram, pH, content of runoff, and the availability of ligands. Copper in the Cu2+ valence state will form compounds and
complexes with a variety of organic and inorganic ligands. Calculations of the bioconcentration factor in fish for copper
have ranged from 10 to 100; however, the majority of copper measurements in fish tissues under environmental
conditions have indicated little, if any, bioconcentration. If released to soil, the majority of copper deposited in the soil is
strongly adsorbed. Movement through the soil is dependent on the presence of organic matter, carbonate minerals clay
minerals, hydrous iron and manganese oxides, the pH, and other physical and chemical parameters The neatest
potential for leaching is seen in sandy soils with low pH. If released into the atmosphere, copper is expected to exist as a
dust paniculate or adsorb to particulate matter. Studies have shown that copper can remain in the atmosphere up to 30
days and be carried long distances.
C-19
-------
APPENDIX C
CHEMICAL SUMMARY FOR ETHYLENEDIAMINE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene diamine are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE DIAMINE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air s 1)
Koc
Log Row
Vapor Pressure
Reactivity.
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Hcnty's Law Constant
Fish Bioconcentration Factor
Odor Threshold
107-15-3
1,2-diamineethane; 1,2-ethanediamine
H2NCH2CH2NH2
colorless, clear, thick, liquid
60.10
8.5 °C
116-1 17 °C
lxl03g/l@25°C
d25'4, 0.898
no data
2 (calculated)
-2.04
12.0mmHg@25°C
volatile w/ steam; absorbs CO2 from air
flammable
1 10 °F (43 °C), closed cup
pKa,= 9.92; pKa2 = 6.86
no data
no data
1.73x10"' atm mVmole @ 25 °C
0.02 (calculated)
100% recognizable @11.2 ppm
Howard and Neal( 1992)
Budavarietal. (1996)
Budavarietal. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Riddicketal. (1986)
Budavari et al. (1996)
Lyman et al. (1990)
Hanschetal. (1995)
Boubliketal. (1984)
Budavari et al. (1996)
Aldrich(1997)
Budavari et al. (1996)
Perrin (1972)
Hine and Mookerjee (1975)
Lyman et al. (1990)
Verschueren (1996)
II. ENVIRONMENTAL FATE
A Aquatic Fate
The dominant environmental fate process for ethylenediamine in surface water is expected to be biodegradation. A
number of biological screening studies have determined that ethylenediamine biodegrades readily under aerobic
conditions (Price et al., 1974; Takemoto et al., 1981; Fitter, 1976 ; Mills and Stack, 1955, as cited in HSDB, 1998).
No data were available for the biodegradation of ethylenediamine under anaerobic conditions. An estimated Koc value
of 2, determined from an experimental log Kow of-2.04 (Hansch et al., 1995) and a regression-derived equation (Lyman
et al., 1990), indicates that ethylenediamine is not expected to adsorb to suspended solids and sediment in water. In
general, organic ions are not expected to volatilize from water or adsorb to particulate matter in water to the degree that
C-20
-------
APPENDIX C
would be predicted for their neutral counterparts. Based on an estimated BCF of 0.02 (Lyman et al., 1990) calculated
from the log Kow, a classification scheme (Franke et al., 1994) suggests the potential for bioconcentration in aquatic
organisms is low. Ethylenediamine is not expected to volatilize from water surfaces (Lyman et al., 1990) based upon an
experimental Henry's Law constant of 1.73xlO'9 atm-mVmole (Hine and Mookerjee, 1975). However, volatilization of
ethylenediamine will be pH dependent and attenuated if it is protonated; very little, about 1%, will be available for
volatilization. Hydrolysis of ethylenediamine is not expected to occur due to the lack of hydrolyzable functional groups
(Lyman etal., 1990). .
B.
Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), ethylenediamine, which has a vapor pressure of 12 mm Hg at 25 °C (Boublik et al, 1984), should exist solely as
a gas in the ambient atmosphere. Gas-phase ethylenediamine is degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 6 hours (Meylan and
Howard, 1993). Due to its miscibility with water, ethylenediamine may also be removed physically from the atmosphere
by wet deposition., Ethylenediamine is not expected to directly photolyze in the atmosphere due to the lack of absorption
in the environmental UV spectrum (>290 nm) (Lyman et al., 1990).
C.
Terrestrial Fate
The major environmental fate process for ethylenediamine in aerobic soils is expected to be biodegradation. A number
of biological screening studies have determined that ethylenediamine biodegrades readily under aerobic conditions
(Price et al., 1974; Takemoto et al., 1981; Fitter, 1976 ; Mills and Stack, 1955, as cited in HSDB, 1998). No data on
the biodegradation of ethylenediamine under anaerobic conditions were located in the available literature. An estimated
Koc value of 2 (Lyman et al., 1990), determined from an experimental log Kow of-2.04 (Hansch et al., 1995), indicates
that ethylenediamine is expected to have very high mobility in soil (Swann et al., 1983). Volatilization of
ethylenediamine from moist soil surfaces is not expected to be important (Lyman et al., 1990) given an experimental
Henry's Law constant of 1.73xlO'9 atm-mVmole (Hine and Mookerjee, 1975), although it may volatilize from dry soil
surfaces based upon a vapor pressure of 12 mm Hg (Boublik et al., 1984). However, at environmental pH's of 5-7,
ethylenediamine will most likely be a salt and volatilization will be attenuated. :
D.
Summary
The dominant removal mechanisms of ethylenediamine from the environment are expected to be biodegradation on the
earth's surface and reaction with photochemically-produced hydroxyl radicals in the atmosphere. In both soil and water,
biodegradation is expected to be rapid; a large number of biological screening studies have determined that
ethylenediamine biodegrades readily under aerobic conditions. If released to air, a vapor pressure of 12 mm Hg
indicates ethylenediamine should exist solely as a gas in the ambient atmosphere. Gas-phase ethylenediamine will be
degraded in the atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction
in air is estimated to be 6 hours. Physical removal of gas-phase ethylenediamine from the atmosphere may also occur
via wet deposition processes based on the miscibility of this compound with water. With a pKa, of 9.92,
ethylenediamine and its conjugate acid will.exist in environmental media in varying proportions that are pH dependent.
If released to soil, ethylenediamine may display very high mobility based upon an estimated Koc of 2. If released to soil
in high concentrations, such as those encountered in a spill, ethylenediamine may travel through soil and reach
groundwater. Volatilization of ethylenediamine from water and moist soil surfaces is not expected to be an important
fate process based upon a Henry's Law constant of 1.73xlO"9 atm-mVmole, although its vapor pressure indicates that
volatilization from dry soil surfaces may occur. However, at environmental pH's,of 5-7, ethylenediamine will most
likely be a salt and volatilization will be attenuated. In water, ethylenediamine is not expected to bioconcentrate in fish
and aquatic organisms based on an estimated BCF of 0.02.
C-21
-------
APPENDIX C
CHEMICAL SUMMARY FOR ETHYLENE GLYCOL
*
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene glycol are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE GLYCOL
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flam inability
Flash Point
Dissociation Constant
Molecular Difiusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
Data
107-21-1
1,2-ethanedioI
C2H602
HOCH2CH2OH
slightly viscous liquid •
62.07
-13 °C
197.6 °C
miscible (1,000 g/1)
l.llg/cm3
2.1
4 (estimated)
-1.36
0.092 mm Hg
no data
combustible
240°F(115°C)
15.1
no data
no data
e.OxlO"8 atm mVmol
10
25 ppm
Reference
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. I;i996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari etal. (1996)
Budavari etal. (1996)
Budavari etal. (1996)
' Riddick etal (1986)
Budavari etal. (1996)
Verschueren(1996)
SRC (1998)
Hansch etal. (1995), as cited in HSDB
(1998)
Daubert and Danner (1989)
no data
" Lewis (1993)
Budavari et al. (1996)
Howard and Meylan (1997)
Howard and Meylan (1997)
HSDB (1998)
ECDIN (1998)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
The dominant environmental fate process for ethylene glycol in water is expected to be biodegradation. A large number
of biological screening studies have determined that ethylene glycol biodegrades readily under both aerobic and
anaerobic conditions (Bridie et al. 1979; Fitter 1976; and Price et al. 1974, as cited in HSDB, 1998). Aerobic
degradation is essentially complete in
-------
APPENDIX C
and a regression-derived equation (Lyman et al., 1990, as cited in HSDB, 1998). Volatilization from the water column
to the atmosphere is not expected to occur (Lyman et al., 1990, as cited in HSDB, 1998) based on a Henry's Law
constant of 6.0xlO'8 atm-mVmole (Butler and Ramchandani 1935, as cited in Howard and Meylan 1997) . Ethylene
glycol is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the environment (Lyman et
al., 1990, as cited in HSDB, 1998). According to a classification scheme (Franke et al., 1994), a BCF of 10 in golden
ide fish (Freitag et al., 1985, as cited in HSDB, 1998) suggests that the potential for bioconcentration in aquatic
organisms is low. ,
B. . Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman 1989,
as cited in HSDB, 1998), ethylene glycol, which has a vapor pressure of 0.092 mm Hg at 25°C (Daubert and Danner
1989), should exist solely as a gas in the ambient atmosphere. Nonetheless, ethylene glycol has been detected adsorbed
onto atmospheric paniculate material (Abdelghani et al., 1990, as cited in HSDB, 1998); the small amount of ethylene
glycol deposited onto particulate material may be physically removed by wet and dry deposition. The predominant
removal process of ethylene glycol from the atmosphere is reaction with photochemically-produced hydroxyl radicals;
the half-life for this reaction in air is estimated to be 50 hours (Atkinson 1989, as cited in HSDB, 1998). Ethylene
glycol may undergo some degradation by direct photolysis; 12.1% of applied ethylene glycol was degraded after 17
hours following irradiation by light > 290 nm (Freitag et al., 1985, as cited in HSDB, 1998).
C.
Terrestrial Fate
The major environmental fate process for ethylene glycol in soil is expected to be biodegradation. A large number -of
biological screening studies have determined that ethylene glycol biodegrades readily under both aerobic and anaerobic
conditions; complete biodegradation was shown in one soil within 2 days and 97% biodegradation in 12 days was
reported for a second soil (McGahey and Bower 1992, as cited in HSDB, 1998). Based on a classification scheme
(Swann et al., 1983, as cited in HSDB, 1998), an estimated Koc of 4, determined from a log Kow of-1.36 (Hansch et
al., 1995, as cited in HSDB, 1998) and a regression-derived equation (Lyman 1990 et al., as cited in HSDB, 1998),
indicates that ethylene glycol is expected to have very high mobility in soil. Percent adsorption to 4 soils (2 clay and 2
sandy clay soils) ranged from 0-0.5% (Abdelghani et al 1990, as cited in HSDB, 1998). Volatilization of ethylene
glycol from moist soil surfaces is not expected to be important (Lyman et al. 1990, as cited in HSDB, 1998) given a
Henry's Law constant of 6.0xlO'8 atm-mVmole (Butler and Ramchandani 1935, as cited in Howard and Meylan 1997).
Ethylene glycol may volatilize from dry soil given its vapor pressure of 0.092 mm Hg (Daubert and Danner 1989); this
may be attenuated by hydrogen bonding to soil materials (SRC 1998).
D.
Summary
If released to air, a vapor pressure of 0.092 mm Hg at 25 °C indicates that ethylene glycol should exist solely as a gas in
the ambient atmosphere; however, experimental results show that at least some ethylene glycol is associated with
atmospheric particulates. Gas-phase ethylene glycol will be degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 50 hours. Adsorption
studies indicate that ethylene glycol is not expected to adsorb to suspended solids and sediments in water. A BCF of 10
in golden ide fish suggests the potential for bioconcentration in aquatic organisms is low. If released to soil, ethylene
glycol is expected to have very high mobility based upon an estimated Koc of 4, and, therefore, it has the potential to
leach to groundwater. Volatilization from water and from moist soil surfaces is not expected to be an important fate
process based upon a Henry's Law constant of 6.0xlO:8 atm-mVmole. Volatilization from dry soil surfaces may occur
based upon the vapor pressure of this compound, although this may be attenuated by hydrogen bonding to soil materials.
Biodegradation is expected, to be rapid and may be the dominant fate process in both soil and water-under non-spill
conditions; a large number of biological screening studies have determined that ethylene glycol biodegrades readily
under both aerobic and anaerobic conditions.
C-23
-------
APPENDIX C
CHEMICAL SUMMARY FOR ETHYLENE GLYCOL MONOBUTYL ETHER
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for ethylene glycol monoburyl ether.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene glycol monobutyl ether are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE GLYCOL
MONOBUTYL ETHER
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
Data
111-76-2
BUGS, butoxyethanol, Dowanol EB
C6H14O2
CH3(CH2)30CH2CH20H
Clear, colorless liquid
118.18
-70°C
171°C, 743 mm Hg
>1000 g/L, 25°C
-------
APPENDIX C
CHEMICAL SUMMARY FOR FLUOROBORIC ACID (fluoride)
This chemical was identified by one or more suppliers as a bath ingredient for the electroless copper and tin-palladium
processes. This summary is based on information retrieved from a systematic search limited to secondary sources. The
only exception is summaries of studies from unpublished TSCA submissions that may have been included. These
sources include online databases, unpublished EPA information, government publications, review documents, and
standard reference materials. No attempt has been made to verify information in these databases and secondary sources.
Very little information on the environmental fate and toxicity of fluoroboric acid or fluoroborates was found in the
available secondary sources. Supplemental information is provided for fluoride which may be a degradation product and
for sodium bifluoride.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of fluoroboric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF FLUOROBORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Koc
LogKow
Vapor Pressure
Vapor Density
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK.)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Difiusivify Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
16872-11-0
hydrogen tetrafluoroborate
fluoboric acid
hydrofluoroboric acid
HBF,
B-F4-H
colorless liquid
87.82
-90°C
130°C (decomposes)
miscible;
sol. in hot water
~1.84g/mL
NA
NA
5.1mmHgat20°C
3.0
strong acid; corrosive
NA
NA
-4.9
NA
NA
NA
NA
NA
Na
Reference
HSDB(1995)
HSDB(1995)
HSDB (1995)
•^ Fisher Scientific (1993)
HSDB (1995)
HSDB (1995)
Fisher Scientific (1993)
HSDB (1995)
HSDB (1995)
HSDB (1995)
Fisher Scientific (1993) -
Fisher Scientific ( 1 993)
HSDB (1995)
HSDB (1995)
The chemical identity and physical/chemical properties of sodium tetrafluoroborate are summarized in Table 2.
C-25
-------
APPENDIX C
TABLE 2. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM
TETRAFLUOROBORATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
K«c
Log Row
Vapor Pressure
Reactivity
Flammability
Hash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Difiusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconccntration Factor
Odor Threshold
Conversion Factors
Data
013755-29-8
sodium fluoroborate
STB
sodium borfluoride
sodium boron tetrafluoride
NaNF4
Na-F«-B
white crystalline powder
109.82
384°C
108g/100mLat26°C
210 g/100 mL at 100 °C
2.470 -
NA
NA
NA
reacts with strong oxidizing
agents; sensitive to moisture
noncombustible
NA
NA
NA
NA
NA
NA
Reference
Lockheed Martin (1994)
Lockheed Martin (1994)
Sigma-Aldrich (1992)
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Sigma-Aldrich (1992)
Sjgma-Aldrich (1992)
Lockheed Martin (1994)
The chemical identity and physical/chemical properties of sodium fluoride are summarized in Table 3.
C-26
-------
APPENDIX C
TABLE 3. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM FLUORIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Data
7681-49-4
sodium hydrofluoride
sodium monfluoride
floridine
NaF
Na-F
crystals
42.00
993 °C
1704°C
4.0g/100mLatl5°C
4.3g/100mLat25°C
2.78
NA
NA
1 mm Hg at 1077°C
stable under normal conditions
nonflammable
Reference
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Budavari etal. (1989)
Budavarietal. (1989)
Budavari et al. (1989)
Keith and Walters (1985)
Keith and Walters (1985)
Keith and Walters (1985)
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffiisivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
The chemical identity and physical/chemical properties of sodium bifluoride are summarized in Table 4.
C-27
-------
APPENDIX C
TABLE 4. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM BIFLUORIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boil ing Point
Water Solubility
Density
Koc
LogKo*
Vapor Pressure
Vapor Density
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diflusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
1333-83-1
sodium hydrogen difluoride
sodium hydrogen fluoride
sodium acid fluoride
' NaHFj
F2-H-Na
white, crystalline powder
62.01
decomposes on heating
NA
soluble in cold and hot water
2.08
NA
NA
NA
NA
aqueous solution corrodes glass
slightly combustible
NA
NA
NA
NA
NA
NA
NA
NA
Reference
HSDB (1995)
HSDB (1995)
Lewis (1993)
HSDB (1995)
Budavarietal. (1989)
Budavari et al. (1989)
Lewis (1993)
Lide(1991)
Lewis (1993)
Budavarietal. (1989)
Lockheed Martin (1990)
H. ENVIRONMENTAL FATE
A. Environmental Release
Fluoroboric acid may be released into the environment in emissions and effluents from facilities involved in its
manufacture or use. It is used primarily in industrial metal plating solutions (60%), in the synthesis of diazo salts (20%),
and in metal finishing (20%) (HSDB 1995). It is used in bright dipping solutions for Sn-Pb alloys in printed circuits and
other electrical components (HSDB 1995).
B.
Transport
No information was found in the available secondary sources on the environmental transport of fluoroboric acid. Its
miscibility with water indicates that transport in aqueous systems is very likely.
C-28
-------
APPENDIX C
C. Transformation/Persistence
FLUOROBORIC ACID:
1.
2.
3.
4.
Air — No information was found in the available secondary sources on the transformation and persistence of
fluorofaoric acid or fluoroborates in the atmosphere.
Soil —No information was found in the available secondary sources on the transformation and persistence .of
fluoroboric acid or fluoroborates in soil. Fluoroboric acid may undergo limited hydrolysis in moist soils
(Budavari et al. 1989).
Water— Fluoroboric acid undergoes limited hydrolysis in water to form hydroxyfluoroborate ions, the major
product is BF3OH' (Budavari et al. 1989).
Biota — No information was found in the available secondary sources on the biotransformation or
bioconcentration of fluoroboric acid or fluoroborates. Rapid urinary excretion of tetrafluoroborates suggests
that these salts would not bioaccumulate.
FLUORIDES:
.1.
2.
3.
4.
Air— Gaseous inorganic fluorides undergo hydrolysis in the atmosphere; however, particulate forms are
relatively stable and do not hydrolyze readily (ATSDR 1993). '
Soil — Fluorides tend to persist in soils as fluorosilicate complexes under acidic conditions and as calcium
fluoride under alkaline conditions. Sandy acidic soils favor the formation of soluble forms (ATSDR 1993).
Water — In dilute solutions and at neutral pH, fluoride is generally present as dissolved fluoride ion. High
calcium carbonate levels may lead to precipitation as calcium fluoride (ATSDR 1993).
Biota— Fluorides have been shown to accumulate in some aquatic organisms (ATSDR 1993). Soluble forms
of fluoride are taken up by terrestrial plants and converted into fluoro-organic compounds (ATSDR 1993).
C-29
-------
APPENDIX C
CHEMICAL SUMMARY FOR HYDROCHLORIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of hydrochloric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF HYDROCHLORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air ** 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flam inability
Hash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Dillusivity Constant
Hcniy's Law Constant
Fish Bioconccntration Constant
Odor Threshold
Data
7647-01-0
muriatic acid
HC1
HC1
filming liquid
36.46
-25.4°C(39.17%soIn)
108.58 "Cat 760 mm Hg
479.1g/l(40%soln)
1.20 g/cm3 (39.11% soln)
1.639g/t
expected to be < 50
expected to be < 1
no data
toxic, corrosive fumes w/H2O or steam
non-combustible
no data
—3
no data
no data
no data
no data
no data
Reference
Budavari et al. (1996)
Budavarietal. (1996)
Budavari et al. (1996)
Budavari et ai. (1996)
Lewis (1993)
Lide (1995)
Budavari et al. (1996)
Budavari et al. (1996)
Weast (1985)
Budavari et al. (1996)
Austin and Glowacki (1989)
SRC (1998)
SRC (1998)
Sax (1984)
Lewis (1993)
Bodeketal.(1988)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If hydrochloric acid is released into the water column at low concentrations, a pKa of- -3.00 (Bodek et al., 1988)
indicates it will dissociate completely into chloride (Cl')"and hydrogen (H+) ions. The amount of gaseous hydrochloric
acid dissolved in water is affected by the pH of the solution. A higher pH allows more aqueous hydrochloric acid to
dissociate, thereby increasing the solubility of hydrochloric acid gas (Bodek et al., 1988). As a result, dilute solutions of
hydrochloric acid are not expected to volatilize from water surfaces or to bioconcentrate in aquatic organisms. Chloride
ions generally do not react with many species in water and are harmless at relatively low concentrations (Manahan,
1991). Hydrochloric acid will protonate amines and other electron pair donators present in natural waters, forming salts;
C-30
-------
APPENDIX C
this will be dependent upon pH. Large releases of the concentrated acid into water, such as may result from,a spill, will
result in a lowering of the pH (Bodek et al., 1988).
B. Atmospheric Fate
If hydrochloric acid is released to the atmosphere, its vapor pressure indicates it will exist as a vapor in the ambient
atmosphere. Wet deposition of hydrochloric acid in rain, snow, or fog is expected to be the dominant fate process in the
atmosphere based upon its high water solubility (Arimoto, 1989).
C.
Terrestrial Fate
If hydrochloric acid is released to soil, it will dissociate into chloride and hydrogen ions in moist soils. Hydrochloric
acid will protonate amines and other electron pair donators present in soils, forming salts; this will be dependent upon
pH. The chloride ion is extremely mobile in soils and almost no soil retention occurs (Bodek et al., 1988). Chloride is
typically the predominant ion in saline soils and the second most abundant anion in sodic soils; thus, it is readily
available for the formation of metal complexes in soil (Bodek et al., 1988; SRC, 1998).
D. Summary
If released into water, hydrochloric acid will dissociate into chloride (Cl~) and hydrogen (H+) ions. Therefore,
hydrochloric acid is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in
aquatic organisms, nor volatilize from water surfaces. Chloride ions generally do not react with many species in water
and are harmless at relatively low concentrations. Hydrochloric acid will protonate amines and other electron pair
donators present in natural waters and soils, forming salts; this will be dependent upon pH. If released to soil,
hydrochloric acid is expected to dissociate into its component ions in moist soils. Because the chloride ion is'extremely
mobile in soils, almost no soil retention occurs. Chloride is typically the predominant ion in saline soils and the second
most abundant anion in sodic soils; thus, it is readily available for the formation of metal complexes in soil.
Volatilization of hydrochloric acid from soil surfaces is not expected to occur. If released to the atmosphere,
hydrochloric acid is expected to exist as a gas. Hydrochloric acid is expected to be physically removed from the
atmosphere by wet deposition based upon its high water solubility.
C-31
-------
APPENDIX C
CHEMICAL SUMMARY FOR HYDROGEN PEROXIDE
This chemical was identified by one or more suppliers as a bath ingredient for the electroless copper, non-formaldehyde
electroless copper, and tin-palladium processes. This summary is based on information retrieved from a systematic
search limited to secondary sources (see Attachment C-l). These sources include online databases, unpublished EPA
information, government publications, review documents, and standard reference materials. No attempt has been made
to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of hydrogen peroxide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF HYDROGEN PEROXIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
K« •
LogKcw
Vapor Pressure
Reactivity
Data
7722-84-1
hydrogen dioxide; hydroperoxide; albone; hioxyl
HA
HA -
colorless, unstable liquid, bitter taste
34.02
-0.43 °C
152°C
miscible
1.463 @0°C
no data
no data
no data
1.97 mm Hg @ 25° C (measured)
strong oxidizer; may decompose violently if traces of
impurities are present
Reference
• Budavari et al. 1989
Budavari et al. 1989
IARC 1985
Budavari etal. 1989
Budavari et al. 1989
Budavari etal. 1989
• Budavari et al. 1989
Budavari etal. 1989
Budavari et al. 1989
CHEMFATE 1995
Budavari etal. 1989
i A r>rf-» moc
Fhmmability
Flash Point
Dissociation Constant
Henry's Law Constant
Molecular Diffiisivity Coefficient
Air Diffiisivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
molecular additions, substitutions, oxidations, reduction;
can form free radicals
not flammable, but can cause spontaneous combustion of'
flammable materials
no data
no data
no data
no data
no data •
no data
odorless
1 ppm = 1.39 mg/m3
1 mg/m3 = 0.72 ppm
30%solnl.lkg/L
anhydrous 1.46 kg/L
IARC 1985
HSDB 1995
Budavari et al. 1989
IARC 1985
Budavari et al. 1989
C-32
-------
APPENDIX C
H. ENVIRONMENTAL FATE
A. Environmental Release
No information was found in the secondary sources searched regarding the environmental release of hydrogen peroxide.
Solutions of hydrogen peroxide gradually deteriorate (Budavari et al., 1989). Hydrogen peroxide is a naturally occurring
substance. Gaseous hydrogen peroxide is recognized to be a key component and product of the earth's lower
atmospheric photochemical reactions, in both clean and polluted atmospheres. Atmospheric hydrogen peroxide is also
believed to be generated by gas-phase photochemical reactions in the remote troposphere (IARC, 1985)
B.
Transport
No information was found in the secondary sources searched regarding the transport of hydrogen peroxide.
C. Transformation/Persistence
1.
2.
3.
4.
Ajr — Hydrogen peroxide may be removed from the atmosphere by photolysis giving rise to hydroxyl radicals,
by reaction with hydroxyl radicals, or by heterogenous loss processes such as rain-out (IARC, 1985). •
Soil — No information was found in the secondary sources searched regarding the transformation or
persistence of hydrogen peroxide in soil, however, .solutions of hydrogen peroxide gradually deteriorate
(Budavari et al., 1989).
Water — Hydrogen peroxide is a naturally occurring substance. Surface water concentrations of hydrogen
peroxide have been found to vary between 51-231 mg/L, increasing both with exposure to sunlight and the
presence of dissolved organic matter (IARC, 1985).
Biota — Hydrogen peroxide is a naturally occurring substance. Endogenous hydrogen peroxide has been
found in plant tissues at the following levels (mg/kg frozen weight): potato tubers, 7.6; green tomatoes, 3.5; red
tomatoes, 3.5; and castor beans in water, 4.7 (IARC, 1985).
C-33
-------
APPENDIX C
CHEMICAL SUMMARY FOR LEAD
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these, databases and secondary sources. The
search identified sources of information for Lead.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of Lead are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF LEAD
Characteristic/Property Data Reference . _^
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Ham inability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
7439-92-1
Pb
N/A
Metal
207.2
327.4°C
1740°C
Insoluble
10.65
no data
no data
no data
1.77mmHg@1000°C
Flammable solid
no data
no data
no data
no data
no data
no data
no data
no data
Howard and Neal( 1992)
Howard and Neal (1992)
Weast(1983)
Weast(1983)
Weast(1983)
Weast(1983)
Weast(1983)
Budavari et al. (1996)
Budavarietal. (1996)
C-34
-------
APPENDIX C
CHEMICAL SUMMARY FOR MALEIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of maleic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY.AND CHEMICAL/PHYSICAL PROPERTIES OF MALEIC ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Roc
Log fCow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Difiusivity Constant
Air DiSusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
1 10-16-7
(Z)-butenedioic acid; toxilic acid
cis-l,2-ethylenedicarboxylic acid
maleinicacid
HOOCCH=CHCOOH
white crystals
1 16.07
130.5°C
no data
441g/lat25°C
1.59g/cm'at20°C
no data
16 (estimated)
-0.34
3.06xlO-smmHgat25°C
stable
combustible
not pertinent
pK, = 1.83;pK2 = 6.07
no data .
no data
no data; estimated to be < IxlO"8 atm mVmol
10-11
no data
Lide (1995)
Budavarietal. (1996)
Budavari et al. (1996)
•Lewis (1993)
Budavari et al. (1996)
Aldrtch (1996)
Budavari etal. (1996)
Budavarietal. (1996)
Lide(1995)
PHYSPROP (1998)
Lide (1995)
Lymanetal. (1990)
Hanschetal. (1995)
Daubert and Danner (1991)
Weiss (1986) .
Lewis (1993)
Weiss (1986)
Howard (1989)
Estimated
HSDB (1998)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If released into water, maleic acid is not expected to adsorb to suspended solids or sediments .in water based upon an
estimated Koc of 16 (Swann et al., 1983), determined from a log Kow of-0.34 (Hansch et al., 1995) and a regression-
derived equation (Lyman et al., 1990). Volatilization from the water column to the atmosphere is not expected to occur
(Lyman et al., 1990) based on an estimated Henry's Law constant of <10"8 atm-mVmole . Maleic acid is expected to be
stable to hydrolysis in the pH range of 5-9 typically encountered in the environment (Lyman et al., 1990). According to
C-35
-------
APPENDIX C
a classification scheme (Franke et al., 1994), a BCF of 10 in golden ide fish (Freitag, 1985, as cited in HSDB, 1998)
suggests that the potential for bioconcentration in aquatic organisms is low. Maleic acid was determined to be readily
degraded in biodegradation screening tests; however, no biodegradation studies were available in environmental waters
(Howard, 1989).
B.
Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), maleic acid, which has a vapor pressure of 3.06xlO~5 mm Hg at 25 °C (Daubert and Danner, 1991), is expected
to exist as both a particulate and vapor in the ambient atmosphere. Because maleic acid has pKa's of 1.83 and 6.07
(Howard, 1989), it is" expected to exist in the dissociated form in the environment and form salts with cations (HSDB,
1998). Removal of maleic acid from the atmosphere by reaction with photochemically-produced hydroxyi radicals
results in an estimated half-life of 2 days (Meylan and Howard, 1993). The reaction of maleic acid with ozone in the
atmosphere results in a gas-phase half-life ranging from 7-13 days (Meylan and Howard, 1993). Maleic acid may
undergo some degradation by direct photolysis; 17% of applied maleic acid was degraded after 17 hours following
irradiation by light > 290 nm (Freitag et al., 1985, cited in HSDB, 1998). Wet deposition of maleic in rain, snow, or fog
is expected to be an important transport process in the atmosphere based upon its high water solubility (Arimoto, 1989).
C.
Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 16, determined from a log Kow of-0.34
(Hansch et al., 1995) and a regression-derived equation (Lyman et al., 1990), indicates that maleic acid is expected to
have very high mobility in soil. Volatilization of maleic acid from moist soil surfaces is not expected to be important
(Lyman et al., 1990) given an estimated Henry's Law constant of <10'8 atm-mVmole. In addition, maleic acid is not
expected to volatilize from dry soil given its vapor pressure of 3.06x10"5 mm Hg (Daubert and Danner, 1991). While
maleic acid is readily biodegradable in screening studies, no degradation data were available for soil systems (Howard,
1989).
. D.
Summary
If released to air, a vapor pressure of 3.06xlO"s mm Hg at 25 °C indicates that maleic acid should exist as both a gas
and particulate in the ambient atmosphere. Gas-phase maleic acid will be degraded in the atmosphere by reaction with
photochemically-produced hydroxyi radicals; the half-life for this reaction in air is estimated to be 2 hours. The reaction
of maleic acid with ozone in the atmosphere results in a gas-phase half-life ranging from 7-13 days. Wet deposition of
maleic acid from the atmosphere is expected to be an important transport process. Screening studies suggest that direct
photolysis if maleic acid may occur. A BCF of 10 in golden ide fish suggests the potential for bioconcentration in
aquatic organisms is low. If released to soil, maleic acid is expected to have very high mobility based upon an estimated
Koc of 16, and, therefore, it has the potential to leach to groundwater. Volatilization from water and from moist soil
surfaces is not expected to be an important fate process based upon an estimated Henry's Law constant of <10"8 atm-
mVmoIe. Volatilization from dry soil surfaces is not expected to occur based upon the vapor pressure of this compound.
Maleic acid was determined to be readily biodegraded in screening studies, although no data were available for
biodegradation in water or soil.
C-36
-------
APPENDIX C
CHEMICAL SUMMARY FOR MALIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of malic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF MALIC ACID
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
6915-15-7
hydroxysuccinic acid; apple acid
COOHCH2CH(OH)COOH
colorless crystals
134.09
100 °C
140 °C, decomposes
592g/lat25°C
1.6g/cm3
no data
5 (estimated)
-1.26
3.28x10-" mm Hg at 25 °C
no data
combustible
no data
3.40 '
. no data
no data
no data; expected to be < lO"8 atm mVmol
' no data; expected to be <1
no data
Lewis (1993)
Lewis (1993)
Budavari et al. (1996)
Lewis (1993)
Lewis (1993)
Budavari etal. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
PHYSPROP (1998)
Lewis (1993)
Lymanetal. (1990)
Hansch et al. (1995)
Yaws (1994)
Lewis (1993)
PHYSPROP (1998)
Estimated
Estimated
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, malic acid is not expected to adsorb to suspended solids and sediments in water based
upon an estimated Koc of 5 (Swann et al., 1983), determined from a log Kow of-1.26 (Hansch et al., 1995) and a
regression-derived equation (Lyman et al., 1990). Volatilization from the water column to the atmosphere is not
expected to occur (Lyman et al., 1990) based on an estimated Henry's Law constant of <1Q-8 atm-rrrVmole . Malic acid
is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the environment (Lyman et al.,
1990). According to a classification scheme (Franke et al., 1994), an estimated BCF of <1 suggests that the potential
for bioconcentration in aquatic organisms is low and not an important fate process. Results from a number of biological
C-37
-------
APPENDIX C
screening tests have shown that malic acid biodegrades relatively fast (Fischer et al., 1974; Malaney and Gerhold, 1969;
Heukelekian and Rand, 1955; as cited in HSDB, 1998).
B. Atmospheric Fate
Accordingto amodel of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), malic acid, which has a vapor pressure of 3.28xlO'8 mm Hg at 25 °C (Yaws, 1994), should exist almost entirely
as a particulate in the ambient atmosphere. Removal of malic acid from the atmosphere by reaction with
photochemically-produced hydroxyl radicals results in an estimated half-life of 2 days,(Meylan and Howard, 1993). Wet
deposition of malic acid in rain, snow, or fog is expected to be the dominant transport process in the atmosphere base4
upon its high water solubility (Arimoto, 1989). Because carboxylic acids are generally resistant to hydrolysis, malic acid,
is not expected to hydrolyze in environmental media (Lyman et al., 1990).
C. Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 5, determined from a log Kow of-1.26
(Hansch et al., 1995) and a regression-derived equation (Lyman et al., 1990), indicates that malic acid is expected to
have very high mobility in soil and may leach to groundwater. Volatilization of malic acid from moist soil surfaces is not
expected to be important (Lyman et al., 1990) given an estimated Henry's Law constant of <10'8 atm-nvVmole. In
addition, malic acid is not expected to volatilize from dry soil given its vapor pressure of 3.28xlO-s mm Hg (Yaws,
1994). Biodegradation screening studies reveal that malic acid biodegrades relatively fast (Fischer et al., 1974; Malaney
and Gerhold, 1969; Heukelekian and Rand, 1955; as cited in HSDB, 1998).
D. Summary
If released to air, a vapor pressure of 3.28xlQ-8 mm Hg at 25 °C indicates that malic acid should exist as a particulate in
the ambient atmosphere. Removal of malic acid from the atmosphere by reaction with photochemically-produced
hydroxyl radicals results in an estimated half-life of 2 days. Wet deposition is expected to be the dominant transport
process of malic acid from the atmosphere. An estimated BCF of <1 suggests the potential for bioconcentration in
aquatic organisms is low. If released to soil, malic acid is expected to have very high mobility based upon an estimated
Koc of 5, and, therefore, it has the potential to leach to groundwater. Volatilization from water and from moist soil
surfaces is not expected to be an important fate process based upon an estimated Henry's Law constant of <10^ atm-
mVmole, also volatilization from dry soil surfaces is not expected to occur based upon the vapor pressure of this-
compound. Hydrolysis of malic acid in environmental media is not expected to occur. Malic acid was determined to be
readily biodegraded in screening studies, although no data were available for biodegradation in water or soil.
C-38
-------
APPENDIX C
CHEMICAL SUMMARY FOR METHANESULFONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of methanesulfonic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF METHANESULFONIC
. ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diflusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold.
75-75-2
methylsulfonic acid
CH,OjS
CH3SO2OH
solid
liquid at room temperature
96.11
20 °C
200 °C; 167 °C at 10 mm Hg
1.0xl03g/Lat20 °C
1.48 g/cm3
no data
1 (estimated)
no data; estimated to be < 1
4.28xlOJ1mmHgat250C
thermally stable at mod. elevated temps
no data
112 °C
-1.86
no data
no data
1.3xlO-8 atm mVmol (estimated)
3 (estimated)
no data
Lide (1995)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Lewis (1993)
Budavari et al. (1996)
Lide (1995)
Lewis (1993); Lide (1995)
PHYSPROP (1998)
Lide (1995)
HSDB (1998)
Estimated •
Daubert and Danner (1991)
Budavari etal. (1996)
ECD1N(1998)
Serjeant and Dempsey (1979)
Meylan and Howard (1991)
Meylan etal. (1997)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, methanesulfonic acid is not expected to adsorb to suspended solids and sediments in
water based upon an estimated Koc of 1 (Swann et al., 1983), determined from a structure fragment estimation method
(Meylan et al., 1992). Volatilization from the water column to the atmosphere is not expected to occur (Lyman et al.,
1990) based on an estimated Henry's Law constant of l.SxlO'8 atm-mVmole (Meylan and Howard, 1991; SRC, 1998).
Methanesulfonic acid is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the
C-39
-------
APPENDIX C
environment (Lyman et al., 1990) According to a classification scheme (Franke et al., 1994), an estimated BCF of 3
(Meylan et al., 1997) suggests that the potential for bioconcentration in aquatic organisms is low. It was determined that
many bacterial types can degrade methanesulfonic acid through diverse routes and at different rates, although specifics
were not given (Baker et al., 1991, as cited in HSDB, 1998). Because methanesulfonic acid has pKa of-1.86 (Serjeant
and Dempsey, 1979),' it is expected to exist in the dissociated form in the environment.
B.
Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), methanesulfonic acid, which has a vapor pressure of 4.28x10-" mm Hg at 25 °C (Daubert and Banner, 1991),
has the potential to exist as both a vapor and particulate in the ambient atmosphere. Because methanesulfonic acid has
pKa of-1.86 (Serjeant and Dempsey, 1979), it is expected to exist in the dissociated form in the environment. Removal
of methanesulfonic acid from the atmosphere by reaction with photochemically- produced hydroxyl radicals results in an
estimated half-life of 58 days (Meylan and Howard, 1993). In the atmosphere, methanesulfonic acid is concentrated in
the smaller size particles, 0.25-2 um in diameter (Kolaitis et al., 1989, as cited in HSDB, 1998). Removal of particulate
methanesulfonic acid from the atmosphere can occur through wet and dry deposition (HSDB, 1998).
C.
Terrestrial Fate
Based on a classification scheme (Swanri et al., 1983), an estimated Koc of 1, determined from a structure fragment
estimation method (Meylan et al., 1992), indicates that methanesulfonic acid is expected to have very high mobility in
soil. Volatilization of methanesulfonic acid from moist soil surfaces is not expected to be important (Lyman et al., 1990)
given an estimated Henry's Law constant of 1.3xlQ-8 atm-nvVmole (Meylan and Howard, 1991; SRC, 1998). In
addition, methanesulfonic acid is not expected to volatilize from dry soil given its vapor pressure of 4.28X10"4 mm Hg
(Daubert and Danner, 1991). It was determined that many bacterial types can degrade methanesulfonic acid through
diverse routes and at different rates, although specifics were not given (Baker et al., 1991,.as cited in HSDB, 1998).
D.
Summary
If released to air, a vapor pressure of 4.28 x 10^ mm Hg at 25 °C indicates that methanesulfonic acid has the potential
to exist as both a vapor and particulate in the ambient atmosphere. Gas-phase methanesulfonic acid will be degraded in
the atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air is
estimated to be 58 hours. Removal of particulate methanesulfonic acid from the atmosphere can occur through wet and
dry deposition. An estimated BCF of 3 suggests the potential for bioconcentration in aquatic organisms is low. If
released to soil, methanesulfonic acid is expected to have very high mobility based upon an estimated Koc of 1, and,
therefore, it has the potential to leach to groundwater. Volatilization from water and from moist soil surfaces is not
expected to be an important fate process based upon an estimated Henry's Law constant of 1.3x10'8 atm-mVmole.
Hydrolysis of methanesulfonic acid is not expected to occur. Volatilization from dry soil surfaces is not expected to
occur based upon the vapor pressure of this compound. Methanesulfonic acid was determined to be biodegraded by
many bacterial types, although specifics were not given.
C-40
-------
APPENDIX C
CHEMICAL SUMMARY FOR NICKEL SULFATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for nickel and soluble salts of nickel, including nickel sulfate and nickel sulfate
hexahydrate. •
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of nickel sulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF NICKEL SULFATE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Difiusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7786-81-4
sulfiiric acid, nickel (2+) salt
NiO«S
NiSQ,
green-yellow orthorhombic crystals
154.757
840 °C, decomposes
no data
293g/LatO°C
4.01 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 C
no data
not flammable
no data; expected to be > 350 °C
no data
no data
no data
no data; expected to be < IxlO"8
no data
no data
Lide(1995)
Howard and Neal (1992)
Budavarietal. (1996)
Lide (1995)
Lide(1995)
Lide (1995).
Lide (1995)
Dean (1985)
Lide (1995)
SRC (1998)
SRC (1998)
Estimated
'Prager(1995)
SRC (1998)
SRC (1998)
H. ENVIRONMENTAL FATE
A. . Aquatic Fate
If released into water, nickel sulfate is expected to dissociate into nickel (Ni2+) and sulfate [(SO4)2"] ions. The
dissociation of nickel sulfate into its component ions indicates that the compound nickel sulfate is not expected to
volatilize from water surfaces. In aqueous solutions, nickel exists as the hexaquonickel ion, [Ni(H2O)62+]; this ion is
poorly absorbed by most living organisms (Sunderman and Oskarsson, 1991). In natural waters, nickel exists both in
the ionic"form and as stable organic complexes (Sunderman and Oskarsson, 1991).. Nickel compounds are generally
C-41
-------
APPENDIX C
soluble at pH values less than 6.5, but at pH values greater than 6:7 nickel exists predominantly as insoluble nickel
hydroxides (Sunderman and Oskarsson, 1991). Shellfish and Crustacea generally contain higher concentrations of nickel
in their flesh than do other species offish (Sunderman and Oskarsson, 1991).
B.
Atmospheric Fate
If released to the atmosphere, nickel sulfate's high melting point (Lide, 1995) and low vapor pressure (SRC, 1998)
indicate that it will exist as a particulate (Bidleman, 1988). Wet and dry deposition of nickel sulfate is expected to be the
dominant fate process in the atmosphere (Arimoto, 1989). The rate of dry deposition will depend on the prevailing
winds and particle size (Bodek et al., 1988). Nickel sulfate's high water solubility (Dean, 1985) indicates that it is
expected to undergo wet deposition in rain, snow, or fog.
C. Terrestrial Fate
If nickel sulfate is released to soil, it is expected to dissociate into Ni2* and (SO4)2' ions in the presence of moisture. Iron
and manganese oxides, clay minerals, and organic matter may be important sorbents of nickel (Bodek et al., 1988) and
will retard its migration through soil. Complexing ligands, such as organic_acids, may reduce the sorption of nickel
(Bodek et al., 1988). Acid rain has a tendency to mobilize nickel from soil and increase leaching into groundwater due
to the high solubility of nickel compounds at pH values less than 6.5 (Sunderman and Oskarsson, 1991). The high
melting point, low vapor pressure, and low Henry's Law constant expected for an ionic salt indicate that nickel sulfate
will not volatilize from either moist or dry soil surfaces (Bodek et al., 1988).
D.
Summary
If released into water, nickel sulfate is expected to dissociate into nickel (Ni2t) and sulfate (SO4)2" ions. Therefore,
nickel sulfate is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic
organisms, or volatilize from water surfaces. In natural waters, nickel exists in both the ionic form and as stable organic
complexes; at pH values greater than 6.7 it exists as insoluble nickel hydroxides. In moist soils, nickel sulfate is
expected to dissociate into its component ions. Ionic nickel may be sorbed by iron and manganese oxides, clay minerals,
and organic matter; acid rain and complexing ligands may reduce the sorption of nickel. Volatilization of nickel sulfate
from soil surfaces is not expected to occur. If released to the atmosphere, nickel sulfate is expected to exist as a
particulate. Nickel sulfate is expected to be physically removed from the atmosphere by wet and dry deposition. The
rate of dry deposition will depend on particle size and prevailing wind patterns. •
C-42
-------
APPENDIX C
CHEMICAL SUMMARY FOR PALLADIUM CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
L CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of palladium chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PALLADIUM CHLORIDE
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammabilify
FlashPoint
Dissociation Constant
Molecular Diffiisivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
' This form of expressing solubility cannot be
7647-10-1
Palladous chloride
Palladium (II) chloride
CI2Pd
PdCl2
red rhombohedral crystals; hygroscopic
177.33
500fC (decomposes)
decomposed at high temperatures
soluble1
4.0 g/cm3
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be <10^ mm Hg
no data
no data
no data
expected to dissociate into Pd2* and Cl"
no data
no data
no data; expected to be <1X10-8
no data
no data
CAS (1998)
Budavarietal. (1996)
Lide (1995)
Budavarietal. (1996)
Budavari et al. (1996)
Lide (1995)
Lide(1995)
Lide (1995)
Budavari et al. (1996)
Dean (1985)
Lide (1995)
SRC (1998)
SRC (1998)
SRC (1998)
SRC (1998)
SRC (1998)
converted into g/L units
n. ENVIRONMENTAL FATE
A. Aquatic Fate
If palladium chloride is released into the water column, it is expected to dissociate into palladium (Pd2+) and chloride
(Cf) ions. The dissociation of palladium chloride into its component ions indicates that palladium chloride is not
expected to bioconcentrate in aquatic organisms or volatilize from water surfaces. Palladium ions may adsorb to
charged surfaces of suspended sediments and humic materials in the water column (Evans; 1989). The chloride ion may
complex with heavy metals in natural waters, thereby increasing their solubility (Bodek et al., 1988). Adsorption of the
C-43
-------
APPENDIX C
chloride ion to suspended solids and sediment in the water column is not expected to be an important fate process.
B. Atmospheric Fate
If palladium chloride is released to the atmosphere, the low vapor pressure expected for an ionic salt indicates that it will
exist as a particulate. Dry deposition of palladium chloride is expected to be the dominant fate process in the
atmosphere (Arimoto, 1989). The rate of dry deposition will depend on the prevailing winds and particle size (Bodek et
al., 1988). Palladium chloride is expected to undergo wet deposition (Arimoto, 1989) in rain, snow, or fog, based upon
its water solubility (Dean, 1985). •
C. Terrestrial Fate
If palladium chloride is released to soil, it is expected to dissociate into its component ions in moist soils. The
dissociation of palladium chloride in moist soils indicates that palladium chloride is not expected to volatilize from moist
soil surfaces. While no specific information concerning the sorption of ionic palladium in soils was available, some
metals adsorb to charged surfaces of soil particles or form inner sphere complexes with variable-charge soilsurfaces
(Evans, 1989). If this occurs with palladium then its rate of migration through soil may be slow. Chloride is extremely
mobile in soils (Bodek et al., 1988). The chloride ion may complex with heavy metals, thereby increasing their
solubility (Bodek etal., 1988) and potential for leaching into groundwater. The low vapor pressure expected for an
ionic salt indicates that palladium chloride will not volatilize from dry soil surfaces.
D.
Summary
If released into water, palladium chloride will dissociate into palladium and chloride ions. Therefore, palladium chloride
is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, nor
volatilize from water surfaces. Palladium ions may adsorb to charged surfaces of suspended sediments and humic matter
in the water column. Adsorption of the chloride ion to suspended solids and sediment in the water column is not
expected to be an important fate process. If released to soil, palladium chloride is expected to dissociate into its
component ions in moist soils. The dissociation of palladium chloride into its component ions indicates that palladium
chloride is not expected to volatilize from moist soil surfaces. Ionic palladium .may adsorb to charged surfaces of soil
particles or form inner sphere complexes with variable-charge soil surfaces. Chloride is extremely mobile in soils. The
low vapor pressure expected for an ionic salt indicates that volatilization of palladium chloride from soil surfaces is not
expected to be an important fate process. If released to the atmosphere, palladium chloride is expected to exist as a
particulate. Palladium chloride is expected to be physically removed from the atmosphere by wet and dry deposition.
The rate of dry deposition will depend on particle size and prevailing wind patterns. -
C-44
-------
APPENDIX C
CHEMICAL SUMMARY FOR PHOSPHORIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL D3ENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of phosphoric, acid are summarized in Table-1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PHOSPHORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Data
7664-38-2
orthophosphoric acid
H30/P
H,P04
unstable, orthorhombic crystals; clear, syrupy liquid
98.00
42.35 °C (crystals); -1 1.8 °C (30% soln)
261 °C (crystals); 101.8 °C (30% soln)
5,480 g/1 at 20 °C (crystals);
354. 1 g/1 at 20 °C (30% soln)
Reference
Lide(1995)
Budavarietal. (1996)
Budavari et al. (1996)
-Budavarietal. (1996)
Budavari et al. ( 1 996)
Budavari et al. (1996)
Gard (1996)
Card (1996)
Weast et al. (1985)
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
, Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
1.86 g/cm3 at 25 °C (crystals);
1.18 g/cm3 at 25 °C (30% soln)
no data
expected to be < 10
expected to be < 1
0.03 mm Hg at 20 °C (crystals);
. 16.3 mm Hg at 20 °C (30% soln)
relatively unreactive at room temperature
no data
no data
pK,: 2.15; pK2: 7.09; pK3:12.32
no data
no data
expected to be < lxlO-8 arm mVmole
no data
no data
Gard (1996)
SRC (1998)
SRC (1998)
Gard (1996)
Gard (1996)
Budavari et al. (1996)
SRC (1998)
H. ENVIRONMENTAL FATE
A. Aquatic Fate
Phosphoric acid is a weak tribasic acid with a pl^ of 2.15 (Budavari et al., 1996) and, if released into the water column
at low concentrations, it will dissociate into dihydrogen phosphate (H2PO4) and hydrogen (H+) ions. Dihydrogen
phosphate then dissociates into hydrogen phosphate ion (HPO4'2; pK2 of 7.09) and orthophosphate ion (PO4'3; pKj of
12.32). As a result, phosphoric acid is not expected to volatilize or bioconcentrate in aquatic organisms. The
phosphates become available in the water column and form salts, thus affecting biological productivity (Bodek et al.,
C-45
-------
APPENDIX C
198.8). Phosphorous, in the form of phosphate, is an essential nutrient to plants in aquatic environments (Bodek et al.,
1988). In addition, the phosphates can complex with metal ions in sediment and water to form insoluble species such as
FePO< and CaHPO4 (Bodek et al., 1988).
B. Atmospheric Fate
If phosphoric acid is released to the atmosphere, its vapor pressure indicates it will exist predominantly as a vapor in the
ambient atmosphere. Wet deposition of phosphoric acid in rain, snow, or fog is expected to be the dominant fate process
in the atmosphere based upon its solubility in water (Arimoto, 1989).
C.
Terrestrial Fate
If phosphoric acid is released to soil, it will dissociate into dihydrogen phosphate and hydrogen ions, ultimately
dissociating to the orthophosphate ion at high pH's. Phosphate added to soil as fertilizer is quickly sorbed and later
"fixed" (probably precipitated) into less soluble forms (Bodek et al., 1988). A similar fate is anticipated for phosphate
species from phosphoric acid. While the exact mechanism of sorption is uncertain, phosphate fixation is appreciable in
all but very coarse-textured soils; only about one-fourth of the fertilizer phosphate is usable by plants, the rest being lost
to the occluded soil fraction (Bodek et al., 1988). Phosphorous, in the form of phosphate, is an essential nutrient to
plants (Bodek et al., 1988).
D.
Summary
Phosphoric acid is atribasic acid in which the first hydrogen is strongly ionizing, the second moderately weak, and the
third very weak. Both acidic and basic salts can be formed from phosphoric acid. .If released into water, phosphoric
acid will dissociate into dihydrogen phosphate (H2PO4) and hydrogen (IT) ions, eventually dissociating into the
orthophosphate ion (PO/3) under the proper conditions. Therefore, phosphoric acid is not expected to adsorb to
suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, nor volatilize from water
surfaces. The phosphates become available in the water column and form salts, affecting biological productivity, and
comptexing with metal ions form insoluble species such as FePO4 and CaHPO4. If released to soil, phosphoric acid is
expected to dissociate into its component ions in moist soils. Phosphate added to soil as fertilizer is quickly sorbed and
later "fixed" into less soluble forms; phosphate fixation is appreciable in all but very coarse-textured soils; only about
one-fourth of the fertilizer phosphate is usable by plants, the rest being lost to the occluded soil fraction. Phosphorous,
in the form of phosphate, is an essential nutrient for aquatic and terrestrial plants. Volatilization of phosphoric acid from
soil surfaces is not expected to occur. If released to the atmosphere, phosphoric acid is expected to exist as a gas.
Phosphoric acid is expected to be physically removed from the atmosphere by wet deposition based upon its water
solubility.
C-46
-------
APPENDIX C
CHEMICAL SUMMARY FOR POTASSIUM AUROCYANIDE
This summary is based on information retrieved from a systematic search limited to secondary sources The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of potassium aurocyanide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF POTASSIUM
AUROCYANIDE'
Characteristic/Property Data Reference
CAS.No. (deleted)
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Difrusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
554-07-4
13967-50-5 '
gold potassium cyanide
potassium dicyanoaurate(I)
C2AuKN2
KAu(CN)2
dihydrate, crystalline powder
288.13
no data; expected to be > 350 °C
no data; expected to be > 500 °C
Approximately 130 g/L2
1 g dissolves in 0.5 ml boiling H2O
. 3.45 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 C
stable in aqueous solution2
not flammable
no data; expected to be > 350 °C
. readily dissociates to K* and [Au(CN)J~
no data
no data
no data; expected to be < IxlO'8 ;
no data
no data
CAS (1998)
CAS (1998)
Budavarietal. (1996)
Budayari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
Budavari et al. (1996)
SRC (1998)
SRC (1998)
Budavari et al. (1996)
Budavari et al. (1996)
Weast(1986)
SRC (1998)
SRC (1998),
SRC (1988)
Cotton and Wilkinson (1966)
ECDIN (1998)
SRC (1998)
Cohn and Stern (1994)
SRC (1998)
1 Both electrochemical and electroless gold plating processes that use potassium aurocyanide under basic conditions may contain potassium cyanide as
a complexing agent (Gmelin, 1998; Cohn and Stem, 1994; McDermott, 1974). The concentration of KCN is typically approximately 6 g/L (0 1 M)
although values as high as 200 g/L (3 M) have been reported (Gmelin, 1988). ,
2 Estimated from a reported solubility of 1 g dissolves in 7 ml H2O (Budavari et al, 1996).
3 Potassium aurocyanide is stable in aqueous solution under both basic and neutral conditions (Cotton and Wilkinson, 1966; Cohn and Stem, 1994).
It is also stable in aqueous solutions under acidic conditions (Cohn and Stem, 1994), although common acids such as HC1, H2SO4, HNO3 and H2S are
known to degrade potassium aurocyanide (Gmelin, 1998) and release HCN and gold monocyanide (Budavari et al, 1996; Gmelin, 1998).'
Concentrated acids and elevated temperatures, or both, are required (Gmelin, 1998). Potassium aurocyanide is commonly used in warm (35-55°C)
acidic plating solutions at a pH of approximately 4 (Gmelin, 1998) and stabilized acidic plating baths containing potassium aurocyanide ihave been
reported down to a pH of 1.5 (McDermott, 1974), yet it is generally considered stable in water above pH 3 (Renner and Johns, 1989). These data
C-47
-------
APPENDIX C
indicate that potassium aurocyanide is expected to be chemically stable in the pH range 5-9 typically found in the environment (Lyman et al, 1990),
but not under highly acidic conditions such as those found in the stomach (pH 1-2).
H. ENVTRONMENTAL FATE
A. Aquatic Fate
If released to water, potassium aurocyanide will rapidly and completely dissociate into potassium (K+) and aurocyanide
([Au(CN)JO ions (Cohn and Stem, 1994). The aurocyanide ion is expected to be stable to hydrolysis in the pH range of
5-9 typically encountered in the environment (Lyman et al, 1990; SRC, 1998). The dissociation of potassium
aurocyanide into its component ions also indicates that it is not expected to volatilize from water surfaces to the
atmosphere, adsorb to sediment and suspended organic matter, or bioconcentrate in fish and aquatic organisms (Biodek
etal., 1988).
B. Atmospheric Fate
If released to the atmosphere, potassium aurocyanide will exist as a particulate. Its atmospheric fate will be dominated
by deposition to the Earth's surface via wet and dry processes, as potassium aurocyanide is not expected to undergo
degradation by the most common atmospheric oxidant, hydroxyl radicals (Lyman et al, 1990; SRC, 1988). The rate of
dry deposition will be dependent on the prevailing winds and particle size; fine particles of potassium aurocyanide have
the potential to be transported significant distances from their original point of release (Bodek et al, 1988). Potassium
aurocyanide is expected to undergo efficient wet deposition in either rain or fog due to its water solubility. Dissolution
in clouds followed by wet deposition may also occur. Potassium aurocyanide is stable to light (Cohn and Stern,. 1994),
and is not expected to undergo degradation by direct photolyis.
C.
Terrestrial Fate
If potassium aurocyanide is released to soil, it is expected to display very high mobility based on its water solubility of
143 g/L (Budavari, 1996). Therefore, it has the potential to leach into groundwater. Its rate of leaching through soil
may be attenuated by the formation of insoluble soil/aurocyanide complexes that can arise from reactions with metals
naturally present in soil (Bodek et al, 1988). The importance of complex formation for potassium aurocyanide in soil is
not known. The very high melting point and low vapor pressure expected for an ionic salt indicates that potassium
aurocyanide will not volatilize from either moist or dry soils to the atmosphere (Bodek et al, 1988).
D.
Summary
If released to water, potassium aurocyanide will dissociate into K* and [Au(CN)J~ ions. Therefore, it is not expected to
adsorb to sediment and suspended organic matter, bioconcentrate in fish and aquatic organisms, or volatilize from water
surfaces to the atmosphere. The aurocyanide ion is expected to be chemically stable and it is not expected to hydrolyze
in the pH range 5-9 typically found in the environment. In soil, potassium aurocyanide is likely to display very high
mobility as a result of its relatively high water solubility and it-has the potential to leach to groundwater. Its rate of
leaching through soil may be attenuated by the formation of insoluble soil/aurocyanide complexes although the
importance of this process is not known. Volatilization from soil surfaces to the atmosphere is not expected to occur. If
released to the atmosphere, potassium aurocyanide is expected to exist as a particulate. Its atmospheric fate'is expected
to be dominated by wet and dry deposition to the Earth's surface. Efficient removal from the atmosphere during rain
events is expected although the rate of dry deposition will be dependent on its particle size and the prevailing wind
patterns. Therefore, fine particles of potassium aurocyanide have the potential to travel significant distances from their
original point of release.
C-48
-------
APPENDIX C
CHEMICAL SUMMARY FOR POTASSIUM PEROXYMONOSULFATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of potassium peroxymonosulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF POTASSIUM
PEROXYMONOSULFATE '_
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammabiliry
FlashPoint
Dissociation Constant
Molecular Diffiisivity Constant
Air Diffiisivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
10058-23-8
Monopotassium peroxymonosulfurate
Peroxymdnosulfuric acid, monopotassium salt
HO5S.K
HOOS(O)(O)OK
no data
153.18
no data
no data
no data
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be <1X10^ mm Hg
no data
no data
no data
expected to dissociate
no data
no data
no data; expected to be <1X10"S •
no data
no data
CAS (1998)
Howard and Neal (1992)
.Howard and Neal (1992)
Howard and Neal (1992)
CAS (1998)
Howard and Neal (1992)
Estimated
Estimated
Estimated
Bodeketal. (1988)
Estimated
II. ENVIRONMENTAL FATE
A. Aquatic Fate
Most potassium salts are highly dissociated in natural waters (Bodek et al., 1988). Therefore, if potassium
peroxymonosulfate is released into water, it is expected to dissociate into potassium (K+) and peroxymonosulfate (SOS~)
ions. The potassium ion is expected to exist predominately as the free ion in most natural waters (Bodek et al., 1988).
Ion exchange processes with suspended solids and sediment in the water column are expected to remove ionic potassium
from solution; however, ionic potassium may be displaced by other cations present in natural waters with a higher
C-49
-------
APPENDIX C
affinity for ion exchange sites (Bodek et al., 1988). Aqueous solutions of the impure potassium peroxymonosulfate, i.e.,
those, containing dipotassium sulfate and monopotassium sulfate, decompose yielding mainly O2 and sulfate (SO42'),
hydrogen peroxide and peroxydisulfate (S2O82~) occur in small amounts (Cotton and Wilkinson, 1980). Sulfate ions may
participate in oxidation-reduction reactions or react with cations present in the water column to form soluble complexes
or insoluble precipitates (Bodek et al., 1988). Sulfate-reducing microorganisms are important mediators in redox
reactions involving this ion (Bodek et al., 1988). Peroxy compounds are short-lived because of the inherent instability of
the O-O bond and are expected to degrade rapidly (U.S. EPA, 1993).
B.
Atmospheric Fate
If potassium peroxymonosulfate is released to the atmosphere, the low vapor pressure expected for an ionic salt indicates
that potassium peroxymonosulfate will exist as a particulate. Wet and dry deposition of potassium peroxymonosulate is
expected to be an important fate process in the atmosphere (Arimoto, 1989). The rate of dry deposition will depend on
the prevailing winds and particle size (Bodek et al., 1988).
C.
Terrestrial Fate
If potassium peroxymonosulfate is released to soil, it may decompose in moist soils; the importance of this process is not
known. The low vapor pressure expected for an ionic salt indicates that potassium peroxymonosulfate will not volatilize
from dry soil surfaces. The uncomplexed potassium ion is expected to be the predominant species in well-drained soils
from pH 4 to pH 10 (Bodek et al., 1988). Ion exchange reactions are expected to attenuate the mobility of the potassium
ion in the subsurface environment, however ionic potassium may be displaced by other cations with a higher affinity for
ion exchange sites (Bodek et al., 1988). Peroxy compounds are short-lived because of the inherent instability of the O-O
bond and are expected to degrade rapidly (U.S. EPA, 1993). .
D.
Summary
If released into water, potassium peroxymonosulfate is expected to dissociate into potassium and peroxymonosulfate
ions. The dissociation of potassium peroxymonosulfate into its component ions indicates that potassium
peroxymonosulfate is not expected to volatilize from water surfaces or bioconcentrate in aquatic organisms. In most
natural waters, the potassium ion is expected to exist predominately as the free ion. Ion exchange processes with
suspended solids and sediment in the water column are expected to remove ionic potassium from solution; however
ionic potassium may be displaced by other cations in natural waters with a higher affinity for ion exchange sites. Sulfate
ions may participate in oxidation-reduction reactions or react with cations present in the water column to form soluble
complexes or insoluble precipitates; sulfate-reducing microorganisms are important mediators in redox reactions
involving this ion. If released to soil, potassium peroxymonosulfate may decompose in moist soils or dissociate into its
component ions. As a result, potassium peroxymonosulfate is not expected to volatilize from moist soil surfaces. The
low vapor pressure expected for an ionic salt indicates that potassium peroxymonosulfate is not expected to volatilize
from dry soil surfaces. The mobility of potassium ions will be retarded by ion exchange processes with charged surfaces
of soil particles. However, since the potassium ion is held weakly by ion exchange processes, it may leach into
groundwater. Peroxy compounds are short-lived because of the inherent instability of the O-O bond and are expected to
degrade rapidly. If released to the atmosphere, potassium peroxymonosulfate is expected to exist as a particulate based
upon the low vapor pressure expected for an ionic salt. Wet and dry deposition is expected to be the dominant fate
process in the atmosphere.
C-50
-------
APPENDIX C
CHEMICAL SUMMARY FOR PROPIONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of propionic acid are summarized in Table 1.
TABLE I. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PROPIONIC ACID
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffiisivity Constant
Air Diffiisivity Constant
Henry's Law Constant
Fish Bioconcentration Factor
Odor Threshold
79-09-4
methyl acetic acid; ethyl formic acid
C,H602
CH3CHZCOOH
oily liquid
74.08
-21.5 °C
141.1 °C
lxlO«g/l@25°C
d25'4, 0.99336
no data
36 (calculated)
0.33
3.53mmHg@25°C
corrodes steel, metal
combustible
136 °F (58 °C), open cup
pKa = 4.88 ' .
no data
no data
4.45x10-' atm mVmole @ 25 °C
0.02 (calculated)
no data
Howard and Neal (1992)
Budavari et al. (1996)
Budavarietal. (1996)
Budavari etal. (1996)
Budavari et al. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavari et al. (1996)
US. EPA (1981)
Budavari et al. (1996)
Lymanetal. (1990)
Hanschetal. (1995)
Daubert and Danner (1985)
Weiss (1986)
Lewis (1993)
Budavarietal. (1996)
Serjeant and Dempsey (1979)
. Butler and Ramchandani (1935)
Lyman etal. (1990)
H. ENVIRONMENTAL FATE
A. Aquatic Fate
Aerobic biodegradation is likely to be the most important removal mechanism of propionic acid from aquatic systems
(Dias and Alexander, 1971, as cited in HSDB, 1998). With a pKa of 4.88 (Serjeant and Dempsey, 1979), propionic
acid and its conjugate base will exist in environmental waters in varying proportions that are pH dependent. Under
neutral and alkaline conditions, propionic acid is expected to exist predominantly as its conjugate base, the propionate
ion (Lyman et al., 1990). In addition, at a pH of 4.88 propionic acid is 50% dissociated; even under mildly acidic
conditions, it will exist predominantly as the conjugate base. In general, organic ions are not expected to volatilize from
water or adsorb to particulate matter in water to the degree that would be predicted for their neutral counterparts. An
C-51
-------
APPENDIX C
estimated Koc of 36, determined from a log Kow of 0.33 (Hansch et al., 1995), indicates propionic acid should not
partition from the water column to organic matter contained in sediments and suspended solids. Similarly, the Kow
indicates that bioconcentration in fish and aquatic organisms is not an important fate process. Propionic acid's Henry's
Law constant of 4.45xlO'7 atm mVmole (Butler and Ramchandani, 1935) indicates that volatilization of propionic acid
from environmental waters should be extremely slow (Lyman et al., 1990). Volatilization will be attenuated depending
upon pH and the amount of propionic acid that is dissociated. Since carboxylie acids are generally resistant to aqueous
hydrolysis (Lyman et al., 1990), it is not expected to be an important fate process for propionic acid. The direct
photolysis (Calvert and Pitts, 1966, as cited in HSDB, 1998) and reaction of propionic acid with photochemically-
generated hydroxyl radicals in water (Anbar and Neta, 1967, as cited in HSDB, 1998) are also not expected to be
important fate processes.
B. Atmospheric Fate
Based on a vapor pressure of 3.53 mm Hg at 25 °C (Daubert and Danner, 1985, as cited in HSDB, 1998), propionic
acid is expected to exist almost entirely in the vapor phase in the ambient atmosphere (Bidleman, 1988). The rate
constant for the reaction of propionic acid with photochemically-produced hydroxyl radicals in air has been
experimentally determined to be 1.22 x 10'12 cmVmolecule-sec at 25 °C (Daugaut et al., 1988, as cited in HSDB, 1998).
This corresponds to an atmospheric half-life of approximately 13 days. Since low molecular weight organic acids have
absorption bands at wavelengths well below the environmentally important range of 290 nm, the direct photolysis of
propionic acid in air is not expected to be important (Calvert and Pitts, 1966, as cited in HSDB, 1998). Extensive
monitoring data (Chapman et al., 1986; Hoffman and Tanner; 1986; Winkeler et al., 1988; Mazurek and Simoneitt,
1986, as cited in HSDB, 1998) has shown that physical removal of propionic acid from the air by wet deposition
(rainfall, dissolution in clouds, etc.) may be an important fate process under the appropriate atmospheric conditions.
C.
Terrestrial Fate
Biodegradation is likely to be the most important rempval mechanism of propionic acid from aerobic soil (Dias and
Alexander, 1971, as cited in HSDB, 1998). With a pKa of 4.88 (Serjeant and Dempsey, 1979), propionic acid and its
conjugate base will exist in varying proportions that are dependent on the pH of the soil. A Henry's Law Constant of
4.45x10'7 atm mVmole (Butler and Ramchandani, 1935) indicates that volatilization of propionic acid from moist soil
should be extremely slow (Lyman et al., 1990). Yet, propionic acid should volatilize rapidly from dry surfaces based
upon a vapor pressure of 3.53 mm Hg at 25 °C (Daubert and Danner, 1985, as cited in HSDB, 1998). Volatilization
will be attenuated depending upon pH and the amount of propionic acid dissociated. An estimated Koc of 36, -
determined from a log Kow of 0.33 (Hansch et al., 1995), indicates that propionic acid may be highly mobile in soil
(Swann et al., 1983). In addition, monitoring data has shown that propionic acid can leach to groundwater (Stuermer et
al., 1982; Burrows and Rowe, 1975; Lema et al., 1988, as cited in HSDB, 1998). Organic ions generally do not
volatilize from moist soil surfaces and do not undergo adsorption to the extent of their neutral counterparts, which is
consistent with propionic acid's potential for displaying high mobility through soils under conditions where rapid
biodegradation does not occur.
D.
Summary
With a pKa of 4.88, propionic acid and its conjugate base will exist in environmental media in varying proportions that
are pH dependent; under typical environmental conditions, propionic acid will exist predominantly as its conjugate base.
A Henry's Law constant of 4.45xlO"7 atm mVmole at 25 °C indicates that volatilization of propionic acid from
environmental waters and moist soil should be extremely slow. Yet, based on a vapor pressure of 3.53 mm Hg,
propionic acid should volatilize rapidly from dry surfaces. However, volatilization of propionic acid will be pH
dependent; if propionic acid is dissociated, very little (about 1%) will be available for volatilization. A relatively low
estimated Koc indicates that propionic acid should not partition from the water column to organic matter contained in
sediments and suspended solids; the Koc also indicates that it should be highly mobile in soil. However, monitoring data
has shown that propionic acid has the potential to leach to groundwater under the appropriate conditions. Propionic acid
is miscible with water and monitoring data has shown that physical removal from air by wet deposition is an important
removal mechanism. Biodegradation is likely to be the most important removal mechanism of propionic acid from
aerobic soil and water. In the atmosphere, propionic acid is expected to exist almost entirely in the gas phase and
oxidative removal by photochemically-produced hydroxyl radicals has a half-life of 13 days. The hydrolysis in water,
C-52
-------
APPENDIX C
photolysis in air, and bioconcentration in aquatic organisms are not expected to be important fate processes for
propionic acid.
C-53
-------
APPENDIX C
CHEMICAL SUMMARY FOR SILVER NITRATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for silver nitrate, other nitrate salts and silver.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of silver nitrate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SILVER NITRATE
Characteristic/Property
Data
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air s 1)
Koc
Log Row
Vapor Pressure
Reactivity
Himmability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant .
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
7761-88-8
silver(I)nitrate
AgNOj
AgN03
colorless, rhombohedral crystals
169.873
212 °C
440 °C decomposes
2,500 g/L water
4.35 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 °C
can explode on contact with soot, organics
not flammable
no data; expected to be > 350 °C
no data
no data
no data
no data; expected to be < IXIO'"
no data
no data
Lide (1995)
Lide (1995)
Budavarietal. (1996)
Lide (1995)
Lide (1995)
Lide (1995)
Lide (1995)
Lide (1995)
Budavarietal. (1996)
Lide (1995)
SRC (1998)
SRC (1998)
Estimated
Renner(1993)
Prager(1995)
SRC (1998)
SRC (1998)
II. ENVIRONMENTAL FATE
A. Aquatic Fate "
If silver nitrate is released into water, it is expected to dissociate'into silver (Ag+) and nitrate (NO3)- ions. The
dissociation of silver nitrate into its component ions indicates that silver nitrate is not expected to volatilize from water
surfaces or bioconcentrate in aquatic organisms (Bodek et al., 1988). Ionic silver may form complexes with hydroxide,
sulfide ligands, halide ligands, and chelating organics (Bodek et al., 1988). Silver-organic complexes may be important
(Bodek et al., 1988). In aquatic systems with high halide concentrations, precipitation of insoluble silver halides may
occur (Bodek et al., 1988). Silver ions may sorb to organic matter and sediment that has high manganese dioxide, iron
C-54
-------
APPENDIX C
oxide, and clay content (Bodek et al., 1988). Nitrate is a minor constituent in natural waters, where its concentration is
limited by biological reactions that consume it (Bodek et al., 1988). In aquatic systems where nitrogen is a limiting
nutrient, high loadings of nitrate into surface waters can cause algal blooms (Bodek et al., 1988). In natural waters with
a low nitrate concentration, complexation with transition metals is not expected to be an important process (Bodek et al
1988).
B.
Atmospheric Fate
If released to the atmosphere, silver nitrate's low vapor pressure indicates that it will exist as a particulate (Bidleman,
1988). Wet and dry deposition of silver nitrate is expected to be the dominant fate process in the atmosphere (Arimoto,
1989). Silver nitrate's high water solubility (Budavari et al., 1996) indicates that it is expected to undergo wet
deposition in rain, snow, or fog. The rate of dry deposition will depend on the prevailing winds and particle size (Bodek
et al., 1988): Pure silver nitrate is not photosensitive (Cappel, 1997); however, trace amounts of organic material
promote its photodegradation (Budavari etal., 1996).
C.
Terrestrial Fate
If released to soil, silver nitrate is expected to dissociate into its component ions in the presence of moisture. Silver may
adsorb to manganese dioxide, iron oxides, clays, and organic matter (Bodek et al., 1988); therefore, its rate of migration
through soil may be slow. The high boiling point, low vapor pressure, and low Henry's Law constant expected for an
ionic salt (SRC, 1998) indicates that silver nitrate will not volatilize from either moist or dry soil surfaces. Inoic silver
may form complexes with hydroxide, sulfide ligands, halide ligands, and chelating organics (Bodek et al., 1988). Nitrate
ions may be converted to gaseous N2 or nitrous oxide (N2O) by microorganisms under anaerobic conditions or may be
assimilated by plants (Bodek et al., 1988). Sorption of nitrate ions by soils is generally insignificant and therefore nitrate
ions are expected to leach into groundwater (Bodek et al., 1988).
D.
Summary
If released into water, silver nitrate will dissociate into silver and nitrate ions. Therefore, silver nitrate is not expected to
adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, or volatilize from .
water surfaces. In natural waters, the concentration of nitrate is limited by biological reactions that consume it. High
loadings of nitrate into surface waters can cause algal blooms if nitrogen is a limiting nutrient. Silver nitrate is expected
• to dissociate into its component ions in moist soils, and ionic silver may adsorb to manganese dioxide, iron oxides, and
clays. Nitrate is highly mobile in soils and therefore may leach into groundwater. Under anaerobic conditions nitrate
may be converted to gaseous N2 or nitrous oxide by microorganisms. Volatilization of silver nitrate from soil surfaces is
not expected to occur. If released to the atmosphere, silver nitrate is expected to exist as a particulate. Silver nitrate is
expected to be physically removed from the atmosphere by wet and dry deposition. Dry deposition will depend on
particle size and prevailing wind patterns. Pure silver nitrate is not photosensitive and will not degrade in sunlight; trace
amounts of organic material promote silver nitrate's photodegradation.
C-55
-------
APPENDIX C
CHEMICAL SUMMARY FOR SODIUM HYDROXIDE
% • ' -
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium hydroxide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM HYDROXIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kdw
Vapor Pressure
Reactivity
Flam inability
Flash Point
Dissociation Constant
Data
1310-73-2
Caustic soda
HNaO
NaOH
white orthohombic crystals; hygroscopic
39.997
323 °C
1388°C
571.9 g/L
2.13g/cm3
not pertinent
no data; expected to be <10
no data; expected to be <1
no data; expected to be <10S mm Hg
when wet, attacks metals such as aluminum, tin, lead,
and zinc to produce flammable hydrogen gas
not flammable
not flammable
readily dissociates into Na+ and OH"
Reference
CAS (1998)
Bodeketal.(1988)
Budavari et al. (1996)
Budavarietal.(1996)
Lide(1995)
Lide (1995)
Lide (1995)
Lide (1995)
Weast(1985)
Lide (1995)
Weiss (1986)
SRC (1998)
SRC (1998)
Weiss (1986)
Weiss (1986)
Weiss (1986)
Weiss (1986)
SRC (1998)
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
no data
no data
no data; expected to be <1X10^
no data
not pertinent
SRC (1998)
Weiss (1986)
H. ENVIRONMENTAL FATE
A. Aquatic Fate ,
If sodium hydroxide is released into water, it will dissociate into sodium (Na+) and hydroxide (OH") ions (Bodek et al.,
1988). The dissociation of sodium hydroxide into its component ions indicates that sodium hydroxide is not expected to
volatilize from water surfaces or bioconcentrate in aquatic organisms. Because it is strongly basic, sodium hydroxide
will react with any protic acids to form salts. Hydroxide is the conjugate base of water; protonation of hydroxide
produces water. The presence of hydroxide in natural waters is entirely dependent on the pH of the water, but massive
amounts of sodium hydroxide may raise the pH of the receiving water.. Metals present in natural waters may form
complexes with the hydroxide ion; complexes with transition metals will result in precipitation of the sparingly soluble
C-56
-------
APPENDIX C
metal hydroxides (Bodek et al., 1988). The sodium ion is expected to exist predominately as the free ion in most natural
waters (Bodek et al., 1988). Ion exchange processes with suspended solids and sediment in the water column are
expected to remove ionic sodium from solution; however, sodium binds weakly to ion exchange sites and is expected to
be displaced by other cations present in natural waters (Bodek et al., 1988).
B.
Atmospheric Fate
If sodium hydroxide is released to the atmosphere, it is expected to exist as a particulate based upon the low vapor
pressure expected for this compound. Wet deposition of sodium hydroxide (Arimoto, 1989) in rain, snow, or fog is
expected to be the dominant fate process in the atmosphere based upon its high water solubility (Budavari et al., 1996);
however, carbon dioxide dissolved in atmospheric water, may react with sodium hydroxide to form sodium carbonate.
C.
Terrestrial Fate
If sodium hydroxide is released to soil, it is expected to dissociate into its component ions in moist soils and react with
any protic acids present in soil to form the sodium salt and water. The low vapor pressure and low Henry's Law constant
expected for an ionic salt indicates that sodium hydroxide will not volatilize from either moist or dry soil surfaces. In
soil, ion exchange processes are important in retarding the mobility of sodium ions, however they may be replaced by
other soil cations since the sodium ion is held weakly by soils (Evans, 1989).
D.
Summary
If released into water, sodium hydroxide will dissociate into sodium and hydroxide ions. The dissociation of sodium
hydroxide into its component ions indicates that sodium hydroxide is not expected to volatilize from water surfaces or
bioconcentrate in aquatic organisms. The hydroxide ion will react with protic acids to form water. Massive amounts of
sodium hydroxide may raise the pH of the water. The sodium ion is expected to participate in ion exchange reactions
with charged surfaces of suspended sediments and sediment in the water column. If released to soil, sodium hydroxide is
expected to dissociate into its component ions in moist soils and react with protic acids to form water. Sodium
hydroxide is not expected to volatilize from moist or dry soil surfaces. The mobility of sodium ions will be retarded by
ion exchange processes with charged surfaces of soil particles. However, since the sodium ion is held weakly by ion
exchange processes, it may leach into groundwater. If released to the atmosphere, sodium hydroxide is expected to exist
as a particulate based upon the low vapor pressure expected for an ionic compound. Sodium hydroxide reacts with
carbon dioxide to form sodium carbonate. Wet deposition in rain, snow, or fog is expected to be the dominant fate
process in the atmosphere based upon sodium hydroxide's high water solubility.
C-57
-------
APPENDIX C
CHEMICAL SUMMARY FOR SODIUM HYPOPHOSPHITE AND
SODIUM HYPOPHOSPHITE MONOHYDRATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium hypophosphite and its monohydrate are summarized in
Tables 1 and 2, respectively.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM
HYPOPHOSPHITE
Characteristic/Property Data •
Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air » 1}
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7681-53-0
Phosphinic acid, sodium salt
H2NaO2P
NaH2P02
colorless, pearly, crystalline plates or white granular
powder
CAS (1998)
Budavarietal. (1996)
Budavarietal. (1996)
Budavari et al. (1996)
Lewis (1993)
Budavari et al. (1996)
87.98
no data
decomposes ' Dean (1985)
approximately 500 g/L' Estimated
no data " ,
no data
no data; expected to be <10 Estimated
no data; expected Jo be <1 Estimated
no data; expected to be < 10* mm Hg Estimated
Explosion risk when mixed with strong oxidizing agents. Lewis (1993)
no data
no data
2.1 (phosphinicacid) Feeetal. (1996)
no data
no data
no data; expected to be <1X10-S , Estimated
no data
no data
Estimated from a repotted solubility of 100 parts in 100 parts at 25°C for the monohydrate (Dean 1985).
C-58
-------
APPENDIX C
TABLE 2. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL
HYPOPHOSPfflTE MONOHYDRATE
PROPERTIES OF SODIUM
Characteristic/Property
Data
Reference
CAS No.
Molecular Formula
Chemical Structure
Physical State
Molecular Weight •
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffiisivity Constant
Air Diffiisivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
10039-56-2
NaPH2CyH20
NaPH202-H20
white, monoclinic
105.99
loses water at 200°C
decomposes
approximately 500 g/L ' •
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be < 10* mm Hg
no data
nodata
no data
2.1 (phosphinic acid)
no data
no data
no data; expected to be <1X10"S
nodata
nodata
CAS (1998)
Dean (1985)
Dean (1985)
Dean(1985)
Dean (1985)
Dean (1985)
Dean (1985)
Estimated
Estimated
Estimated
Estimated
Fee etal. (1996)
Estimated
1 Estimated from a reported solubility of 100 parts in 100 parts at 25 °C (Dean 1985).
II. ENVIRONMENTAL FATE
. A.
Aquatic Fate
Almost all sodium salts are highly dissociated in natural waters (Bodek et al., 1988). Therefore, if sodium
hypophosphite is released into water, it is expected to initially hydrate to form the monohydrate then dissociate into
hypophosphite (H2PO2~) and sodium (Na+) ions. The pKa of phosphinic acid indicates that hypophosphite will exist
mainly in the dissociated state in the environment. The dissociation of sodium hypophosphite into its component ions
indicates sodium hypophosphite will not volatilize from water surfaces or bioconcentrate in aquatic organisms. The
sodium ion is expected to exist predominately as the free ion in most natural waters (Bodek et al., 1988). Ion exchange
processes with suspended solids and sediment in the water column are expected to remove ionic sodium from solution;
however, sodium binds weakly to ion exchange sites and is expected to be displaced by other cations present in natural
waters (Bodek et al., 1988). No information specifically regarding the environmental fate of the phosphinic acid or
hypophosphite ion in water was located in the available literature. Phosphinic acid and its salts are a strong reducing
agents; they are oxidized to phosphonic acid or phosphonate (H3PO3 or HPO32') (Fee et al., 1996). It is unclear how
rapidly this process will occur in the environment.
B.
Atmospheric Fate
If sodium hypophosphite or its monohydrate are released to the atmosphere, it is expected to exist as a particulate based
upon the low vapor pressure expected for this compound. Particulates of the unhydrated salt may also hydrate when
C-59
-------
APPENDIX C
exposed to moisture in the atmosphere to form the monohydrate. Wet deposition of sodium hypophosphite in rain, snow,
or fog is expected to be the dominant fate process in the atmosphere (Arimoto, 1989) based upon its high water
solubility (Betterman et al., 1991).
C.
Terrestrial Fate
If sodium hypophosphite is released to soil, it is expected to initially hydrate to form the monohydrate then dissociate
into its component ions in moist soils. The pKa of phosphinic acid indicates that it will exist mainly in the dissociated
state in the environment. The low vapor pressure and low Henry's Law constant expected for an ionic salt indicates that
neither sodium hypophosphite nor its hydrate will volatilize from either moist or dry soil surfaces. In soil, ion exchange
processes are important in retarding the mobility of sodium ions, however they may be replaced by other soil cations
since the sodium ion is held weakly by soils (Evans, 1989). No information specifically regarding the environmental
fate of the phosphinic acid or hypophosphite ion in soils was located in the available literature. Phosphinic acid and its
salts are a strong reducing agents; they are oxidized to phosphonic acid or phosphonate (H3PO3 or HPO32') (Fee et al.,
1996). It is unclear how rapidly this process will occur in the environment.
D.
Summary
If released into water, sodium hypophosphite and its hydrate are expected to dissociate into sodium and hypophosphite
ions. The dissociation of sodium hypophosphite into its component ions indicates that it will not volatilize from water
surfaces or bioconcentrate in aquatic organisms. The sodium ion is expected to participate in ion exchange reactions
with charged surfaces of suspended sediments and sediment in the water column. If released into soil, sodium
hypophosphite and its hydrate are expected to dissociate into its component ions in moist soils. As a result, sodium
hypophosphite is not expected to volatilize from moist soil surfaces. The mobility of sodium ions will be retarded by ion
exchange processes with charged surfaces of soil particles. However, since the sodium ion is held weakly by ion
exchange processes, it may leach into groundwater. Phosphinic acid and its salts are a strong reducing agents; they are
oxidized to phosphonic acid or phosphonate (H3PO3 or HPO320- It is unclear how rapidly this process will occur in
either soil or water environments. The low vapor pressure expected for an ionic salt indicates that neither sodium
hypophosphite nor its monohydrate are expected to volatilize from dry soil surfaces. If released to the atmosphere, the
low vapor pressure expected for an ionic salt indicates that sodium hypophosphite will exist as a particulate in the
ambient atmosphere. Wet and dry deposition will be the dominant fate process in the atmosphere.
C-60
-------
APPENDIX C
CHEMICAL SUMMARY FOR STANNOUS METHANESULFONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of stannous methanesulfonic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF STANNOUS
' . METHANESULFONIC ACID _____
Characteristic/Property , Data Reference
CAS No.
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air = 1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffiisivity Constant
Air Diffusiviry Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
53408-94-9
C2H8O6S2Sn
[H3CS(O)(O)O]Sn[OS(0)(O)CH3]
no data
310.89
no data
no data
no data
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be <10^ mm Hg at 25 °C
no data
no data
no data
no data
no data
no data
no data; expected to be
-------
APPENDIX C
adsorb to charged surfaces of suspended solids and sediment in the water column, although the importance of this
process in the environment is not known. Limited data indicate that biodegradation of methanesulfonate may be an
important fate process (HSDB, 1998). An estimated BCF of 3 for methanesulfonic acid (Meylan et al., 1997) suggests
the potential for bioconcentration in aquatic organisms is low (Franke et al., 1994).
B.
Atmospheric Fate
If stannous methanesulfonic acid is released to the atmosphere, the low vapor pressure expected for an ionic salt
indicates that it will exist as a particulate. Dry deposition of stannous methanesulfonic acid is expected to be the
dominant fate process in the atmosphere (ArSmoto, 1989). The rate of dry deposition will depend on the prevailing
winds and particle size (Bodek et al., 1988). Wet deposition of stannous methanesulfonic acid may occur (Arimoto,
1989) in rain, snow, or fog.
C. Terrestrial Fate
If stannous methanesulfonic acid is released to soil, it is expected to dissociate into its component ions in moist soils.
The dissociation of stannous methanesulfonic acid into its component ions in moist soils indicates that stannous
methanesulfonic acid is not expected to volatilize from moist soil surfaces. The low vapor pressure expected for an ionic
salt indicates that stannous methanesulfonic acid is not expected to volatilize from dry soil surfaces. Ionic tin may adsorb
to charged surfaces of soil particles or form inner sphere complexes with variable-charge soil surfaces (Evans, 1989)
and therefore its rate of migration through soil may be slow. Methanesulfonic acid has a pKa of-1.86 (Serjeant and
Dempsey, 1979 as cited in PHYSPROP, 1998) indicating it will exist in the ionized form in moist soils in the
environment. Therefore, volatilization of methanesulfonate from moist soil surfaces will not occur. Methanesulfonate
ions may adsorb to charged surfaces of soil particles, however the importance of this process in the environment is
unknown. Limited data indicate that biodegradation of methanesulfonate may be an important fate process (HSDB,
1998).
D.
Summary
If released into water, stannous methanesulfonic acid is expected to dissociate into tin and methanesulfonate ions. The
dissociation of stannous methane sulfonic acid into it component ions indicates that stannous methanesulfonic acid is not
expected to bioconcentrate in aquatic organisms nor volatilize from water surfaces. Ionic tin may adsorb to charged
surfaces of suspended sediments and humic materials in the water column. Methanesulfonate ions may adsorb to
charged surfaces of suspended sediments and humic materials in the water column, however the importance of this
process in the environment is unknown. If released to soil, stannous methanesulfonic acid is expected to dissociate into
its component ions in moist soils. The dissociation of stannous methanesulfonic acid into its component ions in moist
soils indicates that volatilization from soil surfaces is not expected to be an important fate process. Ionic tin may adsorb
to charged surfaces of soil particles or form inner sphere complexes with variable-charge soil surfaces and therefore its
rate of migration through soil may be slow. Methanesulfonate ions may adsorb to charged surfaces of soil particles,
however the importance of this process in the environment is unknown. The low vapor pressure expected for an ionic
salt indicates that stannous methanesulfonic acid is not expected to volatilize from dry soil surfaces. Limited data
indicate that biodegradation of methanesulfonate may be an important fate process. If released to the atmosphere,
stannous methanesulfonic acid is expected to exist as a particulate in the ambient atmosphere based upon the low vapor
pressure expected for an ionic salt. Wet and dry deposition will be the dominant fate process in the atmosphere. The
rate of dry deposition will depend on the prevailing winds and particle size.
C-62
-------
APPENDIX C
CHEMICAL SUMMARY FOR SULFURIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sulfuric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SULFURIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Data
7664-93-9
Battery acid
H204S
H2S04
colorless oily liquid
98.080
10.3TC
337°C
1000g/Lat25°C
1.8g/cm3
not pertinent
no data; expected to be <10
no data; expected to be <1
5.98X10'5 mm Hg at 25°C
very reactive, dissolves most metals; concentrated acid
Reference
CAS (1998) .
Weiss (1986)
Budavari et al. (1996)
Budavari et al. (1996)
Lide (19.95)
Lide(1995)
Lide (1995)
Lide (1995)
Gunther et al. (1968) as cited in
PHYSPROP (1998)
Lide (1995)
. Weiss (1986)
Estimated
Estimated
Daubert and Danner (1987)
Lewis (1993) .
Flammability
Flash Point
Dissociation Constant
Molecular Difrusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
oxidizes, dehydrates, or sulfonates most organic
compounds, often causes charring.
not flammable ' Weiss (1986)
not flammable Weiss (1986)
pIQ, =-3.00, pKn2= 1.99 Bodeketal.(1988)
no data
no data
no data; expected to be <1X10"8 Estimated
no data
greater than 1 mg/m3 Weiss (1986)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If sulfuric acid is released into the water column at low concentrations, a pKa] of -3.00 (Bodek et al., 1988) indicates
sulfuric acid will dissociate into bisulfate (HSO4") and hydrogen (H*) ions. In virtually all natural waters, the bisulfete
ion will also dissociate into sulfate (SO42") and hydrogen ions based upon a pKa of 1.99 (Bodek et al., 1988). Sulfuric
acid will form salts with basic components in water. The dissociation of sulfuric acid into its component ions indicates
that sulfuric acid is not expected to volatilize from water surfaces or bioconcentrate in aquatic organisms. Sulfate ions
C-63
-------
APPENDIX C
may participate in oxidation-reduction reactions or react with cations present in the water column to form soluble
complexes or insoluble precipitates (Bodek et al., 1988). Sulfate-reducing microorganisms are important mediators in
redox reactions involving this ion (Bodek et al., 1988). Large releases of the concentrated acid into water, such as may
result from a spill, will result in a lowering of the pH (Bodek et al., 1988).
B.
Atmospheric Fate
If sulfuric acid is released to the atmosphere, its vapor pressure (Daubert and Danner, 1987) indicates it will exist as a
particulate in the ambient atmosphere. Wet deposition of sulfuric acid in rain, snow, or fog is expected to be the
dominant fate process in the atmosphere (Arimoto, 1989) based upon its high water solubility (Gunther et al., 1968 as
cited in PHYSPROP, 1998). In the atmosphere, SO2 is oxidized to sulfuric acid (Graedel et al., 1986).
C.
Terrestrial Fate
If sulfuric acid is released to soil, it will dissociate into sulfate and hydrogen ions in moist soils and will form salts with
basic soil components. The dissociation of sulfuric acid into its component ions indicates that volatilization from moist
soil surfaces is not expected to occur. Sulfate is generally weakly retained by soils (Bodek et al., 1988) and therefore it
may leach into groundwater. Adsorption of the sulfate ion may be important in humic soils containing Al and Fe oxides
(Bodek et al., 1988), Sulfuric acid's vapor pressure (Daubert and Danner, 1987) indicates that volatilization from dry
soil surfaces is not expected to be an important fate process.
D.
Summary
If released into water, sulfuric acid will dissociate into sulfate (SO42') and hydrogen (H+) ions. Therefore, sulfuric acid is
not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, nor
volatilize from water surfaces. Sulfate ions may participate in redox reactions or react with cations present in the water
column. Sulfate-reducing microorganisms have been identified as important mediators in redox reactions-involving the
sulfate ion. SulfuriC'acid will form salts with basic components in water. If released to soil, sulfuric acid is expected to
dissociate into its component ions in moist soils and will form salts with basic soil components. The dissociation of
sulfuric acid into its component ions indicates that volatilization from moist soil surfaces is not expected to be an
important fate process. In general, sulfate is weakly retained by soils and therefore it may leach into groundwater.
Adsorption of the sulfate ion may be important in soils with high organic matter content or soils containing Al and Fe
oxides. Sulfuric acid's vapor pressure indicates that volatilization from dry soil surfaces is not expected to occur. If
released to the atmosphere, sulfuric acid is expected to exist as a particulate. Sulfuric acid is expected to be physically
removed from the atmosphere by wet deposition based upon its high water solubility.-
C-64
-------
APPENDIX C
CHEMICAL SUMMARY FOR THIOUREA
This summary is based on information retrieved from a systematic s'earch limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
•The chemical identity and physical/chemical properties of thiourea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF THIOUREA
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common -Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
LogKow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
62-56-6
Thiocarbamide
Urea, 2-thio
H2NC(=S)NH2
crystals
76.12
182°C
no data
201g/Lat20°C
1.405g/cm3at25°C
no data
no data; estimated to be 2.8
-1.02
3.1 1X104 mm Hg at 25°C (extrapolated)
no data
no data
no data
no data
no data
no data
no data; estimated to be 1.6X10'7
<0.2 to <2 in carp
no data
CAS (1998)
Lide (1995)
Howard and Neal (1992)
Lide (1995)
Budavari et al. (1996)
Budavarietal. (1996)
Lide (1995)
Lide (1995)
Yalkowsky and Dannenfelser (1992)
Lide (1995)
Meylanetal. (1992)
Hanschetal. (1995)
Daubert and Danner (1992)
Meylan and Howard (1991)
Chemicals Inspection and Testing
Institute (1992)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If thiourea is released into water, an estimated Koc value of 2.8 (Meylan et al., 1992) indicates that thiourea is not
expected to adsorb to suspended solids and sediment in the water column (Swann et al., 1983). According to a
classification scheme (Franke et al., 1994), BCFs'of <0.2 and <2 in carp (Chemicals Inspection and Testing Institute,
1992) indicate that bioconcentration in aquatic organisms is low. An estimated Henry's Law constant of 1.6X10'7 atm
mVmole at 25 °C (Meylan and Howard, 1991) indicates that thiourea is expected to be essentially nonvolatile from water
C-65
-------
APPENDIX C
surfaces (Lyman et al., 1990). Thiourea has been demonstrated to be resistant to biodegradation in a variety of standard
biodegradation tests (HSDB, 1998). Thiourea reached 2.6% of its theoretical biological oxygen demand over 2 weeks
in the Japanese MITI test using an activated sludge seed and an initial chemical concentration of 30 mg/L (Chemicals
Inspection and Testing Institute, 1992). In the OECD-screening test, 3% degradation was observed (Schmidt-Bleek et
al., 1982 as cited in HSDB, 1998) and 17% CO2 evolution was measured in a 5-day German GSF Biodegradation Test
(Rott et al., 1982 as cited in HSDB, 1998). Thiourea is stable to hydrolysis at environmental pHs (Schmidt-Bleek et al.,
1982 as cited in HSDB, 1998).
B.
Atmospheric Fate
If thiourea is released to the atmosphere, an extrapolated vapor pressure of 3.11X10"4 mm Hg at 25°C (Daubert and
Danner, 1992) indicates that thiourea will exist as a gas in the ambient atmosphere (Bidleman, 1988). The rate constant
for the gas-phase reaction of urea with photochemically-produced hydroxyl radicals has been estimated to be 4.2X10""
cmVmolecule-sec at 25° C (Meylan and Howard, 1993); this corresponds to a half-life of 9.2 hours.
C.
Terrestrial Fate
If thiourea is released to soil, an estimated Koc value of 2.8 (Meylan et al., 1992) indicates that thiourea is expected to
have very high mobility in soils (Swann et al., 1983). Thiourea is not expected to volatilize from moist soil surfaces
(Lyman et al., 1990) based upon its estimated Henry's Law constant (Meylan and Howard, 1991) or from dry soils based
on its vapor pressure. Biodegradation of thiourea by soil microorganisms may be an important fate process, although
microfiora activity may be suppressed for extended periods of time by high concentrations of this compound (HSDB,
1998). Degradation of thiourea was also observed in sterilized soils (Kolyada, 1969 as cited in HSDB, 1998) indicating
that abiotic degradation may be an important fate process.
D. Summary
If released into water, thiourea is. not expected to be'adsorb to suspended solids and sediment in the water column.
Bioconcentration in aquatic organisms and volatilization from water surfaces are not expected to be important fate
processes. Several biodegradation tests indicate that thiourea may be resistant to biodegradation. Thiourea is stable to
hydrolysis at environmental pHs. If released to the atmosphere, thiourea is expected to exist as a gas in the ambient
atmosphere based upon its extrapolated vapor pressure. Gas-phase thiourea is expected to be degraded in the
atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air has-been
estimated to be 9.2 hours. If released to soil, thiourea is expected to have very high mobility and therefore may leach
into groundwater. Volatilization from moist or dry soil surfaces is not expected to be an important fate process. Biotic
and abiotic degradation of thiourea may be important fate processes, however, no rates were available for these
processes. High concentrations of thiourea may suppress the activity of soil microorganisms for extended periods of
time.
C-66
-------
APPENDIX C
CHEMICAL SUMMARY FOR TIN
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for Tin.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of Tin are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF TIN
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc, .
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Difiusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
7440-31-5
Tin white
Sn
' Metal
118.69
231.9°C
2260°C
Insoluble
'7.31g/mL
no data
no data
no data
no data
.Flammable solid
no data
no data
no data
no data
no data
no data
no data
no data
Reference
Howard and Neal (1992)
Weast(I983)
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Weast (1983)
Weast(1983)
Weast (1983)
Weast (1983)
Budavari et al. (1996)
C-67
-------
APPENDIX C
CHEMICAL SUMMARY FOR TIN CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of tin chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF TIN CHLORIDE
Characteristic/Property
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
LogKow
Vapor Pressure
Reactivity
Data
7772-99-8
Tin (H) chloride
Stannous chloride
Cl2Sn
SnCl2
white orthorhombic crystals
189.615 ,
247°C
623°C'
approximately 600 g/L '
3.90 g/cm3
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be < lO'6 mm Hg
violent reactions with BrF3, CaC;, ethylene oxide,
Reference
CAS (1998)
Lide (1995)
Lewis (1993)
Sax (1984)
Lide (1995)
Lide (1995)
Lide (1995)
Lide (1995)
Lide (1995)
Estimated
Lide (1995)
SRC (1998)
SRC (1998)
SRC (1998)
Sax (1984)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diflusivity Constant
Henry's Law Constant
Fish Bioconccntration Constant
Odor Threshold
hydrazine hydrate, nitrates, K, Na, H2O2
no data .
no data
expected to dissociate into Sn2* and Cl"
no data
no data
no data; expected to be <1X10-S
no data
no data
SRC (1998)
SRC (1998)
1 Estimated from a reported solubility of 84 parts in 100 parts water (Dean, 1985).
II. ENVIRONMENTAL FATE
A. Aquatic Fate
Water hydrolyzes tin halides (Cotton and Wilkinson, 1980). Therefore, if tin chloride is released into water, it is
expected to dissociate into tin (Sn2+) and chloride (Cl') ions. In waters containing excess chloride ion, tin .chloride is
expected to dissolve, yielding SnCl3' (Cotton and Wilkinson, 1980). As a result, tin chloride is not expected to volatilize
from water surfaces or bioconcentrate in aquatic organisms. Ionic tin may adsorb to charged surfaces of suspended
C-68
-------
APPENDIX C
sediments and humic materials in the-water column (Evans, 1989). The chloride ion may complex.with heavy metals,
thereby increasing their solubility (Bodek et al., 1988). Adsorption of the chloride ion to suspended solids and sediment
in the water column is not expected to be an important fate process.
B. Atmospheric Fate
If tin chloride is released to the atmosphere, the low vapor pressure expected for an ionic salt indicates that it will exist
as a particulate. Dry deposition of tin chloride is expected to be the dominant fate process in the atmosphere (Arimoto,'
1989). The rate of dry deposition will depend on the prevailing winds and particle size (Bodek et al., 1988). Tin
chloride is expected to undergo wet deposition (Arimoto, 1989) in rain, snow, or fog due to its high water solubility
(Dean, 1985).
C.
Terrestrial Fate
Water hydrolyzes tin halides (Cotton and Wilkinson, 1980). Therefore, if tin chloride is released to soil, it is expected to
dissociate into its component ions in moist soils. Ionic tin may adsorb to charged surfaces of soil particles or form inner
sphere complexes with variable-charge soil surfaces (Evans, 1989) and therefore its rate of migration through soil may
be slow. The dissociation of tin chloride into its component ions in moist soils indicates that tin chloride is not expected
to volatilize from moist soil surfaces. The low vapor pressure expected for an ionic salt indicates that tin chloride is not
expected to volatilize from dry soil surfaces. Chloride is extremely mobile in soils (Bodek et al., 1988). The chloride
ion may complex with heavy metals, thereby increasing their solubility (Bodek et al., 1988) and potential for leaching
into groundwater.
D.
Summary
If released into water, tin chloride is expected to dissociate into tin and chloride ions. The dissociation of tin chloride
into its component ions indicates that tin chloride is not expected to volatilize from water surfaces or bioconcentration in
aquatic organisms. Ionic tin may adsorb to charged surfaces of suspended sediments and humic materials in the water
column. The chloride ion may complex with heavy metals, thereby increasing their solubility. Adsorption of the
chloride ion to suspended solids and sediment in the water column is not expected to be an important fate process. If
released to soil, tin chloride is expected to dissociate into its component ions in moist soils. The dissociation of tin
chloride into its component ions in moist soils indicates that tin chloride is not expected to volatilize from moist soil
surfaces. The low vapor pressure expected for an ionic salt indicates that tin chloride is not expected to volatilize from
dry soil surfaces. Ionic tin may adsorb to charged surfaces of soil particles or form inner sphere complexes with
variable-charge soil surfaces and therefore its rate of migration through soil may be slow. The chloride ion is extremely
mobile in soils; it may complex heavy metals, thereby increasing their solubility and the potential to leach into
groundwater. If released to the atmosphere, tin chloride is expected to exist as a particulate in the ambient atmosphere
based upon the low vapor pressure expected for an ionic salt. Wet and dry deposition will be the dominant fate process
in the atmosphere. The rate of dry deposition will depend on the prevailing winds and particle size.
C-69
-------
APPENDIX C
CHEMICAL SUMMARY FOR UREA
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of urea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF UREA
Characteristic/Property
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Mclling Point
Boiling Point
Water Solubility
Density
Vapor Density (air= 1)
Koc '
Log Row
Vapor Pressure
Reactivity
Flammability
Flash Point
Data
57-13-6
Carbamide
Carbonyldiamide
1 CH,N2O
H2NC(=0)NH2
Tetragonal prisms
60.06 •
132.7°C
decomposes
545g/Lat25°C
1.3230 g/cm3 at 20°C
not pertinent
8
-2.11
1.2X10J mm Hg at 25°C (extrapolated)
no reaction with water or common materials
not flammable •
not flammable
Reference
CAS (1998)
Lide (1995)
Budavari et al. (1996)
Lide (1995)
Budavari et al. (1996)
Budavari et al. (1996)
Lide (1995)
Lide (1995)
' Lide (1995)
Yalkowsky and Dannenfelser (1992)
Lide (1995)
Weiss (1986)
Hance (1965) as cited in HSDB (1998)
Hanschetal. (1995)
Jones (1960) as cited in PHYSPROP
(1998)
Weiss (1986)
Weiss (1986)
Weiss (1986)
Dissociation Constant
Molecular Difiusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
no data
no data
no data
no data; estimated to be less than 1X10"8
<10
not pertinent
PHYSPROP (1998)
Freitag et al. (1985) as cited in HSDB
(1998)
Weiss (1986)
II. ENVIRONMENTAL FATE
A. Aquatic Fate
If urea is released into water, a Koc value of 8 (Hance, 1965 as cited in HSDB, 1998) indicates that urea is not expected
to adsorb to suspended solids and sediment in the water column (Swann et al., 1983). According to a classification
scheme (Franke et al., 1994), a BCF of <10 in golden ide (Freitag et al., 1985 as cited in HSDB, 1998) indicates that
bioconcentration in aquatic organisms is low. An estimated Henry's Law constant of <1X10"8 atm mVmole at25°C
(PHYSPROP, 1998) indicates that urea is expected to be essentially nonvolatile from water surfaces (Lyman et al.,
C-70
-------
APPENDIX C
1990). In natural waters, biodegradation of urea is expected to be an important fate process; ammonia and carbon
dioxide have been identified as degradation products (HSDB, 1998). The rate of biodegradation is expected to decrease
with decreasing temperatures; at 8°C, negligible degradation was observed after incubation in river water for 14 days,
while at 20°C complete degradation was observed after 4 to 6 days incubation (Evans and Patterson, 1973 as cited in
HSDB, 1998). The presence of naturally-occurring phytoplankton in water is expected to increase the rate of
biodegradation (HSDB, 1998). Urea is used as an agricultural fertilizer (Lewis, 1993) and will be taken up by plants as
a source of nitrogen. Abiotic hydrolysis of urea occurs slowly yielding ammonium carbamate (HSDB, 1998). At5°C,
0.35% of urea hydrolyzed during a 10-day test period in demineralized/distilled water (Atkinson, 1971 as cited in
HSDB,1998).
B.
Atmospheric Fate
If urea is released to the atmosphere, a vapor pressure of 1.2X-10"5 mm Hg at 25°C (Jones, 1960 as cited in
PHYSPROP, 1998) indicates that urea will exist as both a particulate-and a gas in the ambient atmosphere (Bidleman,
1988). The rate constant for the gas-phase reaction of urea with photochemically-produced hydroxyl radicals has been
estimated to be 2.0X10'12 cmVmolecuie-sec at 25° C (Meylan and Howard, 1993); this corresponds to a half-life of 8.0
days. Particulate-phase urea is expected to be physically removed from the atmosphere by wet and dry deposition
(Arimoto, 1989).
C.
Terrestrial Fate
If urea is released to soil, it is expected to hydrolyze to ammonia through soil urease activity (HSDB, 1998). The rate of
hydrolysis can range from 24 hours to weeks depending upon soil type, moisture content, and urea formulation (Malhi
and Nyborg, 1979 as cited in HSDB, 1998). Urea is used as an agricultural fertilizer (Lewis, 1993) and will be taken up
by plants as a source of nitrogen. While no specific studies were identified in the literature, it is anticipated that urea will
biodegrade rapidly in soil as has been reported in water. A Koc value of 8 (Hance, 1965 as cited in HSDB, 1998)
indicates that urea is expected to have very high mobility in soils (Swann et al., 1983). Urea is not expected to volatilize
from soil surfaces based upon its vapor pressure and estimated Henry's Law constant.
D.
Summary
If released into water, urea is expected to be biodegraded yielding ammonia and carbon dioxide. Biodegradation is
expected to be more rapid in waters containing phytoplankton and during summer months when warmer water
temperatures prevail. Urea will be taken'up by plants and used as a source of nitrogen. Bioconcentration in aquatic
organisms, adsorption to suspended solids and sediment in the water column, and volatilization from water surfaces are
not expected to be important fate processes. If released to the atmosphere, urea is expected to exist as both a particulate
and as a gas based upon its vapor pressure. Gas-phase urea is expected to be degraded in the atmosphere by reaction
with photochemically-produced hydroxyl radicals; the half-life for this reaction in air has been estimated to be 8.0 days.
Particulate-phase urea is expected to be physically removed from the atmosphere by wet and dry deposition. The rate of
dry deposition will depend upon particle size and prevailing wind patterns. If released to soil, urea is expected to
hydrolyze to ammonia through the activity of soil urease as well as biodegrade as is the case in water. The rate of
hydrolysis can range from 24 hours to weeks depending upon soil type, moisture content, and urea formulation. Urea is
used an agricultural fertilizer as a source of nitrogen. Urea is expected to have very high mobility in soils and therefore
may leach into groundwater. Volatilization from moist and dry soil surfaces is not expected to be an important fate
process. • •
C-71
-------
APPENDIXC
REFERENCES
Abdelghani, A.A., A.C. Anderson, G.A. Khoury and S.N. Chang. 1990. Fate of ethylene glycol in the environment.
Louisiana Department of Transportation and Development in Cooperation with the US Department of Transportation
Federal Highway Administration. FHWA/LA-90/228.PB90-264672. Cited in HSDB, 1998.
Aldrich. 1996. Handbook Catalog of Fine Chemicals. Aldrich Chemical Co., Milwaukee, WI.
Aldrich. 1997. Handbook Catalog of Fine Chemicals. Aldrich Chemical Co., Milwaukee, WI.
Anbar M. and P. Neta. 1967. Title not available. Int JAppl Radiation Isotopes 18:493-523. Cited in HSDB, 1998.
Arimoto, R. 1989. Atmospheric deposition of chemical contaminants to the Great Lakes. J Great Lakes Res 15:339-
56.
Atkinson, R. 1971. Some aspects of the use of urea for deicing airfield runways. Water Treat Exam 20: 193-203.
Cited in HSDB, 1998.
Atkinson, R. 1988. Estimation of gas-phase hydroxyl radical rate constants for organic chemicals. Environ Toxicol
Chem 7:435-42.
Atkinson, R. 1989. Title not available. JPhys Chem Ref Data. Monograph No. 1. Cited in HSDB, 1998.
ATSDR (Agency for Toxic Substances and Disease Registry). 1990. Toxicological Profile for Ammonia. U.S.
Department of Health and Human Services, Public Health Service. Contract No. 205-88-0608. Chamblee, GA. 125pp.
ATSDR (Agency for Toxic Substances and Disease Registry). 1993. Toxicological Profile for Fluorides, Hydrogen
Fluoride, and Fluorine (F). ATSDR/TP-91-17. ATSDR, Atlanta, GA.
Austin, S. and A. Glowacki. 1989. Hydrochloric acid. In: Ullmann 's Encyclopedia of Industrial Chemistry, 5th
Edition, Vol A13. VCH Publishers, Deerfield Beach, FL. pp. 283-96.
Baker, S.C., D.P. Kelly and J.C. Murrell. 1991. Microbial degradation of methane sulphonic acid: a missing link in the
biogeochemical sulphur cycle. Nature 350: 627-28. Cited in HSDB, 1998.
Betterman, G., G. Reiss, and T. Hofmann. 1991. Phosphorus Compounds, Inorganic. In: Ullmann's Encyclopedia of
Industrial Chemistry. VCH Publishers, New York, NY. p. 531.
Betterton, E.A. 1992. Henry's Law constants of soluble and moderately soluble organic gases. In: Gaseous Pollutants:
Characterization and Cycling, J.O. Nriagu, Ed. John Wiley & Sons, Inc. pp. 1-50.
Bidleman, T.F. 1988. Atmospheric processes. Environ Sci Technol 22: 361-67. Cited in HSDB, 1998.
Bodek, I., WJ. Lyman, W.F. Reehl and D.H. Rosenblatt. 1988. Environmental Inorganic Chemistry. Pergamon Press,
Elmsford,NY. •
Boethling, R.S., P.H. Howard, W. Meylan, W. Stiteler, J. Beauman and N. Tirado. 1994. Group contribution method
for predicting probability and rate of aerobic biodegradation. Environ Sci Technol 28: 459-465.
Boublik, T., V. Fried and E. Hala. 1984. The Vapor Pressures of Pure Substances: Selected Values of the Temperature
Dependence of the Vapor Pressures of Some Pure Substances in the Normal and Low Pressure Region, Vol. 17.
Elsevier Sci. Publ., Amsterdam, Netherlands.
C-72
-------
APPENDIX C
Bridie, A.L., C. J.M.Wolff and M. Winter. 1979. BOD and COD of some petrochemicals. Water Res 13- 627-30
Cited in HSDB, 1998.
Budavari, S., MJ. O'Neil, A. Smith and P.E. Heckelman (Eds). 1989. The Merck Index, 11th Edition. Merck & Co.,
Inc., Rahway,NJ. p. 1360.,
Budavari, S., MJ. O'Neil, A. Smith, P.E. Heckelman and J.F. Kinneary (Eds). 1996. The Merck Index: An
Encyclopedia of Chemicals and Drugs, 12th Edition. Merck and Co., Inc., Whitehouse Station, NJ. '
Burrows, W.D. and R.S. Rowe. 1975. Title not available. J Water Pollut Control Fed47: 92-93 Cited in HSDB
1998.
Butler, J.A.V. and C.N. Ramchandani. 1935. The solubility of non-electrolytes. Part II. The influence of the polar
group on the free energy of hydration of ah'phate compounds. JChemSoc 952-55. Cited in Howard and Meylan, 1997.
Calvert, J.G. and J.N. Pitts. 1966. Photochemistry. John Wiley & Sons, Inc., New York, NY. pp 427-30 Cited in
HSDB, 1998.
Cappel, C.R. 1997. Silver Compounds. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4* Edition, Vol. 22:
179-95. VCH Publishers, Deerfield Beach, FL.
CAS Registry On-line. 1998.
Chapman, E.G., D.S. Sklarew and J.S. Flickinger. 1986. Organic acids in springtime Wisconsin precipitation samples.
Atmos Environ 20: 1717-27. Cited in HSDB, 1998.
CHEMFATE. 1995-1996. Syracuse Research Corporation's Environmental Fate Data Bases. Syracuse Research
Corporation, Syracuse, NY. .
Chemicals Inspection and Testing Institute. 1992. Biodegradation and Bioaccumulation Data of Existing Chemicals
Based on the CSCL Japan. Japan Chemical Industry Ecology - Toxicology and Information Center. ISBN 4-89074-
101-1 2-50. p. 2-104.
Cohn, J.G. and E.W. Stem. 1994. Gold and Gold Compounds. In: Kirk-Othmer Encyclopedia of Chemical
Technology, 4th Edition, Vol. 12. John Wiley & Sons, Inc., New York. pp. 738-767.
Cote, A.E. 1997'. Fire Protection Handbook, 18th Edition. National Fire Protection Association, Quincy, MA. p. 4-60.
Cotton, F.A. and F.R.S. Wilkinson. 1966. Advanced Inorganic Chemistry. Interscience Publishers, New York, NY.
Cotton, F.A. and G. Wilkinson. 1980. Advanced Inorganic Chemistry. A Comprehensive Text, 4* Edition. John
Wiley & Sons, New York, NY. p. 535.
Daubert, T.E. and R.P. Danner. 1985. Physical and Thermodynamic Properties of Pure Chemicals: Data Compilation.
Design Inst Physical Property Data. American Institute of Chemical Engineering. Hemisphere Pub. Corp., New York,
NY, '-,..-
Daubert, T.E. and R.P. Danner. 1987. Data Compilation Tables of Properties of Pure Compounds. American Institute
of Chemical Engineers, V5._ '
Daubert, T.E. and R.P. Danner. 1989. Physical and Thermodynamic Properties of Pure Chemicals: Data Compilation.
Design Inst Physical Property Data. American Institute of Chemical Engineering. Hemisphere Pub. Corp., New York,
NY. Cited in HSDB; 1998.
C-73
-------
APPENDIX C
Daubert, T.E. and R.P. Banner. 1991. Physical and Thermodynamic Properties of Pure Chemicals: Data Compilation.
Design Inst Physical Property Data. American Institute of Chemical Engineering. HemispherePub. Corp, New York,
NY. 5 Vol.
Daubert, T.E. and R.P. Danner. 1992. Data Compilation Tables of Properties of Pure Compounds. American Institute
of Chemical Engineers, V5.
Daugaut,etal. 1988. Title not available. MJChemKinet 20: 331-338. CitedinHSDB, 1998.
Dean,J.A. 1985. Lange's Handbook of Chemistry, 13th Edition. McGraw-Hill Company, New York, NY.
Dias.F.F. and M.Alexander. 1971. Title not available. Appl Microb 22: 1114-118. Cited in HSDB, 1998.
Dore, M., N. Brunet and B. Legube. 1975. Title not available. TribCebedeau 28:3-11. Cited in HSDB, 1998.
ECDIN (Environmental Chemicals Data and Information Network). 1998. On-lineaddress:
http://ulisse.etoit.eudra.org/Ecdin/E_hinfo.htmI.
EPI (Estimation Program Interface). 1994. Syracuse Research Corporation, Syuacuse, NY.
Evans, W.H. and S. Patterson. 1973. Water Res 7: 975-85. As cited in Environment Canada Tech. Info, for Problem
Spills: Urea p. 48. Cited in HSDB, 1998.
Evans, L.J. 1989. Chemistry of metal retention by soils. Environ Sci Techno! 23:1046-56.
Fee, D.C., D.R. Gard and C-H Yang. 1996. Phosphorus Compounds. In: Kirk-Othmer Encyclopedia of Chemical
Technology, 4th Edition, Vol.' 18. J.I. Kroschwitz and M. Howe-Grant (Eds): John Wiley & Sons, New York. p. 777.
Fischer, W.K., P. Gerike and R. Schmid. 1974. Combination method of successive tests and estimation methods for the
biological degradability of synthetic substances, i.e. organic complex compounds, based on common applicable
summary parameters (BOD, CO2, COD, TOC). (GER). Wasser-und-Abwasser-Forchung 7: 99-118. Cited in HSDB,
1998.
Fisher Scientific. 1985. Sodium citrate. Materials Safety Data Sheet.
Flick, E.W. (Ed). 1991. Industrial Solvents Handbook, 4th Edition. Noyes Data Corporation, Park Ridge, NY.
Franke, C., G. Studinger and G. Berger, et al. 1994. The assessment of bioaccumulation. Chemosphere 29: 1501-14.
Franke, C., G. Studinger, G. Berger, S. Boehling, U. Bruckmann, D. Cohors-Fresenborg and U.E. Joehncke. 1994. The
assessment of bioaccumulation. Chemosphere 29:1501-1514. Cited in HSDB, 1998.
Freitag, D., L. Ballhom, H. Geyer and F. Korte.. 1985. Environmental hazard profile of organic chemicals.
Chemosphere 14:1589-616. Cited in HSDB, 1998.
Gaffney, J.S., G.E. Streit, W.D. Spall and J.H. Hall. 1987. Beyond Acid Rain - Do Soluble Oxidants and Organic
Toxins Interact with SO2 and Nox to Increase Ecosystem Effects. Environ Sci Technol 21:519-23. Cited in HSDB,
1998.
Gard, D.R. 1996. Phosphoric acid and phosphates. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4"'
Edition, Vol 18. John Wiley & Sons, New York, NY. pp. 669-718.
Gmelin. 1998. Gmelin Inorganic Database, GMEL9704. Crossfire Searching Software. Beilstein Chemiedaten und
Software GmbH. Online May 28.
C-74
-------
APPENDIX C
Gordon, A.S. and FJ. Millero. 1985. Micro Ecol 11: 289-98. Cited in HSDB, 1998.
Covers, H., C, Ruepert, T. Stevens and CJ. Vanleeuwen. > 1986. Experimental determination and prediction partition
coefficients of thioureas and their toxicity to photobacterium.phosphoreum. Chemosphere 15-383-93 Cited in
PHYSPRDP.
Graedel, T.E., D.T. Hawkins and L.D. Claxton. 1986. Atmospheric Chemical Compounds: Sources, Occurrence, and
Bioassay. Academic Press, Inc., New York. p. 70.
Grafje, H., W. Kornigand H.M. Weitz, et al. 1985. Butanediols, Butenediol, and Biitylendiol. In: Ullmann's
Encyclopedia of Industrial Chemistry, 5* Edition, Vol. A4. VCH Publishers, Deerfield Beach, FL. pp. 455-62.
Gregory, G.L., R.C. Harriss, R.W. Talbot, R.A. Rasmussen, M. Garstang, M.O. Andreae, R.R. Hinton, E.V. Browell and
S.M. Beck. 1986. Title not available. J Geophys Res 91: 8603-612. Cited in HSDB, 1998.
Grosjean, D. 1992. Formic acid and acetic acid: Emissions, atomospheric formation and dry deposition at two Southern
California locations. Atmospheric Enviuronment 26:3279.
Gunther, F.A., W.E. Westlake and P.S. Jaglan. 1968. Res Rev 20: 1-148. Cited in PHYSPROP, 1998.
Hance, RJ. 1965. The adsorption of urea and some of its derivatives by a variety" of soils. Weed Res 5- 98 Cited in
HSDB, 1998. '
Hansch C, A. Leo, and D. Hoekman. 1995. Exploring QSAR: Hydrophobic, Electronic, and Steric Constants. ACS
Professional Reference Book. American Chemical Society, Washington, DC. Cited in HSDB, 1998.
Hartmann, W.R., M.O. Andreae and G. Helas. 1989. Title not available. Atmos Environ 23- 1531-33 Cited in
HSDB, 1998.
Hemphill, L. and W.S. Swanson. 1964. Proceed of the 18th Industrial Waste Conf, Eng Bull Purdue Univ Lafayette
IN. 18:204-17. Cited in HSDB, 1998. , ''.'-•
Heukelekian, H. and M.C. Rand. 1955. Biochemical oxygen demand of pure organic compounds. J Water Pollut
Control Assoc 27: 1040-53. Cited1 in HSDB,1998.
Hine, J. and P.K. Mookerjee. 1975. The intrinsic hydrophilic character of organic compounds: correlations in terms of
structural contributions. J Org Chem 40: 292-28.
Hoffman, W.A. and R.L. Tanner. 1986. Detection of Organic Acids in Atmospheric Precipitation. BNL-51922. NTIS
DE86 005294. Brookhaven National Laboratory. Environ Chem Div, Dept Appl Sci. pp.21. Cited in HSDB, 1998.
JHoward, P.H. 1989. Handbook of Environmental Fate and Exposure Data for Organic Chemicals, Vol. I. Lewis
Publishers, Chelsea, MI. p. 377-81. '.'..'.
Howard, P.H. and M. Neal. 1992. Dictionary of Chemical Names and Synonyms. Lewis Publishers, Chelsea, MI.
Howard, P.H. and W.M. Meylan. 1997. Handbook of Physical Properties of Organic Compounds. Lewis Publishers,
Chelsea, MI.
HSDB. (Hazardous Substances Data Bank). 1995-1996. MEDLARS Online Information Retrieval System, National
Library of Medicine, National Toxicology Program (via TOXNET), Bethesda, MD.
HSDB (Hazardous Substances Data Bank). 1998.. MEDLARS Online Information Retrieval System, National Library
of Medicine, National Toxicology Program (via TOXNET), Bethesda, MD. June.
C-75
-------
APPENDIX C
JARC. (International Agency for Research on Cancer). 1985. Hydrogen Peroxide. IN: IARC Monographs on the
Evaluation of Carcinogenic Risk of Chemicals to Humans: Allyl Compounds, Aldehydes, Eposides and Peroxides,
Vol.36. IARC,Lyon,France.
Jones, A.H. 1960. Sublimation-pressure data for organic compounds. JChemEngData 5: 196-200. Cited in HSDB,
1998.
Kameya, T., T. Murayama, K. Urano and M. Kitano. 1995. Title not available. Sci Tot Environ 170:43-51. Cited in
HSDB, 1998.
Keene, W.C. and J.N. Galloway. 1984. Title not available. Atmos Environ 18:2491-97. Cited in HSDB, 1998.
Keith, L.H. and D.B. Walters (Eds.) 1985. Compendium of Safety Data Sheets for Research and Industrial
Chemicals. VCH Publishers, Deerfield Beach. .
Khwaja,H.A. 1995. Title not available. Atmos Environ 29:127-39. Cited in HSDB, 1998.
Kolaitis, N., S.F.J. Bruynseel, R.E. Vangrieken and M.O. Andreae. 1989. Determination of methanesulfonic acid and
non-sea-salt sulfate in single marine aerosol particles. Environ Sci Technol 23: 236-40. Cited in HSDB, 1998.
KoIyada,T.I. 1969. Decomposition of thiourea in the soil. VestsiAkadNavukBelarus 3: 36-40. Cited in HSDB,
1998.
Lema, J.M., R. Mendez and R. Blazquez. 1988. Title not available. Water Air Soil Pollut 40: 223-250. Cited in
HSDB, 1998.
Lewis, R.J. 1993. Hawley's Condensed Chemical Dictionary, 12th Edition. Van Nostrand Reinhold Company, New
York, NY.
Lide, D.R. 1995. CRC Handbook of Chemistry and Physics, 76th Edition. CRC Press, Boca Raton, FL.
Lockheed Martin (Lockheed Martin Energy Systems, Inc.). 1994. Material Safety Data Sheet for Sodium
Tetrafluoroborate.
Lockheed Martin (Lockheed Martin Energy Systems, Inc.). 1991. Materials Safety Reference Sheet.
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt. 1990. Handbook of Chemical Property Estimation Methods.
American Chemical Society, Washington, DC. Cited in HSDB, 1998.
Malaney, G.W. and R.M. Gerhold. 1969. Structural determinants in the oxidation of aliphatic compounds by activated
sludge. J Water Pollut Control Fed41: R18-R33. Cited in HSDB, 1998.
Malhi, S.S. and M. Nyborg. 1979. Rate of hydrolysis of urea (in soils) as influenced by thiourea and pellet size. Plant
Soil 51:177-86. Cited in HSDB, 1998.
Manahan, S.E. 1991. Environmental Chemistry, 5* Edition. Lewis Publishers, Chelsea, MI.
MawsonA.J.,R.I.EarleandV.F.Larsen. 1991. Title not available. Wat Res 25: 1549-54. Cited in HSDB, 1998.
Mazurek, M.A. and B.R.T. Simoneitt. 1986. Organic components in bulk and wet-oily precipitation. CRC Critical
Review. 'Environ Control 16:140. Cited in HSDB, 1998.
McDermott, J. 1974., Plating of Plastics With Metals. Noyes Data Corporation. Park Ridge, New Jersey.
C-76
-------
APPENDIX C
McGahey, C. and E.J. Bouwer. 1992. Biodegradation of ethylene glycol in simulated subsurface environments Wat
Sci Technol 26: 41-49. Cited in HSDB, 1998.
Meylan, W.M. and P.H. Howard. 1991. Bond contribution method for estimating Henry's Law constants Environ
ToxicolChem 10: 1283-1293. '
Meylan, W.M., P.H. Howard and R.S. Boethling. 1992. Molecular topology/fragment contribution method for
predicting soil sorption coefficients. Environ Sci Technol 26: 1560-67.
Meylan, W.M. and P.H. Howard. 1993. Computer estimation of the atmospheric gas-phase reaction rate of organic
compounds with hydroxyl radicals and ozone. Chemosphere 26: 2293-2299.
Meylan, W.M., P.H. Howard, D. Aronson, H. Printup and S. Gouchie. 1997. Improved Method for Estimating
Bioconcentration Factor (BCF) from Octanol-Water Partition Coefficient. Third Update Report Prepared for
Boethling, R.S., U.S. EPA.
Mills, E.J. and V.TJ. Stack. 1955. Title not available. Sewage Indust Wastes 27: 1061-1064. Cited in HSDB, 1998.
MSDS. Material Safety Data Sheet Archive at University of Vermont (www.siri.org, site 3).
Ohm, R.F. 1997. Rubber Chemicals. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4th Edition., Vol. 21.
John Wiley & Sons, New York, NY. pp. 460-81.
Osol, A. (Ed). 1980. Sodium citrate. In: Remington's Pharmaceutical Sciences, 16th Edition. Mack Printing Co.,
Easton, PA. p. 769.
Perrin, D.D. 1972. Dissociation Constants of Organic Bases in Aqueous Solution. IUPAC Chem Data Series,
Supplement 1972. Buttersworth, London. , .
PHYSPROP. 1998. Syracuse Research Corporation's Physical Properties Database in ISIS/Base format.
Fitter, P. 1976. Determination of biological degradability of organic substances. Water Res 10-231-35 Cited in
HSDB, 1998. . .
Placak, O.R. and C.C. Ruchhoft. 1947. Studies of Sewage purification, XVII. The utilization of organic substrates by
activated sludge. Sewage Works J19: 423-40. Cited in HSDB, 1998.
Prager, J.C. 1995. Environmental Contaminant Reference Databooic, Vol. I. Van Nostrand Reinhold New York,
NY.
Price, K.S., G.T. Waggy and R.A. Conway. 1974. Brine shrimp bioassay and seawater BOD of petrochemicals. J
Water Pollut Control Fed 46: 63-77. Cited in HSDB, 1998.
Renner, H. and M. Johns. 1989. Gold, Gold Alloys, and Gold Compounds. In: Ullman's Encyclopedia of Industrial
Chemistry, 5th Edition, Vol. 12. VCH Publishers, New York, NY. pp. 499-533.
Renner, H. 1993. Silver, Silver Compounds, and Silver Alloys. In: Ullmann 's Encyclopedia of Industrial Chemistry,
5* Edition, Vol. A24. VCH Publishers, Deerfield Beach, FL., pp. 107-67.
Riddick, J.A., W.B. Hunger and T.K. Sakano. 1986. Organic Solvents: Physical Properties and Methods of
Purification. Techniques of Chemistry, 4* Edition. Wiley-Interscience. New York, NY
Rott, B., R. Viswanathan, D. Freitag and F. Korte. 1982. Comparison of the applicability of various tests for screening
the degradability of environmental chemicals. Chemosphere 11: 531-8. Cited in HSDB, 1998.
C-77
-------
APPENDIX C
Sansone, J.F., C.C. Andrews and M.Y. Okamoto. 1987. Title not available. Geochimica et Cosmochimica Acta 51:
1889-1896. Cited in HSDB, 1998.
Sax,N.I. 1984. Dangerous Properties of Industrial Materials, 6th Edition. Van Nostrand Reinhold Company, New
York,NY. . '
Sax, N.I. and RJ. Lewis, Sr. 1987. Rowley's Condensed Chemical Dictionary. Van Nostrand Reinhold Company,
New York, NY. .
Schmidt-Bleek, F., W. Haberland, A.W. Klein and S. Caroli. 1982. Steps towards environmental hazard assessment of
new chemicals (including a hazard ranking scheme, based upon directive 79/83 I/EEC). Chemosphere 11: 383-415.
Cited in HSDB51998.
Serjeant, E.P. and B. Dempsey. 1979. lonisation Constants of Organic Acids in Aqueous Solution. IUPAC Chem Data
Scr No. 23. Pergamon Press, New York, NY. p. 989. Cited in HSDB, 1998.
Sigma-Aldrich Corporation. 1992. Material Safety Data Sheet for Sodium Tetrafluoroborate. Sigma-Aldrich,
Milwaukee, WI.
SRC (Syracuse Research Corporation). 1998. Estimated value based on structure-activity relationships.
Stuermer, D.H., DJ. Ng and C.J. Morris. 1982. Title not available. Environ ToxicolChem 16: 582-87. Cited in
HSDB, 1998.
Sunderman, Jr., F.W. 1989. Mechanisms of nickel carcinogenesis. Scand J Work Environ Health 15: 1-12. Cited in
ATSDR.1997.
Sunderman, Jr., F.W. and A. Oskarsson. 1991. Nickel. In: Metals and Their Compounds in the Environment, E.
Merian,Ed. VCH Verlagsgesellschaft, New York, NY. pp. 1101-1126. Cited in ATSDR, 1997.
Swann, R.L., D.A. Laskowski, P.J. Mccall, K. Vanderkuy and H.J. Dishburger. 1983. A rapid method for the
estimation of the environmental parameters octanol/water partition coefficient, soil sorption constant, water to air ratio,
and water solubility. Res Rev 85: 23. Cited in HSDB, 1998.
Swindol, C.M., C.M. Aelion, D.C. Dobbins, O. Jiang and F.K. Pfaender. -1988. Title not available. Environ Toxicol
Chem.7:291-99. Cited in HSDB, 1998.
Takemoto, S., Y. Kugeand-M.Nakamoto. 1981. Title not available. Suishitsu Okaku Kenkyu 4: 80-90. Cited in
HSDB, 1998.
TRI93. 1995. Toxics Release Inventory, Public Data Release: Office of Pollution Prevention and Toxics, U.S. EPA,
Washington, DC.
U.S. Air Force. 1990. Copper - Elemental Copper. In: The Installation Restoration Toxicology Guide, Vol. 5. Wright-
Patterson Air Force Base, OH. pp: 77-1 - 77-44.
U.S. EPA. 1981. Treatability Manual I. Treatability Data. -EPA-600/2-82-001A. U.S. EPA, Washington, DC.
U.S. EPA. 1987. Drinking Water Criteria Document for Copper. Environmental Criteria and Assessment Office,
Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH.
ECAO-CIN-477.
U.S. EPA. 1993. Reregistration Eligibility Document (RED): Peroxy Compounds. EPA 738-R-93-030. Office of
Pesticide Programs, Special Review and Reregistration Division.
C-78
-------
APPENDIX C
Verschueren, K. 1996. Handbook of Environmental Data on Organic Chemicals, 3rd Edition. Van Nostrand
Reinhold, New York, NY.
Von Oepen B., W. Koerdel and W. Klien. 1991. Title not available. Chemosphere 22: 285-304. Cited in HSDB
1998. , '
Weast, R.C. 1983. CRC Handbook of Chemistry and Physics. CRC Press, Boca Raton, FL.
Weast, R.C. (Ed). 1983-1984. CRC Handbook of Chemistry and Physics, 64th Edition. CRC Press, Inc., Boca Raton,
Florida, p. B-139.
Weast, R.C., MJ. Astle and W.H. Beyer. 1985. CRC Handbook of Chemistry and Physics, 66* Edition. CRC Press,
Boca Raton, FL. p. D-245.,
Weast, R.C. 1986. CRC Handbook on Organic Compounds. (Cited in HSDS 1959).
Weiss, G. 1986. Hazardous Chemicals Data Book, 2nd Edition. Noyes Data Corporation, Park Ridge, NJ.
Winkeler, H.D., U. Puttins and K. Levsen. 1988. Title not available. Vom Wasser 70:107-17. Cited in HSDB, 1998.
Yalkowsky, S.H. and R.M. Dannenfelser. 1992. Aquasol Database of Aqueous Solubility, Version 5. College of
Pharmacy, Univ. of Ariz, Tucson, AZ. PC Version. .
Yaws, C.L. 1994. Handbook of Vapor Pressure, Vol. 1 - Cl to C4 Compounds. Gulf Publishing Co, Houston, TX.
Zahn, R. and H. Wellens. 1980. Title not available. Z Wasser Abwasser Forsch 13: 1-7. Cited in HSDB, 1998.
C-79
-------
-------
Appendix D
Supplemental Exposure
Assessment Information
-------
-------
APPENDIX D
Technical Memorandum RE: Modeling Worker Inhalation Exposure
D-l
-------
APPENDIX D
D.I Technical Memorandum RE: Modeling Worker Inhalation Exposure
»
TECHNICAL MEMORANDUM
TO: Debbie Boger
PWB Project File, EPA # X823941-01-0
cc: Lori Kincaid, Jack Geibig, Dean Menke, Diane Perhac
FROM: Bruce Robinson, Chris Cox, Nick Jackson, Mary Swanson
DATE: December 22,1995 (Revised 8/96)
RE: MODELING WORKER INHALATION EXPOSURE
I.
INTRODUCTION
This technical memorandum is submitted for review by the RM2 work group. Air transport
models to estimate worker inhalation exposure to chemicals from printed wiring board (PWB)
making holes conductive (MHC) lines are presented here for review and comment. The purpose
is to reach agreement on our technical approach before proceeding with further analysis.
Three air transport models will be required to estimate worker exposure:
• Volatilization of chemicals induced by air sparging.
• Aerosol generation induced by air sparging.
• Volatilization of chemicals from the open surface of MHC tanks.
The total transport of chemicals from the air-sparged baths will be determined by summing the
releases calculated using each of the three models described above. Air-sparged baths include the
electroless-copper baths and some cleaning tanks. Only the third model will be applied to
determine the atmospheric releases of chemicals from unsparged baths. This document includes a
review of the relevant literature, descriptions of the models, and examples demonstrating the
proposed use of the models. The results of the model calculations will be compared to available
occupational monitoring data.
D-2
-------
APPENDIX D
II. VOLATILIZATION OF CHEMICALS FROM AIR-SPARGED PWB
MANUFACTURING TANKS
Mixing in plating tanks, e.g., the electroless copper plating tank, is commonly accomplished by
sparging the tank with air. This is similar to aeration in wastewater treatment plants, and the
volatilization of chemicals from these plants has been the focus of recent research. The
volatilization models used in that research are based on well accepted gas transfer theory,
discussed below.
Background
Volatilization of chemicals from water to air has been investigated by many researchers (Liss and
Slater, 1974; Smith et al, 1980; Roberts, 1983; Peng et al, 1993). In PWB manufacturing,
volatilization due to air sparging of process tanks is expected to be one of the main pathways for
contaminant transfer to the air. In bubble aeration systems, the volatilization rate is dependent
upon the volumetric gas flow rate, partial pressure of the gas, and the mass transfer rate
coefficient (Matter-Muller, 1981). The volatilization characteristics for different diffuser types
and turbulent conditions were evaluated by Matter-Miiller (1981), Peng (1995), and Hsieh
(1994). . . .
Volatilization from aerated systems has been mainly quantified using the two-film theory (Cohen
et al, 1978; Mackay and Leinonen, 1975). This work is discussed below and is used to model
chemical transfer rates from air-sparged PWB process tanks. The main assumption of the theory
is that the velocity at a fluid interface is zero. Molecular diffusion across the interfacial liquid film
is the limiting factor for mass transfer to the air, and it is used to develop a simple equation
relating the overall mass transfer coefficient to the diffusion coefficient of the chemical hi water.
The two-film model of gas transfer was expanded to include mass transfer in diffused aeration
systems (Matter-Miiller et al, 1981). Matter-Miiller et al. assumed that the system was
isothermal, hydraulic conditions were steady, and that pressure and volume changes within the
bubbles were negligible. Further, an overall mass transfer coefficient was applied to represent
transfer of contaminants to the bubble as they rose through the homogeneous liquid volume.
Parker (1993) demonstrated that liquid-phase concentration can be assumed constant during the
rise time of the bubble. Under these assumptions, Matter-Muller et al. derived the following
relationship predicting the mass transfer rate from an aerated system:
I exp
(1)
where:
L y
= mass transfer rate of chemical j> out of the system by sparging (m/t)
= gas flow rate (!3/t)
= dimensionless Henry's constant for chemical y
= concentration of chemical y in bulk liquid (m/13)
= overall mass transfer coefficient for chemical y (1/t)
D-3
-------
APPENDIX D
a =interfacial area of bubble per unit volume of liquid (I2/!3)
VL -volume of liquid (I3)
The overall mass-transfer coefficient is defined as the inverse sum of the reciprocals of the liquid
and gas-phase mass transfer coefficients; but, because molecular diffusion of oxygen and nonpolar
organic substances is 103 times greater hi air than in water (Matter-Muller e? al, 1981), it is set
equal to the liquid phase coefficient only. The mass transfer coefficient of a chemical can then be
related to oxygen using the following equation:
a
(L,<2 •
(2)
where:
D
D
O2
= molecular diffusion coefficient for chemical y hi water (!2/t)
= molecular diffusion coefficient for oxygen in water (!2/t)
= 2.1x1 0'5 cm2/cm @ 25° C (Cussler, 1984)
= overall mass transfer coefficient for chemical y (1/t)
KoL.02 = overall mass transfer coefficient for oxygen in water (1/t)
The value of KoU02 at 25°C in diffused aeration systems can be estimated using a correlation
developed by Bailey and Ollis (1977):
1/3
a
a
(3)
where:
PH20
Patr
g
Hmo
= bubble diameter (1)
= density of water (m/13)
= density of air (m/13)
= gravitational constant (1/t2)
= viscosity of water (m/l-t)
If a measured value of Dy is not available, then it can be calculated from the Hayduk and Laudie
correlation (Lyman et al., 1982):
2 13 .26 xlO
/sec ) =
-5
l.tt
BO
0.59
(4)
where:
Vm = molar volume of solute (cmVmol)
umo = viscosity of water (centipoise)
D-4
-------
APPENDIX D
The mass transfer coefficient can be corrected for the bath temperature (°C) as follows
(Tschabanoglous, 1991):
C 1-024
(5)
Bailey and Ollis (1977) developed a relationship for the interfacial area per unit volume (a) as a
function of the bubble diameter, gas flow rate, and tank geometry:
6 Q n t
G 'b
(6)
where:
h = tank depth (1); and
HO
(7)
Values of Hy are often reported at 25°C. The Henry's constant can be corrected to the bath
temperature using the van't Hoff equation:
' y,T " y,-S"(fxf
273 . IS +T
(8)
where:
s = enthalpy of the chemical in the gas phase (cal/mol)
AHaq = enthalpy of the chemical in the aqueous phase (cal/mol)
R = gas constant (1.987 cal/mol-K)
Matter-Muller (1981) concluded that surfactants do not significantly alter the rate of volatilization
from the water. Some agents did lower the overall mass transfer coefficient, but most showed no
appreciable difference. This was attributed to an increase in the specific interfacial area, a, when
the interfacial energy, or mass transfer coefficient, was decreased. The transfer rate of volatile
organic compounds (VOCs) was found to depend heavily upon the type of aerators used, and the
degree of saturation of the bubbles rising through the liquid.
III. AEROSOL GENERATION FROM BATHS MIXED BY SPARGING WITH AIR
Aerosols or mists have been identified as a major source of contaminants released by
electroplating baths to the atmosphere (Burgess, 1981) and should be investigated as a potential
source of contaminants from electroless baths. At least two sources of aerosols exist in
electroplating baths: 1) aerosols generated due to liquid dripping from parts as they are removed
from the bath (drag-out drips); and 2) aerosols generated due to bursting of the bubbles at the
—., , D-5
-------
APPENDIX D
surface. Drag-out drips are insignificant compared to other sources of aerosols (Berglund and
Lindh, 1987; Cooper etal., 1993). •
Bubbles in electroplating baths can originate from the dissociation of water at the electrode, or
mixing of the bath via air sparging. Bubbles in other plating baths (e.g., electroless plating baths)
can originate from reactions in the bath or mixing of the bath via air sparging. The rate of aerosol
generation per unit bubble volume decreases with increasing bubble size. Bubbles generated, by
water dissociation are typically smaller than those generated by air sparging; therefore, aerosol
generation in electroless plating processes may be less significant than in electroplating operations.
The focus of this memo is aerosols generated by air sparging. Except for the conductive polymer
and non-formaldehyde electroless alternatives, MHC processes hi PWB manufacturing do not use
electroplating and therefore would not dissociate water to form gas bubbles. Information
collection is continuing to allow prediction of aerosol formation in MHC processes that do have
an electroplating step. Importantly, Berglund and Lindh (1987) report that aerosol generation
from electroplating tanks is greatly reduced by sparging; the relatively large air bubbles formed
during air sparging coalesce the smaller bubbles formed by hydrolysis and electroless plating
reactions.
To estimate the emission of contaminants resulting from aerosols, the rate of aerosol generation
and the concentration of contaminant in the aerosol are required. Limited information concerning
the rate of aerosol formation was found in the literature. The following sources were consulted:
• U.S. EPA (1991). Chemical Engineering Branch Manual for the Preparation of
Engineering Assessments.
• Chemical Abstracts, 1986 to date.
• Current and past text books ha air pollution, chemical engineering, and water and
wastevvater treatment.
• Perry's Handbook (1984) related to entrainment hi distillation trays.
• The last five years of Water Environment Research and ASCE Journal of the
Environmental Engineering Division.
• A title key-word search of holdings in the library of the University of Tennessee.
• The ASPEN model commonly used for modeling chemical manufacturing processes. (It
was found that any aerosol formation routines within ASPEN would be relevant to
entrainment in devices such as distillation trays and not relevant to sparging of tanks.)
• The manager of the US EPA Center for Environmental Assessment Modeling in Athens,
Georgia, as well as an expert in the Air and Energy Lab - Emission Modeling Branch in
North Carolina.
D-6
-------
APPENDIX D
In this work, the aerosol formation rates will be predicted based upon limited measurements of
aerosol generation in electroplating (Berglund and Lindh, 1987) and other air-sparged baths
(Wangwongwatana et al., 1988; Wangwongwatana et aL, 1990) found in the literature.
Berglund and Lindh (1987) developed several graphs relating aerosol generation to air sparging
rate (Figure la), bath temperature (Figure Ib), air flow rate above the bath (Figure Ic), and
distance between bath surface and the tank rim (Figure Id). Using Figures la-Id, the following
relationship may be developed:
= 5 .5r/0 (Q G I A \+0 .01
TFA F'D
(9)
where:
RA = aerosol generation rate (ml/min/m2)
QG/A - air sparging rate per unit bath area (1/min/m2)
FT = temperature correction factor
FA = air velocity correction factor
FD — distance between the bath surface and tank rim correction factor
Wangwongwatana et al. (1988) presented figures relating the number of aerosol droplets
generated as a function of air flow rate, bubble rise distance, bubble size, and colloid
concentration (Figure 2). Droplet size distribution measurements by these researchers indicate
volume mean diameters of 5 to 10 urn. The aerosol generation rate can be calculated using the
following equation:
(10).
where:
Cd
Vd
A
droplet concentration (I"3)
droplet volume (1)
bath area (I2)
Contaminants may be present in aerosols at elevated concentration relative to the bath
.concentration. Colloidal contaminants may be collected on the bubble surface as it rises through
the bath. As the bubble bursts, the contaminants on the bubble surface are incorporated into
aerosols. Wangwongwatana et al. (1990) report that in their experiments about one in two
aerosols contain polystyrene latex spheres, compared to about one in 250 expected based upon
the concentration of latex sphere in the bath. Organic contaminants may also partition at the air-
water interface. A correlation for the water-interface partitioning coefficient for nonpolar:
compounds, kiw, defined as the ratio of the mass of contaminant per unit area of interface to the
mass of contaminant per unit volume of water is given by Hoff et al. (1993):
log
-0 .769
log
S'
c w
(ID
D-7
-------
APPENDIX D
where:
Csw = saturated aqueous solubility of the contaminant.
For more polar compounds a more complicated relationship is required:
log
= -7. SOS -hog
-
-------
APPENDIX D
IV. VOLATILIZATION OF CHEMICALS FROM THE OPEN SURFACE OF MHC
TANKS .
Most plating tanks have a free liquid surface from which chemicals can volatilize into the
workplace air. Air currents across the tank will accelerate the rate of volatilization. The model
presented hi the Chemical Engineering Branch Manual for the Preparation of Engineering
Assessments (CEBMPEA) (US EPA, 1991) has potential application in this case. Some
limitations of the model should be pointed out. The model was developed to predict the rate of
volatilization of pure chemicals, not aqueous solutions. The model was also validated using pure
chemicals. As a result, the model implicitly assumes that mass transfer resistance on the gas side
is limiting. The model may fail in describing volatilization of chemicals from solutions when
liquid-side mass transfer controls.
CEBMPEA models the evaporation of chemicals from open surfaces using the following model:
Fy;0 = 2cL>yHyA[Dy>airv/(7uz)]0-5 ' ' (16)
where: .
Fyo = volatilization rate of chemical y from open tanks (m/t)
Dyair = molecular diffusion coefficient of chemical y in air (!2/t)
vz = air velocity (1/t)
z = distance along the pool surface (1)
The value of vz recommended by CEBMPEA is 100 ft-min1'. The value of Dyair can be estimated
by the following formula (US EPA, 1991):
(17)
where:
Dyair = molecular diffusion coefficient of chemical y in air (cm2/s)
T = ah- temperature (K)
M = molecular weight (g/mol)
Pt .= total pressure (atm)
This equation is based on kinetic theory and generally gives values of Dyajr that agree closely with
experimental data.
V. CALCULATION OF CHEMICAL CONCENTRATION IN WORKPLACE AIR
FROM EMISSION RATES
The indoor air concentration will be estimated from the following equation (US EPA, 1991):
Cy = FyJ/(VRRvk)
D-9
-------
APPENDIX D
where:
Rv
k
= workplace contaminant concentration (m/13)
= total emission rate of chemical from all sources (m/t)
= room volume (!3/t)
= room ventilation rate (t"1)
= dimensionless mixhig factor
The mixing factor accounts for slow and incomplete mixhig of ventilation air with room air.
CEBMPEA sets this factor to 0.5 for the typical case and 0.1 for the worst case. CEBMPEA
commonly uses values of the ventilation rate Q from 500 ftVmin to 3,500 ftYmin. Appropriate
ventilation rates for MHC lines will be chosen from facility data and typical industrial
recommendations. •
VI. EXAMPLE MODELING OF FORMALDEHYDE RELEASE TO ATMOSPHERE
FROM AIR-SPARGED ELECTROLESS COPPER BATH .
In the examples below, the values of some parameters are based upon a site visit to SM
Corporation hi Asheville, NC. Except where stated otherwise, final values of the various-
parameters used in the models will be chosen based on the results of the Workplace Practices
Questionnaire, chemical suppliers information, site visits, and performance demonstrations. All
parameter values are based on preliminary information and are subject to change.
Values of site-specific parameters assumed in the example
Tank volume = 242 L
Tank depth = 71 cm
Tank width = 48 cm
Tanklength = 71 cm
Air sparging rate = 53.80 L/min
Tank temperature = 5 l'.67°C
H2CO Concentration hi tank = 7,000 mg/L
Bubble diameter at tank surface = 2.00 mm
Room length = 20 m
Room width = 20 m
Room height = 5 m
Air turnovers/hour = 4 hr"1
Air velocity across tank surface = 0.508 m/s
Dimensionless mixhig factor = 0.5
Site visit to SM Co., Asheville, NC
Assumed
Assumed
Assumed
Midpoint of values given in Perry's Handbook,
1985, pg 19.13
Site visit to SM Co., Asheville, NC
Product data sheets
Assumed
Assumed
Assumed
Assumed
Assumed
Default recommended by US EPA, 1991
Default recommended by US EPA, 1991
Volatilization induced by air sparging
Calculating overall mass transfer coefficient for oxygen in -water:
D-10
-------
APPENDIX D
Ka,o2 •= °-31 *
a
HO D a
where:
PH20
Pgas
g
= 0.0113 cm/sec
= 0.678 cm/min
= 0.2 cm
= 0.997 g/cm3 (Dean, 1985)
= 0.00118 g/cm3 (Dean, 1985)
= 980 cm/sec2
= 0.0089 (g/cm-sec) (Dean, 1985)
= 2.1xlO'5 cm2/sec (Cussler, 1984)
Calculating molecular diffusion coefficient of formaldehyde in -water:
13 .26 x/0
= 1.81xlO-5cm2/sec
= 36.8 cmVmol
where:
m
M-H20 ~ 0.89 centipoise
Calculating mass transfer coefficient of formaldehyde in water:
CLy
(2
a,a
1 .81 I/O
2 . 10 I/O
* 0 .678
= 0.584 cm/min
Correcting mass transfer coefficient for temperature:
KoL,y,5i.67 = KO^^ 1.024(T'25) = 0.584* 1.024(51-67-25> = 1.10 cm/min
Calculating tb:
= 0.291 sec
= 4.85xlO"3min
D-ll
-------
APPENDIX D
where:
• h = 71 cm
Calculating interfacial area per unit volume:
Q r>
G 'b
•• 0.0323 cm2/cm3
where:
QG
VL
= 53,800 cmVmin
= 242,000 cm3
Correcting Henry's constant for temperature:
y,S. .67
298 . 15 273 . 15 +r
where:
DH,,
R
= 1.99xlO"5 (dimensionless)
= IJxlO"7 atm-mVmol (Risk Assistant, 1995)
= 6.38X10"6 (dimensionless)
= -27,700 cal/mol
=-35,900 cal/mol
=1.987cal/mol-K
Calculating mass transfer rate of formaldehyde by air sparging:
I exp
= 7.49mg/min .,
The argument of the exponential function is -8031. This indicates that the formaldehyde
concentration in the air bubbles is essentially in equilibrium with the bath concentration.
Transport in aerosols
The aerosol generation rate will be estimated using data presented by both Berglund and Lindh
(1987) and Wangwongwatana et al. (1988).
Calculating aerosol generation rate using Berglund and Lindh (1987) data:
— D-12
-------
APPENDIX D
where:
Qo/A
FT
= O.Oi87mL/min/m2
= (53.8*10,000)7(71*48) =158 (L/min/m2)
= 0.95 @ 51.67°C (Figure Ib)
= 1.2 @ 0.508 m/s (Figure 1 c)
= 1.0 assumed (Figure Id)
Calculating aerosol generation rate using Wangwongwatana et al. (1988) data:
The air sparging rate used hi the example (53.8 L/min) must be converted to an equivalent rate in
the experimental apparatus using the ratio of the area of the example bath (0.341 m2) to the area
of the experimental apparatus (0.123 m2). The equivalent rate is 19.4 L/min. The bubble rise
distance would be approximately 0.6 m. From Figure 2, it can be inferred that the droplet
concentration is not much greater than 100 droplets/cm3. The aerosol generation rate can now be
calculated:
= 8.27xlO'3mi/m2/min
where:
QG
cd
Vd
dd
A
53800 cmVmin
100 droplets/cm3
(p/6)dd3 = 5.24xlO-10cm3
0.001 cm (upper end of range reported by Wangwongwatana et al., 1988)
0.341m2 " :
The aerosol generation rates calculated by the two methods agree quite well. The model of
Berglund and Lindh (1987) will be used because it gives a slightly greater generation rate and is
easier to use.
Emission rate from bath. If it is assumed that the formaldehyde concentration in the aerosols is
equal to the bath concentration (7 mg/mL) then the formaldehyde emission rate is:
Fy>a = (7 mg/mL) • (0.0187 mL/m2/min) • (0.341 m2) = 4.46xlO'2 mg/min
To determine if accumulation of the contaminant at the air-water interface is significant, k!W must
be estimated using Equation 11. Since formaldehyde is a gas at the temperatures of interest,
interfacial tension data are not available; however, average values of other aldehydes may be used
(Hoff et al., 1993). Calculation of k,w@25°C is summarized below; information was not available
for calculating k]W at other temperatures. *
__
-------
APPENDIX D
where:
k*
Yw
R
= -7 .SOS +log
2-303 KT
k,,
= -6.848
- 1.452 Method 1, page 11-10 in Lyman et al. (1982)
= 9.35xl08 cm2/mol Calculated from: as = 8.45xl07 V/3
= 8.314x10 7erg/molK
= 72 dyne/cm Hoff et al. (1993)
= 21.9 dyne/cm Value for acetaldehyde, Weast, 1980
= 14.6 dyne/cm Average value for n-heptaldehyde and benzaldehyde, Girfalco
and Good, 1957
= 1.418x10'7 cm
Formaldehyde emissions due to aerosols can now be calculated:
Calculating the ratio of contaminant mass sorbed at the air-water interface to mass in gas
volume of bubble:
= 0.2138
Calculating fraction of bubble interface ejected as aerosols:
.KA * "b
Q a 'b
4.35xlO'3
where:
lb =5xlO'7 cm (Rosen, 1978)
Calculating formaldehyde mass transfer rate via aerosols from tank to the atmosphere:
M
M
= 0.00697 mg/min
Volatilization from open tanks
Calculating molecular diffusion coefficient of formaldehyde in air:
D-14
-------
APPENDIX D
Dy>air = 4.09xlO'5 T1'9 (1/29.
= 0.174 cnrVsec
M
-°'33
Pt
where:
T = 298.15 K
M = 30.03 g/mol
Pt =1 atm . • .
Calculating volatilization rate of formaldehyde from open tanks:
Fy>0 = 2cLjyHyA[Dy>airv2/(pZ)f5
= 13.8 mg/min
where:
Dy air = molecular diffusion coefficient of chemical in air (!2/t)
V2' = 0.508 m/sec
z = 0.48 m (shortest tank dimension gives highest mass transfer rate)
The gas side mass transfer coefficient (kg) in the above model is:
= 0.484 cm/sec
Thibodeaux (1979) reports a value of the liquid side mass transfer coefficient (kj) in large water
bodies of about 6x10^ cm/sec for wind speeds of 0.5 m/sec. Although not directly applicable to
the current situation, it can be used as a first estimate to determine the potential for liquid "film
resistance to control the mass transfer rate.
Liquid side resistance =
= 3.3xlO"2 sec/cm
Gas side resistance = l/kg = 2.1 sec/cm
It can be concluded that formaldehyde volatilization from open tanks is controlled by gas-side
mass transfer resistance; therefore, the CEBMPEA equation appears to be valid. It should be
noted that it may be necessary to consider liquid-side mass transfer resistance for chemicals with
larger Henry's constants, hi this case the CEBMPEA model would not be valid.
Surprisingly, volatilization due to air sparging is less significant than that from open tanks.
Although the concentration of formaldehyde hi the bubbles is high (virtually at equilibrium with .
the formaldehyde concentration in the bath), the volume of air sparged is small compared to the
volume of room air flowing over the top of the tanks.
D-15
-------
APPENDIX D
Concentration of formaldehyde in workplace air
Cy = Fyy(VRRvk)
= 0.326 mg/m3
= 0.265 ppmv
where: Fy>T = 7.49 mg/min + 0.421 mg/min + 13.8 mg/min = 21.71 mg/min
VR =20m-20m-5m = 2000m3
Rv = 4 hr-l = 0.0667mm-1
k = 0.5
VH. COMPARISON OF PREDICTED FORMALDEHYDE CONCENTRATIONS IN
WORKPLACE AIR TO MONITORING DATA
In this section, the concentrations of formaldehyde hi the workplace air predicted by the model
are compared to available monitoring data. The purpose of the comparison is not to validate the
model but to determine if the modeling approach gives reasonable values of formaldehyde
concentration. Model validation would require calculation of formaldehyde concentrations using
the conditions specific to the monitoring sites. Such data are not available.
The results of an OSHA database (OCIS) search of monitoring data for formaldehyde (provided
by OPPT) include 43 measured air concentrations for 10 facilities in Standard Industrial
Classification (SIC) 3672 (printed circuit boards). The concentrations range from not detected to
4.65 ppmv. Most of the concentrations (37/42) range from <; 0.04 to 0.6 ppmv, with all but one
less man 1.55 ppmv. Cooper et al. reports formaldehyde concentrations from three electroless
plating operations measured over a two day period. The mean concentrations ranged from 0.088
to 0.199 ppmv. The predicted concentration of formaldehyde in the workplace air was 0.263
ppmv. Thus the predicted value is within the range of concentrations determined by monitoring,
arid less than the OSHA time-weighted-average concentration of 0.75 ppmv. The authors
conclude that the results are reasonable.
D-16
-------
APPENDIX D
REFERENCES
Bailey and Ollis. Biochemical Engineering Fundamentals. New York: McGraw-Hill, Inc.,
1977.
Berglund, R. and E. Lindh. "Prediction of the Mist Emission Rate from Plating Baths." Proc.
Am. Electroplaters and Surface Finishers Soc. Annu. 'Tech. Conf, 1987.
Burgess, W.H. Recognition of Health Hazards in Industry: A Review of Materials and
Processes. New York: John Wiley and Sons, 1981.
Cohen, Y. and W. Cocchio. Laboratory Study of Liquid-Phase Controlled Volatilization Rates in
Presence of Wind Waves. Environ. Sci. Technol, 12:553,1978.
Cooper, C.D., R.L. Wayson, J.D. Dietz, D. Bauman, K. Cheze and P.J. Sutch. Atmospheric
Releases of Formaldehyde from Electroless Copper Plating Operations. Proceedings of
the 80th AESF Annual Technical Conference, Anaheim, CA. 1993.
Cussler, E.L. Diffusion: Mass Transfer in Fluid Systems. Cambridge: Cambridge University.
Press, 1984.
Dean, J. A. (Ed). Lange's Handbook of Chemistry, IS^ed. New York: McGraw Hill, 1985.
Girifalco, L.A. and R.J. Good. "A Theory for the Estimation of Surface and Interfacial Energies:
I. Derivation and Application to Interfacial Tension." J. Phys. Chem., 61(7):904-909,
1957.
Hoff, J.T., D. Mackay, R. Gillham and W.Y. Shiu. "Partitioning of Organic Chemicals at the Air-
Water Interface in Environmental Systems." Environ. Sci. Technol., 27(10):2174-2180,
1993.
Hsieh, C., R. Babcock and M. Strenstrom. Estimating Semivolatile Organic Compound Emission
Rates and Oxygen Transfer Coefficients in Diffused Aeration. Water Environ. Research,
66:206, 1994.
Liss, P.S. and P.G. Slater. Flux of Gases Across the Air-Sea Interface. Nature, 247:181,1974.
Lyman, W.J., W,F. Reehl and D.H. Rosenblatt. Handbook of Chemical Property Estimation
Methods, Washington DC: American Chemical Society, 1982.
Mackay, D. and P.J. Leinonen. Rate of Evaporation of Low Solubility Contaminants from Water
Bodies to Atmosphere. Environ. Sci. Technol, 9:1178, 1975.
Matter-Miiller, C., W. Gujer and W. Giger. Transfer of Volatile Substances from the Water to
the Atmosphere. Institute for Water Resources and Water Pollution Control (EAWAG),
D-17
-------
APPENDIX D
Swiss Federal Institute of Technol., CH-8600 Dubendorf, Switzerland, 15:1271,1981.
Parker, W., D. Thompson and J. Bell. Fate of Volatile Organic Compounds in Municipal
Activated Sludge Plants. Water Environ. Research, 65:58,1993.
Peng, J., J.K. Bewtra and N. Biswas. Transport of High-Volatility Chemicals from Water into
Air. Proceeding of 1993 Joint CSCE-ASCE National Conf. on Environmental Eng.,
120:662,1993.
Peng, J., J. Bewtra and N. Biswas. Effect of Turbulence on Volatilization of Selected Organic
Compounds from Water, Water Environ. Research, 67:000, 1995.
Perry, R.H., D.W. Green and J.O. Maloney (Eds). Perry's Chemical Engineers' Handbook, New
York: McGraw-Hill Book Company, 1984.
Risk Assistant Software. Alexandria, VA: Thistle Publishing, 1995.
Roberts, P.V., P. Dandliker and C. Matter-Muller. Volatilization of Organic Pollutants in
Waste-water Treatment-Model Studies, EPA-R-806631. U.S. EPA, Munic. Environ. Res.
Lab., Cincinnati, Ohio, 1983.
Rosen, MJ. Surfactants and Interfacial Phenomena. New York: John Wiley & Sons, 1978.
Smith, J. H., D.C. Bomberger and D.L. Haynes. Prediction of the Volatilization Rates of
High-Volatility Chemicals from Natural Water Bodies, Environ. Sci. Technol. ,14:1332,
1980.
Thibodeaux, LJ. Chemodynamics: Environmental Movement of Chemicals in Air, Water and
Soil New York: John Wiley & Sons, 1979.
Tschabanoglous, G. and F.L. Burton. Wastewater Engineering: Treatment, Disposal, and
Reuse. New York: McGraw-Hill, Inc., 1991.
U.S. Environmental Protection Agency. Chemical Engineering Branch Manual for the
Preparation of Engineering Assessments. Washington, DC: U.S. EPA Office of Toxic
Substances. February 28,1991.
Wangwongwatana, S., P.V. Scarpino and K. Willeke. "Liquid-to-Air Transmission of Aerosols
from a Bubbling Liquid Surface." J. Aerosol Sci., 19(7):947-951,1988.
Wangwongwatana, S., P.V. Scarpino, K. Willeke and P.A. Baron. "System for Characterizing
Aerosols from Bubbling Liquids." Aerosol Sci. Technol., 13(3):297-307, 1990.
Weast, R.C. (Ed.) CRC Handbook of Chemistry and Physics, 61st ed. Boca Raton, FL: CRC
Press, 1980.
D-18
-------
APPENDIX D
.a
o
s-
'
•
CJ
o
o
ao
E
SJO
E
o
c:
(-UI/U1U1/IUI)
SUO;SSIUK>
c:
4)
. Sb
E
33 -
o
cs'
cs
o
E
D-19
-------
APPENDIX D
o
o
Z=
O
o
o
i 10
Q.
O
Q = 8 Lpm
p-o, -
20
Q = 13
2.6
.40 60 20
Bubble rise distance (cm)
40
60
Figure 2. Effect of bubble rise distance on droplets number concentration. (From
Wangwongwatana et al., 1990)
D-20
-------
Appendix E
Drag-Out Model
-------
-------
Contents
Summary of Non-conveyorized HASL Chemicals in Process Wastewater
Summary of Conveyorized HASL Chemicals in Process Wastewater
Summary of Non-Conveyorized Nickel/Gold Chemicals in Process Wastewater ....;...
Summary of Non-conveyorized Nickel/Palladium/Gold Chemicals in Process Wastewater
Summary of Non-conveyorized OSP Chemicals in Process Wastewater
Summary of Conveyorized OSP Chemicals in Process Wastewater . . • ;. . .
Summary of Conveyorized Immersion Silver Chemicals in Process Wastewater .
Summary of Npn-conveyorized Immersion Tin Chemicals in Process Wastewater
Summary of Conveyorized Immersion Tin Chemicals in Process Wastewater
E-l
E-2
E-3
E-4
E-5
E-6
E-7
E-8
E-9
Prediction of Water Quality From Printed Wiring Board Processes E-l 0
-------
-------
APPENDIX E
to
c
o
to
0)
o
o
O
10
0)
o
I
CO
T3
I
O)
CO
o
(O
1
^
CO
Ul
CO
T3
CD
O
CD
1
t—
O
^
CD
E
CO
Z
S
S
E
a.
'CO
1^5
m
_
^
E
S"
_
CD
•s
a:
CZ
O
1
73
2
Q.
CM
CO
Jxl
CO
CO
CO
CD
0
g
Q_
t«—
0
E_
CO
£1
E
3
Z"
t—
^^
0)
n
B
t_
o
u_
^
^^
c:
CO
Q.
O
o
o
cf
0
co_
co~
T—
s
of
2
1
u_
E '
CO
s
CO
jg
i
1
CD
co
5
CO
M
CD
(J
o
a.
E
ca
co
o
CO
o
75
•5
>>
co
E
E
CO
^ _J
_. •-§ O E
E 2 0- _j-
TO •+-» __. r—
fi g g.|
co §1 i
o o Co
0=5 2
u_ i—
(5 0
•is .92 -S
co 'o °
1 — UJ
II
E E E a
co £ I- E
0 .0
0 5
c «-
.2 -2
2 i _i
a 4—1 o)
CD CO c
o co £=
O.E
» I_
— ^ T3
•2 ® ra
H ™
>
IE-
CO
£ E -a
m co "™
"5.
CO
o:
3
T 32
n* ra
Q
D
—
5
5
D
O
•C—
O
o
CO
o
o
CO
^"
oo
'C
^"~
n
o
oo
:enediol
CO
•c-^
oo
a
c
o
1^
^
o
CO
5
oo
"o
73
D
><
o
<
^5
C
.O
O
1
c
0
co
c\
33
CO
CM
5!
D
s
5
co
D
<
CM
0
o
co
IT
E
CO
m
o
5
13
5
D
5.
<
CM
O
CO
m
a
^~
oo
m
in
CO
o
CO
5
5
^
D
|
D
5
5
D
5.
<
d
c
0
CO
^
o
cq
T-"
CO
CM
5
0
5.
<
^~
o
CO
o
c\
o
rj-
CO
co
s
CM
In
o
^~
•a
0
CO
O
m
c
co
00
CO
C!
O
CO
'
oo
CO
CO
Tf
N
""
co
o
CO
D
5
5
§
D
D
O
O
5
2.
O
CN
O
*
N
CM
O
•<*•
5
D
5.
0
D
O
1
UJ
CO
er
0
CO
^
oo
»
^f
o
o
o
5
j
3)
D
5
LI
^_
O
^
O
CO
10
o
,_
1^-
3
3)
T—
CO
3
^j
T—
5
5
—
2*
5
2
5
|
3)
D
0
UJ
K
O
OO
CO
1
T—
CM
CO
CO
CO
CO
3
0
3
5
3
7!
U-
o
c
o
CO
co
0
CO
oo
CD"
CM
3
O
T—
O
in
co
M
C
^~
o
co
n
T—
3
a
o
3
3
3
3
"*
^
O
O
CO
c
in
CO
n
in
s
CM
0
CO
D
1
D
0
3)
3
—
CO
c
o
CM
CO
(-J
co
CM
o
CO
•o
o
CO
5
f
3
c
c
o
cq
in
^
oo
CM
D
B
5
"5
CO
5
|
3
C
O
O
CO
CO
in
CO
CM
m
CO
CO
o
o
CO
3
CO
o
3
o
L
CM
CO
CO
co
CM
CD
3
3
3
3
CO
CO
T_
•**•
n
3
CM
CO
n
g
o
3
0
3
3
CO
C
0
i
£
ncern cones
o
o
"o
1
o
1
a:
CO
J
-------
APPENDIX E
co
c
o
ys
s
m
0
c
o
o
CD
15
CD
0
rs
s=
3
CO
•a
c
ns
^
1
<1>
«•"•
(0
CO
•s
to
o
•*•*
co
E
•55
UJ
_!
X
.1
O
f
c
o
O
CD'
ca
rocess
D_
oo
0
T™"
^
£
cr
CO
^
TS
OL
_o
"o
T3
2
D_
CM
CO
C
CO
CO
CO
CD
o
CL
•s
CD
ja
Z
CM
oo
^*j
^"
3
3
iH
1
u_
••-*
ca
D_
0
CD
o
CD"
o
co
co~
*~°
-o
-i
at
2
1
u_
£
ca
CD
CO
J_
w
cs
~
^
O
"5
X-
O
&
umma
CO
C.
- "»-*
to ^^ "**^
^— 1 D5
c O E
CD n
o IA- •+£
° .E 2_
}= _o "co
|£^
CO
_^
m i>
£ CD vp
co "o °
CD ST
ul «^
1— UJ
^3 o
(^ O C
S E-c"
— c 2
co § £
f) v^rf
o 2
O i-
j—
"c ^
~ "rtj
T5 > ^J
c £ c
CD co t:
o co
0
_
~c
CD
^ s^
m co a>
D_
CD
a:
CD
E
CO
^^_
Chemical
CO
0
o
o
CD
O
CO
N.
O
CO
CM
CO
0
4-Butenediol
•^
0
CD
O
0
CD
T—
CM
CD
,3.
O>
"o
TJ
CD
1
JO
<
CO
0
o
o
o
m
CO
o
o
o
CD
j^_
"*
Ikylaryl sulfonate
<
CD
OO
0
O
o"
CO
c\i
co
^
Ikylphenol ethoxylate
<
^-
oo
0
o
to
CM
oo
o
CO
m
Ikylphenolpolyethoxyl
<
lO
8
o
0
CD"
CO
o
CD
^j.
co"
"o
CD
Q.
"c5
CO
co
0
o
0
CO
CO
•<*
o
CM
CO
oo
"o
CO
_o
0
CO
CO
0
CD
CO
oo
l»_
CM
o
o
oo
CO
CO
co
.2
re
13
m
"c.
CD
Q.
JD
co
CD
Q.
Q.
o
o
T—
CM
O
o
CO
co
^_
co
,^_
thoxylated alkypheno
UJ
CO
CM
CD"
^^
Is*-
o
CO
"o
"en
CD
_g>
£
UJ
^—
o
0
0
CD
,—
O
CM
CO
CM
CM
»_
CD
Tl>
^
_j
-Q
O
O
£
"o
"en
CD
_CD
UJ
t^_
m
0
0
^.
'*"
CO
co
"o
CO
_o
o
.a
0
El
o
o
CD
o
co
CD"
^3-
£
3
O
(s^
CD
CD
0
O>
CM
^}-
CD
CM
"o
CO
_O
_g
o
2
X
0
co
CD
CD"
CD
CO
o
CO
o
o
CO
0
o
ydrogen peroxide
X
T^-
o
0
o"
co
o"
o
ydroxyaryl acid
X
to
CM
O
0
CD
to
o"
CO
CO
ydroxyaryl sulfonate
X
oo
CM
CD
m
oo
to
CO
!•--
co
hosphoric acid
D-
co
0
0
0
CO
^_
to
0
T—
oo
o
CM
oo
s.
CO
o
otassium oeroxvmon
Q_
o
o
0
o
o
t—
CM
CD
CO
CD
CD
W
cz
odium benzene sulfo
CO
o
o
0
0
o
co
CD
TJ-
odium hydroxide
to
T—
^~
|
co
eo
o
CM
T—
to
T—
"o
CO
_0
a
=1
to
.
Ijj
"c
O
c
o
o
£
0)
o
1
"o
__
s
1
cr
«
o
£
to
9
0)
•s
£
"c
-------
APPENDIX E
ions
Concen
ickel/Gold
«. z
ate
ater and Surface
/as
9
o>
(a
Q
•s
in
I
at
Ul
tl
§
tr C3) CO in O
Is gs
g- "g"
S «-
I
£
I
s
in
J?
o
o
m
T5
o
O
"5
«u
CO
C • 1 *
••§>_! ° C
^5 |E ^1 O C
I 0 £
C rn ^
O e ®
0 -g E
£ -2 S
CO "o £
£ u. 1-
35
« * ' S 8
IB
nj Q)
£ "o
K£
.i j ssssssicfilsiii^I
|t °§°°§°°°8°°°2°°§
ID ~
0 C
CO •
o EH
'^ i =
^ to
E £
S "~
£ o
CO 5
^0, ™CO lO-^^^CMgeOUJ^I-_g(0
3 ^~
5 jB
5 |
11
s| ^"U^cicis^tfsls5!!
f" ^ a
-
&
•g co^SS<\i!?S"iooc8S!c30-cDC»
a> ^ ci ^~ ^~ m
-C £ -r.
! Si
tn to "
a.
CO
t£
•5 cocMSc»§mtDo0-!§§cocM?!oo
"3
o
Q «•
to <0 1 S
— "o *o 9 _
: roto >,m<1>5'^
i .2 .9 g TJ „, -a £ S
1 c5
vs * ">> ">• ^ ^ £~ a^-rr-ri
j^^. o
C
o
CN CO 00 "S
a>
o
•i
§
3S§cisl5s^"^°°°^^° I
BoS§§2§§0|§§0§°§§|-
3 § ° § . ° °" °" 1
o
"to
3
co"
CO
(0
J O ° O O ^~ x
ci ci ' *
1
X.
C
CD
"5
?|^ps^fe|-"|p|c1^ J
-2
CO
to
3:
Jg^:gWO^Sfc[2^ScDoS^3 i=
o" o" ° §,
f-
o
Is
; ; to.
0" 0 : •*- CM "
£
"co
TJ
to m
"s -J
"S, '
•*- 2
i'co'ro c "g 2 _^" ^ S
III 2 T, 1 & 5 S • S ^^5
«•§•§ •2^"8Si^o^lT,'|S =
* ^ *- ^- «£ - .52 c C >» >* ~CS ~^ "O p- '
^^.52.52-§^EoS§-c:-Fw-§too^ '
Ililiiflllllilll i
-------
APPENDIX E
i
o.
Id
o
ui
•s s s s
ui Q. D. z a. co
£
I
ca
co
01
I
n.
_
s
I
o
"5
"o
£
E
o
H ° ^
C CB g
£ o TS
CO 13 H
£11.1-
55
- l£
m g:
CO CD
2 .52
•2 *
Si •£
— "c 2
co S E
O "t:!
o £
f^
JO _
ca o
u 3)
^ to
o :g
C «
s H
t3
ca
I
E
m g
"i.
O
a:
l
"5
9
ra
Q
CD
Chemical Nam
co
8
ci
CD
s
CM
oi
in
(Aliphatic add B
8
d
CD
CO
CD
CD
CO
•e—
CO
(Aliphatic add E
I
ci
oi
CO
•a-
CM
<(•
(Aliphatic dlearboxyllc add
1
ci
CO
in
in
CM
CO
O
(Aliphatic dlearboxyllc add
CO
8
ci
°
CM
in
CD
in
lAlkylamlno acid B
CO
8
ci
0
1
°
CO
CO
~G
-D
£
CD
8
ci
o
S
CM
CO
S
CO
lAlkylpolyol
i>-
8
o"
£
CM
CM
2
CM
lAmlno add salt
in
g
o
o
CM
CO
CO
CM
0
lAmlno carboxyllc add
o
ci
^
,
in
CD
CD
CO
in
lAmmonla compound A
.0000871
o
CD
O
ci
CM
o"
CM
o
lAmmonla compound B
CM
in
o
ci
in
in
CD
CO
CO
in
CD
lAmmonlum hydroxide
o
5
ci
~L"
§
in
CM
*
(Citric add
8
o
CO
CO
o
o
in
CO
in
8
^
£
(Copper sulfate pentahydrs
0.00081 1
in
CO
o"
^Z
in
CO
oi
lEthoxylated alkylphenol
o
d
CM
§
in
o
9
lEthylenedlamlne
•«—
ci
CM
CM
S
g
CM
^~
I Hydrochloric acid
CO
0
o
ci
o
a>
•«
CO
o
o
CO
CO
CM
in
S
S
CO
(Hydrogen peroxide
0.00023
CM
ci
o
co"
s
ci
in
CM
(Hydroxyaryl acid
CD
CM
8
O
o"
CO
10
o
0
o"
in
CD
CO
CO
co"
m
1
CO
Is
E
.0
i
o
c
in
o
o
o"
co
in
fe
55
o
CM
(Malelcadd
O)
o
ci
o
CM
CO
s
CO
CD
in
in
1 Malic acid
in
in
o
o"
CM
o
o
s
CD
in
CO
to
8
(Nickel sulfate
o
8
o'
co"
1
2
s
(Palladium salt
CO
8
CD"
°
1
"
CO
(Phosphoric acid
8
CD"
o>
8
§
S
(Potassium compound
CO
o"
CM
8
^
CO
(Potassium gold cyanide
CO
5
o"
CD
CM
T—
Si
(Proplonlcadd
0.00016
o"
ci
ci
CO
r—
(Sodium hydroxide
S
o
ci
CO
CO
CO
8
1
CD
"io
^
y
o
c
o
(Sodium hypophosphlte rm
CO
CD
in
CO
•sr
CD
CD
CO
•sr
in
(Sodium salt
o
CD"
o
CO
2
o
o
ft
g
CO
CO
CO
CO
o"
-
CM
CD
ci
o
(Urea compound B
c
_o
TS
8
8
c
o
u
0
"Jo
g-
CO
-------
APPENDIX E
a_
co
O
73
CD
N
'o CO CO t- O
S CO CO CZ)
CD CD CD O
1 ~§
9 S-
o "~
• - %}
—•: C 73
E^^3^-
» S is
••-sS if
g ro o § .2
i-Si*
»5 2|c:
S S5^.|
mil
Q- Q. Z Q. CO
c.
— :>
1° & 1
Ig E
0 "- <-r
C O) c
O c 0
111
•£3
CO
-4-« _
c: >»
QJ O
£ 111
.O a
| 1
co "m -i-r
CD i-i CI
j= -t: c
03 § E
° "S
o 2
O 1-
c
c ^~
.0 £
5 g =i
^ *•• C
SCO <=
CO
§5
O
j-T
^ "co
si ?n ^
=J -J*i O)
i-|
>
• , :•
CD
^ CD "^
Q CO D>
Q.
«o
-3 O
O O
n CM
3 V.
T ?
E> Jx^
~> M
- «3-
) O
r *a-
CD
O
vl
CO
3
g
—
"5.
CO
<
CM
C
0
{£
f>
ir
in
in
o>
S
^
3
73
3
5.
_CD
i
I
i
^
3
<
(O
C
o
CO
co
V
°
c
CM
r
5
[•*»
M
•^r
3
3
CD
H.
3
o
o
CO
c
o
CO
co
en
i
m
in
o>
^
co
CM
^
O
i
CO
f
i
§
O
CO
03
10 r^-
Tf Ot
° c
D **
vj f>
O CD
o ^~
in CM
CM "3-
M
CM
o -3-
?o
>
a
3
^ "o
1 s
D •§
s —
D OT
S CD
) ">1
i; o
J 111
I§ ^
° c
O T-;
n tr
»
a> oo
-
i) "NT
o""
I
0,
>
>
- E
3 J
±1 C9
oo
CD
O
o
33 OO ^T
D O §
0 CM 0
- X- If
*~ T-
0 CO CM
•3 'CM (*.
M CM
3 CO CM
- CO T-
> 00
> in o
I CM CM
i in
CD
l-o
2 ?
S -
^B
CD S?
O O
73 73
^ ^
~*t •*- t*- C
l°"l C
q i CM o c
\i in ^~- c
V1 T-1 U
D T- CO C<
"> in CM c*
in •*
- ^ co T
$ in ic
C33 I--
3 r^- -xj- f
t o T- cc
CO T-
£\
•
:
"si
S|-o
"§-"=«
^ ^I O "5
CL CO CO
^
i
1
o
1
s
5
3
a
S |
« 1
si
>< o
» S
I S
CO
II
^
1 Q.
S 0
" T?
3 £
f i
tl
3 °
E (D
s 1
i (O
'. 03
1 |
n)
old indicate tt
it not evaluati
.£ T3
E-5
-------
APPENDIX E
Q.
CO
O
o
O
O CO CM O
CO O
CM O
O
O
CO
C0~
D_ CO
0)
13
flj
(0
(O
(0
Cfl
(11
u
S
Q.
vn
ra
_o
E
a>
O
75
•s
£•
OS
E
E
CO
c
o
'ro > =i
^ O E
0 C? J2
0 •§ E
1= S
CO o «--
f« LL h-
35
c >;
CO O
E C o
to "«3 °
CO U=
L— **-
1- UJ
O
0
•*-•"
CO
£ « "O
CO o "ra
m to ra
Q.
0)
0)
E
CO
Chemical N
CM
CM
0
CM
O>
CO
CO
CM
T3
'o
CO
o
"S
o
CO
o
CD
o
c»
09
0
in
h~
§
CM
lAlkylaryl imidazole
CO
CM
O
CD
0
0
CO
ja
ts
3
2
JAromatic imidizole p
^j-
CO
o
0
o
0
T—
0
CO.
"o
CD
c
&
m
CM
0
o
CO
CO
00
T—
o
m
CM
CM
.0
CO
Q.
a.
o
o
o
o
0
o
o"
CO
00
0
in
o
o
o
CM
CO
O
"ro
CO
CD
Q.
Q.
o
o
in
o
0
CO
CO
CO
CO
o
T—
T—
in
CO
CO
•^j*
•«-*
2
CO
^"
"o
o
"ra
0)
CD
>>
LU
^>
o
o
o
CO
o
00
E
ZJ
CD
in
CD
s
s
o
CM
1 Hydrochloric acid
in
T~"
o
o"
o
0)
m
o
5
CO
o>
T" *
1 Hydrogen peroxide
CD
o
0
Q
CO
O
CO
CM
JHydroxyaryl acid
^~
co
0
o
0
2
in
CO
CO
1
CO
CO
B
•o
X
CM
CO
CD
CO
CO
CM
^i»
^^
Tj"
-a
"o
CO
o
0
a.
CO
0
Q.
^3-
0
o
CD
in
CD
CO
CD
TO
•a
E
'•&
o
CO
T—
CM
m
t--
00
^•—
CM
T3
'o
CO
O
a
CO
c
i
s
§
o
c •
o
o
c
0>
o
§
o
I
to
if
<* •£
-a °
,»-;
g-f
11
0 ?
0 CO
0 CO
E 5
S £
P CO
••-• r"
CO C
T3 CO
a> as
?
-------
APPENDIX E
to
«
5
•5
CO
0)
*»-»
I
111
I
CO
c
o
"52
. CD
|
CD
1^1
o
>
C
o
O
CD"
CO
^
CO
CO
CD
O
e
.0.
co
CO
73
E
cr
w_
to
a:
c
.0
1
OL
"*
CO
-2£
cr
•CO
CO
CD
o
2
Q.
it-
CD
CD
-Q
E
'Z.
CO
CO
0
GO
«-
•*-•
2
I
U-
»^
>
"c
CO
Q-
o
o
0
o~
o
CO
co"
.*"
s
CD"
2
£
o
u_
E
CO
CO
Jj
1
ts
i
in
(A
u
o
Q.
JO
to
o
CD
O
75
•5
2"
umma
CO
c
.0
"c O E
o °- *f
o c ®
jz .0 to
1=5,2
2 u- i-
•«— '
CO
_
c >»
CD CJ
E C
*-• ® ~5
co "O °
CD u=
H1 UJ
_O ™
E -—
10 "S? *J"
CD £ C
j= -ti CD
CO c5 E
c «
o 2
O h-
c
"~~ _
0 ®
"-^ CO _,
2 5 >=
c B |>
CD to c
O CO •
0 ^
o
-•-*"
CD
.c E
m co °*
0.
0)
CD
E
Chemical Na
CT)
CM
o
o
o"
o
o>
o>
CM
o
o
CO
^^"
o
CD
CO
,4-Butenediol
•^~
CM
^^
0
CO
CO
CO
o
CO
T™
T3
"o
CO
0
E
-CO
<
CO
CM
O
O
O
o
U5
O>
fs.
^j*
o
o
0
fv_
J^
CM
CO
C.
E
CO
to
u_
CD
CM
0
O
O
CD
CM
0
co
CM
•sf
CM
CO
NT
CO
(D
ydrogen Peroxid
31
•c —
CM
o
o
IO
CO
T—
CO
CM
/
"O
o
CO
c:
CD
CO
2
-z.
CD
CM
0
o
CO
•sr
to
CO
-0
"c
onionic Surfacta
•z.
CM
CN
o
CO
U)
CO
T—
0)
CO
CM
<5
CO
o
o
Q.
CO
o
.c
CL
tn
CM
o
0
o
0
0
CO
O)
CO
CO
o
0
o
o
o
-^J-
oo"
CD
1
i
CO
|s^
^-
o
CD
fs«»
^•^
T—
CM
CO
odium hydroxide
CO
T_
"j—
p
o
fv_
^^
^_
5;
73
0
CO
0
"5
CO
.0
s
c
§
c
f—
0
o
*o
•2
^~
<0
f 1
(D "c
To -*=
"S
I I.
^ ^"
5 o
, - -Jg
u
i 1
C "^
1 1
(D -X
B B
O (O
U (0
E 5
to g>
£
•«> £
T3 tD
Ol IO
11
ffi =
Ol ••—
Ol T3
f- Ol
^ "S
Mumbers in bold Indicate
This Ingredient not evalu
a .0
E-7
-------
APPENDIX E
tn
_o
I
§
u
o
O
ca
I
o
I
co
T3
CO
CU
•5
I
^^
to
JO
9
_c
t-
I
E
TJ
CD
18
O
o
c
o
• TT O
CM CO
n°-
-
CO
~ J?"R E = o
^ £ S 3 J2 &
w Q. a. Z o. co
CD
1
B
IO
CO
ca
to
CD
u
o
s
0)
ca
•5
CO
E
tn
C D> _I
"~" C ^^
^ •— o
_o > c
•*;{ ^— ^? -
5 o p c
F E ° E
CD C f), C
o £ S
o 35 i-
.»*
§3 c?
£ CD vP
ca "5 °
£ t
111,
c £ ==
CD J^ 5?
o co £
C £-
0 —
o ,
1 I
03 CO i
£; _ 5 -~~
CD ~~ £ C
o to c
c ca
o >
o >
-1
« |5
°^ ^
1- ca
CD
j= E
•»-« CD TJ
CD co
a.
CD
o:
"3
l«
Q
CD
£
ca
*^>
Chemical
35
b
d
CM
CM
CD
CM
to
CO
CO
co-
•<*•
Q
•a
u
CO
o
IS
.e
CD
CD
>s
.e
m
10
CO
o
o
CO
1
CO
CO
CO
2
o
CO
—
o
o
u_
CM
2
0
CO
CO
CO
CM
CO
CO
CM
ydrochlorlc acid
X
CO
o
•<*•
1
CO
CO
CO
CO
CO
T—
ydroxy carboxylic acid
X
CO
CO
o
CM
CO
CO
CO
0
*
CD
CO
CD
in
73
u
CO
o
c
£
~3
CO
CD
CO
in
CO
0
CO
00
0
CO
CO
IO
T—
•<*•
CO
hosphoric acid
a.
CO
CO
o
CO
o
CO
CO
CO
o
CM
0
CM
0
00
10
DO
CO
CO
CO
CO
CD
s
3
CO
otassium peroxymono
D-
^
CO
o
o"
o
CO
o
o
$
CO
oo
CD
CO
CM
CM
CD
CO
CD
TJ
^
]E
o
E
_3
c
uantenary alkylammo
a
CM
o
CO
o
o
o
CD
o
o
CD
CO
o
o
0
IO
o
CO
>
CO
o
CO
0
o
CM
CO
CD
IO
CO
CO
£
CO
odium benzene sulfon
CO
CO
CM
CO
to
CO
o
h-
co
CO
CM
IO
CO
!*??
odium phosphorus sal
CO
CM
^
o
••*•
m
co
o
CO
CO
o
CO
•<*•
00
oo
CM
CM
IO
CO
"5
CO
"c
o
tannous methane surf
CO
00
CO
o
CD
•<*•
CD
IO
*"
10
"•
CD
CO
CM
O
T— ,
ulfuric acid
CO
0
CO
o
o"
o
CO
o
CO
o
T"
o
10
o
to
CM
CO
o>
CO
CO
£
3
K
CO
CM
o
CO
o
^"
s
o"
IO
CM
0
CO
IO
CO
CO
10
In chloride
H-
eo
o"
?
CO
o
CO
n
CO
nspecified tartrate
oo
CM
d
CO
IO
to
CO
CO
CO
CM
CO
IO
CO
CO
£
ZJ
CM
CO
0
o
o"
0
CO
CM
g
o
IO
CO
o
CO
CO
to
3)
o
TJ
3
o
0.
E
o
u
CO
£
5
CD
CO
0
0
CM
CM
CO
CM
IO
CO
CO
CO
3>
-------
APPENDIX E
0)
E
CO
CM
i I
at
u
o
O
CD
U
0)
•o
i
s
I
«•—
o
in
UJ
CD
O
O
CO O
CD CD
t- p
CD
O
CO
"I
I
IO
m
(A
in
i
0> O
£ S^
« 0 ^
o
o
0
0
o
m
o
CM
kylimine dialkanol
<
CM
in
o
o
p
CD"
in
GO
0
0)
CD
kylphenol ethoxylate
<
0.0000035
CD
o
o"
i
o
smuth compound
m
CO
m
0
o
s
N
T3
O
CO
O
0
o
§
0
CD
T—
CM
O>
CD
T3
E
CO
o
5
CJ
o
o
0
CD
b
CD
in
10"
"o
c
CD
Q.
"co
TJ
CD
s
X
O
c
UJ
CM
CO
O
O
O
0
CO
CO
CD
CO
JQ
O
C
o
E
~o
a
CD
CD
Lti
CO
CO
CD
O
CD"
m
in
in
•»*•
uoboric acid
L.
O
0
O
o
CO
CO
/drochloric acid
X
r—
m
0
o
o
••a-
o>
o
-a
0
CO
o
1
Q
CO
o
1
I
in
in
p
o
o>
n
TJ
O
CO
O
O
"5
CO
CD
CO
5
S
CM
CO
o
o
o"
CO
3
T5
O
CO
O
O
Q.
•f)
a.
^
CD
C
0
O
o>
p
o
CO
CO
n
CD
S
stassium peroxymonosi
Q.
CM
CO
o
o
o"
CO
in"
CO
co
CD
•J3
o
I
0
E
CO
2:
CO
CO
c:
CD
c
CO
a
.000000521
o
CO
x>
o
^
a>
o
D
O
5
5
CO
0
o
o
o
0
q
in
o
CD
)dium benzene sulfonat
CO
CM
CD
N
CO
o
)dium phosphorus salt
CO
in
0
CD
m
CM
S
CM
2
CO
o
annous methane sulfon
CO
o
p
CD
in
CM
•o
0)
13
u
CO
o
3
CO
CO
p
CD
^
O
IS
5
1-
o
o
p
o
CO"
in
CM
n chloride
i-
Tf
CO
o
p
CD
CO
in"
m
ispecified tartrate
=>
CN
O
o"
N
CO
O
CO
.
>
c
Q
2
§
tic toxicity con
01
cr
CD
tt)
-C
to
H
p
D
U
S
5
3
i
a
i
3
E-9
-------
APPENDIX E
PREDICTION OF WATER QUALITY
FROM PRINTED WIRING BOARD PROCESSES
Final Report to the University of Tennessee Center for Clean Products and
Clean Technologies and to the U.S. Environmental Protection Agency
Part of the Verification of Finishing Technologies Project
EPA Grant X825373-01-2 (Amendment No. 2)
By
Dr. R. Bruce Robinson
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
Office: 865/974-7730, FAX: 865/974-2669, E-Mail: rbr@utk.edu
Dr. Chris Cox
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
Office: 865/974-7729, FAX: 865/974-2669, E-Mail: ccox9@utk.edu
Jennie Ducker
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
August 6,1999
E-10
-------
APPENDIX E
TABLE OF CONTENTS
INTRODUCTION
Objectives
LITERATURE REVIEW
Pollutant Generation Rate and Waste Generation Volume
Drag-out Tests at Micom, Inc.
Other Published Drag-out Estimates
Discussions with Experts in the Surface Finishing Industry
Summary of Drag-out Studies
Drag-out Prediction Equations
Rinsing Theory
Other Rinsing Theory Studies
Printed Wiring Board Pollution Prevention and Control Technology
Water Use Rates from Survey of MHC Facilities
RESEARCH APPROACH
LABORATORY DRAG-OUT EXPERIMENTS
Apparatus
Procedure
Quality Assurance and Quality Control (QA/QC)
Results and Discussion
DRAG-OUT MODEL DEVELOPMENT
PWB WASTEWATER MODEL
COLLECTION AND ANALYSIS OF FIELD SAMPLES
, Process Characterization
Sample Collection
Temperature
pH
Conductivity ' -. •
Viscosity
Specific Gravity
Surface Tension
Metals Analysis
Quality Assurance and Quality Control (QA/QC)
Results from Analysis of Field Samples
DYNAMIC MASS BALANCE MODEL FOR INTERPRETATION OF FIELD DATA
E-ll
-------
APPENDIX E
MpDEL VALID ATION
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
Conclusions
Recommendations
REFERENCES
LIST OF SYMBOLS
E-12
-------
APPENDIX E
INTRODUCTION
The Design for the Environment (DFE) Project Printed Wiring Boards (PWB) Cleaner
Technologies Substitutes Assessment: Making Holes Conductive (MHC) was performed by the
Center for Clean Products and Clean Technologies (CCPCT) at the University of Tennessee. The
project and results were well received by industry and the U.S. Environmental Protection Agency.
However, all parties agreed that one weakness in the project was the evaluation of impacts of
chemicals in the wastewater discharges of bath solutions from the MHC plating tines. Evaluation
of these impacts was more difficult than anticipated partly because of insufficient information from
surveyed facilities on the water quality of their discharges. Attempts at a mass balance to predict
chemical discharges were also unsatisfactory due to insufficient data on chemical use and ultimate
fate.
An estimate of the pollutants in the raw wastewater from PWB plating processes is needed in •
order to evaluate health risks, impacts on the environment, impacts on municipal wastewater
plants, and overall manufacturing costs which includes treatment/disposal costs. The main source
of pollutants in the raw wastewater is the drag-out from the baths. Hence^ drag-out is the key
variable for determining pollutant mass.
PWB facilities analyze at most only a couple of chemicals in their wastewater, and the facilities
generally have insufficient data to calculate chemical mass balances. Therefore, a different
approach is required to estimate the pollutant loads and wastewater quality of the PWB
wastewater discharges. This report discusses the development, validation, and use of predictive
tools to satisfy this need.
Objectives:
The objectives of this research were:
• Develop tools and methodologies to predict, but more importantly to compare the mass of
pollutants in the raw wastewater discharges from PWB plating processes. .
• Validate these tools and methodologies against data available hi the literature and against
samples collected at PWB facilities.
E-13
-------
APPENDIXE
LITERATURE REVIEW
Literature was identified through a computerized search on several key words. Additional papers
were found from the references in papers and from a manual search of recent Chemical Abstracts
(1998).
Pollutant Generation Rate and Waste Generation Volume
The sources of the pollutants in the wastewater generated in the MHC and surface finishing .
processes for PWB manufacturing are the chemicals that escape from the process baths and from
other processes such as stripping racks of plating deposits. Our assumption for estimating the
pollutant mass generation rate, e.g., kg Cu/day, is that the source of the pollutants is
predominantly the drag-out from the process baths. Whatever chemicals are drug out of the
process tanks by solution adhering to the surface of the boards and racks will be removed in the
rinse tanks and ultimately end up in the raw wastewater discharge before any treatment or metals
recovery. This is consistent with the literature (Mooney 1991) and is expressed in a simple mass
balance:
/'mass of pollutants^ _ fmass of pollutants^
V in drag-out ) \in rinse discharge^
Eqn 1
As discussed later, the etchant process baths themselves are generally not dumped into the
wastewater at the end of their useful life, but are typically sent off-site for processing. Other
process baths are apparently not sent off-site and do need to be accounted for in the waste
generation. Although pollutants from the stripping of racks may be significant at times, the
average mass pollutants originating from this process should be less than that contributed by drag-
out. Therefore, an estimate of the expected drag-out from various process tanks under differing
conditions is critical for estimating the waste mass generation rate. The arrangement of the rinse
tanks and the rinse flow rates will not change the total mass of contaminants released, only the
concentration and the volume of wastes. The waste generation volume primarily depends on the
rinse flow rates since this is the main source of wastewater discharge. If certain assumptions are
made, then conventional rinsing theory may be used to estimate the volume of waste based on the
drag-out and needed final rinse water quality. Importantly, the primary goal of this work is a
methodology that can be used to compare the relative amounts of wastes.generated from
alternative PWB surface finishing manufacturing processes.
There are many references giving advice on minimizing drag-out and rinse water. Factors that
will reduce the drag-out include slow withdrawal from the process tank, longer drainage times,
tilting the boards so that the liquid drains to a corner, using drip shields, using drag-out/drag-in
tanks, as well as others. Solution density, viscosity, which depends on the bath chemistry arid
temperature, and surface tension also affect how well the liquid drains off the boards, and hence
affects drag-out. Because of the number of variables which have complex relationships with drag-
out, estimating drag-out for a series of baths is a difficult, unsolved problem. The following
E-14
-------
APPENDIX E
sections review what is known about estimating drag-out, including several references that include
predictive equations and experimental measurements.
Drag-Out Tests at Micom, Inc.
The MnTAP/EPA Write study (Pagel 1992) at Micom, Inc. evaluated the ability of two
modifications to reduce waste from PWB surface finishing processes. At the time of the study,
Micom produced 92 - 111 nrVday of double-sided and multilayered PWBs with the average board
being 0.46 m by 0.53 m and having 8000 holes. Micom had already implemented several waste
reduction measures, including countercurrent rinses, flow restrictors, softened water in the rinses
(softened water improved the rinsing and increased the efficiency of the ion exchange waste
treatment system), and air and mechanical agitation. However, Micom evaluated whether
changes to the way PWBs were transferred from process baths to the rinse tanks could further
reduce the amount of waste by reducing the drag-out.
Two processes were tested at Micom, Inc. in their MHC line: 1) a micro-etch bath and the
countercurrent rinse tanks following it; and 2) an electroless copper bath and the countercurrent
rinse tanks following it. The PWBs were moved from tank to tank in racks. The racks were 0.86
m high by 0.50 m wide by 0.33 m deep and could hold 24 boards. Typically, the operator
controlled a hoist and allowed the rack to drain for 3-5 seconds before going into the next tank.
The residence time was about 75 seconds in the micro-etch tank, 30 minutes in the electroless
copper tank, which held two racks at a time, and 2-3 minutes in each rinse tank.
The modifications evaluated at Micom were: 1) slowing the withdrawal rate of the racks from the
process bath; and 2) using an intermediate rack withdrawal rate combined with a longer drain time
over the process bath before going into the rinse tanks. .Slowing the withdrawal rate was
achieved by lowering the speed of the motor on the mechanical hoist used to-move the racks.
Installation of new equipment prohibited matching the withdrawal rates used in the first
modification with tests on the second modification, hence the designation of "intermediate"
withdrawal rate. Withdrawal time was defined as the time it took to raise the boards from the
bath to a height needed to clear the tank walls, a total of 0.91 m. Increasing the drain time was
achieved by the operator simply waiting longer before placing the boards in the next bath. Drain
time was defined from the moment that the rack cleared the water surface until half of the rack
was over the adjacent rinse tank. Measurement of drag-out was accomplished by shutting off the
rinse water and then measuring the increase in copper concentration after a known quantity of
boards had been rinsed. Copper was measured by atomic absorption spectrophotometry. The
electroless copper samples were preserved with a hydrochloric/nitric acid mixture rather than just
nitric, because copper precipitated out of solution as the solution cooled when nitric acid alone
was used. There were some analytical difficulties of unknown origin in that the copper
measurements done by an outside laboratory showed about 1800-2200 mg/L of copper whereas "
Micom's laboratory analyses showed about 2400 mg/L.
E-15
-------
APPENDIX E
Baseline drag-out measurements were made over a twelve day period using 136 samples for 12
paks.of racks. The first modification experiments were also made using 136 samples for 12 pairs
of racks, and the second modification experiments used 109 samples for 9 pairs of racks.
The results of the experiments are summarized in Tables 1 and 2. It should be noted that the
values for drag-out, withdraw rate, and drain time are averages of a rather broad range of values
grouped by relative magnitude by Pagel.
Table 1. Drag-Out Test Results on the Microetch Bath at Micom, Inc.
Parameter ' ' >
Drag-out, mL/m2
Withdrawal time, sec
Withdrawal rate, m/sec
Drain time, sec
Total time, sec
Surface area/rack, m2
Water flow rate, 1pm
Baseline
129
1.7
0.51
3.4 •
5.1
8:2
9.8
Slow Withdrawal Rate
t **
72.1
14.9
0.056
,2.5
17.4
7.7
—
Intermediate Withdrawal Rate &
Longer Drain Time
76.4
4.3
0.20
12.1
16.4
8.6
—
Table 2. Drag-Out Test Results on the Electroless Bath at Micom, lac.
Parameter
Drag-out, mL/m2
Withdrawal time, sec
Withdrawal rate, m/min
Drain time, sec
Total time, sec
Surface area/rack, m2
Water flow rate, Ipm
Baseline
64.6
1.8
0.48
5.2
7.0
15.7
12.5
Slow Withdrawal Rate
32.3
13.9
0.061
3.2
17.1
15.0
—
Intermediate Withdrawal Elate
& Longer Drain Time
31.4
4.3
0.175
11.9
16.3
16.3
—
For the micro-etch bath, the first modification reduced the drag-out by 45% while the second
modification reduced drag-out by 41%. For-the electroless copper bath, the reductions were 50%
and 52%, respectively. Because it was easier for Micom to control the drain tune than the
withdrawal rate, they implemented a longer drain time.
It should be noted that reducing the drag-out from the micro-etch affects the bath. This bath
removes copper until the etchants are exhausted. Make-up chemicals may be added to replace
etchant solution is lost in drag-out. Reducing drag-out may mean that the entire bath must be
replaced more frequently, because of increased copper build-up in the bath. However, Micom.
preferred to retain the copper in the bath and replace the bath, because there is greater
E-16
-------
APPENDIX E
opportunity to recover metals in the etchant bath than in the rinses. For the electroless bath, drag-
put reduction helps retain the chemicals in the bath and increase its life, providing that build-up of
impurities does not offset this advantage. Reduction of drag-out from upstream baths would help
in this regard.
Other Published Drag-Out Estimates
, Sup (1990) evaluated several ways to minimize drag-out, including the effect of the inclination
angle during drainage,,the withdrawal rate, and the drainage time. Several experiments focused
on the inclination angle in the design of electroplating product holders and its effect on drag-out.
The holders were not for PWBs but apparently for a variety of electroplated products. The
holders typically had horizontal cross-braces or struts. Su|3 noted that the drag-out from the
holder could be as much as 50% of the total drag-out in these cases. Sup experimented with
holder designs that had struts of different angles and showed that drag-out could be reduced
significantly. The effect of the inclination angle of the struts on drag-out is shown in Table 3.
Struts tilted at a 45° angle to horizontal had only 36% of the drag-out as a horizontal one.
Table 3. Effect of Inclination Angle of the Product Holder Strut on r»rap-Oiit
Angle to Horizontal ,
0°
15°
30°
45°
90°
Dragrbut
mL/m2
44
35
25
16
22
% of Maximum
100
80
57
36
50
Sup (1990) also experimented with chromium plated sheets suspended from the holders to
determine the effect of drainage time and inclination angle of the sheet. The experiments used
either 19-20 g/L or 240-250 g/L CrO3 electrolytes. The effect of drainage time and inclination
angle is shown in Table 4. (Note: the data reported in Table 4 were read from two graphs in Sup
(1990) and include representative data, but not all the data.). As seen in the table, a 45°
inclination angle had about 33% less drag-out at short drainage times compared to a horizontal
angle and nearly 50% less drag-out at long drainage times. An increase in the drainage time
greatly reduced drag-out up to about 20-30 seconds, but had a relatively small effect for longer
times. Further experiments were conducted on the effect of withdrawal rate and inclination angle
of the sheet. The effect of withdrawal rate is shown in Table 5. Slower withdrawal rates reduced
the drag-out, but not as much as inclination angle. A plate withdrawn at 60 m/min had roughly
25-30% more drag-out volume than a plate withdrawn at 6 m/min. The drag-out volumes
reported by Sup are approximately a factor of two less than the drag-out volumes reported in the
Micom study (Pagel 1992) discussed above. One explanation for the difference may be that the
boards in the Sup study did not contain holes but the boards used in the Micom study did. It
should be noted that Sup was not clear how the drag-out was calculated. It appears to be
American practice to report the drag-out in terms of the area of one side of the board It is
possible that Stip calculated his drag-out based on the area of both sides of the board, leading to
E-17
-------
APPENDIX E
numbers which are half as large. If this were the case, then to be comparable to American
practice, his drag-out volumes should be doubled. However, in a later paper, Sup (1992) used an
equation which was developed for drag-out on the basis of one side of the board. It is likely that
he was aware of the assumptions built into the equation, and considering that his values are
comparable to the Micom study, we will assume, that Sufi's drag-out volumes are directly
comparable to other values. In either case, the trends are the same.
Table 4. Effect of Drainage Time and Inclination Angle on Drag-Out.
Drainage
Time, s
0
10
20
30
45
60
, " , Drag-Out, mL/m* . • '
280-320 g/L CrO3,
0* angle, 40°C
57
28
22
20
19
19
280-320 g/L CrO3,
45° angle, 40°C
•_
21
13
11
—
10
20g/LCrO3,
0° angle, 20°C
64
33
28
25
21
19
20g/LCrO3,
, 45° angle, 20°C
—
24
19
15
13
11
TableS. Effect of Withdrawal Rate on Drag-Out.
Withdrawal Rate,
m/min
3.6
6
9
18
36
60
Drag-Out
240-250 g/L CrO3
(40±1°Q
mL/m2
17
22
24.5
26.5
27
28
19-20 g/L CrO3
(20±1°Q
mL/m2
21
26
29
, 32
33
33
In a second paper, Sup (1992) evaluated two drag-out prediction equations by comparing
measured volumes of drag-out to predicted values. The first equation was from Kushner (1951):
7=0.02.
Eqn2
or:
E-18
-------
APPENDIX E
where:
f
V
h
P
Eqn3
film thickness, cm
dynamic viscosity of electrolyte, g/(cm-s)
height of metal sheet
density of electrolyte, gm/cm3
withdrawal time, s
kinematic viscosity, cmVs
withdrawal rate of metal sheet, cm/s
The second equation was:
Eqn4
where:
••dr
gravity, 981 cm/s2
drainage time, s
Experiments were conducted on 21.0 x 21.4 cm metal sheets which had no holes. The sheets
were withdrawn from the bath at 20 cm/s and allowed to drain for 10 seconds.
Neither of the two equations predicted the measured values very well. Sixteen different
electrolytes were tested with concentrations ranging from 17 to 300 gm/L of material, densities
ranging from 1.015 to 1.562 g/cm3, dynamic viscosities ranging from 0.713 to 8.6 cP, and
temperatures ranging from 18 to 59.5°C. The average measured drag-out was 47.4 mL/m2 with a
standard deviation of 16.3 mL/m2. The average predicted drag-out and standard deviation
predicted by equation 3 were 96.8 and 17.8 mL/m2, respectively, while equatiotr4 had average
predicted drag-out and standard deviation of 15.6 and 2.06 mL/m2, respectively. A linear
regression of measured versus predicted drag-out volumes gave an r2 of 0.021 and 0.012 for
equations 3 and 4, respectively. Taking an average of the two equations yielded no better results.
A scatter plot of the measured drag-out and the predicted drag-out is shown in Figure 1.
E-19
-------
APPENDIX E
Measured drag-out vs predicted
& 160 -
•5 i/tn
^ "
*3, 120 •
3 1AA
o 100 •
cj 80 •
^ 60 •
^3
•g 40"
1 20-
-£* A
•
A
•
•
A
5~"
•
V*
*v
£*
^%A
•A*
•
A
•
*Eqnl
nEqnS
& Ave Bqn 1&3
B* 0 i :
0 20 40 60 80 100
Measured drag-ont, mL/sq.m
Figure 1. Measured Versus Predicted Drag-Out for Results by Sup (1992).
Sfip commented that the equations do not account for electrolyte that adheres to the surface and
bottom edge even after long drain times, i.e., there is a minimal film thickness left. This becomes
increasingly important for rougher surfaces. Sup. recommended that drag-out estimations for use
in recycling procedures and wastewater treatment should be based on measurements rather than
calculations. Part of the reason that poor correlation was found between Slip's measured drag-
out and the predictive equations is that Slip's drag-out showed little variation with viscosity as
shown in Figure 2.
Viscosity vs Drag-out by Su'p's data
100 -
E «n
a> 80 -
5 60
E 60 -
o 4U -
M
K
ft 20 -
* **
*
•»
* * *
•
*
* * *
»
-1 1 —
0 0.5 1 1.5 2 2.5
Kinematic viscosity, cSt
Figure 2. Measured Drag-out as a Function of Kinematic Viscosity for
Results of Slip (1992).
E-20
-------
APPENDIX E
McKesson and Wegener (1998) at RD Chemical Company experimentally measured the amount
of drainage from PWBs as a function of time. They pointed out that longer "hang" or drainage
times allows more liquid to drain from the PWB with consequently less drag-in into the rinse
tanks and thus more efficient rinsing. However, too long of a drainage time may result in lower
PWB quality due to drying and tarnishing. McKesson and Wegener tested two outer layer boards
with solder mask and solder plated and one inner layer board with no holes. A typical result is
shown in Figure 3. (This figure is reconstructed from a figure in McKesson and Wegener.)
50
100
150
200
250
300
350
Hang time, secoofc
Figure 3. Drainage vs Hang Time (McKesson and Wegener 1998).
The results for all three PWBs lay virtually on top of each other in Figure 3. The authors chose to
report just the percentage of liquid that remains on the board rather than mass or volume. This
allowed the authors to see the great similarities in drainage among varying conditions. The figure
shows two drainage phases. For short times, the liquid drains very quickly followed at longer
times by a much slower drainage rate. The authors concluded that 30 seconds appeared to be an
optimal drain time. The authors also studied the effect of surfactants and found very little
difference. They also tested canting the boards at about a 15-20° angle and saw only minor
differences. -
It appears that the most influential reference for typical drag-out volumes is the Electroplating
Engineering Handbook (Pinkerton 1984). These values seem to go back to work by Soderberg
published in 1.936. Typical drag-out volumes are given in Table 6 as reported by Pinkerton.
E-21
-------
APPENDIX E
Table 6. Drag-Out per Unit Area (Pinkerton 1984).
Condition
Vertical parts, well drained
Vertical parts, poorly drained
Vertical parts, very poorly drained
Horizontal parts, well drained
Horizontal parts, very poorly drained
Cup shaped parts, very poorly drained
Drag-Out mL/m2
16.21
82
160
32
410
320-980
1 Suggested by Pinkerton as being the absolute minimum for drag-out on a vertical sheet.
Hanson and Zabban (1959) discussed the design of a wastewater treatment plant at an IBM plant.
To design the plant, an estimate of the wastewater quality was needed. Because a primary source
of the contaminants was the plating lines, the drag-out was estimated based on information
published by Graham in the Electroplating Engineering Handbook. (Note: the data given are the
same as that in a more recent version of the Handbook given by Pinkerton [1984] and
experimental data from another IBM plant which showed drag-out volumes ranging from 100 to
160 mL/m2^ For design, a drag-out value of 200 mL/m2 was used.
Yost (1991) studied the effect of various rinsing arrangements on the costs of cadmium
electroplating wastewater costs. In doing the calculations, Yost arbitrarily assumed drag-out of
200 mL/m2 with no reference for the value.
Chang and McCoy (1990) used a drag-out value of 160 mL/ft2 to evaluate waste minimization for
PWB manufacture. No source was given for their drag-out value, but this value appears to be
commonly used by several researchers.
Discussions with Experts in the Surface Finishing Industry
Contacts were made with several experts in the surface finishing industry. One expert source
(Sharp 1998) had the following comments on drag-out:
• CH2M-Hill did a drag-out study for Merix Corporation sometime in the mid-80s (our
efforts to obtain the report from Merix were unsuccessful). CH2M-Hill used a bath tank
and One rinse tank and dipped the boards in the bath and rinsed them sequentially and
monitored the conductivity of the rinse tank. The boards were vertical and had no holes
(interlayer boards about 20 mils thick), but the hang time and other variables can only be
found in the original report. The amount of drag-out was il/z gallons of process bath
liquid per 3,000 ft2 (102 mL/m2) of board area (one side only).
• Holes make a difference for drag-out since the holes are small enough that the liquid does
not drain out of them very well. "Hang time" also affects the drag-out.
E-22
-------
APPENDIX E
• Horizontal lines have drag-out of about 2-5 gallons per 3,000 ft2 (39-66 mL/m2) of board
area (one side only) for boards with no holes. The drag-out is lower for horizontal lines
compared to vertical lines because of the rollers used to squeegee the water off. Vertical
boards are the older process, and the trend is to go to horizontal boards. Currently, the
industry is about 1Avertical and Vz horizontal.
• One vendor has suggested that the drag-out is about 15 gallons per 3,000 ft2 (200 mL/m2)
of board area (one side only). However, this appears too high because the experts's mass
balances on his own plating line didn't work out using this number.
• Based on the mass balances on the expert's surface finishing line, i.e., accounting for the
amount of chemicals added, consumed, and those in the waste, etc., the drag-out ought to
be about 7 gallons per 3,000 ft2 (95 mL/m2) of board area (one side only) for circuit
boards with holes, and about 3 gallons per 3,000 ft2 (41 mL/ft2) for interlayer boards.
• There are not any available computer models that could be used to predict wastewater
concentrations, flows, etc. for plating lines.
Most of the baths used at the expert's facility (Sharp 1998) have a specific gravity of about 1.08,
but the .the viscosity and surface tension are unknown. .The expert thought that chemical supply
companies know the viscosity or surface tension of the process baths, but it is nearly impossible to
get those data from the suppliers.
Summary of Drag-Out Studies
Table 7 summarizes the reported drag-out quantities from researchers and practitioners.
E-23
-------
APPENDIX E
Table 7. Summary of Reported Drag-Out Volumes in the Literature.
Board
Orientation
Vertical
»
ii
it
n
ti
Vertical
Horizontal
Vertical
n
n
ti
»
n
Not specified
n
Vertical
Vertical
Bath
Microetch
n
n
Electroless
n
it
Not
specified
n
it
n
n
n
it
ti
Not
specified
n
19-20 g/L
& 240-250
g/LCrO3
Various
electrolytes
Conditions/Description
-t~ v/ ^"! < 'J T
Baseline
Slow withdrawal rate
Intermediate withdrawal rate & longer drain
time
Baseline
Slow withdrawal rate
Intermediate withdrawal rate & longer drain
time
CH2M-Hill study
Based on experience
Boards with holes
Interlayer boards without holes
Vertical parts, well drained
Vertical parts, poorly drained
Vertical parts, very poorly drained
Rack plating (used to estimate metals in
wastewater for design of wastewater
treatment system)
Drag-out value assumed in order to compare
costs of rinsing alternatives
Drag-out value assumed to evaluate waste
minimization
Studies at varying drainage angles, drainage
times, and withdrawal rates
Experimental determinations to test
theoretical equations
Drag-Out,
rnL/m*
130
72
76
65
32
31
103
27-67
95
41
161
82
160
203
162
160
12-65
18-94
Reference
Pagel 1992
tl
II
It
11
Sharp 1998
"
n
"
Pinkerton 1984
n
H
Hansan&
Zabban 1959
Yost
Chang &
McCoy 1990
Sup 1990
Sup 1992
Suggested by Pinkerton as being the absolute minimum for drag-out on a vertical sheet.
Drag-Out Prediction Equations
Kushner (1951a) was one of the first researchers to study drag-out in detail. Kushner
distinguished two stages in the generation of drag-out The first stage is the "withdrawal" stage
in which the work piece is moving out of the liquid but. is still in contact with it. The second stage
is "drainage" in which the work piece is completely out of the liquid, but is still over the bath and
liquid is still running off the piece. Kushner considered the withdrawal stage the more important,
because the withdrawal determined the thickness of the adhering liquid film. The factors that
E-24
-------
APPENDIX E
control the film thickness are the velocity of withdrawal, viscosity of the liquid, density of the
liquid, and surface tension of the liquid although he believed surface tension was a minor factor.
Using dimensional analysis, Kushner derived the following equation:
f=K
where:
f
K
V-
n
p
g
m
v
Eqn5
film thickness
unknown constant determined by experiments
velocity of withdrawal
viscosity
density
acceleration of gravity
unknown exponent determined by experiments
Based on experimental work of others, Kushner concluded that the best fit equation was equation
3 presented earlier:
/ = 0.02,
Eqn3
Note that although equation 3 was derived by dimensional analysis, it does not appear
dimensionally consistent, because the acceleration of gravity is dropped as a term. This is also the
equation referenced by Pinkerton and Graham in the Electroplating Engineering Handbook
(1984). Importantly, this equation is for work pieces with smooth surfaces, unlike PWBs which
have many small holes. This equation will tend to underestimate drag-out for PWBs. Notably,
this is one of two equations tested by Sup (1992) and discussed above. The equation performed
poorly in predicting drag-out for a variety of electrolytes.
Kushner (1951b) argued that equation 3 gives good drag-out predictions for short drainage times,
but increasingly overestimates the drag-out with longer drainage times, because it does not allow'
for the liquid that drains off the work piece. Conceptually for a rectangular sheet, the volume of
liquid that drains off the sheet is: .
Eqn6
volume of liquid that drains from the rectangular sheet
area of the sheet
thickness of the film that drains off the sheet
function describing a relationship between the independent variables and
thickness of the film that drains from the sheet
where:
AV
A
E-25 .
-------
APPENDIX E
Oft = surface tension of the liquid .
ta. . = drainage time
Hence, the net film thickness or the drag-out volume per unit area after any drainage time, t4, is:
Eqn?
The volume of liquid that drains from the board is a complex process and Kushner was not able to
develop a predictive equation. He did, however, make qualitative statements about the effect of
several variables. Kushner believed that viscosity was the most important property of the plating
solution. Higher viscosities tend to increase the liquid adhering to the sheet as it is withdrawn
from the bath and tend to decrease the liquid that drains. Some chemicals in particular are surface
active and have molecular structures that increase viscosity. These chemicals may cause a
"surface viscosity" that give higher drag-out. Higher densities tend to decrease the liquid
adhering to the sheet and increase the drainage. However, the increase in density due to a higher
concentration of chemicals hi solution is usually outweighed by the increase in viscosity. Kushner
gave an example of increasing a sucrose solution from 20% to 60%. This increases the density by
18% while the viscosity increases by 2700%. Lower surface tension will thin the film thickness as
the sheet is withdrawn and also increase the drainage as well as reducing the volume of the bead
of liquid along the bottom edge of the sheet. Of course, wetting agents are surface active and will
concentrate in the drag-out, and hence will be removed at a higher rate than other chemicals.
Longer withdrawal times and drain times will reduce drag-out, but Kushner believed that it is
better to have a longer withdrawal time than a longer drain time. His rationale was to start with
the smallest volume on the work piece to begin with. He also referenced work by Soderberg that
drainage times beyond 60 seconds have little effect. Finally, Kushner recommended that work
pieces be oriented to minimize the drainage distance and that the pieces be tilted.
Rinsing Theory
The primary source of the quantity of wastewater generated is rinse water. Most process baths
are followed by two rinses, but sometimes just one rinse and sometimes three rinses. The
development of rinsing theory can be traced at least as far back as Kushner (1949). Pinkerton and
Graham (1984) summarized some of the fundamental mathematical relationships for rinsing. For
a non-running rinse tank and assuming that ideal, instantaneous mixing occurs, the concentration
of a contaminant hi the rinse tank is given by:
i _
Eqn 8
where:
C,
concentration of contaminant in rinse tank after t min
concentration of contaminant solution being drug into rinse tank
E-26
-------
APPENDIX E
Vt
D
n
volume of rinse tank
volume of drag-over or drag-out on rack and work rinsing operation
number of rinsing operations in t min
Most rinse operations at larger facilities use multiple countercurrent cascade rinses. In this case,
the concentration in the effluent from the rfc rinse tank is:
Eqn9
where:
Cr
Q
t
r
concentration of contaminant in the effluent of the r* rinse tank
rate of fresh water flow
time interval between rinsing operations '
number of rinse tanks in series
Talmadge (1968) presents equations similar to the above but with an extra term to account for
imperfect mixing, i.e., imperfect removal of the contaminant from the work piece.
An approximate equation for multiple, countercurrent rinses has apparently been used by some
(Hanson and Zabban 1959; Mohler 1984):
1/r
EqnlO
Mohler (1984) discussed how rinsing equations can be used in practice. In general the rinse must
not cause a loss in product quality. There is, then, a maximum allowable concentration in the final
rinse called the "contamination limit." The ratio of the concentration in the drag-in, C0, into the
first rinse tank (or drag-out from the process bath) to the concentration in the final rinse, C^ is the
dilution factor or "rinsing ratio," CJCt. Either the contamination limit or rinsing ratio can be used
to calculate the required rinse flow rate if the other parameters are known. For example, assume
that the rinsing ratio is 5,000, there are two countercurrent rinse tanks, the drag-out volume is
100 mL/m2 of PWB, each rinse cycle rinses 15 m2 of PWBs, and the time interval between
operations is 3 minutes. Then:
cycr = 5000
D = (100mL/m2)(15m2)=1.5L
t =3 minutes
r = • 2 tanks
Solution of equation 10 yields the required rinse flow rate, Q = 35.41pm.
E-27
-------
APPENDIX E
The approach above is consistent with Kushner (1949). Kushner observed that the purpose of the
rinse tanks are to "stand guard between baths to keep one solution from mixing with another and
contaminating ft." The rinse water flow rate partially determines the concentration of carryover
into the next plating tank and thus the plating quality. Kushner believed that each rinse system in
a facility would have its own unique rinsing ratio, CJCr Kushner suggested several values for the
rinsing ratio as listed in Table 8. These values would not be valid to use for PWB manufacturing
because ft is a different system than what Kushner dealt with and Kushner gave these criteria as
approximations based on only limited data, but probably on the conservative side.
Table 8. Kushner's (1949) Suggested Rinsing Ratios.
Type of Rinse Tank
Rinse after alkaline cleaner
Rinse after acid dip
Rinse after cyanide dip
Rinse after cyanide copper
Rinse before drying (better work)
Rinse before drying (cheaper work)
Rinsing Ratio
5000 - 7000
2000 - 3000
3000 - 5000
1500 - 2500
10,000
5,000
Kushner (1979) observed that the theoretical rinsing equations as discussed above assume ideal
mixing. Kushner cited work by Talmadge showing that if mixing is very poor so that mixing is by
diffusion only, then the equations based on ideal mixing can not be used. However, Kushner
stated that experience had shown for most practical applications that the ideal mixing equations
were more accurate than equations based on diffusion as the dominant mixing mechanism.
Indeed, Talmadge and Bufiham (1961) stated that if the primary concern is to estimate the
amount of contaminants that enter the wastewater, then rinsing equations based on complete
mixing would be adequate and provide conservative answers.
Although using rinsing ratios and the rinsing equations is an interesting approach to calculating
the volume of rinse water, ft is apparently difficult to do this in practice. The contamination limits
are apparently not readily known and are influenced by upstream processes. This was also
pointed out by McKesson and Wegener (1998) who stated that.there is not standard for rinsing
that can be used to determine "manageable" concentrations of contaminants remaining on the
work. What is manageable would need to be determined for each specific process and would
depend on:
• "The type of contaminant."
• "The tolerance of the following process step for the particular contaminant in question."
• • "The effect the residual contaminants have on the work."
E-28
-------
APPENDIX E
Other Rinsing Theory Studies
Several other rinsing theory studies have been conducted by various researches. Some of these
have focused on how well the drag-out is dispersed into the rinsing tank. While interesting, these
studies are not applicable to this project, because sufficient rinsing is used in practice such that
most of the drag-out ends up in the rinse water and thence the wastewater. For example,
Talmadge and Sik (1969) developed equations to describe the dispersing of the bead of liquid at
the bottom of a plate into the rinse water. They extended previous work that used diffusion
theory to predict the residual contaminant on a plate in a rinse tank. Talmadge and Buffham
(1961) and Talmadge et al. (1962) made detailed investigations of rinsing effectiveness in the
absence of mixing or agitation other than the flow of rinse water in the tank, i.e., molecular
diffusion is the dominant mass transfer mechanism. They found in such cases that about 10% of
the contaminant is left in the film a flat sheet as compared to typically less than 0.1% when using
ideal mixing rinse equations. However, the situation is not typical of practice, and as mentioned
above, using the ideal complete mixing equations gives a conservative estimate of contaminant in
the wastewater, i.e., less contaminant is left on the board.
PWB Pollution Prevention and Control Technology: Analysis of Updated Survey Results
As part of an EPA funded project, a questionnaire survey form on pollution prevention was sent
to 400 PWB shops in 1995 and 40 shops responded. A shortened survey was sent in 1997 to 250
PWB shops in California and 45 responded for a total of 85 between the two surveys. A
summary of information relevant to this project follows (U.S. EPA 1998).
Wastewater generation. Most of the wastewater generated is from rinsing. The best estimate of
water usage is 10 gallons/(layer-ft2 of production) or 410 1/m2 which is the "wetted" surface area
and was "calculated based on the total surface area of all layers of boards manufactured." This
value is the mean of the 20 largest shops. Large shops had the most reliable data. Smaller shops
were encouraged to estimate their data if they did not know, and this made their data suspect.
Recycle, recovery, and bath maintenance The survey revealed several practices for recycle,
recovery, and bath maintenance:
• Nearly all shops responding to the survey reported using off-site recycling for one or more
of their spent process baths although the percentage recycled for each bath type was not
reported. The most common bath sent for recycle was spent etching because the baths
have high copper concentrations of about 150 g/L. About 80-85% of the responders used
an ammoniacal etchant and most of the rest used cupric chloride. The volume of spent
ammoniacal etchant solutions generated was 1 gallon per 30 ft2 (1.41/m2) of inner- and
outer-layer panels. Other types of spent baths were far less likely to be sent off-site for
recycle. Tin and/or tin-lead stripping solutions were the next most common spent bath
sent off-site and was reported by 20% of the survey responders. Approximately 50% of
the responders used a tin outer-layer etch resist and 50% used a tin-lead etch resist. Only
10% of responders indicated that spent rack stripping solutions are sent off-site. This
E-29
-------
APPENDIX E
stripping solution results from removing plating deposits from racks used to hold the
PWBs. This solution can be a significant waste. Based on the survey report, we will
assume that only spent etchant baths are sent off-site for recycle.
• The use of various technologies to recycle and recover baths and waste streams on-site
varied. Ion exchange was used by 45% of the responders to treat and recover discharges,
but many times this was part of their waste treatment system.
• The volume of wastes generated from spent baths was estimated as shown in Table 9.
Wastewater treatment. Wastewater treatment systems removed the metals by conventional
precipitation systems, ion exchange, or a combination of the two. Wastewater treatment sludges
generated are typically (88% of responders) sent off-site for recycle rather than disposed of in a
landfill. Sludge generation data were few. The three largest facilities reporting data had sludge
generation rates of 0.02, 0.31, and 0.24 kg/m2. The smallest number, 0.02 kg/m2, came from a
facility making only single sided boards whereas the other two had a larger mix of products which
generated more waste.
Drag-out reduction practices. Table 10 shows the drag-out reduction or recovery practices used
by the responders.
Drag-out reduction can reduce pollution, but it can cause problems for the process baths due to
greater build-up of contaminants in the bath. One or more bath maintenance techniques may be
required.
Water Use Rates from Survey of MHC Facilities
As part of a U.S. EPA sponsored research project, the University of Tennessee CCPCT (1997)
surveyed MHC PWB plating facilities. Part of the survey addressed water use for various MHC
process alternatives. Table 11 shows the estimated water consumption for MHC alternatives
based on the survey data and normalizing assumptions.
These water consumption rates are of the same order of magnitude as those from the U.S. EPA
(1998) survey discussed earlier which estimated water usage to be 10 gallons/(layer-ft2 of
production) as the mean of the 20 largest shops.
E-30
-------
APPENDIX E
r .Process
Etching, inner and outer layers
Dry film resist developer
Dry film resist stripper
Tin-lead stripper
Soldermask developer
Microetch; inner and outer layers
Sulfiiric acid dips
Electroless copper
Board trim
Waste
-a
Spent etchant
Spent developer
Spent stripping solution
Spent stripping solution
Spend developer
Spent micro-etchant
Spent sulfuric acid baths
Waste electroless Cu bath
Waste copper-clad material
~ Volume1
(per 1,000ft2 of
4 layer boards)
140 gallons
200 gallons
6 gallons
17 gallons
60 gallons
16 gallons
12 gallons
26 gallons
187.5 ft2, 42.9 IbsCu
Volume1
(per m2 of
4 layer boards)
5.7 liters
8.1 liters
0.24 liters .
0.69 liters
2.4 liters
0.65 liters
0.48 liters
.1.1 liters
0.1 875m2, 19.6 kg
a) Ammoniacal etchant used for both inner- and outer-layers, 70% of copper foils etched, 1 oz. copper used on all'
layers, and 20 oz/gal carrying capacity of etchant.
b) 50% of film developed (30% outer, 70% inner), developer carrying of 3 mil-ft2/gal, and 1 mil film is used
throughout.
c) 50% of film stripped (70% outer, 30% inner), stripper carrying capacity of 100 mil-ft2/gal, and 1 mil film is used
throughout.
d) 30% metal area, tin-lead resist is 0.3 mil thick and stripper capacity of 15 oz/gal of metal.
e) 30% of mask developed, 1 mil thickness, 10 mil-f^/gal carrying capacity.
f) Oxide, electroless Cu, and pre-pattern plate microetches (50%, 100%, arid 30% of surface area etched, respectively)
considered. Many facilities may employ additional baths.
g) Microetches average etch and 4 oz/gal carrying capacity.
h) Bath life of 1 gallon/500 ssf, 3 sulfuric dips (oxide, electroless copper, and pattern plate lines).
I) 18x24 panels with 0.75 inch thief area and 0.25 inch spacing of 6 step-and-repeats, outer layer 2 oz copper (80%
trim area), inner layer 1 oz copper (50% trim area).
E-31
-------
APPENDIX E
Table 10. Drag-out Reduction or Recovery Practices Used by the Responders,.
Drag-Out Redaction or Recovery Practice
1 "i • * *s ,- — -,
s, , \
Allow for long drip times over process tanks
Have drip shields between process and rinse tanks
Practice slow rack withdrawal from process tanks
Use drag-in/drag-out rinse tank arrangements
Use drag-out tanks and return contents to process baths
Use wetting agents to lower viscosity
Use air knives to remove drag-out
Use drip tanks and return contents to process baths
Use fog or spray rinses over heated process baths
Operate at lowest permissible chemical concentrations
Operate at highest permissible temperatures
PWB Responders
Using, %x
76.3
60.5
52.6
34.2
34.2
31.6
26.3
10.5
10.5
7.9
. 5.2,
Plating Shops
Using, %2
60.43
56.9
38.13
20.83
61.03
32.4
2.23
27.03
18.93
34.6
17.9
1 Data from PWB survey.
2 Data from 1993-1994 survey of for the metal finishing industry.
3 Data are for manually operated methods, which are the predominant type for the plating operations surveyed during
the NCMS/NAMF project
Table 11. Water Consumption Rates of PWB MHC Alternatives.
.Process Type
Electroless copper, non-conveyorized
Electroless copper, conveyorized
Carbon, conveyorized
Conductive polymer, conveyorized
Graphite, conveyorized
Non-formaldehyde electroless copper, non-conveyorized
Organic-palladium, non-conveyorized
Organic-palladium, conveyorized
Tin-palladium, non-conveyorized
Tin-palladium, conveyorized
Water Consumption1
(gal/ft3)
11.7
1.15
1.29
0.73
0.45
3.74
1.35
1.13
1.80
0.57
(t/m2)
476
46.8
52.5
30
18
152
54.9
46.0
73.2 '
23
Based on wetted board surface area.
E-32
-------
APPENDIX E
RESEARCH APPROACH
The objective of this study was to develop and validate methods to predict the quality of waste
water generated from PWB manufacturing processes. The methods can then be used to compare
alternative manufacturing processes in the PWB industry. In the DFE studies, industrial and
environmental exposure and risk are evaluated on a chemical-specific basis for individual .
manufacturing operations. Wastewater data collected during routine regulatory sampling are
inadequate for these purposes because data are collected for only a few specific pollutants and the
samples contain wastewater from the entire plant rather than an individual process line. For these
reasons, a mass-balance calculation is the most suitable approach to estimating the load of each
pollutant emanating from a given process line.
The literature review revealed that drag-out was the source of most of the contaminants in the
wastewater from a given process. Process-specific waste loads originating from drag-out can be
estimated by the product of the drag-out volume and the chemical concentration in the process
baths. The latter are determined as an existing component of the DFE process. However,
according to the literature review, drag-out volume from PWBs and other flat, vertical pieces can
vary between about 10 and 120 mL/m2. Drag-out was affected by variables such as bath
chemistry, board withdraw rate, drain time, and orientation of the boards during withdraw.
Board surface characteristics and the number and geometry of holes drilled in the board may also
be significant, but these variables have not been systematically investigated to date. Equations
presently available in the literature fail to accurately predict the volume of drag-out from vertical
plates (Sup 1992).
The MHC process was selected as the basis of the research because a significant data base already
existed for this process as a result of the previously concluded DFE project. Also, the research
team was most experienced and familiar with this process line. The results of this work apply to
other PWB processes that employ process baths in which the boards are vertically oriented.
The specific steps in the research plan were:
• To conduct limited laboratory drag-out experiments for the purpose of supplementing
existing data in the literature.
• To identify or develop an accurate and comprehensive drag-out model for PWB using a
data-base that includes data developed in this study and by others.
• To develop a computer model to predict wastewater quality and quantity from a PWB
processes that incorporates the new drag-out model.
• To validate the model using data from process bath and rinse water samples collected
from three MHC process lines.
E-33
-------
APPENDIX E
LABORATORY DRAG-OUT EXPERIMENTS
Laboratory drag-out experiments were conducted to supplement existing drag-out data in the
literature. Existing drag-out equations do not accurately predict the effect of fluid properties on
drag-out from vertical flat pieces such as PWBs (Siip 1992). While some studies have
investigated the effect of viscosity, another parameter that may exert significant influence, surface
tension, has received virtually no attention. The scope of this study did not allow a
comprehensive evaluation of the effect of these parameters. Instead, an alkaline cleaner bath was
selected as a bath that was more difficult to drain and a microetch bath was selected as one that
would be relatively easy to dram. During the study, viscosity and surface tension would be
measured to gain an indication of the relative influence of these parameters on drag-out.
The procedures for the laboratory drag-out experiments were devised to simulate conditions
occurring in the PWB manufacturing process. The drag-out volume was measured
gravimetrically as the boards were withdrawn from the process tanks. Experiments were
conducted using two heated process baths to determine the range of expected drag-out volumes
under various conditions. Because the alkaline cleaner/condition and microeth baths have
significantly different chemical compositions and properties, these baths were chosen for the
experiments to provide a realistic range of drag-out volumes. The board size was 0.457 m by
0.610 m. Experimental conditions that were studied were the orientation of the board during the
drain time, the length of the drain time, the board withdraw rate from the bath, and shaking the
board at the beginning of the drain period. Withdraw rates of 0.076 m/sec and 0.305 m/sec were
tested, and the boards were drained with the long edge horizontally, vertically, or at a 45° angle.
Drain periods of 10 seconds, 20 seconds, and 30 seconds were studied. The basic operating
conditions (BOC) for the majority of the tests were: 0.076 m/sec withdraw rate, 10 second drain
time, no shaking after board withdraw, 45° drain angle, and the board oriented with the long edge
horizontal. Nine sets of experiments were conducted on each bath for a total of eighteen drag-out
experiments. Several additional experiments were conducted with the microetch bath for a drilled
board with a different hole density and design. The matrix of experimental conditions that were
tested for each of the two baths is presented in Table 12.
For the alkaline cleaner/conditioner experiments, generally five repetitions were made for each
condition, with the circuit board remaining submersed in the bath for one minute on each test.
Since the etching process changed both the properties of the circuit board and the chemical
composition of the bath, only three repetitions for each condition were performed and the boards
were only allowed to remain submersed for 30 seconds. These conditions were taken into
account by assuming that the copper etch rate would remain constant over the duration of the
experiments. This assumption was verified by weighing the boards before and after the tests to
determine the mass of copper etched from the board.
E-34
-------
APPENDIX E
Table 12. Experimental Matrix for Laboratory Study of Drag-out Volumes for
Each Bath Type.
Experimental
Conditions
0.076 m/sec withdraw
45° drain angle
10 sec drip time
no shaking
0.076 m/sec withdraw
long edge horizontal
10 sec drip time
no shaking
0.076 m/sec withdraw
long edge vertical
10 sec drip time
no shaking
0.076 m/sec withdraw
45° drain angle
20 sec drip time
no shaking
0.305 m/sec withdraw
45° drain angle
30 sec drip time
no shaking
1.0 ips withdraw
45° drain angle
10 sec drip time
no shaking
0.076 m/sec withdraw
45° drain angle
10 sec drip time
shake board
Drilled Board
• •
•
•
•
•
- •
•
Undrilled Board
•
Drilled, Etched Board
•
Apparatus
10 cm by 61 cm by 76 cm high density polyethylene (HDPE) tank, supported and
stabilized to prevent tipping. •
Magna-Whirl Constant Temperature Water Bath, Model MW-1140A-1.
Pump, ITT Jabsco Self-Priming, Model 12290-0001, 115 volt, 3.3 amp, with thermal
overload protection.
6 m of 1.3 cm diameter stainless steel tubing, coiled to fit inside bottom of HDPE tank.
1.3 cm I.D. Nalgene tubing, lab/food grade, with connection clamps.
48 liters bath solution (Alkaline Cleaner/Conditioner or Microetch).
Mettler Toledo Electronic Analytical Balance, Model PR5002, Maximum 5100 grams,
with cardboard air current shield.
E-35
-------
APPENDIX E
0.457 m by 0.610 m circuit boards (copper clad with holes; copper clad without holes;
etched, with holes).
Plastic bags, 0.50 mil, 1101 capacity.
Whittner Taktell Super-Mini Metronom, Model 886051, set at 120 beats per i
Laboratory clamps and clips.
5.
6.
7.
8.
• minute.
Procedure
For the first set of experiments, the Alkaline Cleaner/Conditioner bath was prepared
according to the manufacturer's specifications by filling the HDPE tank with 24 L of
deionized water. Next, 2.88 L of Electro-Brite ML-371 were added, and the tank was
brought to a volume of 48 L with deionized water to produce a 6% (by volume)
concentration. The solution was gently mixed. For the second set of experiments, the
Microetch bath was prepared according to the manufacturer's specifications by filling the
HDPE process tank with 24 L of tap water and adding 720 g of copper sulfate
pentahydrate (CuSO45H2O) and 8.64 L of 66° Baume sulfuric acid (H2SO4). The acid was
added very slowly, taking care that the temperature of the mixture remained below 54° C.
A laboratory thermometer was inserted into the mixture to monitor temperature. Next,
3.34 L of Co-Bra Etch Inhibitor Makeup were added, and the mixture was brought to a
volume of 48 L with tap water.
The stainless steel heating coil was placed into the HDPE tank containing the simulated
bath. The coil inlet was connected to tubing from the water bath (with the in-line pump),
and the coil outlet connected to tubing discharging back to the water bath. The
experimental set up is presented as Figure 4.
The Magna-Whirl water bath was filled with approximately 95 liters of hot tap water. The
water bath heater and pump were turned on, allowing the bath to equilibrate to 57° C for
the alkaline cleaner/conditioner, and 52° C for the microetch bath. The water bath
thermostat was set, and a thermometer was placed in the bath to monitor the bath
temperature.
The bath temperature, pH, and density were measured in-situ in the tank. Conductivity,
viscosity, and surface tension were measured on a sample collected from the tank.
Analyses were performed as described later in the section entitled: COLLECTION AND
ANALYSIS OF FIELD SAMPLES.
The circuit board was cleaned with tap water and detergent, and thoroughly rinsed with
deionized water. The board was dried using compressed air to ensure no moisture
remained entrapped in the holes.
The board was centered on the analytical balance, and the weight was recorded to the
nearest 0.01 g.
A clean new plastic bag was weighed on the analytical balance, and the results recorded to
the nearest 0.01 g.
The plastic bag was opened, and carefully attached to the outside of the HDPE tank using
small laboratory clips.
E-36
-------
APPENDIX E
9. The metronome was turned on, and two laboratory clamps were attached to the circuit
board to serve as handles. The circuit board was slowly lowered into the tank so the
entire surface was completely submerged in the bath. The board was agitated slightly to
remove entrapped air bubbles, and then allowed to remain submerged for approximately
one minute in the alkaline cleaner/conditioner bath or 30 seconds in the microetch bath.
The process was timed by counting ticks on the metronome.
10. The board was, removed vertically at the appropriate withdraw rate, stopping several
inches above the bath surface. Depending on the experiment, the board was then either
held steady or given one quick shake, and the board held so that its edge was either level
or at a 45° angle during the allotted drain time. The appropriate withdraw rates, drain
positions, and drain times were specified in the Table 12. Both the withdraw rate and drip
time were timed by ticks of the metronome.
11. The board was immediately placed into the plastic bag attached to the tank. Extra care
was taken to ensure that any drips after the specified drain period fell into the bag, and
that the sharp corners of the board did not puncture the bag.
. 12. The clamps were removed from the board, along with the clips holding the bag to the
tank. The bag was carefully sealed, removing as much air as possible.
13. The sealed bag containing the circuit board and drag-out was centered on the analytical
balance and weighed, the results were recorded to the nearest 0.01'g.
14. The circuit board was carefully removed from the bag, and the process was repeated,
beginning with weighing a clean new plastic bag.
15. After the specified number of runs were completed for each set of conditions, the bath
temperature, pH, and density were again measured in-situ in the tank. Conductivity,
viscosity, and surface tension were measured on a sample collected from the tank.
Analyses were performed immediately after collecting the sample, and the results were
recorded. - •
16. The drag-out volumes were calculated.
Before the actual drag-out experiments were conducted using PWB bath chemicals, a series of
four preliminary tests were conducted to validate the proposed methodology and to verify that the
drag-out could be measured accurately and precisely. The preliminary tests also served as
practice runs, and allowed for any necessary adjustments to the procedure and apparatus. The
coefficients of variation for the first two tests were 0.039 and 0.056, for eleven and nine trials,
respectively. The coefficients of variation in the third and fourth tests improved to 0.007 and
0.008, respectively, for series of seven trials each. Since preliminary tests were not designed to
cover the full range of operating variables, the following representative variables were selected:
1) ambient temperature tap water was used to simulate bath chemicals; 2) a 0.265 m x 0.457 m
drilled etched board was used in the first two preliminary tests, and a 0.457 m by 0.610 m drilled
copper clad board was used for the third and fourth tests; and (3) the circuit board was withdrawn
at 0.15 m/sec, given one quick shake after removal, and allowed to drip for 10 seconds.
E-37
-------
APPENDIX E
Quality Assurance and Quality Control (QA/QC)
Prior to the experiments, all laboratory equipment was thoroughly cleaned with detergent
followed by a thorough deionized water rinse. The analytical balance used for weighing the
boards was allowed to warm up for at least 30 minutes before any measurements were made. The
balance was calibrated using calibration weights at the beginning and end of each laboratory
session, to ensure the instrument had not drifted. A large shield was placed around the balance to
decrease the effects of drafts while weighing the board.
Prior to mixing the actual baths, 500 ml batches of the solution were prepared per the
manufacturers' product information sheets. Measurements of viscosity, specific gravity, surface
tension, conductivity and pH were compared between the 500 ml batches and the full bath
volume. Temperature was monitored continuously during the drag-out experiments in the baths
by suspending a laboratory thermometer in the tank. Before the tests, the timing of the
metronome was checked with a clock to ensure proper timing. The tank was positioned in front of
a fume hood for adequate ventilation, and a large strip of tape was affixed to the fume hood shield
at a 45° angle from the horizontal to use as a guide during drain periods. Personal protection
equipment such as safety goggles, gloves, and aprons were used whenever feasible. All waste
material including plastic bags contaminated with the drag-out chemicals and the used bath
solutions were stored for proper disposal. All laboratory experimental information and data were
recorded in a laboratory notebook, with carbon copies given to the principal investigators upon
test completion.
Results and Discussion
Results of the laboratory drag-out volume experiments are presented in Tables 13 and 14 for the
alkaline cleaner/conditioner and microeth baths, respectively.
Table 13. Drag-Out Results for Alkaline Cleaner/Conditioner Bath
Test
BOG
BOC, board edge horizontal
BOC, board edge vertical
BOC, 20 sec. drip time
BOC, 30 sec. drip time
BOC, 1 fps withdraw
BOC, with shake
BOC
BOC
Board Type
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
undrilled
drilled, etched
Drag-Out (ml/sq.m)
77.8
75.6
81.3
68.2
64.5
98.7
77.8
38.6
89.2
Coeff. of Variation
0.032
0.015
0.021
0.040
0.047
0.013
0.032
0.016
0.038
Note: Design 1 , 56 1 9 holes; Design 2, 7824 holes.
E-38
-------
APPENDIX E
Table 14. Drag-Out Results for Microetch Bath.
Test
BOC (2/2/99)
BOC (2/1 3/99)
BOC (2/13/99)
BOC, board edge horizontal
BOC, board edge vertical
BOC, 20 sec. drip time
BOC, 30 sec. drip time
BOC, 1 fps withdraw
BOC, with shake
BOC
BOC, etched board
BOC, etched board
Board Type
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
Drag-Out, ml/sq m
108.9
107.8
93.4
120.9
113.0
98.1
94.4
133.1
111.9
69.8
112.3
118.3 -
' Coeffi of Variation
0.043
0.023
0.038
0.006
0.006
0.015
0.007
0.016
0.021
0.038
0.022
0.021
Note: Design 1,5619 holes; Design 2,7824 holes.
The drag-out volume for each experimental condition was calculated using the mean drag-out
weight from the group of tests for the specific condition. This was generally five runs for the
alkaline cleaner/conditioners and three runs for the microetch. In addition to calculating the mean
drag-out weight (in grams), the standard deviation and the coefficient of variation of the
measurements were checked for each condition. The coefficient of variation was less than 0.05
for all experiments.
The mean drag-out volume for all experimental conditions for the alkaline cleaner/conditioner was
74.7 ml/m2, which is approximately 30% less than the mean drag-out volume of 108 ml/m2 for the
microetch bath. The mean drag-out for all experimental conditions for both baths combined was
91.1 ml/m2, and was calculated using only data from the same board hole design so as not to skew
the results. It appears that drain time has an affect on drag-out volume, as reflected in the
decreasing drag-out volumes as drain time increased. It also appears that the drag-out volume
increases as the board withdraw rate decreases. Board tilt and orientation did not appear to affect
the drag-out volume; however, drilled boards had more drag-out than undrilled boards, as
expected. .
Results from the microetch experiments compare favorably to those performed at Micom, Inc.
(Pagel 1992), although a direct comparison was difficult since operating conditions were different.
Board hole density for both tests were similar, with Micom boards having 33,000 holes/m2
compared to 28,000 holes/m2 for the boards used in the microetch experiments in this study.
Pagel's drag-out volumes appear to be less than those measured in this study. At a withdraw rate
of 0.20 m/sec and drain time of 12.1 sec, Pagel reported a drag-out volume of 76.4 mL/m2.
Under similar conditions, specifically a withdraw rate of 0.305 m/sec and a drain time of 10
seconds, this study resulted in a drag-out of 130 mL/m2. Other differences in experimental
E-39
-------
APPENDIX E
procedures that could affect drag-out volumes include: 1) a 45° drain angle used in this study,
compared to a 0° angle used by Pagel; 2) Pagel's experiments included drag-out associated with
the racks; and 3) drag-out was measured by completely different approaches; specifically, Pagel
used a concentration approach whereas this study used a weight approach.
Analyses of parameters for the alkaline cleaner/conditioner and microetch simulated baths were
performed before the drag-out tests were run, and again after the tests were completed. Results
of the tests are presented in Tables 15 and 16.
Parameter
pH
Conductivity mS/cm
Specific Gravity
Surface Tension, dynes/cm
Viscosity, cP
Before Experiments
8.65 @ 58°C
0.21 @ 35°C
8.65 @ 57°C
34.7.
0.85
After Experiments
8.47 @ 57°C
0.23 @ 35°C
0.995 @ 57°C
34.7
0.87
Table 16. Microetch Bath Analyses^
Parameter
PH
Conductivity mS/cm
Specific Gravity
Surface Tension, dynes/cm
Viscosity, cP
Before Experiments
-0.42 @ 53°C
1374@22°C
1.175 @53°C
71
1.44 @49°C
After Experiments
-0.62@55°C
1562@22°C
1.205@57°C
60
0.87 @ 50°C
As expected, there was no significant variation in the bath parameters for the alkaline
cleaner/condition bath comparing values before and after the drag-out tests. There were,
however, significant variations in the microetch bath characteristics, as expected. Conductivity,
specific gravity, hydrogen ion concentration and viscosity all increased, possibly due to the
increase in copper in the bath as a result of etching from the PWBs during the drag-out tests.
E-40
-------
APPENDIX E
DRAG-OUT MODEL DEVELOPMENT
As stated previously the goal of this project was to develop and validate methods for predicting
the quality of wastewater generated during PWB manufacturing. Drag-out and bath dumps are the
two major sources of process wastewater. The literature reports drag-out rates for flat panels and
PWBs ranging from 10 to 160 ml/m2. Currently-available models utilize solution viscosity and
. withdraw rate as the primary independent variables. Siip (1992) has demonstrated that drag-out
rates predicted using these models are poorly correlated with results from experiments., Clearly
there is a need for a more a more accurate means of predicting drag-out for PWB manufacturing.
In addition to the drag-out data collected as part of this study, three data sets containing extensive
drag-out data for PWBs or flat panels were available in the literature (Sup 1990; Sii(3 1992;
Pagel 1992; Ducker). An attempt was made to develop regression models to predict drag-out
volumes as a function of PWB manufacturing practices. Possible model variables that were either
recorded or varied in each study are summarized in Table 17.
Table 17. Potential Variables for PWB Drag-Out Prediction Model.
Board Size
Withdraw Rate
Drain Time
Board Orientation
Board Angle
Board Surface
Holes
Shaking or Vibration
Bath Type
Kinematic Viscosity
Surface Tension
Sup 1990
•
• •
•
•
•
1 -
•
Sftp 1992
'
•
•
•.
•
. •
Pagel 1992
• -
•
-
•
- •
This Study
- - *
'
• •
•
•
•
•
•
•
Of the variables listed in the table above, not all were evaluated for inclusion in the model. Board
surface (etched or unetched) and shaking were not included in the parameters to be evaluated
because the little data that were available for these parameters indicated they have a minor effect
on drag-out volumes. Board orientation during draining was also not considered because
relatively few data were available and it is not one of the waste minimization practices commonly
practiced. We hypothesized that kinematic viscosity and surface tension were two fluid properties
that may be most significant in determining drag-out volumes. However, Slip (1992) showed that
drag-out volume was poorly correlated with kinematic viscosity. Furthermore, Pagel's data set
did not include data for either kinematic viscosity or surface tension of the baths and Slip's data
did not include any surface tension data. It was judged that the quantity of data and range of
values for these two variables were insufficient to justify their inclusion in the model.
E-41
-------
APPENDIX E
In the data base used to develop the model, board size (m2), withdraw rate (m/sec), and drain time
(sec) were treated quantitatively by using the numerical value of the variable. Three other
variables were treated qualitatively using indicator variables having values of 1 or 0. The
indicator variable for board angle was assigned a value of 1 if the board was angled and a value of
0 if the board edge was kept horizontal. Similarly, the indicator variable for holes was assigned a
value of 1 if it contained holes and a value of 0 if the board did not contain holes. The hole
density for the drilled boards in the data base ranged from 20,000 to 33,000 holes/m2; however,
data needed to further quantify the effect of drilled holes, such as hole diameter and aspect ratio,
were not available. Three different indicator variables were included to specify bath type:
alkaline cleaner, micro-etch and electroless copper. The obvious disadvantage of this approach is
that the model can make bath-specific predictions only for these three bath types, but insufficient
viscosity and surface tension data are available to make the model more general.
The data set was not ideal for development of the model. The work of Sup (1990, 1992) was not
specific to the PWB industry; therefore, he did not use standard PWB process baths, his boards
were smaller than those often used in the PWB industry, and his boards did not contain drilled
holes. As a result, variables describing board size and holes were strongly correlated (0.904),
making it difficult to distinguish between the effects of these two parameters. Also, Sup did not
use actual PWB process baths, thus bath type and board size were also correlated. During model
development, it was necessary to be aware of the effects that these peculiarities may have on the
developed model.
Both a linear regression model and a multiplicative regression model were tested. The linear
model was:
WR
—
a6HOLES+a7 ANGLE + asALK + a^MICRO + awELCTRS
where:
DO
SIZE
WR
DT
HOLES
ANGLE
ALK
MICRO
ELCTRLS
drag-out volume, mL/m2
board area, m2
withdraw rate, m/sec
drain time, sec
1 if the board is drilled and = 0 for undrilled boards
1 of the board is tilted during draining and = 0 if the board is kept-
horizontal
1 if the bath is an alkaline cleaner bath and = 0 otherwise
1 if the bath is a micro-etch bath and = 0 otherwise
1 if the bath is an electroless copper bath and = 0 otherwise
The multiplicative model was:
E-42
-------
APPENDIX E
DO=aQ- SIZE* • WR** • DT"3 • a6HOULS • a7
ANGLE
•a,
ALK
a,
MICRO
a
ELCTRLS
10
Eqn 12
which was rewritten in linear form for analysis by linear regression:
logZX? = Ioga0 + aj logSIZE +
ANGLEloga7 + ALKlogas
a3 logDT+ HOLESloga6 +
9 + ELCTRLS\agalo
Eqn 13
Both models were evaluated using stepwise regression (SSPS ver. 9). This procedure adds or
removes independent variables to the model based on criteria related to the reduction in the sum
of squares achieved by inclusion of the variable. The final model includes only the variables that
result in a statistically significant reduction in the sum of squares error. The stepwise regression
procedure yielded an r2 = 0.883 for the linear model and 0.814 for the multiplicative model. The
linear model was:
WR
DO = 3.63 + 694-SIZE - 180-ELCTRLS + 89.6- —
- 755-ALK + 38.6-HOLES + 29.9-WR - 0.443-DT-127 • MICRO
Eqn 14
The statistical package did not include the variables of ANGLE and WR-DT in the model because
they were not statistically significant. Inspection of this equation reveals that .all three bath-type
coefficients are relatively large negative numbers, which would cause it to predict an erroneously
large drag-out for large boards (ca. 0.25 m2) with bath-types not explicitly accounted for in the
model. For small boards (ca. 0.05 m2) used with the bath-types accounted for in the model., it
could predict negative drag-out values. These anomalies were the result of correlation of the
independent variables, as described earlier. To correct this problem it was necessary to eliminate
one of the three bath types as a variable in the model. Each of the three bath types was evaluated
for elimination, the best fit was given by eliminating MICRO as a variable (R2 =0.852). The final
drag-out model was:
DO = 18 + 201-SIZE - 60.1-ELCTRLS + 73-
WR
DT
-20.9-ALK + 26.0-HOLES + 26.1-WR - 0,355-DT
: Eqn 15
A comparison of predicted and measured drag-out volumes is shown in Figure 5. The groups of
vertically-aligned data points occur when the model predicts a near-constant drag-our for
conditions in which the measured drag-out is variable. While some of the variability is random
error, some is also the result of variation of the independent variables, indicating that the model is
not able to accurately account for all the variables that affect drag-out. A more comprehensive
E-43
-------
APPENDIX E
data base in which the independent variables are systematically varied is needed if more accurate
predictions of drag-out from PWB manufacturing processes are desired.
iou -
160 -
2 40 -
20 -
n -
• Ducker, 1999 (Alkaline Cleaner)
O Ducker, 1999 (Microetch)
T Pagel, 1992 (Microetch)
v
w
v Pagel, 1992 (Electroless Coppeer) ^ VT v
• Sup, 1992
0 Sup, 1990
8
• A
v vv 1
1^7 V
r~j
tA
^
T '
TJ8* T
QW^
1
^
D*^
20
40
60
80 100 120 140. 160 180
Predicted Drag-out (mUm2)
Figure 5. Comparison of Measured and Predicted Drag-Out Volumes.
E-44
-------
APPENDIX E
PWB WASTEWATER MODEL
Given the volume of drag-out from and chemical composition of each bath, it is possible to
calculate the mass of each contaminant that would enter the waste stream for a given PWB
process line.. A computer model was developed to facilitate such calculations. The model was
based on the following assumptions:
1. Contaminants in wastewater are from drag-out from process baths and from dumping of
some baths at the end of their useful life. Contaminants from the stripping of racks from
deposits are ignored.
2. Essentially 100% of the drag-out ends up in the wastewater, i.e., very efficient rinsing.
3. Predictions are for vertical boards only.
4. Various predictive equations reported in literature are of limited value for estimating
absolute values of drag-out as evidenced by the results of Sup's work comparing
predicted versus measured drag-out. Equation 15 was used to estimate drag-out in the
model here.
5. Insufficient information exists to include surface tension as a variable although the authors
recognize that it may be an important variable.
6. The estimate of drag-out of contaminants in g/d is based on the PWB production rate,
chemical composition of each bath, and the estimated drag-out from each bath, according
to the following equation:
kg/dof
contaminant i
fiombathj ,
'PWB production^
,rate, m2/d J
f Concentration of ^ ( drag - out from j
a in bath j, mg / LJ tbafli j, mL / m2J
Eqnl6
The model is coded in an Excel Spreadsheet and utilizes a Visual Basic Macro. The user is
required to enter information in a separate spreadsheet defining the operating conditions of the
process line and the chemical composition of the baths. The effect of bath dumps on the overall
pollutant load can be included by specifying their frequency. The model calculates the mass of
contaminants coming from each process tank, together with the contaminant mass and
concentration for the entire process line. A user's manual is included in the Appendix.
E-45
-------
APPENDIX E
COLLECTION AND ANALYSIS OF FIELD SAMPLES
Samples of plating baths and rinse waters were collected from the MHC process line from three
different PWB facilities for the purpose of verifying the drag-out model. Three process baths at
each plant were selected for sampling: microetch, electroless copper, and Anti-Tarnish. Sodium
or potassium were selected as tracers for each bath because they are common ions in PWB baths,
and they tend to be relatively stable in solution. The relative amount of sodium and potassium in
the bath and downstream rinses can be used to estimate the drag-out from each tank and to verify
the overall mass balance approach to modeling wastewater quality from PWB facilities. In
addition to sodium and potassium, fluid properties (viscosity, surface tension and specific gravity)
that might effect the quantity of drag-out were measured. Routine measurements of conductivity
and pH were taken too. The project QA/QC plan (Robinson and Cox 1998), submitted to and
approved by EPA, was followed except where field conditions necessitated minor changes.
Process Characterization
Operating practices affect the amount of drag-out and the concentration of contaminants in the
rinse-tank effluent. Extensive data characterizing the operating practices used at each site were
collected during the site visits. Operating practices potentially affecting the amount of drag-out
or the rinsing process are summarized in Tables 18-20. These data were later used to predict the
drag-out from each process bath using equation 15 and to independently calculate the drag-out
via a dynamic mass balance approach described later.
Table 18. Summary of MHC Operating Practices for the Field Sites.
Plant 1
Plant 2
Plants
Cycle Time, mih
30
37
27
•Withdraw Rate, m/sec ,
0.173
0.163
0.234
Board Tilt, ^degrees
.5
0
0
Hole Density, #/m2
100,000 to 570,000
NA
50,000
Table 19. Summary of Drip Times for Process Baths at Field Sites.
Bath
Plant 1 ME
Plant 1 EC
Plant 1 AT
Plant 2 ME
Plant 2 EC
Plant 2 AT
Plant 3 ME
Plant 3 EC
Plant 3 AT
Drip Time,
sec .
5
25
5.
10
15
10
5
10
5
E-46
-------
APPENDIX E
Table 20. Summary of Rinsing Practices Used at Field Sites.
Plant 1 ME Rinse 1
Plant 1 ME Rinse 2
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Plant 1 AT Rinse 1
Plant 1 AT Rinse 2
Plant 2 ME Rinse 1
Plant 2 EC Rinse 1
Plant 2 AT Rinse 1
Plant 3 ME Rinse 1
Plant 3 EC Rinse 1
Plant 3 EC Rinse 2
Plant 3 AT Rinse 1
Rinse Time
(min:sec)
1:20
1:00
2:10
1:00
3:20
2:00
2:05
8:00
3:55
1:15
2:00
4:20
6:04
Rinse Tank
Vol (I)
832
832
832
832
832
832
415
415
415
892
892
892
892
Rinse flow
Rate 0/min)
7.6
7.6
7.6
7.6
7.6
7.6
3.8
3.8
3.8
9.8
7.6
7.6
7.6
Rinse Water
Source
MERinse2
city
EC Rinse 2
city
AT Rinse 2
city
city
AT Rinse 1
city
H2SO4 rinse
EC Rinse 2
AT Rinse 1
city
Mixing1
1,2
1*2
u
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1
1 Mixing: • 1 = Board Agitation; 2 == Aeration. ,
Sample Collection
Samples were collected for analyses from the laboratory drag-out study tanks in the U.T
laboratory and from actual process baths and rinse tanks during the PWB industry site visits. For
the laboratory drag-out study in the UT laboratory, grab samples were collected for surface
tension and viscosity. The samples were collected directly from the experiment tank in a clean
beaker, and the analyses were immediately performed.
Samples were collected during the PWB site visits from the microetch (ME), electroless copper
(EC), and anti-tarnish (AT) process baths and their succeeding rinse tanks in the MHC process
line. Grab samples were collected using either a plastic measuring cup or a sampling beaker,
which consisted of a plastic beaker with a long handle attached. The sampling container was
thoroughly rinsed with the sampling fluid prior to sample collection. The grab sample was then
immediately transferred from the sampling cup or beaker into a clean 500 ml HPDE sample bottle
and capped. Before the sampling event, pre-printed labels were prepared in duplicate, with
one label pre-attached to the sample bottle. After the sample was collected, the remaining label
was attached to the Sub-Unit Data Collection Log, and the sample description, person taking the
sample, time of sample, sample volume, and method of preservation was recorded in ink.
Duplicate samples taken in identical manner were collected at plants 1 and 2. At plant 3, the two
samples were taken at different times in the board cycle. The first sample was taken just prior to
the boards entering the rinse tank while the second was taken just after the boards were removed.
Replicates were taken for approximately 20% of the samples. The sample bottles were sealed
E-47
-------
APPENDIX E
with color-coded tamper-proof tape (to identify the sampler and establish chain-of-custody)., and
placed in plastic lined containers for transport to the UT laboratory.
Temperature
Temperature was measured in-situ in the laboratory drag-out tanks. In the field, temperature was
measured on grab samples collected from the process and rinse tanks. Measurements were made
immediately after collection.
pH
pH was measured in-situ in the laboratory drag-out tanks. In the field; pH was measured on grab
samples collected from the process and rinse tanks. Measurements were made immediately after
collection.
Apparatus . .
• Orion Digital Portable pH Meter, Model 250A.
• Orion Triode™pH Electrode, Model 91-57BN.
Procedure for pH Measurements
I. After the meter was calibrated, the electrode was placed into the laboratory drag-out tank
or sample and agitated slightly.
2. When the pH display was stable, the pH was recorded on the Sub-Unit Data Collection
Log.
3. The electrode was rinsed with deionized water, and the process repeated.
The pH meter was calibrated prior to taking measurements for each sub unit. A two buffer
calibration was performed using the 4.01 and 7.00 buffers for the acid sub units, and 7.00 and
10.01 buffers for the alkaline sub units. The first measurement in a sub unit was made in the
samples from the last rinse tank, and the measurements progressed up-line, with the last
measurement made on the process bath sample.
Conductivity
Conductivity measurements were performed both in the UT laboratory and at the PWB site visits.
The instrument automatically compensates for temperature effects to a certain degree, except for
acids. Since many of the PWB baths and rinses were acids, and temperature could have a
significant effect on the conductance of these solutions, it was determined that all conductivity
measurements should be made at the reference temperature of 25° C. The conductivity
measurements originally made in the field at the PWB sites were re-analyzed on samples in the UT
laboratory at a controlled temperature of approximately 25° C. At the beginning of each lab
session, the conductivity meter was checked against a solution of known conductance to verify
accuracy.
E-48
-------
APPENDIX E
The conductivity measurements of the rinse tanks were within the meter range of 0.0 to 199.9
mS/cm; however, as anticipated, the values of some of the process baths were higher. Since
conductivity is a nearly linear function of total dissolved solids (Snoeyink and Jenkins 1980), a
1:10 or 1:100 dilution with deionized water was performed on the sample if the initial reading was
above the highest range on the meter. The measurement was then taken on the diluted sample,
and the meter reading multiplied by the dilution factor.
Two temperature and conductivity readings were taken on each sample, with the mean values
reported.
Apparatus
• Orion Conductivity/Temperature Meter, Model 122.
Viscosity
Viscosity was measured on site from grab samples collected from the rinse tanks, process baths,
and laboratory drag-out tanks. . .
Apparatus
• Gilmont Falling Ball Viscometer, size 1, with stainless steel ball, range 1 to 10 centipoise.
Procedure
1. The temperature of the rinse tank or process bath was taken using the laboratory
thermometer.
2. A grab sample was collected from the tank using a 2000 ml beaker. The viscometer,
stainless steel ball, and thermometer were immediately submerged into the sample for
approximately one minute to allow the laboratory equipment to equilibrate to the liquid
temperature.
3. The inside of the viscometer was rinsed with the sample, then slowly filled with rinse or
process bath liquid, making sure no air bubbles adhered to the sides of the viscometer.
4. The temperature of the liquid in the beaker was checked and compared with the tank
temperature. In general, if the temperature difference was more than approximately 5°C,
the beaker was emptied and a new sample collected.
5. The viscometer was held vertical in the center of the 2000 ml beaker. (The beaker still
contained the rinse or process liquid, which acted as a temperature bath for the
viscometer.) The stainless steel ball was carefully placed by hand into the filled viscometer,
making sure no air bubbles stuck to the ball.
6. A stopwatch was used to time the descent of the ball between the fiducial lines on the
viscometer. The time was recorded on the Sub-Unit Data Collection Log.
7. The viscometer and beaker were emptied, and the process repeated.
Using the mean descent time, the viscosity was calculated as follows:
Eqn 17
E-49
-------
APPENDIX E
where:
m
K
r
t
viscosity, centipoise
viscometer constant (0.257 with stainless steel ball, based on laboratory calibration
tests using deionized water and sucrose solutions, described below)
density of ball, mg/1 (8.02 for stainless steel ball)
density of liquid, mg/1
time of descent, minutes
The viscosity was recorded on the Sub-Unit Data Collection Log.
The viscometer, stainless steel ball, and beaker were thoroughly rinsed with deionized water prior
to the next test.
Before viscosity measurements were made in the field and on the laboratory drag-out tanks, a
series of tests were performed to establish the viscometer constant, K, for the falling ball
viscometer. The constant was obtained by measuring the time of descent of the stainless steel ball
in standard solutions of known viscosity, and was calculated using the following relationship:
-_ /f.
EqnlS
Three solutions were used in the investigation: 30 percent sucrose (by weight), 40 percent
sucrose (by weight), and deionized water. Before the sucrose solutions were prepared, the
sucrose was dried in a desiccator, and all glassware was cleaned and completely air dried. A 1000
ml volumetric flask was weighed on an electronic analytical balance, and the weight recorded to
the nearest 0.01 gram. The appropriate amount of sucrose was weighed on the analytical balance
(338.10 g and 470.60 g for the 30 percent and 40 percent solutions, respectively), and added to
the clean, dry volumetric flask. Approximately 500 ml of deionized water was added to the flask,
and the mixture agitated by swirling. Additional deionized water was added slowly, while being
swirled, until the sucrose was completely dissolved and the bottom of the meniscus reached the
1000 ml reference line on the volumetric flask. The solution was allowed to rest to allow any
entrapped air bubbles to rise. The volumetric flask containing the solution was weighed on the
analytical balance, and the temperature was measured with a laboratory thermometer; both
measurements were recorded in a laboratory research notebook.
The density of the sucrose solutions and the deionized water was calculated using the following
relationship:
Eqnl9
E-50
-------
APPENDIX E
where:
D
m
v
density, g/ml
mass of solution = mass of flask and solution - mass of flask, g/L
volume of solution, ml
Prior to the experiments to determine the viscometer constant, the sucrose solutions were gently
stirred to ensure a homogeneous mixture. A laboratory thermometer was used to measure the
temperatures of the sucrose solutions and deionized water, and the results were recorded in a
laboratory research notebook. The same procedure as described above was used except the
constant temperature bath was not needed because the experiments were done at ambient
temperature. Instead, the filled viscometer was held vertical in a 50 ml glass cylinder. The
viscometer constant, K, was determined to be 0.257 by fitting equation 17 to the experimental
time and literature values of viscosity.
Specific Gravity
Specific gravity was measured in-situ in the laboratory drag-out tanks. In the field, specific
gravity was measured on grab samples collected from the process and rinse tanks. Measurements
were made immediately after collection.
Apparatus
• Hydrometer, Fisherbrand, range 0.890 to 1.000.
• Hydrometer, Fisherbrand, range 1.000 to 1.600.
• 500 ml glass cylinder (optional).
Before the hydrometers were used for measurements for the rinse tanks, process baths and
laboratory drag-out tests, the accuracy of the instruments was verified. Hydrometer readings
were taken on deionized water and a 40 percent (by weight) sucrose solution. The temperature of
the water and sucrose solution was measured with a laboratory thermometer, and the specific
gravity measurements were compared with published values. Results of the verification for
deionized water resulted in a value 0.15% higher than the expected published value of 1,000 at
20° C, and 0.5% less than the published value of 1.176 for the 40 percent sucrose solution at
20° C.
Surface Tension
Surface tension was measured in the UT laboratory on grab samples collected from the rinse
tanks, process baths, and laboratory drag-out tanks.
Apparatus
• „ Fisher Surface Tensiomat, Model 21, with platinum-iridium ring.
• 5 cm inch diameter glass vessel, approximately 1.3 cm deep.
• Magna-Whirl water bath.
E-51
-------
APPENDIX E
Procedure
1. ,' A water bath was prepared to simulate the temperature of the rinse tank or process bath as
measured in the field and recorded on the Sub-Unit Data Collection Log.
2. The rinse tank or process bath sample bottles were placed in the water bath, and allowed
to equilibrate to the bath temperature. The water bath and sample temperatures were
intermittently monitored using the thermometer. The sample bottles remained in the water
bath until used for the surface tension measurement.
3. The clean platinum-indium ring was placed on the hook on the lever arm of the tensiomat.
4. A clean 5 cm diameter glass vessel was filled with a portion of the sample (transferred •
immediately from the water bath) and placed on the sample table inside the tensiomat.
5. The sample table was raised until the ring was immersed in the liquid to a depth of
approximately 3 mm.
6. The torsion arm on the tensiomat was released, and the instrument was adjusted to a zero
reading by turning the knob on the right side of the case until the index and its image were
in line with the mark on the mirror. Care was taken to ensure the ring remained in the
liquid by adjusting the height of the sample table. The knob on the front of the case
beneath the main dial was adjusted until the vernier read zero on the outer scale of the
dial.
7. The sample table was lowered until the ring was at the surface of the liquid. At the same
time, the knob on the right side of the case was adjusted to keep the index hi line with the
mark on the mirror. The two simultaneous adjustments were continued until the distended
film at the surface of the liquid broke.
8. The reading on the scale at the breaking point (surface tension hi dynes per centimeter)
was recorded on the Sub-Unit Data Collection Log.
9. The liquid was emptied from the glass vessel, and the process was repeated.
10. Both the platinum-hidium ring and glass vessel were rinsed with deionized water prior to
the next test.
Prior to the surface tension tests,, the calibration of the tensiomat was checked and the platinum-
iridium ring was thoroughly cleaned.
To verify the calibration according to the instrument's instruction manual, the ring was placed on
the lever arm and the instrument was adjusted to a zero reading. A 600 mg piece of aluminum foil
was placed on the ring, and the knob on the right side of the case was adjusted until the index and
its image were hi line with the mark on the mirror. The dial reading was recorded, and compared
with the calculated surface tension:
S =
Mg
2L
Eqn 20
where:
S
M
g
dial reading = apparent surface tension hi dynes/cm
weight (0.6 grams)
acceleration of gravity (980 cm/sec2)
E-52
-------
APPENDIX E
L = mean circumference of ring (6.00 cm)
The platinum-indium ring was cleaned per the manufacturer's instructions: the ring was: 1)
soaked in concentrated nitric acid for approximately 2 minutes, then rinsed with deionized water;
2) rinsed with acetone, followed by deionized water; and 3) flamed with a Bunsen burner.
Before surface tension measurements were made, the surface tension of deionized water was
checked at 20°C to verify accuracy. Seven measurements were made, with a mean value of 74.96
dynes/cm, a standard deviation of 2.03 dynes/cm. This mean value is 4.2 percent higher than the
expected value of 72 dynes/cm for the deionized water.
Metals Analysis
Sodium and/or Potassium analyses were conducted in the UT laboratory on grab samples
collected from the process baths and rinse tanks.
Apparatus
• Allied Analytical Systems Atomic Absorption Spectrophotometer, IL Video 12, Serial
Number 1857.
Sartorius Analytical Balance, Model AC 120S, UT ID Number 427286.
Reagents ,
• Sodium calibration standard, Fisher Scientific, 1000 mg/L.
• Potassium calibration standard, Fisher Scientific, 1000 mg/L. .
• Potassium chloride (KC1), Fisher Scientific, certified grade.
• Lanthanum chloride (LaCl 6H2O), Fisher Scientific, certified grade.
Procedure
1. Stock potassium chloride solution was prepared by dissolving 23.84 g. of potassium
chloride in 250 ml of deionized water in a volumetric flask. This produced a solution of
50,000 mg/L as K, which was used as an ionization.suppressant for the sodium samples.
A stock solution of lanthanum chloride was prepared by dissolving 12.72 g. of lanthanum
chloride in 100 ml of deionized water in a volumetric flask. This produced a solution of
50,000 mg/L as La, which was used as an ionization suppressant for the potassium
samples.
2. Sodium and potassium standards were prepared by diluting the Fisher Scientific calibration
standards with deionized water to achieve the desired standards concentrations.
3. The samples were prepared by performing dilutions with deionized water to get the
anticipated analyte concentrations within the linear range of the instrument. Volumetric
pipettes and volumetric flasks were used, and the samples were transferred to new, clean
125 ml HOPE sample bottles. Samples were acidified with ultrapure nitric acid, and
ionization suppressants were added to achieve a concentration of 2000 mg/L as K for the
sodium samples, and 1000 mg/L as La for the potassium samples.
E-53
-------
APPENDIX E
4. The appropriate lamp was inserted in the atomic absorption spectrophotometer, and a
safely check of all settings was performed. The instrument electronics were turned on and
allowed to warm up for approximately 30 minutes.
5. The instrument printer, compressed air, and acetylene were turned on. The pilot was lit,
the flame adjusted, and the sampling tube was placed in a fresh beaker of deionized water.
6. The instrument was calibrated with the appropriate sodium or potassium standards. A
standards curve was printed, and a linear regression performed to check linearity of the
curve. If the value of r2 value was below 0.9950, the instrument was re-calibrated with
fresh standards.
7. The prepared samples were analyzed, beginning with the rinse samples and progressing
up-line to the process tank. Approximately ten analyses were run per sample, each lasting
approximately eight seconds. Results were printed and transferred to an Excel
spreadsheet.
8. The method of standard additions was performed on process bath samples to reduce
matrix effects. The samples were diluted 1:1 with known standards and analyzed in the
absorption mode. Generally, 0, 50,100 and 200 mg/L standards were used for potassium
analyses, and 0,20, 50 and 100 mg/L standards were used for sodium analyses; however
there was some variation since it was necessary to keep concentrations within the
instrument's linear range. A plot of absorption verses concentration of added standards
was then prepared, from which the actual concentration in the sample was derived. If
necessary, standard additions were performed on the succeeding rinse tanks, as described
later in this section.
Before and during the atomic absorption analyses, all laboratory glassware and sample bottles
were acid washed in accordance with Standard Methods.
The analyte (sodium or potassium) was determined based on process bath composition, as
provided by either industry representatives, manufacturers' material safety data sheets, or
previous research conducted by the University of Tennessee's CCPCT.
Because of the extremely high anticipated concentration of analyte in some of the process baths,
along with the wide range of anticipated concentrations between the process baths and rinse
tanks, atomic absorption analyses were conducted using the least sensitive wavelengths (330.2 nm
for sodium, and 404.4 nm for potassium) whenever possible. Dilutions were still necessary on
many of the samples. For sodium samples with very low sodium concentrations, it was necessary
to use the most sensitive wavelength of 589.0 nm.
The instrument was calibrated at the beginning of each lab session by using generally five
calibration standards within the linear range of the instrument, including a zero standard. The
standards used for the least sensitive wavelength for sodium (330.2 nm) were usually 0,20, 50,
100, and 150 mg/L; however these occasionally varied depending on the anticipated concentration
of the sample. In all cases, the standards were chosen to best bracket the sample concentration.
Standards used for the most sensitive sodium analyses (589.0 nm wavelength) were usually 0,
0.25 0.50. 0.75,1.0 and 1.25 mg/L. Calibration standards for the least sensitive wavelength for
E-54
-------
APPENDIX E
potassium (404.4 nm) were usually 0, 50, 100, 200 and 600. As with the sodium analyses,
standards were chosen to best bracket the sample potassium concentration. Standards checks
were performed during the measurements to ensure the instrument had not drifted. The checks
usually were performed after every four or five measurements, but always after ten measurements
were taken.
The samples were prepared for analysis by dilution with deionized water to achieve an anticipated
analyte concentration within the linear range of the instrument. The anticipated concentrations
were based on previous research conducted by the University of Tennessee's CCPCT. Alkali
salts were added to the samples and standards as an ionization suppressant. Potassium chloride
was added to sodium samples at 2000 mg/L, and lanthanum chloride at 1000 mg/L was added to
the potassium samples. Process and rinse tank samples and standard solutions were acidified to
pH < 2 in accordance with Standard Methods, using ultrapure concentrated nitric acid.
Electroless copper samples were not acidified due to the possibility of the baths containing
cyanide.
As an interference check, a standard additions analysis was performed on one sample for each
process bath, and compared with analysis results performed without standard additions.
Whenever there was a difference greater than 10 percent between the two measurements, a
standard addition analysis was performed on the duplicate bath sample, and the standard addition
results were used. If standard additions were necessary for the process bath samples, the
succeeding rinse tank samples were also checked, to determine if standard additions should be
used.
Quality Assurance and Quality Control (QA/QC)
Prior to the site visit to collect the samples, the 500 ml new HDPE sample bottles were
thoroughly cleaned with detergent, triple rinsed with deionized water, and allowed to air dry.
Field blanks were used to monitor any contamination from the bottles. The field blanks were pre-
labeled and filled with deionized water in the UT laboratory prior to the site visits. During the
visit, the bottles were opened for approximately, two minutes, then re-sealed.
All laboratory equipment transported to the site was thoroughly cleaned according to Standard
Methods prior to leaving the UT laboratory, and was again thoroughly cleaned between sites. All
laboratory equipment, including reagents and deionized water was transported from the UT
laboratory, including cleaning supplied. The samples remained in the custody of the sampling
team until arrival back to the UT laboratory, where they were placed in a limited access, locked
cold room until analyses.
Results from Analysis of Field Samples
Mean values of temperature, specific gravity, viscosity, conductivity, surface tension for each of
the field samples are summarized in Table 21.
E-55
-------
APPENDIX E
Measurements of conductivity, specific gravity, surface tension, viscosity were all completed in
duplicate. The coefficients for all measurements were all excellent (conductivity 0.04, surface
tension 0.005, specific gravity 0.001% and viscosity 0.073).-
Sodium and potassium concentrations are summarized in Table 22. Replicate samples at plants 1
and 2 were taken in identical manner, and the results were averaged and reported as a single
value. At plant 3, two samples were taken at different times in the board cycle time. Samples
labeled "A" were taken just prior to the boards entering the rinse tank and should normally
correspond to the lowest concentration present in the rinse tank. Samples "B" and "R" were
taken just after the boards were removed from the rinse tank and should be near the maximum
concentration in the rinse cycle. The individual samples from plant 3 were not averaged, but
reported individually. Details of the analytical procedure used for each sample are summarized in
the Appendix.
E-56
-------
APPENDIX E
Table 21. Temperature, Specific Gravity, Viscosity, Conductivity, Surface Tension for
Field Samples.
Sample Name
Plant 1 ME Process
Plant 1MB Rinse 1
Plant 1 ME Rinse 2
Plant 1 EC Process
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Plant 1 AT Process
Plant 1 AT Rinse 1
Plant AT Rinse 2
Plant 1 FB
Plant 2 ME Process
Plant 2 ME Rinse 1
Plant 2 EC Process
Plant 2 EC Rinse 1
Plant 2 AT Process
Plant 2 AT Rinse
Plant 2 FB
Plant 3 ME Process
Plant EC Process
Plant 3 EC Rinse 1
Plant 3 EC Rinse 2
Plant 3 AT Process
Plant 3 AT Rinse
Plant 3 FB
Temp.,
°C
30
20
20
45.5
21
20
19
20
20
NA
37
15
38
20
19
16.5
NA
29
54
27
30
25
30.5
NA
Specific
Gravity
1.110
1.005
1.004
1.170
1.003
1.005
1.004
1.002
1.002
NA
1.175
1.004
1.110
1.002
1.005
1.005
NA
1.145
1.115
1.002
1.003
1.005
0.994
NA
Viscosity,
cP
1.140
1.112
1.142
1.218
.977
1.097
1.172
1.097
1.022
NA
1.246
1.172
1 .421
.932
1.202
1.037
NA
1.340
1.139
0.992
NA
1.127
0.798
NA
Conductivity ,m
S/cm,2S°C
304,000
1,935
213
224,000 .
1,043
224
341
229
223
1.8
477,000
2,170
119,600
676
353 •
256
1.9 •
168,400
261,000
736
155
543
156
1.8
Surface Tension,
dynes/cm
76.2
75.9
75.6
73.2
76.0
76.3
72.2
74.4
76.2
76.2
78.0
77.0
51.2
73.2
75.0
76.3
76.1
77.6
56.2
74.0
75.4
72.2
73.6
75.0
Table 22. Metals Concentrations Measured in Field Samnles.
Sample Name
Plant 1 ME Process
Plant 1 ME Rinse 1
Plant 1MB Rinse 2
Plant 1 EC Process
Sodium, mg/L
67,750
Potassium, mg/L
20,380
77.4
<7.5
Method
Standard Additions
Standard Curve
Standard Curve
Standard Additions
E-57
-------
APPENDIX E
, .Sample Name - . ' ,-
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Plant 1 AT Process
Plant 1 AT Rinse 1
Plant 1 AT Rinse 2
Plant 1 Makeup water
Plant 1 FB
Plant 2 ME Process
Plant 2 ME Rinse 1
Plant 2 EC Process
Plant 2 EC Rinse 1
Plant 2 AT Process
Plant 2 AT Rinse
Plant 2 Makeup water
Plant 2 FB
Plant 3 ME Process
Plant 3 ME Rinse 1-A
Plant 3 ME Rinse 1-B
Plant 3 ME Rinse 1-R
Plant 3 EC Process
Plant 3 EC Rinse 1-A
Plant 3 EC Rinse 1-B
Plant 3 EC Rinse 1-R
Sample Name
Plant 3 EC Rinse 2-A
Plant 3 EC Rinse 2-B
Plant 3 AT Process
Plant 3 AT Rinse 1-A
Plant 3 AT Rinse 1-B
Plant 3 AT Rinse 1-R
Plant 2 Makeup water
Plant 3 FB
Sodium, mg/L
242
24.5
2.8
20.15
63,450
128.6
30.8
34.5
31.36
<0.01
41,550
173.6
242
289
72,950
109.3
173.5
191.7
Sodium, mg/L
24.3
24.4
111
19.1
19.1
23.2
23.1
0.1
Potassium, mg/L
94
<7.5
<7.5
<7.5
<7.5
62,300 ,
98.8
<7.5
<7.5
<7.5
Potassium, mg/L
<7.5
Method
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Additions
Standard Additions
Standard Additions
Standard Additions
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Method
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Curve
Standard Curve
E-58
-------
APPENDIX E
The pooled instrumental relative standard deviation for potassium was.determined to be 0.77%,
based on eighteen potassium samples with a mean sample concentration of 113.6 mg/L, and a
pooled instrumental standard deviation of 0.87 mg/L. The pooled instrumental relative standard
deviation for sodium was determined to be 1.6% based on seventy-three analyses with a mean
concentration of 60.6 mg/L. The pooled instrumental standard deviation was 0.97 mg/L. Data
on which these calculations are based are included in the Appendix.
The relative standard deviation for duplicate potassium samples ranged from 0.17 to 6.95% for
tests run with no standard additions, with a pooled standard deviation of 3.46 mg/L. There were
no duplicate or replicate analyses for potassium using the method of standard additions. The
relative standard deviation for duplicate sodium measurements without standard additions ranged
from 0.11 percent to 18.94 percent, with a pooled standard deviation of 8.05 mg/L. The relative
standard deviation for duplicate sodium analyses performed with standard additions ranged from
0.52 to 6.13%, with a pooled standard deviation of 2.76'mg/L. Data for duplicate samples from
which these results were determined are listed in the Appendix.
E-59
-------
APPENDIX E
DYNAMIC MASS BALANCE MODEL
FOR INTERPRETATION OF FIELD DATA
The field data collected at the PWB manufacturers was used to validate the drag-out component
of the wastewater generation model. The output from the model is the average mass rate of
contaminant hi the rinse water from a particular process bath; the model can also calculate
average concentrations in the rinse tank effluent by dividing by the rinse flow rate.
However, the average concentration predicted by the model does not correspond directly to the
contaminant concentrations measured in the field samples. The MHC process is dynamic in that
the concentrations of contaminants hi the rinse effluent change as a function of time. The
operation cycle of a given rinse tank consists of a short period of time hi which a board is
immersed in the tank, followed by a longer period of time during which no boards are in the tanks.
Contaminants are continually flushed from the rinse tank during the entire operation time of the
bath. As a result of this operational practice, the rinse-tank concentration history will be a
periodic saw-tooth wave ftmction. In the field, instantaneous grab samples were collected from
the rinse tanks, usually immediately after removal of the board. Clearly, the concentrations in the
instantaneous grab samples may not be directly comparable to the average concentration
calculated by the model; therefore, a means of verifying the model is needed. A dynamic material
balance model was used to compare the concentration of contaminant hi the grab samples with the
average concentration of contaminant predicted by the models.
The following material balance equation describes the concentration of contaminant hi a
completely-mixed rinse tank:
Eqn21
where:
Q
v
c
C0
t
at
flow rate through the tank, L3/t
tank volume, L3
concentration of contaminant hi the tank, M/L3
concentration of contaminant in the feed water to the tank, M/L3
time, t
The concentration of contaminant in the tank as a function of time can be determined by
separating the variables hi equation 21 and integrating using 'appropriate boundary conditions.
Assume that when the line is first started (before the first board is dipped hi the tank) that the
contaminant concentration in the tank is equal to the feed water concentration. Also assume that
at t=0 a rack of boards, containing mass of contaminant M, instantly releases all of its contaminant
to solution. Under these conditions, the concentration hi the tank at t=0 is:
Eqn22
V
E-60
-------
APPENDIX E
The solution to equation 21 describing the concentration history after removal of the first board is
then given by:
dC
Eqn23
Eqn24
v v
As time progresses additional boards will enter the rinse tank. Assume that additional boards
enter the tank at a constant period of 1. It is convenient to redefine t as:
t = n A, + 9
Eqn 25
where
n
number of cycles completed since t = 0
time elapsed in the current cycle, t
The effluent history during the rinsing cycle for the second board processed after start-up would
be given by:
V ~
dC
-C
Eqn 26
Eqn 27
This result can be extended to represent the effluent history for the rinsing period after the nfc
board is rinsed:
Eqn 28
Steady-state is defined to occur when n = «=. Substituting
E-61
-------
APPENDIX E
Eqn29
yields an expression concentration history for a single rinse tank, operating at steady-state:
Af /Q0Y 1
V '1-exJ-y-
EqnSO
Example:
A rinsing tank receives a rack containing 60 ft2 of boards every 30 minutes. The drag-out rate is
10 ml/ft2 and the contaminant concentration in the process tank is 3000 mg/L. The rinse rate is 2
gpm and the tank is 220 gallons hi volume. The feed water contains 40 mg/L of the contaminant.
Calculate the effluent concentration history during the 30 minute cycle period under steady-state
conditions:
ft2 L
1
=
22QgaI\) \ 3.7ZL)
2l.6mg/L Eqn3l
: = 40+2.16extJ-
"~\ 220;. f-2-30
l~^~m
Eqn32
Equation 32 is plotted over the course of one process cycle in Figure 6.
E-62
-------
APPENDIX E
time (min)
Figure 6. Example Concentration History of Rinse Tank Effluent During One
Plating Cycle.
E-63
-------
APPENDIX E
MODEL VALIDATION
The purpose of the field samples was to validate the drag-out prediction model 'and the overall
mass balance approach to predicting wastewater quality from PWB facilities. The dynamic
material balance model for the rinsing process was developed in the previous section to facilitate
this comparison. First, equation 30 was solved for the mass of contaminant in the drag-out:
= (C-C0)F
-exp
V
Eqn33
The volume of the drag-out could then be calculated by dividing the mass of contaminant in. the
drag-out by the bath concentration:
drag-out
M
"bath
Eqn34
The drag-out volumes calculated-from the field data and the dynamic mass balance (equations 33
and 34) are compared to those predicted using the drag-out regression model (equation 15) in
Table 23. Replicate samples at the plants 1 and 2 were taken in identical manner, and the results
were averaged and reported as a single value. At plant 3, the duplicate samples were taken at
different times in the board cycle time. Samples labeled "A" were taken just prior to the boards
entering the rinse tank and should normally correspond to the lowest concentration present in the
rinse tank. Samples "B" and "R" were taken just after the boards were removed from the rinse
tank and should be near the maximum concentration in the rinse cycle. Samples 3MER1-A and -B
were taken soon after the MHC line had been shut down for a short period of time and may have
been erroneously low. The individual samples from plant 3 were not averaged; separate
calculations were made for each one. Sodium and potassium concentrations in the anti-tarnish
rinse tanks were too low to accurately calculate either the mass of contaminant in the drag-our or
the drag-out volume.
The drag-out volumes calculated from the field data are consistently less than those predicted by
the drag-out model. They are also significantly less than those measured both in the laboratory
experiments performed as a part of this work and the data collected by Pagel (1992). For
example, the drag-out volumes from Microetch baths calculated from our field data ranged from
22.8 to 53.6 mL/m2, compared to a range of 76 to 122 mL/m2 in this study and a range of 57 to
145 mL/m2 in Pagel's work. Similarly, the drag-out volumes from the Electroless baths calculated
from our field data ranged from 9.73 to 32.9 mL/m2, compared to a range of 20.4 to 81.8 mL/m2
in Pagel's work. A possible explanation is that the drag-out volumes calculated from the field data
were based on the assumption in the dynamic mass balance model that all the contaminant was
released instantaneously from the PWB and that the rinse tank was perfectly mixed. The rinsing
tanks used in PWB plants may not approximate this ideal behavior. Rinse water typically enters
the bottom of the rinse tank and flows over a weir at the water surface. As the board enters the
tank, it is likely that a significant fraction of the pollutant flows over the weir prior to being mixed
E-64
-------
APPENDIX E
throughout the tank. Fluid shear may contribute to the loss of contaminant near the water surface
of the tank as the board enters the tank. The grab samples were generally collected immediately
following removal of the board from the rinse tank. We hypothesize that the short-circuiting
process described above may have caused a large fraction of the contaminant to be removed from
the rinse tank prior to the time that we collected the sample. Our laboratory drag-out study, and
the work of Pagel (in which the rinse water flow rate was set to zero during the sampling) were
not subject to this influence.
Table 23. Comparison of Drag-Out Volumes Calculated from Field Samples to Those
Predicted by Regression Model.
Sample Description
Plant 1, Microetch
Plant 1, Electroless Copper
Plan 2, Microetch
Plant 2, Electroless Copper
Plant 3, Microetch A
Plant 3, Microetch B
Plant 3, Microetch R
Plant 3, Electroless A
Plant 3, Electroless B
Plant 3, Electroless R
Drag-Out Volume "Calculated
from Field Data, mL/m2
53.6
32.9
22.8
23.2
28.2
41.0
37.9
9.73
6.83
10.9
• "Drag-Out Volume Calculated-
from Regression Model, mL/m2
127
59.1
102 .
39.9
98.2
982
98.2
34.7
34.7
34.7
A regression equation was fitted to the 'data in Table 23, resulting in the following relationshipOr2
= 0.71):
Vfield = 036Vpredicted + 0.68
Eqn35
where:
V,
'field
predicted
drag-out volume calculated from the field data
is the drag-out volume predicted by the regression model
The slope of the regression equation suggests that about 2/3 of the total mass of contaminant
flows over the weir prior to being mixed. The relatively good correlation coefficient indicates that
the field and predicted drag-out volumes were comparative on a relative basis. This suggests that
the drag-out regression model and overall mass balance approach may be valid for making relative
comparisons between process alternatives.
E-65
-------
APPENDIX E
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
Conclusions
• Contaminant mass in PWB process wastewaters can be expressed as a mass balance in
which the mass of contaminant in the wastewater is equal to the mass of contaminant
released via drag-out from the process baths (which ultimately ends up in the rinse tanks),
periodic dumping of process tanks into the wastewater, and stripping deposits from racks;
Drag-out is generally considered to be the major contaminant source. Data quantifying •
composition of the process baths, the volume of wastewater produced, and the frequency
of bath dumps are usually collected during the course of the DFE process. For example,
this information was collected for the MHC process during a previous study by the
University of Tennessee CCPCT (Kincaid et al. 1997).
• Very little data exists quantifying the rate of drag-out from PWB processes, i.e., the mass
or volume of drag-out per unit surface area of PWB, e.g., mL/m2. A study reported by
Pagel at Micom, Inc. is the only readily available study on PWB facilities. Limited drag-
out research has been conducted on flat pieces, most notably by Sup. However, the.
numerous small holes present in PWBs renders application of drag-out volumes measured
from non-PWB pieces problematic. Practitioners tend to use rules-of-thumb or
historically accepted values for drag-out. This one-size-fits-all approach ignores process
specific information such as bath type, viscosity, surface tension, board withdrawal rate, or
drain time. Drag-out rates reported in the literature for vertically-oriented flat pieces.
range from 10 to 160 mL/m2,
• Commonly-cited equation found in the literature offer predictions of the drag-out rate as a
function of kinematic viscosity and board withdrawal rate. Sup showed that this equation
does not predict drag-out very well for the rectangular flat pieces that he studied. There
was no relationship between kinematic viscosity and drag-out, and two previously
proposed predictive equations performed poorly.
• Several variables have been shown to affect the drag-out rate. Studies at Micom, Inc.
reported by Pagel (1992) showed the importance of longer drainage time and slower
withdrawal rate in reducing drag-out. Siip (1990,1992) also found that these variables
are important as well as the angle of the board during drainage. No research was found
that addressed the effect of surface tension. Based on the present study, surface tension
may be an important variable.
• Considerable literature exists on rinsing theory which appears highly developed and well
studied for ideal mixing situations. While rinsing theory is not as well developed for non-
ideal mixing, previous researchers have concluded the assumption of ideal mixing-is valid
for estimating long-term-average wastewater concentrations because nearly all of the
drag-out ultimately reaches the wastewater effluent.
• Laboratory studies conducted as part of this research expanded the data base of drag-out
rates for two PWB process baths and several operating conditions. The experimental
procedures showed good reproducibility, and the data were consistent with previous
research.
E-66
-------
APPENDIX E
• A regression model for predicting drag-out volume was developed using the available data
bases of Sup (1990,1992), Pagel (1992), and the present study. .The dependent variables
were a choice of two types of process baths (plus a default for any other type of bath),
board withdrawal rate, drain time, board size, and presence of drilled holes. The model
had an R2 of 0.852.
• The regression model for predicting drag-out rate was incorporated in a computer model
for predicting contaminant mass loading and mean pollutant concentrations from PWB
manufacturing process lines. The model was written as a Visual Basic macro within an
EXCEL spreadsheet. Input variables included facility production rate, board size, number
and types of process baths, bath composition, frequency of bath dumps, and wastewater
production rate.
• Samples were collected from three PWB facilities in order to validate the drag-out model.
Samples were collected from various process and rinse tanks and analyzed for
temperature, specific gravity, viscosity, surface tension, conductivity,-and potassium or
sodium concentration. Since it was not convenient to collect composite samples from the
rinse tanks, grab samples were collected at various times after a board was inserted into a
rinse tank. An equation was developed to relate the time-dependent concentration of
potassium or sodium in the rinse tank to the drag-out volume. Unfortunately, it appears
that poor mixing in the rinse tanks led to unrepresentative sampling. Although the
apparent drag-out rates were about 1/3 of the predicted rates, a comparison of drag-out
rates between process tanks showed a correlation (r2 = 0.71) with the previously
developed regression model, and in that sense lend support to it:
Recommendations ' . .
• The authors believe that this work has resulted in a more universally applicable method for
estimating the mass and concentration of contaminants hi a PWB process wastewater than
currently exists, especially for relative evaluations. However, much can still be done to
improve the model since the existing data are so limited. Previous work has not studied
the effect of surface tension, but the laboratory studies in this work showed that surface
tension may be an important variable. Indeed, one of the drag-out reduction best practices
is to use a wetting agent which would reduce surface tension. The effect of viscosity was
previously thought to be important, but neither Sup nor this work found it to be
significant. There has also been only one previously reported study of an actual PWB
facility. The authors believe that a better quantitative understanding of the variables
affecting drag-out could lead to the development of a better prediction equation. The first
phase of such research should focus determining the effect of bath fluid properties and
operating variables under controlled laboratory conditions. Expansion and testing'of the
model could be accomplished by samples collected at PWB facilities during a second
phase of the study.
• Beyond determining the wastewater quality emanating from PWB manufacturing
processes, there is a need to assess the fate of the chemicals in the PWB wastewater both
in the onsite treatment processes at PWB manufacturing facilities and at Publicly-Owned
Treatment Works (POTW). Information of the effect of chemical speciation on the fate of
E-67
-------
APPENDIX E
pollutants is especially needed. For example, metals are one of the primary pollutants of
concern in PWB wastewater, and it is likely that many of the metals are chemically
complexed in PWB wastewater. On-site treatment processes are likely to preferentially
remove the least stable metal complexes, while the most stable complexes are discharged
to the POTW. Standard removal efficiencies for metals in activated sludge processes are
probably not applicable to these highly complexed metals and the potential for release of
the metals to the aquatic environment may be underestimated.
A third issue needful of better understanding is the volatilization of chemicals from tanks
and baths such as in PWB plating processes and other manufacturing processes. The
volatilization models used hi the previous assessment of emissions for the MHC process
and the present assessment of surface finishing assume gas-side control of the mass
transfer, i.e., volatilization, of chemicals from the process baths. In the MHC, and
presumably in the surface finishing process, there were several chemicals whose mass
transfer appeared to be liquid-side controlled. The mass transfer model used does not
apply for this situation and could lead to an overestimate of the emission and consequent
risk for these chemicals. It would be productive to research the literature to find more
appropriate liquid-side control mass transfer models and applicable constants for various
types of manufacturing process tanks. For example, there is a body of literature available
on surface renewal theory models which would be more appropriate for liquid-side mass
transfer control. This literature search could be completed within a year and a decision
made at that time as to whether any lab based research is warranted.
E-68
-------
APPENDIX E
REFERENCES .
American Chemical Society, Chemical Abstracts, American Chemical Society, Washington, D.C.,
1998.
Chang, L., and McCoy, B.J., "Waste Minimization for Printed Circuit Board Manufacture,"
Hazardous Waste & Hazardous Materials, 7, No. 3,293-318(1990).
Hanson, N. H., and Zabban, W., "Design and Operation Problems of a Continuous Automatic
Plating Waste Treatment Plant," Plating, 909-918 (August, 1959).
Hatschek, Emil, "The Viscosity of Liquids" D. Van Nostrand Company, New York, 1928.
Kincaid, L.E., Geibig, J.R., Swanson, M.B., and PWB Engineering Support Team, Printed
Wiring Board Cleaner Technologies Substitutes Assessment: Making Holes Conductive, Center
for Clean Products and Clean Technologies, University of Tennessee, Knoxville, Tennessee
(1997). "
Kushner, J.B., "Rinsing: I. With Single-Compartment Task," Plating, 36, August, p. 798-801,
866(1949).
Kushner, J.B., ccWhere Do We Go from Here? Part m - Water Control," Metal Finishing, 47,
No. 12, 52-58,67 (1951). '
Kushner, J.B., "Dragout Control - Part I," Metal Finishing, 49, November, 59-64 (1951).
Kushner, J.B., "Dragout Control - Part TL" Metal Finishing, 49, December, 58-61,67 (1951).
Kushner, J.B., "Rinsing Techniques," in Metal Finishing: 47th Guidebook-Directory Issue 1979,
Vol. 77, No. 13, (January, 1979).
McKesson, Doug, and Wgener, M.J., c
-------
APPENDIX E „___
Pinkerton, H.L., and Graham, A.K., "Rinsing,'''in Electroplating Engineering Handbook,
Lawrence J. Durney, Ed., van Nostrand Rheinhold, New York, 1984.
Robinson, R.B. and Cox, C.D., OA/QCPlan for Verification of Hot Solder Finishing: Prediction
of Water Quality from Printed Wiring Board Processes, University of Tennessee, Knoxville, TN,
1998.
Sharp, J., Teradyne Corp., Nashua, NH., Personal communication (September, 1998).
Sup, Von M., 'Technologische Mapnahmen zur Minimierung von Ausschleppverlusten
(Technological Measures for Minimizing Drag Out)," Galvanotechnik, 81, No. 11, 3873-3877,
(1990).
Sup, Von M., "Bestimmung Elektrolytspezifischer Ausschleppverluste (Detemination of
Electrolyte Specific Losses Due to Drag-out)," Galvanotechnik, 83, No. 2, 462-465, (1992).
Talmadge, John A., "Improved Rinse Design in Electoplating and Other Industries," Proceedings
of the Second Mid-Atlantic Industrial Waste Conference, p.217-234, (1968).
Talmadge, John A., Buffham, Bryan A, and Barbolini, Robert R., "A Diffusion Model for
Rinsing," AIChE Journal, 8, No. 5, 649-653, (1962).
Talmadge, John A., and Buffham, Bryan A, "Rinsing Effectiveness in Metal Finishing," J. Water
Pollution Control Federation, 33, No. 8, 817-828, (1961).
Talmadge, J. A, and Sik, U.L., "A Drop Dispersal Model for Rinsing," AIChE Journal, 15, No. 4,
521-526, (1969).
U.S. Environmental Protection Agency, Printed Wiring Board Pollution Prevention and Control
Technology: Analysis of Updated Survey Results, EPA/744-R-98-003, Design for the
Environment Branch, Economics, Exposure, and Technology Division, Office of Pollution
Prevention and Toxics., Washington, D.C. (1998).
Yost, K.J., "The Computer Analysis of Waste Treatment for a Model of Cadmium Electroplating
Facility," Proceedings Third International Cadmium Conference, p. 56-59 (1991).
Zaytsev, Ivan Demitrievich, and Aseyev, Georgiy Georgievich, Editors, Properties of Solutions of
Electrolytes, M. A. Lazarev and V. R. Sorochenko, Translators, CRC Press, Boca Raton, 1992.
E-70
-------
APPENDIX E
Cr
Ct
D
f
f
•F*
g
h
K
m
n
Q
r
t
T
tdr
V,
AV
v
P
LIST OF SYMBOLS
area of the sheet
mass content of the component is kg of component per kg of solution
concentration of contaminant solution being drug into rinse tank
concentration of contaminant in the effluent of the r* rinse tank
concentration of contaminant in rinse tank after t min
volume of drag-over or drag-out on rack and work rinsing operation
coefficient calculated as shown below for each component for use in the
method given by Zaytsev and Aseyev
empirical coefficients chosen for each electrolyte component from a table
for use in the method given by Zaytsev and Aseyev
film thickness, cm
thickness of the film that drains off the sheet
function describing a relationship between the independent variables and
thickness of the film that drains from the sheet
gravity (981 cm/s?)
height of metal sheet
unknown constant determined by experiments
unknown exponent determined by experiments
number of rinsing operations in t min
rate of fresh water flow
number of rinse tanks in series
time interval between rinsing operations
temperature of solution, °C
drainage time, s
withdrawal.time, s
velocity of withdrawal
withdrawal rate of metal sheet, cm/s
volume of rinse tank
volume of liquid that drains from the rectangular sheet
kinematic viscosity, cm2/s
density of electrolyte, gm/cm3
dynamic viscosity of electrolyte, g/(cm-s)
viscosity of water, Pa-s
surface tension of the liquid
E-71
-------
-------
Appendix F
Supplemental Performance Demonstration
Information
-------
-------
APPENDIX F
F.I Modeling the Test Results
General linear models (GLMs) were used to analyze the test data for each of the 23 electrical
circuits in Table 4.1 at each test time. The GLM analysis determines which experimental factors or,
when possible, combinations of factors (interactions) explain a statistically significant portion of the
observed variation in the test results.
A GLM used to analyze the test results with respect to sites, flux type, and their interactions
(where possible) is expressed as the following 22-term equation:
Y = Po + p-iDi + p2D2 + p3P3 + p4D4 + PsD5 + p6D6
+ P12D12 + Pi3Dis + pi4D14 + PisD15 + Pi6Di6
p7D7 + p8D8
Pi7D3D16 +
P21D12D16
P22D15D16
p19D6D16 + p2oD10D16
p9D9 + p10D10
(Main effects)
(Two-factor interactions)
(F.1)
The coefficients in the GLM (J30, Pi, Pa, - . .) are estimated using ordinary least squares regression
techniques. The dummy variables, DI to Die, are set equal to 1 to identify type of surface
finish/manufacturing site and type of flux that are associated with individual test results. Otherwise,
the dummy variables are set to 0. The following dummy variables can be used to represent the
experimental variables for each test environment for each electrical response variable. '•
DI = 0 if surface finish is not HASL- Site 2
= 1 if surface finish is HASL - Site 2
D2 =0 if surface finish is not HASL -Site 3
= 1 if surface finish is HASL - Site 3
D3 = 0 if surface finish is not OSP- Site 4
= 1 if surface finish is OSP - Site 4
D4 =0 if surface finish is not OSP -Site 5
= 1 if surface finish is OSP - Site 5
D5 = 0 if surface finish is not OSP- Site,6
= 1 if surface finish is OSP - Site 6
D6 = 0 if surface finish is not immersion Sn - Site 7
= 1 if surf ace finish is immersion Sn - Site 7
D7 = 0 if surface finish is not immersion Sn- Site 8
= 1 if surface finish is immersion Sn - Site 8
D8 = 0 if surface finish is not immersion Sn- Site 9
= 1 if surface finish -is immersion Sn— Site 9
D9 = 0 if surface finish is not immersion Sn- Site 10
= 1 if surface finish is immersion Sn - Site 10
DIG = 0 if surface finish is not immersion Ag - Site 11
= 1 if surface finish is immersion Ag - Site 11
Du = 0 if surface finish is not immersion Ag - Site 12
= 1 if surface finish is immersion Ag - Site 12
D12 = 0 if surface finish is not Ni / Au - Site 13
= 1 if surface finish is Ni/Au- Site 13
DIS = 0 if surface finish is not Ni/Au -Site 14
= 1 if sutface'finish is Ni/Au - Site 14
Di4 = 0 if surface finish is not Ni / Au - Site 15
= 1 if surface finish is Ni / Au - Site 15
D15 = 0 if surface finish is not Ni / Pd / Au - Site 16
= 1 if surface finish is Ni / Pd / Au - Site 16
Die = 0 if flux is not water soluble
= 1 if flux is water soluble
F-l
-------
APPENDIX F
The "base case" is obtained by setting all DI = 0. Note that the surface finish/manufacturing site is
0. Likewise, if Die = 0, the flux is low-residue. Thus, the base case is HASL / Site 1 with LR flux.
Note the GLM in Equation F. 1 contains six interactions terms that represent the last six sites in
Table 4.2 (5, 6,7, 11,13, and 16) for which both LR and WS fluxes were used.
The GLM approach provides a tool for identifying the statistically significant experimental
variables and their interactions. That is, all terms in the model that are significantly different from the
base case are identified through tests of statistical hypotheses of the form:
HO: Pi = 0 versus HI : p i ^ 0 for all i
If the null hypothesis is rejected, then the coefficient of the corresponding term in the GLM is
significantly different from 0, which means that the particular experimental conditions represented by
that term (surface finish or flux type) differ significantly from the base case. If the null hypothesis is
not rejected, then the coefficient of the corresponding term hi the GLM is not significantly different
from 0 and, therefore, the experimental conditions represented by that term do not differ significantly
from the base case. Such terms are sequentially eliminated from the GLM (see Iman, 1994, for
complete details).
The GLM approach is quite flexible and easily adaptable to a variety of requirements. For
example, if the focus is on surface finishes and not sites; the GLM hi Equation F.I would be replaced
by one of the following form:
F.2
This model contains only main effects where the dummy variables are defined as follows.
Dt
D2
=0 if surface finish is not OSP
= 1 if surface finish is OSP -
- 0 if surface finish is not immersion Sn ,
- 1 if surface finish is immersion Sn
Da - 0 if surface finish is not immersion Ag
= 1 if surface finish is immersion Ag
D4 =0 if surface finish is not Ni/Au
= 1 if surface finish is Ni / Au
DS = 0 if surface finish is not Ni/Pd/Au
— 1 if surface finish is Ni / Pd / Au
Dfi =0 if flux is not water soluble
- 1 if flux is water soluble
As before, the "base case" is obtained by setting all Dj = 0, which is HASL with LR flux. Note
that the base case associated with the GLM in Equation F. 1 was also HASL with LR flux, but also
required Site 1 . That requirement is not part of the latter model since sites are not included in the
model in Equation F.2.
As a final illustration of the flexibility of the GLM approach consider a subset of the data base that
only includes the results for Sites 1, 4, 5, 7, 1 1, 13, and 16 in Table 4.2. These sites were selected
because their surface finish was processed with both LR and WS fluxes, which allows an interaction
term to be added to the model in Equation F.2 for each surface finish and flux combination. However,
by excluding the other sites, the number of data points is reduced from 1 64 to 92.
. . F-2
-------
APPENDIX F
Example of GLM Analysis
The data base for the electrical responses incorporates the dummy variables used to define the
experimental parameters for each measurement. The data base contains 164 rows (one for each PWA).
Sample data base entries for the GLM in Equation F.2 for leakage measurement on the 10-mil pads
(response number 18 in Table 4.1) in logio ohms could appear as follows:
Row
OSP
ImmSn Imm Ag
Ni/Au
Ni/Pd/Au
Flux
Leakage
1
2
3
4
0
1
0.
0
0
0
1
0
0
0
0
.0
0
0
0
0
0
0
0
1
0
1
0
1
12.8
11.9
12.1
11.8
The interpretation of these data base entries is as follows. The first row has zeros for OSP,
immersion Sn, immersion Ag, Ni/Au, and Ni/Pd/Au. This implies that the surface finish is HASL.
The surface finishes for rows 2, 3, and 4 are OSP, immersion Sn, and Ni/Pd/Au, respectively. Water
soluble flux is used on rows 2 and 4. The leakage measurements are given in the last column. The
above table would be expanded to include other experimental parameters or products (interactions) of
the experimental parameters depending on the requirements of the GLM such as given in Equation F. 1.
The above table would also include columns containing the other 22 electrical measurements.
Computer software is used with the entries in the data base to find the least squares estimates of
coefficients in the GLM. For example, such an analysis for the GLM hi Equation F.2 could produce
an estimated equation such as the following for leakage for the 10-mil pads.
Y = 12.5 - 0.200 OSP + 0.192 Immersion Sn - 0.164 Immersion Ag + 0.006 Ni/Au - 0.292 Ni/Pd/Au - 1.04 Flux
Note that the least squares process has simply solved a set of equations to determine an estimated
coefficient for each term appearing in the GLM in Equation F.2. However, it does not necessarily
follow that each of the terms in this estimated model makes a statistically significant contribution
toward explaining the variation in the leakage measurements. Rather, this determination is
accomplished by subjecting the coefficients in the fall model to the following hypothesis test in a
sequential (stepwise) manner to determine if they are significantly different from 0:
HQ: Pi = 0 versus Hi: pi*0
If the coefficient is not significantly different from 0, it is eliminated from the model. Thus, the
only terms remaining in the model at the conclusion of this sequence of tests are those that are declared
to be significantly different from 0. This stepwise process eliminates some of the terms from the
model and the least squares calculations are repeated without those terms, which produces a reduced
model such as:
Y = 12.35 - 0.34 OSP - 0.38 Immersion Ag - 0.24 Ni/Pd/Au - 1.06 Flux
The intercept in this model, 12.35, is the estimated resistance for the base case—FIASL processed with
LR flux. Mean predictions for other combinations of the experimental parameters can be made by
substituting the appropriate dummy variables into the model. For example, the mean prediction for a
OSP (D,=l, D2=0, D3=0, D4=0, D5=0) PWA processed with WS flux (D6=l) is found as:
F-3
-------
APPENDIX F
Y = 12.35 - 0.34 (1) - 1.06 (1) = 10.95
F.2 Overview of Test Results
Table F.I Anomaly Summary by Surface Finish after Exposure to 85/85
MSN Site Flux
Circuit
Test Technician Comments
083-2
WS 7 HFPTHSOMHz OpenPTH
8 HFPTHf(-3dB) OpenPTH
9 HFPTHf(-40dB) OpenPTH
ipsf
056-4
lllftMftmbtt.Ss
030-4
032-4
086-2
102-4
^IfirkC' *W A*W*ft IV ^. *
082-2
094-4
" ;S-O.^
5
b,^.^
9
8
7
' 10
i"-^ ^
11
12
, j,ys--s"
LR
>1T!§II^
WS
LR
WS
WS
5'"-x *•"
LR
WS
x " ""-^4 v -" x^^^^^^j*.
Circuit
12 HFSMTf(-40dB)
7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
10 HF SMT 50MHz
11 HFSMTf(-3dB)
12 HF SMT f(-40dB)
10 HF SMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
10 HF SMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
23 Stranded Wire 2
5fEl ^ s^W Wr* ", " "V
12 HFSMTf(-40dB)
10 HF SMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
^ •• ^ •• % ss ; ^* s •• •-, f •••• ^^i-
Test Technician Comments
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH .
OpenPTH
Open PTH
OpenPTH
Waveform shifted
Distorted Waveform (does not quite go to -40dB, reads at-
3dB)
Minor
;,, *,. z *,,,s. - , - - " .Y»-.^..
Distorted waveform (goes to 40db but flattens and crosses
beyond 900mhz
Open PTH on coil
Open PTH on coil
Open PTH on coil
F-4
-------
APPENDIX F
014-3
056-2
056-4
058-1
060-1
060-2
ilH&ti&i&itoii'!
028-2
030-4
032-4
033-2
037-2
084-1
086-2
087-3
088-3
089-1
089-2
089-4
090-2
102-4
•Uar,
071-1
072-1
5'
5
5
5
5
5
9
9
8
8
9
7
7
7
7
7.
7
7
7
10
&8.n
11
11
LR
LR
LR
WS
WS
WS
- """..,,^-..
LR
LR
LR
LR
LR
LR
WS
WS
WS
LR
WS
WS
WS
WS
WS
-^ £...*..-•
LR
LR
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
7 HFPTHSOMHz
8 HFPTHf(-3dB)
7 HFPTHSOMHz
8 HFPTHf(-3!ffi)
9 HFPTHf(-40dB)
10 HFSMTSOMHz
11' HFSMTf(-3dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
12 HFSMTf(-40dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
12 HFSMTf(-40dB)
>.. "..»..."... ,'-: =>"~- ^-- "
10 HFSMTSOMHz
12 HFSMTf(-40dB)
4 HVLCSMT
7 HFPTHSOMHz
8 HFPTHf(-3dB)
17 HFTLCRNR
5 HSD PTH
10 HF SMT SOMHz .
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
5 HSD PTH
12 HFSMTf(-40dB)
7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HF SMT f(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
17 HFTLCRNR
"•; %-\-,™<,x »v.v J
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HF SMT f(-40dB)
7 HFPTHSOMHz
OpenPTH .' ,
OpenPTH
OpenPTH
Open PTH
Open PTH
2 open PTHs
2 open PTHs
2 open PTHs
2 open PTHs
2openPTHs
2 open PTHs
OpenPTH
OpenPTH
OpenPTH
Open PTH
•~\x xk^rrtz^^,- " ^Vx^vr^; " %%% ^ ""-L^, ^
OpenPTH
OpenPTH
Burnt etch (visual)
OpenPTH
Open PTH
Likely component failure
OpenPTH
OpenPTH
OpenPTH
OpenPTH
OpenPTH ,
OpenPTH
Likely component failure
Distorted Waveform
High resistance on coil (acts like open PTH)
High resistance on coil (acts like open PTH)
High resistance on coil (acts like open PTH) .
High resistance on coil (acts like open PTH)
OpenPTH
OpenPTH
Open PTH
OpenPTH
Open PTH
OpenPTH
OpenPTH .
High resistance on coil (acts like open PTH)
High resistance on coil (acts like open PTH)
High resistance on coil (acts like open PTH)
Open PTH
OpenPTH
OpenPTH
Open PTH on coil
Open PTH on coil
Open PTH on coil
r.* .'\ - " -- """ -- --'" v v- -« % -% ; , ^ -,-
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH
F-5 , ,
-------
APPENDIX F
8 HFPTHf(-3dB)
9 HFPTHfMOdB)
073-3 11 LR 7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
15 HRTLClGHz
082-2 11 WS 12 HFSMTfMOdB)
085-1 12 WS 7 HFPTH 50MHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTfMOdB)
085-2 12 WS 7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
091-4 12 WS 12 HFSMTf(-40dB)
094-1 12 WS 7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HF SMT f(-40dB)
094-4 12 WS 7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
^S®BL__ , "* ^^^MfJl^'^^S.^^'-: ..;-,
013-1 13 LR 6 HSDSMT
OpenPTH
Open PTH
OpenPTH
OpenPTH
OpenPTH
Burnt etch
OpenPTH
OpenPTH , •
Open PTH
OpenPTH
OpenPTH
OpenPTH
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Burnt Etch, High Resistance PTH, and Open PTH
Burnt Etch, High Resistance PTH, and Open PTH
Burnt Etch, High Resistance PTH, and Open PTH
Burnt Etch, High Resistance PTH, and Open PTH
Burnt Etch, High Resistance PTH, and Open PTH
Burnt Etch, High Resistance PTH, and Open PTH
Open PTH
Open .PTH
OpenPTH
v ' x>" * % £' •.•• •• s s " •• >. „ % •. ' \ % ; ••••. v^ 5 ,
Device failed, U3
015-2 14 LR
7 HFPTH 50MHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
Open PTH on coil
Open PTH on coil
Open PTH on coil
055-1 13 WS
7 HFPTHSOMHz
8 HFPTHf(-3dB)
9 HFPTHf(-40dB)
OpenPTH
OpenPTH
OpenPTH
036-1 16 WS
6 HSDSMT
Likely component failure
Table F.3 Anomaly Summary After Mechanical Shock
(shaded entries signify carry over TS anomalies)
HSS& "
MSN
039-2
046-1
046-2
046-4
076-1
076-2 .
079-4
Site
2
2
2
2
1
1
1
xl^Dtr-
Flux
LR
LR
LR
LR
LR
LR
WS
;-f v^--|c? f ;S^lSf f ^
Circuit
12 HFSMTf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB),
12 HF SMT f(-40dB)
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
1 HCLVPTH
12 'H£WMf40dB>
X^^v -; «-o% _x,**^xv" -^"-f^ , '"--""' Y
Test Technician Comments
Waveform distorted
OpenPTH •
OpenPTH
Distorted waveform
High resistance
Waveform does not go to -40dB
F-6
-------
APPENDIX F
080-4
083-2
096-4
098-2
098-3
098-4
099-1
099-4
100-3
^Br^,,
006-4
007-3
009-2
010-1
010-2
010-4
014-1
014-3
056-1
056-2
056-3
056-4
057-1
058-1
060-1
060-2
060-4
061-4
lv
1
3
3
3
3
3
3
3
••MX-- ,,
6
6
6
4
4
4
5
5
5
5
5
5
5
5
5
'5
5
4
WS
WS
WS
WS
WS
WS
WS
WS
WS
t S'^TS
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
WS
WS
WS
WS
WS
WS
12 HF SMT f(-40dB)
7TJrfc?"'3E>(T*u *«3h% &13W %
iCif£ Jri O:-,v3s*iV3EJClZ- -. c
7 HFPTHf(-3dB)
10 HFSMTSOMHz
12 HFSMTf(-40dB)
10 JGFSMT 50MHz
12 WsMrfH-OdlB) **
12 HF,S*NfflM6$B> •*••"*
12 HFSMTf(-40dB)
12 HFSMTf(-40dB)
"S3Z^^^±^U^™^™*
12 ln?^W£4MB) '
12 HF SMT f(-40dB)
1 f\ f?ft£ ££& jpy -4fVSjStip-i( ••
i\J alG £f&f$£, J\fj.y3^J.-£i. svw viVk
1 HCLVPTH
12 HFSMTf(-40dB)
12 HF SMT f(-40dB)
14 HFTLC 500MHz
10 HFSMTSOMHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
1 HCLVPTH
12 HFSMTf(-40dB)
1 HCLVPTHw_
9 'riFsMT soMnz5"1" v
10 HFSMTf(-3dB)
12 HFSMTf(-40dB) •
12 HFSMTf(-40dB)
7^3*t? "itvyiry -4ifvX ^tV-f-^- •••• *•
•£** +T *'-JvJ.-.'*j< v^MtRwl •. "•
9 liiFFXH 3t(-^Q^) "
10 HFp4Tf^MKfe.-o-
11 HF^IT^t^feV ^
12 HF SMT f(-40dB)
11 WsMTf(-3dB)
1.2, .tCli^ uuN^tj. Js*^rvWA5r \
7T3T? -Ot^TE* -^iTtSv jSCF-w
JtXt^ 39 Jfwt J "• juHt7:£ Vt±3t>£, v
12 HFSMTf(-40dB)
12 HF SMT f(-40dB)
OpenPTH
Open PTH, distorted waveform
OpenPTH
OpenPTH
Waveform shifted
Distorted waveform .
Distorted waveform
Distorted waveform
\v»c * s * ^« ~ $. \ \ ," % ,«* ^ ^;s
Distorted waveform
Open PTH
Distorted waveform
Open etch
OpenPTH
Waveform does not go to -40 at the correct frequency
OpenPTH
Waveform shifted
Open PTH - 2 places
Waveform does not go to -40dB
OpenPTH
Distorted waveform
Open PTH
Distorted waveform
F-7
-------
APPENDIX F
062-1
062-4
065-1
J06S4
026-4
028-2
029-1
029-2
030-4
032-4
033-2
037-2
040-3
084-1
084-2
084-4
086-2
087-1
087-3
087-4
088-3
089-1
089-2
089-4
4
4
4
4
VkSAWWV • - IV *.V
gill -gal ^
"" T v
9
9
9
9
8
8
9
8
7-
7
7
7
7
7
7
7
7
7
7
WS
WS
WS
..JVS
""££"""
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
WS
WS
WS
WS
LR.
WS
WS
WS
12 HFSMTfMOdB)
12 HFSMTfMOdB)
12 HFSMTfMOdB)
^ 12 HF SMT f(-40dB)
™ HSDFTH
11 ~Hf s1^1(3dB) ~ "'
1 HCLVPTH
17 HFTLCRNR.
9 HFPTHfMOdB)
9 'HFPTH'fMOdBf^
' 17 "^BltC^I^^. „
11 J^IM? f{-i^j- "-
12 *H^^C? f[^8tJB2 --
9 HFPTHfMOdB)
11 sWsslr H&ify ^ %
12 HFSMTfMOdB)
9 HFPTHfMOdB)
•10 HFSMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTfMOdB)
10 HFSMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTfMOdB)
15 HFTLClGHz
1 HCLVPTH
12 "HFliit f-40l§V ^
J.A, ^£. 44, ^s^W.Vjt t _^'"fcT\*W^>^
12 HFSMTfMOdB)
S^?&£^VW&«WM<&S«HK!'& .&>f.vSv.
10 HFSMT 50MHz
1 -I TJT7 ClVyTP ff "3/TR^
12 HFSMTfMOdB)
"
1 (\ < 1-3T? •CJhrf1^ ^*/tKjft3w
i\J ^ JtJj^ ^>LVx^k jJfcftiyU-I-jG,
1 1 '«^P?T?%'C3Sj3*T* f?" ^^JiV ** **
7 HFPTHSOMHz
o H r* Jt^ JL Jtl I(** j d_*j )
n^sl^t^f-
Distorted waveform
Waveform shifted
High resistance
Bad HSDPTH device
Open etch
Burnt etch (visual)
OpenPTH
Open etch
Distorted waveform
OpenPTH
OpenPTH
OpenPTH
Distorted waveform
Open PTH 2 places SMT & PTH
Distorted waveform
OpenPTH
OpenPTH
Waveform does not go to -40dB
Open PTH
Open PTH -2 places
090-2
WS 7 tHJPi>TH50M££>:
8 HFfT$f(-3dB>
10 HFSMT50MHz
11 HFSMTf(-3dB)
Open PTH 2 places SMT & PTH
F-8
-------
APPENDIX F
102-4
104-4
113-1
072-1
072-2
.072-4
073-3
075-2
075-3
082-2
082-3
085-1
085-2
091-4
094-1
10
10
10
11
11
11
11
11
11
11
11
12
12
12
12
WS
WS
WS
"• "• °" "• \
LR
LR
LR
LR
LR
LR
WS
WS
ws
ws
ws
ws
12 HFSMTf(-40dB)
17 HF.SSsG^lR, "" ,'
12 HFSMTf(-40dB)
10 HF SMT 50MHz
11 HFSMTf(-3dB)
12 HFSMTf(-40dB)
< /, T^C~x>~P " , "*
9TA'fe' Bt*i*j-y>^ii^L.'ifij
12 HFSMTf(-40dB)
12 HF SMT f(-40dB)
8 .JS&$&%j%3$® ^
9 BFFT^f<^(3^.)x-- .
12 HFSMTf(-40dB)
13 HFTLC 50MHz
10 HFSMTJOMHz
13 HFTLC 50MHz
12 HFSMTf(-40dB)
1 HCLVPTH _
8 ^HFPlH3f(-3dB^ J,
1 HCLVPTH
10 HF SMT 50MHz
11 Jff SMTf(-3dB)
7 fiSFf^^ler
10 H£ j$iM 5PMFfz ^
n% ¥':rt? 'tfikji't^ $•&" "^t t$&*i: ^
*3* 7w3j[i-JC. Jt* NrW**f "• "•
1 o Y-fy? ^J^j^|!? j^. jt^yyfex
13 "HFTLC 50MHz' ""
Open PTH
•". ~w^™*wv%™~, ^ ' ' v\ "•, " % '•'•'•'• '••" % ' \ ^ %^ f
OpenPTH
Waveform shifted
Waveform does not go to -40dB
OpenPTH
Distorted waveform
OpenPTH
Open PTH, distorted waveform
Open PTH -2 places
OpenPTH
Open etch
Open PTH - 2 places
094-3
12 WS
9 HF PTH f(-40dB)
12 HFSMTf(-40dB)
13 HFTLC 50MHz '
17 HFTLCRNR
Waveform distorted
094-4
12 WS
- 095-4
12 WS
1 HCLVPTH
7
9
10
11
12
13
1
Open PTH - 2 places
HF SMT 50MHz
HFSMTf(-3dB)
HF SMT f(-40dB)
HFTLC 50MHz
HCLVPTH
Open etch
F-9
-------
APPENDIX F
013-1
• 015-2
t -.
051-2
054-4
055-1
v.~v
13
14
13
13
13
•^ \ •.""'v^l;!
LR
LR
WS
WS
WS
r^--
6
9
8
8
7 :
;:;v"F-^r-
--,. .^ - ^n - ^
-USD -SMT JT + HSD device fail
^Fl*iSI-.^MIfe ^ __ Open etch
HFPTHf(-3dB)
HFPTHf(-3dB)
^SfeFIHSfflMife*'
Opei
letch
055-4 13 WS 12 HFSMTf(-40dB) Waveform distorted
036-2 16 WS -12 HFSMTf(-40dB)
F.3 HCLV Circuitry
Pre-test measurements and deltas were analyzed with the GLM in Equation F. 1 for the main
effects site and flux and their interactions. These data were also subjected to a second GLM analysis
based on Equation F.2 for the main effects surface finish and flux. The base case for the GLM hi
Equation F. 1 is defined as HASL at Site 1 and processed with LR flux. The base case for the GLM in
Equation F.2 is defined as HASL processed with LR flux.
Tables F.4 and F.5 summarize the results of these GLM analyses for HCLV PTH and HCLV
SMT. The upper portion of these tables contain the GLM results for Equation F. 1 while the lower
portion of these tables contain the GLM results for Equation F.2. The rows labeled "Constant" hi these
tables contain the least squares estimates of PO in Equations F. 1 and F.2 for each test time. The
numbers hi the columns beneath the "Constants" are the estimated coefficients of the terms in
Equations F. 1 and F.2 that are significantly different from the base case. Shaded cells signify that the
corresponding term in the GLM is not significantly different from the base case.
The rows labeled Model R2 in Tables F.4 and F.5 show the percent of variation in the voltage
measurements explained by the respective estimated model. This value can range from 0% to 100%.
The model R2s for Equations F. 1 and F.2 for the HCLV circuitry' are summarized as follows for each
test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish and Flux
HCLV PTH
HCLVSMT
HCLV PTH
HCLVSMT
2:0%
4.2%
0.7%
1.5%
2.3%
7.7%
1.3%
0.3%
3.7%
10.9%
1.7%
9.8%
19.1%
2.1%
7.7,%
0.7%
These R2s are all quite small, which indicates that the experimental parameters: surface finish,
site, and flux do not significantly affect the HCLV voltage measurements at Pre-test nor do they affect
the changes in the voltage after exposure to each of the three test environments. That is, the HCLV
measurements are robust with respect to surface finish, site, and flux. The results for the two GLMs
used in the analysis are now examined hi more detail.
GLM Results for Site and Flux
The uppermost portion of Table F.4 for HCLV PTH shows that only two experimental factors
(Site 2 and Site 8) are significantly different from the base case for the GLM in Equation F. 1. The
F-10
-------
. APPJEJVPIX:F
estimated GLM at Pre-test for Equation F. 1 is obtained from the estimated coefficients in the second
column of Table F.4 as:
Y = 7.14 + 0.06 Site2 + 0.07 Site 8
where Y represents the voltage response. The predicted voltage from this estimated model is 7.14V
for all site and flux combinations except Sites 2 and 8. The predictions for these two sites are 7.14V +
0.06V = 7,20V and 7.14V + 0.07V = 7.2IV, respectively. Note that even though these two terms are
statistically significant, they represent very small changes from the base case voltage and, as such, are
not of practical interest. Moreover, the model R2 is only 2.0%, which has no practical value. Similar
comments hold for the GLM analyses at Pre-test for HCLV SMT.
Columns 3 to 5 in Tables F.4 and F.5 give the HCLV.PTH and HCLV SMT GLM results for
Delta 1, 2, and 3, respectively. Note that these latter three analyses are based on changes in the voltage
measurements from Pre-test. The model R2 values after 85/85 and TS are also quite small, which
implies that the experimental parameters did not influence the HCLV measurements after exposure to
the 85/85, TS, and MS test environments.
In spite of the lack of significant experimental .parameters in the HCLV GLMs, there is one very
interesting aspect of the model for HCLV SMT at Post MS. Note that the estimate of the constant term
in the last column of Table F.5 is 2.48, whereas, the estimated constants at Post 85/85 and Post TS
were 0.04 and 0.05, respectively. This is an increase of approximately 2.43V. The explanation of this
increase requires a review of the HCLV circuit, which is given in Section F. 10. In particular, Section
F.10 explains that the HCLV circuit has seven 10H resistors, RI, R2,..., R? in parallel. The overall
circuit resistance, Rtotai, is the parallel combination of these seven resistors, which is given as:
-^
ion
(F.3)
ion
(F.4)
Since a current (I) of 5A was applied to the circuit, Ohm's Law gives the resulting voltage (V) as
= 7.J4JX-
(F.5)
During the MS test, it was noted that one to three of the resistors frequently fell off the board. In fact,
158 of the 164 PWAs were missing at least one of these resistors. If a single resistor is missing,
Equation F,5 would be revised as follows:
(F.6)
Likewise, two missing resistors increase the voltage to 10V. Next consider the following dotplot of
voltage measurements at Post MS.
F-ll
-------
APPENDIX F
+ + + + + + Voltage-
7.20 7.80 8.40 9.00 9.60 10.20
Note how the voltages are lumped around the points at 7.14V, 8.33V, and 10V, which
corresponds to the loss of no, one, or two resistors. Thus, the constant term in the GLM represents an
average increase in voltage of 2.48V over the nominal expected value of 7.14V, which is between one
and two missing resistors.
GLM Results for Surface Finish and Flux
The lower portion of Table F.4 for HCLV PTH shows that only one experimental factor
(Ni/Pd/Au) is significantly from the base case at Pre-test for the GLM in Equation F.2. The estimated
model is:
Y = 7.15-0.04 Ni/Pd/Au
where Y represents the voltage response. The predicted voltage from this estimated model is 7.15V.
for all surface finish and flux combinations except for Ni/Pd/Au processed with either flux, in which
case the prediction is decreased by 0.04V or 7.15V- 0.04V = 7.11V. As was just discussed with the
previous GLM, even though the coefficient for Ni/Pd/Au is statistically significant, it actually
represents a very small change from the base case and, as such, is not of practical interest. Moreover,
the model R2 is only 0.7%, which has no practical value. Similar comments hold for the GLM
analyses at Pre-test for HCLV SMT.
These low R2 values imply that the experimental parameters do not differ significantly from the
base case in terms of their impact on the voltage of the HCLV PTH and HCLV SMT circuits. That is,
there is no practical difference from the base case voltage measurements due to surface finish or flux
type. This result is to be expected since there were no difference among sites for these circuits in the
GLM analysis based on Equation F. 1.
Columns 3 to 5 in Tables F.4 and F.5 give the HCLV PTH and HCLV SMT GLM results for
Delta 1,2, and 3, respectively. The model R2 values at Post 85/85, Post TS, and Post MS are also quite
small, which implies that the experimental parameters did not influence the HCLV measurements after
exposure to the 85/85 and TS test environments. However, as just explained for the Site and Flux
model, the constant term in the last column of Table F.5 is affected by the missing resistors.
F-12
-------
APPENDIX F
Table F.4 Significant Coefficients for the Two GLM Analyses by Test Time for HCLV PTH
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
7.14
0.04
0.05
0.14
Flux
Site 2
SiteS
,'-.' _;-- - -
0.06
-0.17
V
Site 4
SiteS
Site 6
^
Site?
SiteS
Site 9
0.07
'
&••? >% \-w.\v c- "" (
X," ""-•*"\- *
\^ ^ -X% ^ %% % v. f
'< -- "-^^^^
Site 10
Site 11
Site 12
0.13
0.80
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
-0.16
Model 1C
Standard Deviation
2.0%
0.13
2,3%
0.18
3.7%
0.17
19.1%
0.36
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au .
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
7.15
r:v ;^t;^
:^«-. ^> --Xv. v
i \ v %«»
^ " '•>" ^- ^^
1 v "*\ •-<-'• ^%
• '•> ^ ^ "\ %%
-0.04
•. :• -. w.vi •
-- „.•.-?"•' '--
0.7%
0.10
85/85
(Delta 1)
0.03
-,* - - - ^ >•.
0.07
?, - -" "'^^% ,-
^•^^^ ' "1
1.3%
0.10
Thermal Shock
(Delta 2)
0.04
V- " s% %v,;™ -: "-
• ' % %sss^ •-
'"% "'0.07
^\ - ^
<•'• V"1" "•'""• "• -.-.•'•'
"• y-
"• % *ss VXV
•• ^ •* s
1.7%
0.17
Mech Shock
(DeltaS)
0.13
•."• --\ ^ ••x^^'- "• % ^
* : *:> ^ «*S*.
0.34
-sx-^x % *,
•••.•. x--- ^
v s "" "" '•
• 7.7%
0.38
F-13
-------
APPENDIX F
Table F.5 Significant Coefficients for the Two GLM Analyses by Test Time for HCLV SMT
GLM from Eq. F.I; Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
Site 8
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
OitA 1 <
one ID
Site 16
Site 4* Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
7.26
;;^\*;u^r,
-Ok,
;%f> f ^--*-
* V*^' *
V? -. " -.-X f
*<& "-^ e
% /»»»
;**? .
1 s?5-> ' ; - -
""""oTbT
^c;c^ "s»*
,>•$.•••• > *
-0.07
S^TX;-^ ,
, t>.^.- ;, , ,
-V5^, „
\^i/ * ^* *
v - x "'- /
,~ ' '-3: *
U«^ '*'
>* i. .>''•
;- * "^|^«?
- \ v
C<^W*.« -
f- ^ " "
??*$
^ , **^\V
, S^f ,
-.\ v "^ "• X$v. -^ 5s
4.2%
0.09
85/85
(Delta 1)
0.04
;'-- -^- ^ , \
^ * ,>-^->'\"-v
s,- % " - v» -.- *
•• , -
X* 5 - V x
" "" To.bV """
i-XWWV\^sJ
"* ^ f f *" ;
^ ^^«.^V.s\%% ^fff x \
f \ ;
"" ' ;
, %-w. .i ^ -.-. ^ %-h.Xv ;
-0.14
4\->^v, N % \
7.1%
0.12
Thermal Shock
(Delta 2)
0.05
: ""•s, "• "•
; - " %
-0.10
L "^Nf-
•>
f
, ^ .
•••«,•.. ^^ /«, 0
«N '•^ « ™
0.11
* ^'' ^0 ~
% o-- 7 ^ > ,
»
% %% ff ff
,,-,-f c -l"
>> % ^•.^••.
% •. % \
-0.11
10.9%
0.13
Mech Shock
(Delta 3)
2.48
-0.48
* >..
%% •• % O- *^ "• :
} ,% *^*^;=
o-%% v * % ••
^>
$
\f. "" ^s --s\ %x\ ^ 1
"j* ,, * *" """ ;
' s V '
s% •. *,
^ •. ^ % •• v^ i
•>
\ ^
, *s
*«• '""% *x^^*^l
f s
^ "• ssx % %1
2.1%
0.70
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
7.26 .
^v-x-s ••x-^^v VssX-->--s
^>\ ^-ix., -. > ^
c\r- ?%xi:-.-\i
s*^ %5\o-- > *. ••;
i-ss-^w,^*! ••"••-,
i-^S4 sW ^
"'%"-- ^- V1
^K^-c;;- •»
•S^t^Vv 5 *^
^^i^x^VfeS- /' . •-\'\-
-0.02
1.5%
0.09
85/85
(Delta 1)
0.03
"•.ff^C "V-.-v.-,---'';
3. •.••:: ^ %ss s
S *V3* ^X-\^ s ^o\^
^• '.. •• \ ^ ^™? w\ Xl , x
", ',;• ^-\ --
-0.10
x •• "^ "•-. \\"' •-. "
f -\ ••'••. •.
9.8%
0.13
Mech Shock
(Delta 3)
2.49
* " % ^ 7^" :
-o.is"
:v. :v,v. •. -. ^ -. x vX-v. % f s.^
s "• ~*f •.•."•'
•. ^\ x:^^:
- *:: --
^ - - , ;^x- ,;
v\<.v--
0.7%
0.70
F-14
-------
APPENDIX F
F.4 HVLC Circuitry
Results of the GLM analyses for HVLC PTH and HVLC SMT circuits are given in Tables F.6 and
F.7, respectively. Columns 3 to 5 in these tables give the GLM results for 85/85, TS, and MS,
respectively. The model R2s for Equations F.I and F.2 for the HVLC circuitry are summarized as
follows for each test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux.
Surface Finish and Flux
HVLC PTH
HVLC SMT
HVLC PTH
HVLC SMT
13.3%
20.9%
7.6%
14.0%
5.2%
14.0%
2.5%
15.3%
0.0%
18.7%
2.6%
12.9%
3.2%
NA
3.2%
NA
These model R2 values are generally higher that those observed for the HCLV measurements.
However, the magnitudes of the coefficients were too small to be of practical significance relative to
the JTP acceptance criteria, which indicates that these parameters do not influence the HVLC
measurements. To further explain this point, consider the coefficients for site and flux in Table F.6 at
Pre-test where the constant term is 5.018uA. The largest coefficient at Pre-test is -0.008(j,A for the
interaction of Site 4 and Flux. Thus, this interaction can decrease the constant term to 5.018(j,A -
O.OOSuA = 5.010|aA, which is so far from the lower and upper limits of 4uA and 6^A that it is not of
practical interest. Note that there are no R2 values listed for HVLC SMT at Post MS. This is due to
resistors coming off the PWA during the MS test, which caused the HVLC SMT circuit to give a
constant response for reasons that will now be explained.
Boxplot Displays of Multiple Comparison Results
Figures F. 1, to F.8 give boxplots for the HVLC PTH and SMT circuits. It is important to keep the
vertical scale in mind relative to the acceptance criteria when viewing these boxplots. That is, the
acceptance criteria indicates that the current should be between 4jjA and 6|jA. These boxplots are
centered close to 5|jA and the total spread is on the order of 0.02uA for the PTH circuits and
approximately 0.5uA for SMT circuits. Hence, even though there are some statistically significantly
differences, they are not likely to be of practical concern. Note the boxplots in Figure F.8 for HCLV
SMT at Post MS. These values are all either OuA for very close to it, reflecting the fact that the
resistors came off the PWA during the MS test.
F-15
-------
APPENDIX F
Table F.6 Significant Coefficients for the Two GLM Analyses by Test Time for HVLC PTH
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site 7
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4* Flux
Site 5* Flux
Site 7* Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
5.018
4XJX, ""x%.,x..
"^jfe-;:
0.007
' rrT^ix^.7
^X^-xV "
, x^^-ist-^-^
0.005
0.004
v "* v.% \
0.004
^^t*''*; * '
"" V, *"
-0.008
;^'JV^";"\-
s x3i%:'\** '.
:r^r^rr
~ M**^^ -
13.3%
0.005
85/85
5.004
^ 5-, ^ :
* ^ ' , ,' , ' „.
\ ff "•
<. ^ s "" f *-, ''•
-" s *^^
s % % ^ *• :
- ' s ><-, ;
xx>«»« S "*••••'„
, \ ;^\* I
0.006
" ' , "
-, f
f f •. % %% % s« v^i
^X-sj
*, , , *^ ^ - :
0.006
„' X , v <;X -' % j
5.2%
0.006
Thermal Shock
4.999
1 %
; $^A "*
: v % ^ f f
\ " - ^s ' ' %*^
; ' " ^^^
X-.x\
[ N \ % % % x, \ e ,-
L "%v"'
:
1 ^••••s %
% •• s« ss ^
' ^^x v ^ %
X ^%
«. x -.
%% X
%
\ *^
"• % X
^% ,XN'^
0.0%
0.006
Mech Shock
4.998
* ,
•.•.••,
ff
-.
--, --s--.
*% v % ^ \
' ^ "• ,
-x- •>^-
^ *
- ' , % 's
^•- \ %
-0.005
V -X
v- ' WJ«, ^
- %% , :\.
^\J ^
X •• sis--% -.
\
3.2%
0.006
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP ' .
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
5.018
,A^- Xv***\
.^.-. -.-uX •.. ».'•'••, --^
0.003
0.003
^ x Vv^^^^W^Vx
^^^^
1.6%
0.005
85/85
5.004
s«V \ ° Xl---
s "•"<• ^f ^ '•
t\^ •.•$•. ^ •• ^ A^
0.003
s-'S-^Ns'-.XXX^^ ^s_s\s ^ s
ff S S S S S %
••s N s s ff f
5 » s %
•"-••A-V-VS^ '•V%'-^ •• ^ S
^ f\ ^
2.5%
0.006
Thermal Shock
4.998
0.002
's
•*• *• s f * I
f •
-. ff '•
<*. f :
••<.«• :
2.6%
0.006
Mech Shock
4.998
x :S1V -
-0.003
.3.2%
0.006
F-16
-------
APPJENJMX F
Table F.7 Significant Coefficients for the Two GLM Analyses by Test Time for HVLC SMT
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
5.038
5.034
5.039
Flux
Site 2
Site 3
Site 4
Site5
Site 6
Site?
SiteS
Site 9
0.172
- ™/;™^PX;^V
, ;-»*s/s4«^ • -
0.173
SST'T^'" T-
0.170
Site 10
Site 11
Site 12
0.111
0.122
0.111
^-.w^%S-a:
0.125
0.109
' 0.120
Site 13
Site 14
Site 15
0.125
v xv ; --, -- - ~
•*S. % - -*-- - -v
^X ^ 0*-^f^ ^
0.126
0.125
Site 16
Site 4* Flux
Site 5-* Flux
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
V...A..A.;.
Model R2
Standard Delation
20.9%
0.100
21.5%
0.100
18.7%
0.112
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
5.032
JL %--x^x\
0.095
0.087
' ' v x V " s \.
y .?-,; ^,|
14.0%
0.100
85/85
5.027
,, , *,;;•".
0.100
0.090
% •••••:
'• ^ \« iW>/^ •>• *"•> -A^-1 ••
A«^s «v -.-.? --?^ W, C \
*'"' '•'•'', %\ * *" *" Xs ~"\ ":
15.3%
0.100
Thermal Shock
5.033
^ % - -.•.*•-..•- % % \
'"•-'•.•.-.
0.097
0.085
- , X s
"- ^*"". ^ "
-i ^ -S^ w»^
12.9%
0.110
Mech Shock
: % v.>s -.-.-.x % f •. J.yk ^ ;
:-, - , ^V-^-^-Tt-
:-- •.- -- -..V ^- -X'-^--'
I ,,_ wo-y^
; s^ -. N^t " ^%
-------
APPENDIX F
F.5 HSD Circuitry
(
The complete results of the GLM analyses are given hi Tables F.8 and F.9, respectively. Columns
3 to 5 in these tables give the GLM results for 85/85, TS, and MS, respectively. Note that these latter
three analyses are based on changes in total propagation delay from Pre-test. The model R2s for
Equations F.I and F.2 for the HSD circuitry are summarized as follows for each test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish and Flux
HSDPTH
HSDSMT
HSDPTH
HSDSMT
5.1%
6.1%
0.9%
1.0%
9.8%
6.4% .
1.6%
0.3%
4.3%
0.0%
1.8%
0.8%
9.5%
2.3% '
6.7%
0.2%
All these model R2 values are quite small at each test time, which indicates that the experimental
parameters under evaluation do not influence the HSD total propagation delay measurements.
Boxplot Displays of Multiple Comparison Results
Figures F.9 and F.10 give boxplots of Pre-test measurements of total propagation delay for the
HSD PTH and HSD SMT circuits, respectively. Note that most total propagation delays in Figure F.9
for HSD PTH are a little over 17 ns with a range of about Ins. Figure F. 10 shows that the total
propagation delays for HSD SMT have a range of about 0.4ns and are centered about 9.2ns. The
percentage changes in the total propagation delay measurements were small and well within the
acceptance criteria so boxplot displays of these measurements are not presented.
F-18
-------
APPENDIX F
Table F.8 Significant Coefficients for the Two GLM Analyses by Test Time for HSD PTH
GLM from Eq. F.I; Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Deltas)
Constant
17.13
0.55
0.98
0.37
Flux
Site 2
Site 3
•>%-K>-. •
**
-0.46
2.60
Site 4
Site5
Site 6
0.14
0.61
-1.00
Site 7
SiteS
Site 9
1.89
- ^..^
^"V %% ** *^
"-^x, *;
"- " V s v\\
"- ---^
1.6%
1.00
Thermal Shock
(Delta 2)
0.88
^- ^V&x --X--^
. %-. \\ •- ^^ fff.. ._ ;
\ ^ •• •••".'•Jv •• ••'•-"• "•
% %'v^^^^-> :
* "" ^ * * % ^ ™ =
v*,% -J-^ --^- !
'-0.36
1.8%
1.30
Mech Shock
(Deltas)
0.52
• *• •• ^ ., *•••••••&•& vb%
" - " Vx-^-
•. ^ % •• '>•••.•• •• v \
-2.89
-. •••• "•"" "" % ^ "-1"
"• \ "•'•"'
wvVX - "•.-
,,o \^--
6.7%
3.5
F-19
-------
APPENDIX F
Table F.9 Significant Coefficients for the Two
GLM from Eq. F.
GLM Analyses by Test Time for BSD SMT
1: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
9.23
*:^~x ,*%^
i^x,;? :vc;'/
v \ •. s
K,-, /, r ;., °
f- XXT-XV-X.^ j. ^ ^ ^ ^ ^
^X^ •.'
,r < " "'/,'
* v-^s'-w ,_ **• s
•. •• ? x,x<>x ^ tiff-- x*-
/:H""
•i^&vVV'l
vf *1W
-x>.™ - * ^
s^./ *y>'t«i'*, '
v^vvv«t .. ^ ^.$y& „
V^xS- ,.*•»«•* •. v\i!x> ' x
0.12
«5 ° vi':?*?** \xSfc %"
•.'''• o
, . . -, . \ f\ V. .
-0.10
v o% %s " %^ s
N^\- x^s 1 , \"w,, « s ,
w~^--\ ^^ „
,^N s- * « ^ ••
6.1%
0.13
85/85
(Delta 1)
0.94
**> , ••*
-1.59
• xs%- •• •
-XV
•j. ,
-.•^.•- f f
. " \ %% ^
*.•. v
'^
"'*----x ' '' '
5- ' ^- ^--
%v>^v s •.
',
V.S ss s JV ^ -.«.
"-i.it
*• ' % '•••'•x vs%
',''" % ^ -^
NS?- •> *
-Xs^ t
•• v> % S
^VV
\^
6.4%
1.65
Thermal Shock
(Delta 2)
1.16
%*&y x\^ ^
^^r/V^'.,^
^ , **• '- ~
^^ """ "" '
^ '' -
•• X; %"" ••%
f •>•*
^ \\ ,.
'-
-
X ^ v-s
' ,' '
S*«* %* "
^ % •."
f •.
0.0%
1.99
Mech Shock
(DelltaS)
-0.002
f
'*'* •>'<%s ••
•••.
•, •, f
-".. .v:.. ?*x
-1.60
•x-\
' % v x' * ^ ,
, * s-"- ,
•• ""s s
j ^X-\%s
\ %
^ * ^
•=••• «, * ,
% •• i%%\. x
2.3%
2.25
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
9.21
O ^. \| "•* ^ •."\v.\-.\ ^ "• •- -v v "if v-x
Xw'-'''"^' ^ - - v v^^\*' f -
^CfX^ ^ ** •***^ ^\ ^i
^•^x--1 ^A-. o •. •.% ^
if'-X.'"- - •• <• -O s ••
^V>s %-v *»sN
fete « s% % % •. " »% v
SS.^r'' v» w ™v.;^^lx^
^^s™^.
1.0%
0.10
85/85
(Delta 1)
0.77
••••x^ %% -.-.\''\ s
^ % S"-'' % -.
\ \V- - %
••\x^ x ~ * s ^ ;
0.35
?**( /• v ••;" f •. ••
•"*\\. 0 V" -"-" - i
0.3%
1.00
Thermal Shock
(Delta 2)
1.23
•• •*.
•.X "• - c-^c-^ *•"•
•/"^ -'x'-
% — •• xO ^ "x- J
-0.56
" r " , X °-
- - '$ ••
-.-. - , ••?,
\ " - \-" "
0.8%
1.90
Mech Shock
(Deltas)
-0.04
V> v > ,* -.
'• * 0 ^ %^
: ;; "v* "*"* 4> \ "~
\ x- *\-yi-. ..
: "• \ >
-0.25
•••• *. ••••••
•• % %x, *x ^ % •. % ^.
,, 5* <«>- +
\ --'X ..""X-x
0.2%
2.2
F-20
-------
APPENDIX F
F.6 HF LPF Circuitry
Pre-test measurements for all HF LPF circuits were subjected to GLM analyses, as were the deltas
after 85/85, TS, and MS. The results of the GLM analyses are given in Tables F. 10 to F. 15. Columns
3 to 5 in these tables give the GLM results for 85/85, TS, and MS, respectively.
Note that these latter three analyses are based on changes from Pre-test measurements. The model
R2s for Equations F. 1 and F.2 for the HF LPF circuitry are summarized as follows for each test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish, and Flux
PTH 50MHz
PTHf(-3dB)
PTHf(-40dB)
SMT 50MHz
SMT f(-3dB)
SMTf(-40dB)
PTH 50MHz
PTHf(-3dB)
PTHf(-40dB)
SMT50MHz
SMT f(-3dB)
SMT f(-40dB)
20.6%
7.1% •
14.3%
3.9%
8.8%
5.3%
4.3%
7.8%
4.5%
2.7%
0.7%
5.2%
29.5%
10.8%
. 9.6%
10.3%
10.5%
2.3%
2.3%
0.2%
1.8%
0.6%
1.5%
0.3%
24,1%
10.2%
7.6%
21.1%
19.1%
16.1%
0.3%
1.6%
1.6%
0.8%
5.0%
4.9%
20.5%
23.4%
13.5%
32.2%
. 14.3%
29.4%
8.1%
10.9%
10.9%
6.1%
3.0%
14.4%
The model R2 values are quite small at Pre-test, which indicates that the parameters under
evaluation do not influence the HF LPF measurements. The same is true at Post 85/85. The model R2
values are also quite small at Post TS and Post MS. However, the test measurements contained many
extreme outlying observations at both of these later two test times, which greatly increases the sample
variance and in turn hinders the interpretation of the GLM results. As indicated in Tables F. 1, F.2, and
F.3 there were many anomalous HF LPF test measurements (171 at Post MS).
Boxplot Displays of Multiple Comparison Results
Boxplot displays of all test results for HF LPF circuits have been created to aid in the
interpretation of the results. Figures 4.9 to 4.15 in Chapter 4 show the boxplots for the analyses with
significant differences or values not meeting acceptance criteria. Figures F. 11 to F.27 show .all
remaining boxplots associated with the HF LPF results.
F-21
-------
APPENDIX F
Table F.10 Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH 50 MHz
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Hux
Site 2
SiteS
Site 4
SiteS
Site 6
Site 7
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4* Hux
Site 5 * Hux
Site 7* Hux
Site 11 * Hux
Site 13 * Hux
Site 16 * Hux
Model R2
Standard Deviation
Pre-Test
-0.721
~^VVN^ ffc:
"!V^| %: ~H
""• %C" \-" %% :
I£':H
383
-0.180
^ N 0**^ > X *"
V. ^ x. f f
StS^
6.160
f f r~i ''ff' tt '
20.6%
0.055
85/85
(Delta 1)
-0.034
0, , **^^
S -x- » s* '-'
^\ , ; s
X- •- 'I\ s
4.
' ^ x
X % •• ""* %
^x^ ^ v w
S ^\ ^
0.197
' "•'•^ ™'
°x ;
-0.206
29.5%
0..048
Thermal Shock
(Delta 2)
-0.002
" -- x^->Vxx
-
f f
•• ^ ^ <«. s * % "•' s
"• s %O\ '•^•^ **
' *XX^'^^
0.192
-0.073
% X Cs *^s ""
•."•s "• ^ tf"-% %%t ^.-^ ^ •.
-0.180
.24.1%
0.063
Mech Shock
(Delta3)
-2.666
f f
^ •" %
> - *"*'
•. •. X-
' -28.1
^X; * X^*^^* S
« \
-18.5
20.5%
14.1
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/PoVAu
Hux
Model R2
Standard Deviation
Pre-Test
-0.720
-0.034
4.3%
0.060
85/85
(Delta 1)
-0.034
^>!:Zfl
0.023
Y" "'"" ^ "~.C •! "•
2.3%
0.050
Thermal Shock
(Delta 2)
0.003
-0.010
-.-. -. -,:•. ; vj.s ^.ws ^ -.x-X^-- -^
•• "• s ^
' cj,r^7^
0.3%
6.072
Mech Shock
(DeltaS)
-3.28
* s ' % ^ S % s
-13.6
'.•-.. •• ....
8.1%
15.00
F-22
-------
APPENDIX F
Table F.ll Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH f(-3dB)
6LM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
,85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
283.0
-0.9
0.5
-1.05
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
-2.2
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
-116
Site 13
Site 14
Site 15
-1.8
-1.5
Site 16
Site 4 * Flux
Site 5 * Flux
0.7
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16* Flux
-1.2
-68
X-.v
-79
Model R^
Standard Deviation
1.1%
2.0
10.8%
0.9
10.2%
1.5
23.4%
58.5
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
283.0
"• Sff f V. ''
£. :-. •;:•..:• ^.•v*:-.J-'
'':?•"• ^-- -\ •• 0
•• , ;?*•• •••• •".• •• i ^
-1.6
•^ " \ % "
%i% __••-. ^% %% %-.-.\ %••%%
7.8%
2.0
85/85
(Delta 1)
-1.0
0.1
^••v* -.-.-... ^ ^•-^.v.i.-vX'X*
T ;V ^ i'^^
%v v f^^y* "•
;,,,- * <•-•,- ,„ ,
«• '•, % •• •- y.* •• " ,
'^'V* % •*•"
%V5^ if. •-. %
0.2%
0.9
Thermal Shock
(Delta 2)
0.5
-0.5
T"" '^"^VTT - %
-; -•'• " s*4s v - ,
,;. . /^->»,
- - x"1 --' %\v \\
1.6%
1.5
Mech Shock
(Delta 3)
4.19
••"•., "••.v-,,^'>*>
*• "• 0% \% \
%\ "• s : ^ *%SX^*^
"-53.0
•• •• »••' , v i ' o •.•.•. •- %'
- •. v ^-- ' ^- \
-- ---- - - ----v xv-
-23.8
10.9%
62.0
F-23
-------
APPENDIX F
Table F.12 Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH f(-40dB)
GLM from Eq. F.I; Sites and Interactions with Flux
, Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
"sifp T3 * Fluv
Site 16 * Flux
Model R2
Standard Deviation .
Pre-Test
472.9
p?;::
-> *%-Jf^v- "• "* ^
Xivw^ ..v. •• •• x
-3.8
ssO
SfiH
• -5.1
^-4.5™'
., '"•. v *• s % 1"vt,
^2T^ s^
5;^ ^v
14.3%
5.1
85/85
(Delta 1)
-0.2
: % ^s ••••••X %%
•» s ^ N% - v; ~\
0.9
^ \ ^ x%
-1.5
r^^ s^*v
X , - "-S, - <--^
* v4U^.%% v" ' %
^
9.6%
1.2
Thermal Shock
(Delta 2)
-0.2
"" \ * %
-1.8
v.\ ^ ^^
^ ^ "• ""••• v. f
"" %S s •. f f
", ," % v%" " -\-- ^
2.6
%^^- \%
- * *-
7.6%
1.5
Mech Shock
(Delta3)
-11.7
i
, N ""V'
•X %
•• ^ \ \ s V>"^%
_140"
^ ^ ^:'Xw^
" "' ^HV:
,x« ^ - v ""••..'•.:
^
13.5%
77.1
GLM from Eq. F.2; Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
472.2
^"ov^7"^ j
<• ••*' ; ( \^o^1
-4^ "•"^>" '
c.\vC-^%% ^ •• •••••• ':
-3.2
\ ••AN'-.w* \ v-\'- s \ ;
•••••>% x %"*- '
:5^3Ti: !
•. ^ ^^ :
4.5%
5.0
85/85
(Delta 1)
-0.1
?
"~ ™-a4™""
' ^ *%•>
*•"•;;•.«.
i N i*»*
"% '^\v ,%*;
\ ^ ' f •.
1.8%
1.0
Thermal Shock
(Delta 2)
-0.3
\ s %
\ s
0.71
1.6%
1.5
Mech Shock
(Delta 3)
-8.41
% \
-83.0
S% v
^ s x-
•C-. :
10.9%
78.0 •
F-24
-------
APPEMMXF
Table F.13 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT 50 MHz
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-0.733
-0.018
0.005
-3.1
Flux
Site 2
Site 3
-0.112
-19.2
Site 4
SiteS
Site 6
-13.5
Site?
SiteS
Site 9
-0.126
-49.7
-0.049
Site 10
Site 11
Site 12
0.031
-31.4
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
0.021
Site 7* Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
->\\^"> •- s S\SJ.S
^047"""*
25.0
Model R2
Standard Deviation
3.9%
0.039
10.3%
0.037
21.1%
0.069
32.2%
17.2
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-0.733
-" -»*- , ,,
• > "'•."" y-'-^'''»vVv-f
'--.<.'•'••- ^ v :
: ^ .. %% •. %
0.020
?>% s v % * •• ~ :
••, ••, " , i ' •
^ %
2.7%
0.030
85/85
(Delta 1)
-0.023
: "^ *
rv-- - ^-** *
"•^•- ,^ ^f J s
S*> f M "• ff f
v.v - -, -=:•••• % •,
0.8%
0.077
Mech Shock
(Delta3)
-5.62
v % \ 0 ^ «. % :
''''•.%'' v*- ••••% '
-10.6
-10.7
w.1; % -, •- ^ ^ .. ••••*
% "% ^^••'^V^ %
^ •.•.%•. •-•.•. -,%\% -^ -.-.%
6.1%
20.0
F-25
-------
APPENDIX F
Table F.14 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT f(-3dB)
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
SiteS
Site 6
Site?
Site 8
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5* Flux
Site 7* Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
319.8
""^•S.*.^ ,. :
11 -. ^'^'"^V.vX •,•• '
•."•^ v* '•.'*''" :
^ww%% fs :
*" *v - -* ""!
\ ••«< i>x ,, <' , c :
. V
%x ^ ,- , ' '•
3.7
_v*j;.v ^
*^^.^^.^
' \ v« % ••;
8.8%
1.9
85/85
(Delta 1)
-1.3
Jo" "
\ '
*,"%:V ,--
1x*»*,^*^ x , ^
1.5
\
•^> ,.:
-2.2
% f
- 10.5%
1.1
Thermal Shock
(Delta 2)
0.7
••
^s, .% «
-15.3
'-4.0™"
•A ," ^^
'_23~~ "
\™
-3.7
11.9
19.1%
4.7
Mech Shock
(Delta 3)
-15.5
%%\^
~108 " ™"~
^ »
\ "• "^
% ^ %%' X
-143
•. •.'
'
-102
"" .. S ^^ V
14.3%
112
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/PoVAu
Flux
Model R2
Standard Deviation
Pre-Test
319.7
0.4
.w.\X-v^v«ss>VKC^-'^v'kX^''--^'-;
>^k, - % "
- •• --A^-i^\ v s
^ v •fsSv*****. --\
"!^xtx^,,
v-\>^sX^^-,--^-
, ^A» M, s,Y
.^>^^
0.7%
2.0
85/85
(Delta 1)
-1.3
,x^ v--
0.5
,\\^ i ,,W
\
*• •. ••> •• ^v. -c- %. %
1.5%
1.0
Thermal Shock
(Delta 2)
0.4
.,•. ?.. %s %v»l »*»
"-2.8
V N %v «S %
5.0%
5.0
Mech Shock
(Delta 3)
-1.98
•• x
%t ' -.
-4l.o"
3.0%
11.0
F-26
-------
APPENDIX JF
Table F.15 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT f(-40dB)
GLM from Eq. F.I; Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta!)
Mech Shock
(Delta3)
Constant
865.5
1.7
-8.1
-80.3
Flux
Site 2
SiteS
-244
Site 4
SiteS
Site 6
-10.7
-171
Site 7
SiteS
Site 9
4.9
-430
Site 10
Site 11
Site 12
2.2
&>>?£*
V*t •"">
-19.7
-365
Site 13
Site 14
Site 15
Site 16
Site 4* Flux
Site 5 * Flux
\
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
-23.7
Model R2
Standard Deviation
5.3%
21.0
2.3%
7.6
16. l'
9.1
29.4%
221
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
861.2
^ " "f: : '• ' ,
: VV~ ••
"•&>'•• f>V- '• 2
13.4
"• wS-Xii "~ f '•
\ -!•"• "~\\^ -.\ % ;
5.2%
21.0
85/85
(Delta 1)
2.0
"• ,_"" ^ ^ s %-wwwS* -C- %•.-.•, •,
- --- ,'- \, - H
1.0
, ,..»„, ••• 11
"• ^ \ ^\x •- •• •• \
0.3%
7.0
Thermal Shock
(Delta 2)
-6.8
5™,^ ,. v.v - v " A v>" -.x %
xS % 5 w"
^ v 0«* ^\%
" "" '- ' " ,*: - ^•>
r ^- ^ ,-
-4.4
4.9%
9.7
Mech Shock
(Delta 3)
-146.2
! ^"^ ^" "-" Vo
u :\ ^-J-^.^
•• •• ^v-\s •• ^-
192.0
171.0
^117.0
14.4%
24.0
F-27
-------
APPENDIX F
F.7 HFTLC Circuitry
Pre-test measurements for all HF TLC circuits except RNF were subjected to GLM analyses, as
were the deltas after 85/85, TS, and MS. The results of the GLM analyses are given in Tables F. 16 to
F.20. Columns 3 to 5 in those tables give the HF TLC PTH and HF TLC SMT GLM results for 85/85,
TS, and MS, respectively. Note that these latter three analyses are based on changes from Pre-test
measurements. The model R2s for Equations F.I and F,2 for the HF TLC circuitry are summarized as
follows for each test time, except for HF TLC KNF, which gave a constant response.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
50MHz
500MHz
IGHz
RNF
RNR
62.3%
10.7%
13.2%
••
2.7%
6.7%
8.1%
10.9%
^
8.2%
0.0%
0.0%
6.1%
IH
2.4%
14.7%
8.1%
7.9%
mm
6.2%
Surface Finish, and Flux
50MHz
500MHz
IGHz
RNF
RNR
48.1%
2.5%
0.9%
••
3.6%
6.6%
0.9%
2,8%
0.6%
5.0%
1.8%
4.1%
mm
3.5%
9.1%
1.4%
0.7%
•i
2.0%
The model R2 values for HF TLC are all quite small at Pre-test except for those at 50MHz, which
are of moderate size. The .small R2 values indicate that the experimental parameters do not influence
the Pre-test HF TLC measurements. The moderate sized R2 values for the 50MHz ease are examined
in further detail below (repeated from Chapter 4).
The predicted response at Pre-test for HF TLC 50MHz for the base case (HASL at Site 1
processed with LR flux) based on the Site & Flux GLM was -47.43dB. The predicted differences from
the base case are given in Appendix F in Table F.21. The results show that the sites that produced
Ni/Au and Ni/Au/Pd (#13-16) have predicted increases of less than 3dB. While statistically
significant, this change is rather small compared to the base case value and is probably not of practical
utility. Overall, some of the sites differ from the base case by approximately -1.5dB to 2.9dB. These
changes again may not have any practical significance since the important concept is not so much the
magnitude of the response, but rather its stability when subject to environmental stress conditions,
which is the basis for the acceptance criteria.
The predicted response at Pre-test for HF TLC 50MHz for the base case (HASL processed with
LRflux) based on the Surface Finish & Flux GLM was -46.73dB, which is almost identical to that for
the Site & Flux GLM. The predicted differences from the base case are given in Appendix F in Table
F.22. These predictions are consistent with those in Table F.21 and show that immersion Sn and
immersion Ag are approximately 1 .OdB lower than the base case and Ni/Au and Ni/Pd/Au are
approximately 1 to 2 dB higher than the base case. Again, these differences are most likely not of
practical utility. ,
Boxplot Displays of Multiple Comparison Results
HF TLC 50MHz. A boxplot display of the Post MS test results is given in Figure 4.16. Boxplots
for the other three test times are displayed in Figures F.28 to F.30.
HF TLC 500MHz. A boxplot display of the Post MS test results is given in Figure 4.17.
Boxplots for the other three test times are displayed in Figures F.31 to F.33.
F-28
-------
AfFENUJXF
HF TLC IGHz. Boxplots displays for are not given for the HF TLC IGHz test results to
conserve space. The total variation at Pre-test for HF TLC IGHz was only 2dB and there was only
one slight anomaly of-5dB at Post MS, which is not of concern.
HF TLC RNR. A boxplot display of the Post MS test results is given in Figure 4.18. Boxplots
for the other three test times are displayed in Figures F. 34 to F. 3 6.
Table F.16 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 50 MHz Forward
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
SiteS
Site 6
Site 7
Mte o
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
-47.43
X <.x.x*""""
0.98
1.19
1.48
-1.51
V.V. v. *••
0.90
% vfw, ••
-1.40
2.90
2.69
2.05
2.19
,"""7 - v
-1.37
- " i* 5" ~ " - •
V'V „ -•>
XX --x i- % ^
-1.50
62.3%
1.00
85/85
(Delta 1)
0.22
V% * " N-.-. <• .. S
•• s
<,^^ *
-. ' ^ *% X-
""*"°V"-\ Nxv
HMik/
v*. %
^t \
^X 0
-1.17
y« •»^--;
-x vx.x -^Xx^.,^
""-;" -**•• -
0.96
^^ , , „ ^^^
^X»s
^
1.41
"x- •>
6.7%
1.0
Thermal Shock
(Delta 2)
-0.08
^Ss* „ ,,-, "
^ ^*^,:* \
' -^ "V; . -"v
^>XHx^,X^ N-ift
-**H , "V
•'^x*x '^:
% % %x % *"
x- ^xx-^x.^ ' •- % sxxx :
•• " * - o-\
• ^ x%
•-•• J\
-VV ^X <.
"V- v
"• •, ^% -.v.-.
^ %Ss \ %^ x
"^ X"^ s* % "%N %
%A^*v% i^*
s % s ""
^***-
•.•. -. X
\« 'X * '
0.0%
1.01
Mech Shock
(Deltas)
0.04
X
% v^ X -.-v. \ ^
4.40
'" v: ^u
--- "^.^^
--\^^-
Xs ^
«~ s-"
3.20
7.60
"S
-,x. %X^" ^
' xx " X V-
•» ^" *\-^\
- % ^\ i
s -x "XH i
- - H , :
"• x" ""^^x- ^
- ~%V - "
" —--^ "^
14.7%
4.80
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-46.73
"*"; ;•>"---, ^
-0.71
-0.97
2.24
1.19
-0.59
48.1%
1.00
85/85
(Delta 1)
0.09
V J.-.-.-.-^ 5. •. v ^ 1
% •. •• % •• \\^ ., *" :
% -. •- •• •-••x*^
, ^••••"a % ••
C---- «v*
-0.45
s x •••--.
' * s %* « S^ w !
•AV.-.-X •.•X«-K>X.>.W'fr- f. '. w? •.•.i v.:-f. •
0.48
. 6.6%
1.00
Thermal Shock
(Delta 2)
-0.30
v,vV^.r -. , •. *^ ^ ^
" v° s% •• % % ••••> V. ••••
4»\ Xxwv. - % *'-•
••x %'''r:.-.-.^ '•'%'' ^
<... \ - V
"• "• •• v v "•"•''
^ "" V, -. V> % •>• X-.-.
0.45^
5.0%
0.99
Mech Shock
(DeltaS)
0.29
*• -•% - ^%iv% ^™-S%
"• ^ "• ^ x^ x ^^:
^ -...X.-.X. 4 ••••:
5s« x^. \. v. n, „ f, v
•-•. s -. v*- ^ i^-.OXO''^ •-•.%%
4.7
^ s ^'•^^x ^X' ?••••!
" ^ % ^^ ™t X v»*. ws
^ •> v.* \ ••v^.X'- ^% i
^ t ^\^ :
9.1%
4.9
F-29
-------
APPENDIX F
Table F.17 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 500 MHz Forward
GLM from Eq. F.I: Sites and Interactions with Flax
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Sits 11
Site 12
Site 13
Site 14
Site 15
Q«*A 1£
bite lo
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
-17.48
0.64
,^^V--5v^""^ ,
0.45
0.53
^^c
.0.56
i 0¥ J^-^^S" ^v**> "°-.
y- Xyy^JSV.W'^ •• % •• ;
,.. Al" ""-''' ',"
S •> "VX^:
^i&^c^ ^«
KOH", ^
« .&» **c
**ss *
10.7%
0.66
85/85
(Delta 1)
0.06
"w^^V "
!^ ^ 5 ™-- ^ ^X I
*. % :
| *?*; V* "j
;'J,vix^
-1.13
-V ^ '-^ \- ' -
*• ' 1
1.35
' % ' ""•
8.1%
0.62
Thermal Shock
(Delta 2)
-0.23
: % » <•>••
f ~~ "^f^ *• %%:
: % f ';
\ ^^ ^ /|
•V. % ;
\ f :
\'f ^"'" - --- \*M
^ ,',J
0.0%
0.60
Mech Shock
(Delta 3)
-0.14
% . '"*x^%
-1.32
s-^l
- ''''/ f ''
:"" ^ \^
° " 08^"""*
s % * .
y y
..50
" *\-"-X
8.1%
0.93
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-17.41
0.27
wwqgflWC8fl*wwwvw. -vs^.-- ;
f f
•^^•Sssvi*"^* ' % , ^.
^ s^ , v
,-\^> ^ V O^v ,'
s«w^w «• '<•'">•*' r f ;
?5^. *s^
' x ^ sv-^ '" ' s
*. ' i
•• %•-•> •- %%
N*\^w
f f * %
0.23
-~^^ : "
0.9%
0.60
Thermal Shock
(Delta 2)
-0.28
0.20
^% >
" *""-
1.8%
0.59
Mech Shock
(Delta3)
-0.09
j \-s s"% ^S %
% * % %-. ^
-0.22
1.4%
0.96
F-30
-------
APPENDIX F
Table F.18 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 1 GHz Forward
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-14.11
0,11
-0.39
-0.22
Flux
Site 2
SiteS
-0.16
-0.30
0.37
Site 4
Site 5
Site 6
0.21
Site?
SiteS
Site 9
-1.26
Site 10
Shell
Site 12
0.46
-0.51
Site 13
Site 14
Site 15
-0.46
^. f^' sv -."
-0.35
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7* Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-14.16
0.09
pS; ^X'^^VX^r™ -. ^
ff "•• f' •>-f^'f., "~ Xv *" "*"
: "• -.^ %" ••••'--. -.
^d-{ f '.
'-••J.-,^ ;\,* f- ,-
f * vt* •. •. -. f .,
~- ^ •. "**••* ". ~~ *•*• •.•• '••.^
j. -.i-X-^ ^v~>f"-y~.v. •. v
^;< ^x^-^H
0.9%
0.30
85/85
(Delta 1)
0.11
* t ;, , - -" ^
s -A " -"X^ -XX^^-. < f
^ - ^ %TS - <-Xs
:' ^^HV i-
-0.15
^••^rr^J^ \ "*^
: --"- , ""^ i
2.8%
0.30
Thermal Shock
(Delta 2)
-0.38
v. » s s ••
--^ ---- ; -\s™.
-0.33
" <.' •. i * «•
""; *<.> -'- ~ -
%\ ^ , ^*™*^~N.'* v; •• .v.
%^J \ •• ^ ™s:
4.1%
0.52
Mech Shock
(Deltas)
-0.30
0.14
'OTjw.'-S ^ wxjjmjft««5»»~».v
fv; - *^ri^
' w % 5 •• i
%""" x \\ \ ^ •
^ ?c«|5er*~^
0.7%
0.71
F-31
-------
APPENDIX F
Table F.19 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC Rev Null Freq
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16 '
Site 4* Flux
Site 5 * Flux
Site 7* Flux
Site 11 * Flux
Site 13* Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
^ % : "\
*&-V«^~*
Vs* *vs / <
\***SS *VvO°
X •. <: 5 ^^
V5 •• <• ^ •• "• i
\^^^s^\5" ^ .^f50"^
%Cv ijvSyvV; *• ^"W:
vXv ^"*l«v>," \
•• s%
: VV>^ ^"v^ '-,,,
10™ ^;\s x.. ••
^^^i;J^
i rJ\- ^ ^
,-, 5 f o -
>% % ,. - Jl."^
V»*^X^ ^},
• st^ki*"?^"^
•••.•• V>W - s/»*^» \
>;o,; vv -^ ^^v
H^rW?'?^:^:
% •. s ^ *<•>••
K^V\^ ^
» - VN ',
N \
f V ~~S'S'S %•• ••
^S-^XN/C, *^%
^ ^ ^->VA-<. ^%%
^SS^-^W ^ %s w
s^ ^ - •« » h
'vV ** % ' •"«! S '•^ %
^^W»xtX>*^ 5
, ^ > - - &
^a , *^-v •.™-.^i-
^V«8lJ^!^y*»v(«« ^
85/85
(Delta 1)
^
^ "»v % ;/^\"
J*" ^ ^s
<•"*
^ xV""1 ^0%
'•• \
•. *
•• <^
.£
^** « -
f f ^, ""
--^T^ -^ ,v*sv
,- /'
^^*""\, *
^ *%«.
,^^ ^
- * T
f •> % % % s.1-
^*" •• \
* s ^ x "• %
<• % ^
s ^ s
-.^ v> *
.. ., --*. .
Thermal Shock
(Delta 2)
,
_-^^V."/
""<" '^ \- ,-"
"• \ ••
\
- ?\ " "
f f
*°>.% *vJ> A •.
%\J
' ' "x^^
' "'^ W
^X-s '
••X-XN "• -f ^ -v •. •*
V „ V - -
f %% s t
s \
" - x
Mech Shock
(Deltas)
f
f
"• X;*-"- «Ss% %% i
-Nv.,-$^< i
-^--^-t.- i
%x
*
" N"v**''%
% ' > %<
•• \f f f
•, 5 ' *• v
;. % ^vv-.i
^ ' v % ', .
vi ^
^ Vs
••\
f % % % <
^ -- ^ k
s % N
•„ "\
••••.•. v\
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
OW^v^^V^"
vv C
^^Itl^l'o
s, *v>^%% - s;iX
f ^4^^^ v~ ^ « ^
*•' •• ^ x' C ^
85/85
(Delta 1)
<•
f \
%••
F-32
-------
APPENDIX F
Table F.20 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC Rev Null Resp
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-33.90
0.20
-0.05
0.02
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
1.13
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
\ ^^ •- -c
*"""" rF.elT"
-3.50
Site 13
Site 14
Site 15
-3.23
Site 16
Site 4* Flux
Site 5 * Flux
-1.25
,,,, s >, ;>.
^ \fv!;^-
Site 7 * Flux
Site -ll* Flux
Site 13 * Flux
Site 16 * Flux
3.60
Model R2
Standard Deviation
2.7%
1.40
8.2%
1.70
2.4%
2.20
6.2%
3.56
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP .
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-33.70
•• •• v "^ % %% s-l" -1 •-
™ ^^"gg-
,,,--^:;v ^ ^ -
•,•. ••x'" •."•••• %
J-.-.-.-.V ^V"* * v. ••
"• %v t-- ••
_, N \^\ ^ _._. ^
"•"•""••• •. -x <'v^ -.
s ^VwwA v. -.«
3.6%
1.00
85/85
(Delta 1)
0.07
,, is •> % %% •* %
"o?34 ~^
^
'••••.'•'•
s\ s ^ - 1 ••
--" d^\ -' -
- T~",r
v s *• \
0.6%
1.00
Thermal Shock
(Delta 2)
0.03
1 ;f~> v\
""'- *yt^\J^
-1.26
,^""-\"
•. •- ^ \v, '- •(
3.5%
2.1
Mech Shock
(Delta 3)
-0.74
^- "- - i^r,--:^
.-. •• % -,w,v,-,v,w. x-V- '^•••••••M^'X •••<
v, •>;.. \i^-% * i
,% ^ %K-S s>sv % %^%<.^sX-<:
: -{^ ,,-^-U^^
' %\ '• ''"'"' '•'•'<•'• '•?'?•'''* -,^''\ :
' "• * •••• •• -... •. ^\ \
'"i.-os"
2.0%
3.6
F-33
-------
APPENDIX F
Table F.21 Predicted Changes from the Base Case at Pre-test for HF TLC 50MHz for the GLM in
Equation F.I
LR Flux WS Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
0.98
1.19
1.48
-1.51
0.90
-1.40
2.90
2.69
2.05
2°.19
0.98
-0.18
1.48
-1.51
0.90
-1.40
2.90
2.69
2.05
0.69
Table F.22 Predicted Changes from the Base Case at Pre-test for HF TLC 50MHz
for the GLM in Equation F.2
LRFlux WSFlux
OSP .
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
-0.71
-0.97
2.24
1.19
-0.59
-1.30
-1.56
1.65
0.60
F.8 Leakage Measurements
The results of the GLM analyses are given in Tables F.23 to F.26. Columns 3 to 5 in these tables
give the GLM results for 85/85, TS, and MS, respectively. The model R2s for Equations F. 1 and F.2
for the GLM analyses of the leakage measurements are summarized as follows.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish and Flux
10-MilPads
PGA-A
PGA-B
Gull Wing
10-MilPads
PGA-A
PGA-B
Gull Wing
85.6%
88.4%
89.4%
55.4%
74.8%
81.3%
88.7% •
48.2%
22.7%
3.9%
5.6%
3.3%
1.9%
2.0%
5.6%
1.9%
10.8%
9.7%
15.5%
2.8%
3.4%
9.7%
16.0%
2.8%
8.6%
9.0%
12.5%
1.7%
1.7%
6.3%
6.7%
2.6%
It is of interest to note that the model R2 values at Pre-test for all but the Gull Wing are all quite
large. However, these values decrease to close to zero after exposure to the 85/85 environment. These
results are now examined in detail for each of the four leakage circuits.
Tables F.27 and F.28 give the predicted changes from their respective base cases for all leakage
measurements at Pre-test for the GLMs in Equations F. 1 and F.2, respectively.
F-34
-------
APPENDIX F
Table F.23 Significant Coefficients for the Two GLM Analyses by Test Time for 10-Mil Pads
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
12.20
13.29
14.45
14.76
Flux
Site 2
SiteS
0.74
-0.97
1.02
Site 4
SiteS
Site 6
0.93
0.85
Site?
SiteS
Site 9
-1.24
-0.95
-0.84
Site 10
Site 11
Site 12
1.00
0.91
Site 1-3
Site 14
Site 15
-0.89
-0.75
0.98
0.23
0.55
Site 16
Site 4 * Flux
Site 5 * Flux
-0.76
Site 7 * Flux
Site 11 *Flux
Site 13 * Flux
Site 16 * Flux
0.85
1.06
1.95
1.74
Model R2
Standard Deviation
85.6%
0.42
22.7%
0.51 .
10.8%
0.70
8.6%
0.59
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
11.75
0.73
0.33
0.48
^V, ^ •.•a"
"• "• ••
"~ *• •- ^ •**•*• ,. 5-% "•"• -...
-.w-.. ^vVx x •.-.-. - v.-x
1.77
74.8%
0.50
85/85
13.21
-.-••.'••.'••'• . ""••
V*- %-.-^v-^
' ,, •x'4",- ---•. : -
•. "• %.>%>S>%%%1 •• % % ^A-^ *•
0.21
v -.v - m C ''••
1.9%
0.50
Thermal Shock
14.30
W-vVT^/
% •>•••. x"**^^ % ^'
<*" ' •. % * ;
•> % *«%i^
«-*~\s, ' &>^ ,,'V
0.27
3.4%
0.72
Mech Shock
14.69
<••• . ••••:•:•••. "^
* - '"-- "\V v *
^ •..«• ,>•••• --y£^ ^ jjf.
- -\ -*„ ; ^x
••'••.
•• *V.«x i^
•. -i . ^
0.31
-.•.-. •. -.-.•.^ •.-•. •. :•.:•. ;# : v.*.-.;:;^
\V.V. V.-, ^ ^ W, X V. *"•
1.7%
0.61
F-35
-------
APPENDIX F
Table F.24 Significant Coefficients for the Two GLM Analyses by Test Time for PGA-A
GLM from Eq. F.I; Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
11.88
12.50
13.66
13.69
Flux
Site 2
SiteS
0.348
0.22
Site 4
SiteS
Site 6
-0.54
Site?
SiteS
Site 9
-0.81
Site 10
Shell
Site 12
-0.34
Site 13
Site 14
Site 15
-0.64
-0.94
Site 16
Site 4 * Flux
Site 5 * Flux
-1.14
X V. •> V,\ -.
"o.scT
0.63
Site 7 * Flux
Site 11 *Flux
Site 13 * Flux
Site 16 * Flux
o.64
Model R2
Standard Deviation
88.4%
0.40
3.9%
0.71
9.7%
0.52 .
9.0%
0.49
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
11.38
0.35
^ts^iSxS* k-,.,v.-. >,f.
^ ^-oTsT^""
2.05
81.3%
0.5
85/85
12.41
0.25
%^ ••-• st* f ^., '•'•'•v
* '•.••''' "
"-;i; **^-:' -••::;:''::
% N% •- "• %"^ ;-'..
s^,i^,;^-^^v ^
"" %^%x % ^ % X v
S V % w. V
% "X^: ^7
•. *• %1"- -X- %\\^ ff ''~~ ,
0.34
9.7%
0.51
Mech Shock
13.66
s% % % % ^
•. -S s% %'^
^ X-.-^
?.-\\V*
v "" ^.^^ %-
% "- " " ""
;, ">"A^>
*s, ^^\;v
0.256
6.3%
0.49
F-36
-------
APPENDIX F
Table F.25 Significant Coefficients for the Two GLM Analyses by Test Time for PGA-B
GLM from Eq. F.I; Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
- 10.71
12.52
13.69
13.83
Flux
Site 2
Site 3
2.77
0.40
•\
-0.49
Site 4,
SiteS
Site 6
-0.41
-. -, •*.™^;; %%5*ssw<''rt%
™v* Jw r-''L "• Xw.
-0.63
-0.42
Site 7
SiteS
Site 9
*, :'•. v. •. vgSsA.
V37
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
-0.34
-0.61
0.69
Site 7 * Flux
Site 11 *Flux
Site 13 * Flux
Site 16 * Flux
0.72
Model R2 ,
Standard Deviation
89.4%
0.47
8.0%
0.53
15.5%
0.56
12.5%
0.50
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
10.77
y $r,*y -->"
^yi\s.O •>-^-K>^-'-' --v *• "•""
v.-.-. >V \ "" ^ ' :
A-.^ "w. ^% "• %-. j vfff. '. "'.'.'.:• j
-0.38
2.71
88.7%
0.4
85/85
12.55
-0.23
^O •- >v'f v' •.-.
---s-«- ^ -" -
- "-\" > •* ?
•• •• •• ,\%0 •* \ :
- r\s\ ,\™ ' i
""" "-0.40-"'""
?s/- v :•> ','•'• 5-%
-. >^ ^ x •. \ ••
<••*"* "• •? ~* '
5.6%
0.50
Thermal Shock
13.72
-0.33
%^N«% ^.^ %5 ^ s^ ••••.
- -x^-% - "' V%
" ^ - « ! -- - ^
•••.-.;' -.
^ ^v^ •• •••.•. % ^
v ^ •- s ^ ^XN v^
•• ^ «Sv ^ •.„
/\&3ST: - "'
16.0%
0.56
Mech Shock
13.70
-0.21
''•• -."-I " \i-.-*---^, ••*,
^ -> - ,- ,-.1-
i.^0-- ,
% i*>^ Sx*
. ' 0.20
6,7%
0.51
F-37 '
-------
APPENDIX F
Table F.26 Significant Coefficients for the Two GLM Analyses by Test Time for the Gull Wing
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4* Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
Site 13* Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
11.72
0.81
j^^vX^^^v.- ^ -V
1 N. O •. *• •,
' vP -. vv.*.1*^ •* "Xs
A vw^ ^ vXv-V •*#•<• •» '
S&AVW S S -WS 4- V . %
0.37
4. v * '•^_< yf
j^K^SSSVu %*~
£ssl ^ * > " ' },'/
l^^x^"^-- v
0.47
-0.65
0.54
•• ^ v ^ s •:<••:
^ y% % ' j'Vx-
~%*sH\ %-% vi*
N!%% V •. ,
5 *K - ' , - *"\ -
x-".5 •• •. •• •sss-
£4f^>J; -""-^ "
•s^s^s^sX^ , v'
\ *%^x£\ 5'*J.V
0.47
1.61
"v-x' V^ '-"
^•^, ^ % ;':
55.4%
0.54
85/85
12.59
O^^^v*1**^ '"^ '
•^ ?•- ^^ " '
' ' * % %
X\w>* ••
s-x-5-.
L:*«i»-^ » ,
%' N-J V-- •• 0
^s s ^ •- •.
c > ^ •.- ^
.•«*#• •• ' *
f f
^
f f
- ^ " "" "^
% s %% " ^ >•
SS f
*"- %•>
> V % ••
,: - % - -^
-0.64
•• •• % ;-.
-• * f •• -,
%\""-
*^ ' i ss"i
^ f
•• ^ '•-.-^ ^ <
\
- %% * ^ »
s- ! ts ^
^
\ %
^ , VKX«:
1.7%
1.06
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion. Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
11.55
0.30
0.27
(SSJOcTn^^T1^
£$^>..,;sfL.*.p;
I'^&T' *s™4* ••••'' %^
l^to^v""^
iS §^-?% C" \\ . ^-av \i.<'
1.09
48.2%
0.50
85/85
12.62
v? ••> ^. •.-. -. -, %
%s* •>"• * •• ^ '. \.v. :
SjV^^Of ^ V
X Vj^s %A^«« i % V %
0.63
; '. ; -.•., ••••••••••!••• "•• \ • \v \v
1.9%
1.00
Thermal Shock
13.76
-"\s, v CJ: ^- ,
^\,v.* ..-, - , ^
^ %s ^ x •••^ •• ••••%
XVV^^' "^
tx-!..t. ,.\.\ ^.^
•- ^ \\v,
- •. , -\\- , -
•^•xl ..
V •• *
^ '••v'^% .% -• .. %%
-0.37
2.8%
1.10
Mech Shock
13.22
•> "* .. ^ s %
'f- * ox %"
- !,-. , -- v,^
X. - , „ v» X"
J"". * ,
0.46
-,-.-.% -. -. -.-. -.^•. •.-.-. •.•.-.-. -.^
^ S ^*-
x%\^X ^
2.6%
1.0
F-38
-------
APPENDJXJ1
Table F.27 Predicted Changes from the Base Case at Pre-test for the Leakage Measurements for the GLM in
Equation F.I
10-MiIPads PGA-A PGA-B Gull Wing
LRFIux WSFlux LRFlux WSFlux LRFlux WS Flux LRFlux WS Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
-0.97
1.02
0;93
0.85
0.91
•-0.89
-0.75
0.98
•-0.76
-0.23
1.76
1.67
1.59
0.74
1.59
0.74 .
0.74
1.74
1.80
1.65
1.80
-0.01
1.72
1.72
-1.19
-0.81
-0.34
-0.64
-0.94
-1.14
0.39
1.58
1.58
1.58
1.58
1.58
1.58 0.57
0.77
1.58
1.24
1.58
1.85
0.64
1.58
1.78 * -0.34
2.77
2.77
• 2.77
2.77
2.77
2.77
3.34
2.77
2.77
2.77
2.77
2.77
2.77
2.77
2.43
0.81
0.81
0.81
0.37 1.18
0.81
1.28
0.81
0.81
0.47 1.28
-0.65 1.77
0.54 1.35
0.81
0.81
0.81
0.81
Table F.28 Predicted Changes from the Base Case at Pre-test for the Leakage Measurements for the
GLM in Equation F.2
10-MilPads PGA-A PGA-B Gull Wing
LRFlux WSFlux LRFlux WSFlux LRFlux WSFlux LRFlux WSFlux
OSP
ImmSn
1mm Ag
Ni/Au
Ni/Pd/Au
0.73
0.33
0.48
2.50
2.10
2.25
1.77
1.77
0.35
-0.35
2.40
2.05
2.05 '
2.05
1.70
2.71
2.71
2.71
2.71
-0.38 2.33
0.30
0.27
1.39
1.36
1.09
1.09
1.09
10-Mii Pads
Examination of the GLM results in Table F.27 for 10-mil pads shows an effect due to flux of
approximately 0.74 orders of magnitude (see column 1 in uppermost portion of Table F.23). There is
also evidence of site-to-site variation and some interaction between site and flux that affects resistance
either positively or negatively by up to an order of magnitude. Sites applying the OSP surface finish
(Sites 6, 7, 8, and 9) as will as Sites 10 and 11 with immersion Sn do not differ from the base case
when LR flux is used.
Table F.28 shows a flux effect of approximately 1.77 orders of magnitude when sites are dropped
from the GLM and replaced by surface finishes. These results show slight increases in resistance over
the base case for OSP, immersion Sn, and immersion Ag.
The differences in the model R2s for both GLMS essentially disappear after exposure to the 85/85
test environment. This result is not unusual and may be due to a cleansing effect from the 85/85 test
environment that removes residues resulting from board fabrication, assembly, and handling. This
same phenomenon was observed for the other three leakage circuits.
Boxplot Displays of Multiple Comparison Results. Boxplot displays of the Pre-test and Post
85/85 test results are given in Figure 4.19 and 4.20. Boxplots for the other test times are displayed in
Figures F.37 and F.38. There are not great changes in the leakage measurements at Post TS and Post
MS as shown in the boxplots.
• F_39
-------
APPENDIX F
PGA-A
Examination of the GLM results in Table F.27 for PGA-A shows an effect due to flux of
approximately 1.58 orders of magnitude. There is also evidence of site-to-site variation and some
interaction between site and flux that affects resistance either positively on negatively by up to an order
of magnitude. Nine of the sites do not differ from the base case when LR flux is used.
Table F.28 shows a flux effect of approximately 2.05 orders of magnitude when sites are dropped
from the GLM and replaced by surface finishes, but no meaningful differences due to surface finishes.
As was the case with the 10-mil pads, the differences in the model R2s for both GLMS essentially
disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given in Figure 4.21. Boxplots for the other three test times are displayed in Figures F.39 to F.41.
PGA-B
Examination of the GLM results hi Table F.27 for PGA-B shows a strong effect due to flux of
approximately 2.77 orders of magnitude. Thirteen of the sites do not differ from the base case when
LR flux is used and the other two only differ slightly. Table F.28 also shows a strong flux effect of
approximately 2.71 orders of magnitude when sites are dropped from the GLM and replaced by
surface finishes, but no meaningful differences due to surface finishes.
As was the case with the 10-mil pads and PGA-A, the differences in the model R2s for both
GLMS essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given hi Figure 4.22. Boxplots for the other three test times are displayed in Figures F.42 to F.44.
Gull Wing
Examination of the GLM results in Table F.27 for the Gull Wing shows a moderate effect due to
flux of approximately 0.81 orders of magnitude. There is evidence of modest site-to-site variation and
some interaction between site and flux. Eleven of the sites do not differ from the base case when LR
flux is used and the other two only differ slightly. Table F.28 shows a flux effect of approximately
1.09 orders of magnitude when sites are dropped from the GLM and replaced by surface finishes, but
no meaningful differences due to surface finishes.
As was the case with the 10-mil pads, PGA-A, and PGA-B the differences in the model R2s for
both GLMS essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given in Figure 4.23. Boxplots for the other three test times are displayed in Figures F.45 to F.47.
F-40
-------
APPENDIX F
F.9 Stranded Wires
Pre-test measurements for the stranded wire circuits were subjected to GLM analyses, as were the
deltas after 85/85, thermal shock, and mechanical shock. The results of the GLM analyses are given in
Tables F.29 and F.30. Columns 3 to 5 in these tables give the results for 85/85, TS, and MS,
respectively. ^Note that these latter three analyses are based on changes from Pre-test measurements.
The model R2s for Equations F. 1 and F.2 for the stranded wire circuitry are summarized as follows for
each test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish and Flux
St. Wire 1
St. Wire 2
St. Wire 1 .
SL Wire 2
3.6%
8.6%
1.8%
0.8%
6.5%
8.2%
1.6%
0.9%
12.5%
8.2%
4.5%
7.4%
11.7%
4.1%
2.1%
2.2%
The model R2 values are all near zero at each test time, which indicates that the experimental
parameters do not influence the stranded wire voltage measurements.
Boxplot Displays of Multiple Comparison Results. Boxplots displays of the Pre-test voltage
measurements (mV) for both stranded wires are displayed in Figures F.48 and F.49.
F-41
-------
APPENDIX F
Table F.29 Significant Coefficients for the Two GLM Analyses by Test Time for Stranded Wire
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site 7
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4* Flux
Site 5 * Flux
Site 7 * Flux
Site 11* Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
12.90
0.55
\ "• s % 1
I^;'S
& iS&^wiWXs % ^ %XX ..X-
* '% i ' ' '
% •"
.v v *•
^ y&"°V$||JSS$ ^c. & & » %X -
sH^ti"^"^^^ *$• i
i^ ^ •. *•
".^^ %^*.^
5 v5 - ^ X
%"' 5 ••
^^ X ^
"^ ^ % ^ ^
^^*XJ --^V-"\\:
3.6%
2.57
85/85
(Delta 1)
0.000
V- * :
*^ ^ x- •• •
-0.001
-0.001
1 ,.. '*^ --1
'• '.^ ^ ^ •. •. .
' - ' '
f ^^''X0' i
"d.ooV
K---; I
i i
6.5%
0.002
Thermal Shock
(Delta 2)
0.001
;);,
x*i
"X" '"'^ ^1
" % - - :
"•"• %
\, ^ - ' j
" 0.024
%"i
, -S - sws j
'•'• ,x •. ,% i
--X v.% ws_ % j
12.5%
0.014
Mech Shock
(DeltaS)
0.005
: "*s %'=' 1
h- ,_s ^
: ^ % % %
- ;- -«,
_,
s ^
6.042"
x^' " •
, ,, -
-5 ;
0.079
11.7%
0.041
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
12.94
~ « « '^ ' t^ ^*~x^
J sJ^X^" %X\i^ :
":L06^^ '
J ^»; ',-x¥|^"!
$• -,?• X.-5 x 5¥f>iiO ••:
siXjs V '•'So wt^iW tC:
^^•"^^
1.8%
2.00
85/85
(Delta 1)
0.000
-0.001
^sw.-.w^sw.^p^.s,^ •,•.-.•.-.
i,%%\\ •> '•'' s<\\^f
'$fa&& X i' ^%ss %
\ <, •*>.% vi% •-
: ^ ::* ,, \ "• v. "•
•.-.•.'•••'• •.•• % . •• ^ f
^ -. 5 * % ?§•' "• ff '
^^ ^^/>
7>V3^ ^.
1.6%
0.001
Thermal Shock
(Delta 2)
0.001
: ff •.'••.'' ""•? -"\% ••
'• s ' *% f f \ *•
"oloio
i "•* ^ * •• , *' %
•". ** *•
4.5%
•0.014
Mech Shock
(Delta 3)
0.006
"• "•"*•.•. v>'"'
f •. "" A"" \
0.019
s % L -.N
s %\ •"> f<
\ t^-^f
•• . ••<••••• Vj. •&•
: s V •> % % -.^
^;**
^ ^* <-*^ "
2.1%
0.043
F-42
-------
APPENDIX F
Table F.30 Significant Coefficients for the Two GLM Analyses by Test Time for Stranded Wire 2
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
23.44
-.000
0.011
0.033
Flux
Site 2
Site 3
''v.X' VV.%
?X% % f ff : v
0.003
Site 4
Site 5
Site 6
Site?
SiteS
Site 9
>'
"^•••5.
Site 10
Site 11
Site 12
-1.56
c*c
0.077
Site 13
Site 14
Site 15
, ";-" 5^ ,:
>. VI-. SA-X;\% %
Site 16
Site 4* Flux'
Site 5 * Flux
X **"< sv ¥ » ••
t "**•**#•• "•
^-"%i,:
— _2-Jj-
Site 7 * Flux
Site 11 *Flux
Site 13 * Flux
Site 16 * Flux
-0.002
0.074
0.130
Model Rz
Standard Deviation
8.6%
1.90
8.2%
0.003
8.2%
0.067
4.1%
0.098
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
. 23.34
^-0.43
•• -,-. VC-vivs,
"* * Vk ^
w ^%0 %vv -=--x-x.
v % f ~--~ ^
V, MftHSfv+f, ^\tsX"
%%^--%%rs^ -% •-
^•X^v. f * •*••-.
._'' -.-. \^~ "*•*> j-y-sss
0.8%
2.00
85/85
(Delta 1)
0.000
i. •^'•••V '•^'•'^ •* •• :
-b.ooi '
•.
% ••-•<%
" s ^^ X vww. s
''P •. % V
0.9%
0.002
Thermal Shock
(Delta 2)
-0.001
•- * % %v j.^
•A . . '-••\ ••
\ % ^ > X'"'"'^-.-' ^
"• •• ^ f •& -.^ s -. v"- ^
0.038
-5 , ^ - \^ v v « vA
vv ^^ ••„, %
% 0.026
7.4%
0.067
Mech Shock
(Delta 3)^
0.021
7" -" " " ^- "''•
0%v». S^ 'w.vl
% \ :
% ^ :
'"• " -." " C* ^v
%\ v««»^ _
0.029 %
2.2%
0.099
F-43
-------
APPENDIX F
Pre-Test
HVLCPTH
Q.
Q
5.03 —
5-02 •
5.01 -
5.00-
SiteFlux
HASL
Boxplots of HVLC PTH by SiteFIux
(means are indicated by so lid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
1
1
t
j
i
1 1 li1
1 if m l
if i
i
i
i
i
* *
* *
1 1
S S p*
I''
1
A ftl
* fy & s
I an
1 T
i
:|:
|
i
;i;:
|
;:;
•;•:
t
{
1
1>
1 *
1 i
1
T
rfe
1
§
1
|
;i
1
T
s t
i i
1 $ i
\ 1 1
his
i1 A 1 lM
1' i MM i
t|j i s s i 1 ;
!| T 1 1| i i
i f [r
! J i
S i
i $
i $
1 5
{ i
I I
co o>
WS WS WS WS
WS
WS
WSWS
WS
WS
CM
WS
Figure F.I Boxplot Displays for HVLC PTH Measurements QjA) at Pre-test by Surface Finish
(Acceptance Criterion =4jiA o T-
WS WS WS WS
WS
WS WSWS WS WS
WS
Figure F.2 Boxplot Displays for HVLC PTH Post 85/85 - Pre-test Measurements (|iA) by Surface Finish
(Acceptance Criterion = 4^A< X <6(oA)
F-44
-------
APPENDIX F
Post Thermal Shock
HVLC PTH
o
5.01 —
5.00 —
4.99 —
4.98 —
SiteFlux
HASL
Boxplots of DTHVLC P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
i
. 1
>
I
T- ex
l
c
1
|:j
f
T
i
•q
!J
i :
I i
< '>
! :
i
; ;
r:
i
i
$
i
p tr
I
;
<£
it
* PI
t s
: • 8
i 1
i 1
: S
r . T
i i r
> 1^- CO
o
j i
;
i r
f I
) C
I
I
1
1
1
*!i
! ! 1
i ;?
.:"
i
i
i
i
t
ii
) T- CM C
J CM CM f
1 •
)
J
ws
ws
ws ws
ws
ws wsws
ws
ws ws
Figure F.3 Boxplot Displays for HVLC PTH Post TS - Pre-test Measurements (^A) by Surface Finish
(Acceptance Criterion = 4jaA< X <6|iA)
Post Mechanical Shock
HVLC PTH
Q.
O
5.01 -
5.00 —
4.99 —
4.98 —
SiteFlux
HASL
Boxplots of DMHVLC P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
*
*
•
\
i
1 1
— i—
'r
i|
* :
! f
L V
8
:
J SOL
| i
i i
: i
: ' t
f: t
|.
|
1
1 .
;
«
J
1 ' I
.
:
i
i
i
:
i
I
V
1
I
1
j
:
f
, f
i 1
i i
: :
! !
— r-
1 *!
i
i
»
>
s «
: «. 2
7 9 J
S
i
i
i i
L
i
1 !
*f 1
1 I i
i ! !
; i i
• 1 i
' 1 '
(
I i
1 !
;
;
f ! I
1 I I
i i I
5 i I
i 11
'
•I i
1
ii
i
j
g
L
i
•
5
I
|
I
i
|
f f 1 'f
1 1 1
(
l
|
1 1
.
*— CMCO^TtnCDt^COCnOT— CM CO'TinCOf^-COCDOT— CMCO
ws ws ws ws ws ws wsws ws ws ws
Figure F.4 Boxplot Displays for HVLC PTH Post MS - Pre-test Measurements (uA) by Surface Finish
(Acceptance Criterion = 4uA< X <6(iA)
F-45
-------
APPENDIX F
Pre-Test
HVLCSMT
Boxplots of HVLC SMT by SiteFlux
(means are indicated bysolid circles)
HASL OSP
5.4 —
5.3-
5.2-
t
CO
0 M
>
^ 5.0-
4.9 —
4.8 —
i
I
l
I
: f
i 1 i
j * J *
S| 1
1 R] || ,
g] I y H H g
§ ig T B T
1 )€
\
l
\
1 1 1 I 1 1 1 1 1
ImmSn ImmAg Ni/Au Ni/Au/Pd
1
1
»
s
s
£
1
|
1
T
1
1
SitaFllIX T-tMcoTtncot— cocnoi-
r « 1
1
m
•;•.
?
|
i
:|:
>:;
i:i
•* ;•;
S
1
! ! !
5
1
J
I
i
1
i
|
i
|
li
Ij
i i
ill i
p j i i
i i •*• t.
i
i
i
;
h
\\
i i
'IB 1 l 1 1 ± i
i § U i Y
i T i
• IT ! i
i ! I
! f - I
1 1 1 1 1 1 1 1 1 1 1 1
cMcoTtncor-^ COOOT-CMCO
WS WS WS WS
ws
ws wsws ws ws
ws
Figure F.5 Boxplot Displays for HVLC SMT Measurements (pA) at Pre-test by Surface Finish
(Acceptance Criterion = 4}iA< X <6jaA) .
Post 85/85
HVLC SMT
Boxplots of DPHN/LC S by SiteFlux
(means are indicated bysolid circles)
5.4 —
5.3-
fc 5-2~
CO
Q 5.1 -
Q 5.0 —
4.9-
4.8 —
SiteFlux
3
*
I
HASL OSP
• •
1
1
1
5
i
\ J
1 ;
K i % i
1 1 | x A I t
LJ n 0 55 0 T s
T ta f ' y T
i
1
J
1 1 1 1 1 1 1 1
O
f
I
\
I
I
I
I I
'• "
i!
!i 1
\ f '
I
\
I
]
J
T- (NJ CO
CM CM CM
WS WS WS WS
ws
ws wsws ws ws
ws
Figure F.6 Boxplot Displays for HVLC PTH Post 85/85 - Pre-test Measurements (jxA) by Surface Finish
(Acceptance Criterion = 4jaA< X <6|jA)
-— : :
-------
APPENDIX F
Post Thermal Shock
HVLCSMT
•HASL
Boxplots of DTHVLC S by SiteFiux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
CO
o
SiteFiux
5.5-
5.4 —
5.3-
5.2 —
5.1 —
5.O.—
4.9 —
4.8 —
: s : 1 : 1
i * i
I 1
i I
1 '
1 i
.;. J
;•: , i
r F
* * \ *•
y i
}
1
TwTiraTTraT T
1 i
1 1
,
i
•
i
1
>;
i
|
':
:
:
i
i
i
|
i
i
i
i
T 1 i
1 j
[
i
0
1 I
* I \
1 1 1 1 1 I I I
i n
i 1 1
i
JX T-ojco^m'coi^coojOT-csicOTrtocDt
1
i
|:|
•:•
T
i
s»
-J
I
$
I
i
i
1
;
i
I
i
l W
i '
i
{
f
1 1
CO O>
i
£
Z
1
3
WS WS WS . WS WS WS WSWS WS
f
1
j
li
j-«
: •
: !
ii i
| T T
J
i
j
!
1 1
T- C*J CO
ws ws
Figure F.7 Boxplot Displays for HVLC PTH Post TS - Pre-test Measurements (^A) by Surface Finish
(Acceptance Criterion = 4|iA< X <6pA)
Post Mec
HVLC SM
0.05 —
0.04 —
| 0.03-
O
X 0.02 —
Q
0.01 —
0.00 —
SiteFiux
hamcal Shock Boxplots of DMHVLC S by SiteFiux
(means are indicated bysolid circles)
HASL OSP.1 ImmSn ImmAg Ni/Au Ni/Au/Pd
» » •-»• — 1
i S l *i
1 - ' ' I ! }
! i 1 I
I II I
i i ! 1
I i 1 I
i II j
i i i I
1 ' _ 'I 1 I
i K \ I i
I i 1 I '
I II I
1 K K 5,5 i *
1 ' A ' ' .•' :
1 » ' 1 l l
& ® » ®|«» «• S «• S «. _ 43. .B » | « '@ •»!«'« « ~f~ *
1 1 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 1 i 1 1 1 1
T-CNlCO^tnCDI^COOTOT-CMCO^lOCDI^-OOOTOT-CMCO
WS WS WS WS WS WS WSWS WS WS WS
Figure F.8 Boxplot Displays for HVLC PTH Post MS - Pre-test Measurements by Surface Finish
(Acceptance Criterion = 4pA< X <6}iA)
F-47
-------
APPENDIX F
Pre-Test
HSDPTH
18.0 —
17.5-
17.0 —
SiteFlux
HASL
I
CM
Boxplots of HSD PTH by SiteFlux
(means are indicated bysolid circles)
OSP Imm Sn Imm Ag
i
CO
r
CO
Ni/Au Ni/Au/Pd
I
O>
O T- CM CO
i
CO
i
CD
ws ws ws ws
WS'
WS WSWS WS WS
ws
Figure F.9 Boxplot Displays for HSD PTH Measurements (nsec) at Pre-test by Surface Finish
Pre-Test
HSD SMT
9.5 —
9.4-
9.3-
9.1 -
9.0 —
8.9 —
SiteFlux
HASL
Boxplots of HSD SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
T-
I
o
1
j
i r
i
> ••a
— I—
j
1
j
j
i
«
j
I
I
I
j
IT
L
cc
L
1
:
.
) l>
[
:
r
- cc
1
:
i
•
I
\
f
c
1
1
X
1
D
i
1
O f
II
i T
|
3
1
CM C'
1
;
> *a
— i —
I
1
P
1 1 f.
: , ••
'• \ '•
': *
i i
M
i
i
\.
\
r ir
L i
-
i
r
tc
J —
!
i
'; 1 *
1 1
\
\
ll
i
i
1
1
h- cc
L|
:
i
'•
\
•
3
\ 1
) a
) C
||
i
3 T-
—s —
1
1
|i[
I
!
{
i
i
i 1
i
i
i
c
!
•
^
'•
I
I
1
J C
1
i
'•
r
5
WS WS WS WS
ws
ws wsws ws ws ws
Figure F.10 Boxplot Displays for HSD SMT Measurements (nsec) at Pre-test by Surface Finish
F-48
-------
APPENDIX F
Post 85/85
HF PTH 50MHz
HASL
Q.
LU
D.
Q
0.5 —
0.4 —
0.3 —
0.2 —
0.1 —
0.0 —
-0.1 —
-0.2 —
-0.3 —I
Boxplots of DPHF PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
S'rteFIux
Ni/Au Ni/Au/Pd
I . , , J_
5 I !
i i !
1 I I
1 . II
i II
' f IS
«. IS
i I 1
i II
i * , 1 1
E3 K3 f™t A 1 •&! r— . 4gj| ga _ Eg] -*- ra gL, E3 i A A E3 i
jg* » gg] gg "S tjjjl M ^r* S ™* ^ ^ ^ *** *¥•
)t H ™^*| i
i T * ' i *" ' I
: "•"' . -» ' '
I If
* • 1 If
i if
i i i n n ~i i i TI i i i TI i — r^
L !
! '
i • 1
I
j
I
1
j
1
t i i j^ .
(fa pi ra *» ra
* I
1
1
1
1
1 1 II 1 " '
CO G>
ws ws ws ws
ws
ws ws ws
ws
O •<- CM CO
CM CM CM CM
ws
ws
Figure F.ll Boxplot Displays for HF PTH 50MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = ±5dB of Pre-test)
Post Thermal Shock
HF PTH 50MHz
Boxplots of DTHF PTH by SiteFlux
(mea,ns are indicated bysolid circles)
0.5 —
O
£
t °-°-
u_
{ —
Q
-0.5 —
SiteFlux
HASL . OSP ImmSn ImmAg
i , j ^ j —
l • • l J
l 1 I
! IS
i 5 5
i 11
i i i
1 IS
^.$4>A|i* i|@ffl|$fi|ifi^<
j| T f j 1 j
i if
i * - if
i i i
s si
! Si
i Si
! i J
1 1 1 1 ! 1 1 1 1 1 1 1 1 1 i 1 1
i~ CM ' CO ^" IO CD t^ CO' CD O *r- CM CO ^ IO CD f*^ 0
WS WS \NS \NS WS WS WSWS
Ni/Au Ni/Au/Pd
. ,
! j
! J
i - i
!
i
^ & | §5'| ®
- .1
1
i
1
S
i
1
* j
) O O T- CM CO
WS WS WS
Figure F.12 Boxplot Displays for HF PTH 50MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = +5dB of Pre-test)
F-49
-------
APPENDIX F
Pre-Test
HF PTH f(-3dB)
HASL
Boxplots of HF PTH-3 by SiteFlux
(means are indicated bysolid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
285-
280 —
275-
SiteFlux
^1
1
a
1
r
c>
1
i
r
1 1
i f
:
I
\
r
i
i •<
i
i
5
y.
1
T
T
«•
T"
I
i
fr
I
*
! i
it,
ll
I
1
I
I
1
1
1
1
1
\
IT
ll
PI
i ty
„
I 1
co r-
u
1
i
1
f
1
c
1
g:
T
X
T
o
c
a
1
i;!
1
I)
pi
1 i
O T-
1
a .
i
i
i i
Cv
'
1 i
CO 'q
— I 1
f !
L '
i j is . pj j r|
I !'l
i I T
I f
! i
i i
1 i
i !
1 {
1 II
r tn co r- co o)
1
1
I .
f 'ii v
f I' T 1
Ii
1
1
1
1
i
1
1 11 i
a s a a
WS WS WS WS
WS
WS
WSWS
WS
WS WS
Figure F.13 Boxplot Displays for HF PTH f(-3dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
Post 85/85
HF PTH f(-3dB)
Boxplots of DPHF PTH by SiteFlux
(means are indicated bysolid circles)
0 —
a.
LL
Q -5 ••—
-10 —
SiteFlux
HASL OSP
1 *
i
f ^
i ^ ®
J *
I
1
j
1
i
?
i i i i i i i i i
WS WS WS WS
Imm Sn Imm Ag Ni/Au Ni/Au/Pd
IS i
! i i
1 ! ^
i ! !
* ' i i
i i i
i i !
i i !
! i !
1 i !
! I I '
* . 1 5 i
i 5 i
1 1 1 1 1 I 1 1 1 1 1 1 1 1
WS WS WSWS WS WS WS
Figure F.14 Boxplot Displays for HF PTH f(-3dB) Post 85/85 - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
F-50
-------
APPENDIX F
Post TTiermal Shock
HF PTH f(-3dB)
HASL
5 —
0 —
D.
U.
-10-
SrteFlux
Boxplots of DTHF PTH by SiteFlux
(means are indicated by so lid circles)
OSP Imm Sn imm Ag
Ni/Au Ni/Au/Pd
1 * .
r
j
i
i i ' f rt
mm® * irt i j_ Pi r
flTfli^ilF
* *
i
.1
i
s . " i
i [
»
i
i
^n i i i~\ i i i
T-CMCOTtnCDr*-(oo3c
ws ws wsws ws
1 —
I
i
1
{ *
?$ p]
j 63
I-
1
i
1
i
I
I
J
I
\ 1 ? i
T T- CM
M CN CN
WS
[$)
Ip
1
to
04
ws
Figure F.15 Boxplot Displays for HF PTH f(-3dB) Post TS - Pre-test Measurements (Mhz) by Surface Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
Pre-Test
HF PTH f(-40dB)
HASL
Boxplots of HFPTH-40 by SiteFlux
(means are indicated by solid circles)
OSP ImmSn 1mm Ag
Ni/Au Ni/Au/Pd
WS
ws
ws
ws
ws
ws wsws
ws
485 —
475 —
D
a-
C
L
L
465 —
455 —
SiteFlux
i
j
|
i
C
I
1
T
i
M
r
i
-; \
r
M
f
i
I
. I .
I? i
r
i
i
i
i
i »
i
i
tn co
I
1
t
f
JL
|
1
j
1
1
T
c
A
|
i
|
I
I
T
0
CT
1
;i
!
!
i
; •
I
r t;
O T-
1
I
I
i
c\
u
r
i
1
lit- p
1 i IP 1
I i I $
I I
i T
r * 1
i
i
I
5
) ^T in CD
1
1
1
~T
i
1
$
i
i
I
]
i
I
I
I
I
I
(
1 "
I 1
1 1 II 1
111
i i f
1
T — i — i — r
O G3 O T-
j
i
I
1 §
i
I
I
I
I
5
i
CM C*
I
•
;
i
>
ws
ws
Figure F.16 Boxplot Displays for HF PTH f(-40dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
F-51
-------
APPENDIX F
Post 85/85
HFPTHf(-40dB)
HASL
Ou
Q
5-
0 —
-5 —
-10 —
SiteFIux
i
CM
i
CO
Boxplots of DPHFPTH- by SiteFIux
(means are'indicated bysolid circles)
OSP ImmSn . ImmAg
Ni/Au Ni/Au/Pd
i
*
I
I
I
I
! *
Iff
!
I
I
I
I
I
\
I
I
j
II I I
in
-------
APPENDIXF
Pre-Test
HFSMT 50MHz
HASL
Boxplots of HF SMT50 by SiteFlux
(means are indicated by so lid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
-0.6 —
-0.7 -
o
to
'|g>
CO
U.
X -0.8 —
-0.9-
SiteFlux
v*
\ J
!•
•
$ ?
:
• ;
1
c\
v\
1
I
L § §!
1 1 5
* i 1
' I
j
I
1
i
!
i
1 1
CO T If
IS WS
1
II 1
i i *
•;• $
* T
f
•
1 1 1 i r~l 1
CO t~- 05 CB O r-
WS WS WS
1 1 —
i !
• A-
, A M |.|j
1 1 ! i 1 1 Is
1 1 £ T
1 1 1 i ' ' 1
! i
i !
1 I
! ;
i !
i f
n i TM i rn
CM co T in to i^ «
WS WSWS
i
1
: {
i |!fl 1
1 1 til f 1 1
raj 3; ra f |j
3 Ip ij; pi
i
i i
i
i
i
j
i i i~i — i —
). O5 O T- c*l CO
WS WS WS
Figure F.19 Boxplot Displays for HF SMT 50MHz Measurements (dB) at Pre-test by Surface Finish
Post 85/85
HFSMT 50MHz
HASL
Boxplots of DPHF SMT by SiteFlux
(means are indicated by solid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
o
LO
CO
n
I
Q.
Q
S'te
0.1 —
0.0 —
-0.1 —
-0.2 —
-0.3 —
' -0.4 —
i Flux
l i ; 1 j j 1
I * I 1 j
i *l i I
ii*Ai«ii**'lA8-?i* '»'• A a a!n
IT *" i if P® I^J^L
i i i
*« !
i i
j * CM* ^r3 sJ * 133
: f I
' : :.
1 j i *
: §
' i • $ i r
! Ill
i if!
i r i i
i ill
i i i i.
i K i i i
{ I I j
I I i T~^ I 1 I T~l I 1 1 111 T^ 1 1 1 I
i
CO
CM
WS WS
WS WS
WS
WS
WSWS
WS
WS
WS
Figure F.20 Boxplot Displays for HF SMT 50MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = ±5 dB of Pre-test)
F-53
-------
APPENDIX F
Post Thermal Shock
HF SMT 50MHz
HASL
0.1 -
0.0-
-0.1 -
0
|2 -0.2 —
i
to
Uj -0.3 -
Q
u -0.4 —
-0.5 —
-0.6-
I
i 1 f
T
Boxplots of DTHF SMT by SiteFlux
(means are indicated bysolid circles)
I
*
I I I I
SiteFlux T- eg « -a-
OSP
5 Fl A
\m | 1 I 1
{
i _
{ *
f
: I
i
r
i
i
i
i
\
Imm Sn Imm Ag
1 r~
JL{ }
?S i
I
i i
! i
x i i
! J
i 1
i !
I I
I I
I \
i I
I I
i i
1
!
|
I
1 1 1 1 1 1 1 1 1 I 1 I
mOT-cjcOTint0N.co
WS WS WS WS
WS WS WSWS
Ni/Au Ni/Au/Pd
. j
1
^ | 1 1 I i
f f 1
i
*i
|
i
1
I
1
I
!
;
I
I
i i I i I
en o i- CN co
WS WS WS
Figure F.22 Boxplot Displays for HF SMT f(-3dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
F-54
-------
APPENDIX F
Post 85/85
HF SMTf(-3dB)
HASL
Boxplots of DPHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
DPHF SMT-3
en o oi c
1 1 1 1
SiteFlux
» i f 1 5 —
5 1*11
1 . ill
* * 1 i i
I \ I i
i i s 1 I
! ' 1 $ 1 • 1 *
! -I |T. !
i r i i i i i i i i i i i i i i i i i j | ' | -\ — '
•"-CMCOTinajr^eoOJOT-CMO-'l-tOCDI^COCDO-r-CMCO
ws ws ws ws
ws
ws wsws ws ws
ws
Figure F.23 Boxplot Displays for HF SMT f(-3dB) Post 85/85 - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
Post Thermal Shock
HFSMTf(-3dB)
HASL
Boxplots of DTHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn . ImmAg
Ni/Au Ni/Au/Pd
10 —
0 —
o
1
CO
£ -10-
X
Q
-20 —
-30 —
SiteFlux
: t : 1 —
j
{
* • 1
1
?«"•»««» 1® dl * [3] p
' • B f
* * i
! :
i i
i i
5 ?
1 ;.
* •
t ¥ •
f ** .
. • * :
i
!
1 1 1 1 ~l 1 1 i T"^
WS WS WS WS
I I . • «
i * i ^ !
5 1 1 ' I
1 $ | ! *
cja y g ca!^ 19 @S| @ [j} &l f& g
B * " B ' ' ' * '
1 * r i | 8 | *
1 1 i
i i :
i i i
i i j
I ! i
i ! i
1 1 !
» 1 I !
i ! - I
1 j , j
i i i — r^ — i — r~^ — i — i — iii — '
ws ws wsws ws ws ws
Figure F.24 Boxplot Displays for HF SMT f(-3dB) Post TS - Pre-test Measurements (MHz) by Surface Finish
(Acceptance Criterion = +50Mhz of Pre-test)
F-55
-------
APPENDIX F
Pre-Test
HFSMTf(-40dB)
HASL
Boxplots of HFSMT-40 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
950-
900 —
a
5
850 —
800 —
SHeFlux
{
i
1 i -
i
1 1 i 1 1 i
f 1 1 t'll ill'8
1
1
t
1 1 1 1 1 1 1 1 1 1
•r-cMcOTin«jt-.coo)0
' WS WS WS WS
I
* I
I
1
i
f
* !
i, 1 i 1
[ i *
i i
t
i i i i i i i
•*- CM (0 TT 10 (0 f~
WS WS WSWS
I
1
!
i
i
I
I
I
$
I
j
5
f
a
^
1
j
I
:•:
i
) O>
WJ
I
n
c
t
1
i
1
*f
f
{
1 f
j
]ljll
1
i
1 t 1
3 T- CM C«>
WS WS
Figure F.25 Boxplot Displays for HF SMT f(-40dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
Post 85/85
HFSMTf(-40dB)
HASL
Boxplots of DPHFSMT- by SiteFlux
(means are indicated bysolid circles) •
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
DPHFSMT-40
50-
40 —
30 —
20 —
10 —
0 —
-10 —
SiteFlux
f i * ! 1
i i 1 \
} I 1 i
i * „, I 1 i
i i I * i
: i $ i
5 * i i I
i i i i
i III
I * , i 1 i
1 1 l 11 l l J,1
'IT 1 ' ' 1 1 i T
I i ! i
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 !
T-CMCOTU3
-------
APPENDIX F
Post Thermal Shock
HFSMTf(-4QdB)
Boxplots of DTHFSMT- by SiteFIux
(means are indicated bysolid circles)
HASL OSP Imm Sn Imm Ag Ni/Au
40 —
30-
20 —
10-
o
T o-
,H> A n
U_
JE -20 —
Q
-30 —
-40 —
-50 —
-60 —
< i I
r s * i
I i • i
* II
* * i i
* i I
i II
(il Jal rl 111 ¥*[ iH n ^ HI * * ' KM r$i ^ £ PI
S rflf 1 fl i ! r
ill' £ 1 I Ifi
i • 1 1T
i I 1
I S !
* E-3 »
!_
1
! '
L ^ii f r f
I il ' i
; i^ f
r i '
: !
: *
\ I
* {
! if
i ( ;
i i i i i i i i i i n i i — i — i
1 n r
SiteFIux ^— O4COTmcot^-co o> OT-oJco^tocoh-cooo'T—
WS WS WS WS WS WS WSWS WS WS
Ni/Au/Pd
T
{
1
I
I
! *
j
! !*!•
1 pi a
* si
i
f
c
*
1
1
i
^~J 1 —
OJ CO
Oi CN
WS
Figure F.27 Boxplot Displays for HF SMT f(-40dB) Post TS - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = ±50Mhz of Pre-test)
Pre-Test
HF TLC 50MHz
HASL
Boxplots of HF TL 50 by SiteFIux
(means are indicated bysolid circles)
OSP. ImmSn ImmAg
Ni/Au Ni/Au/Pd
-42 —
-43 —
-44 —
-45 —
g -46-
I-. ^7
I
-48 -
-49-
-50 —
-51 —
SiteFIux
I 'I
1 |
1 IK] f!\ V
1 * 1
1 •
T
1
1 1 1
^ M CO ^"
1
i
r
I j
i I
! @ JL ¥ A
i | 1
i
! 1
1
1
i 1 1 1 I
.
J, E
. 1 *
g 1
H
]
"I i 1 1
;
i
i
i
li
i Pi '
fi
) i
i
j
j
i
!
1
Tl
I I
. ! f^ I i
! ^ ! 1
L -If HI •
j .5 j 1
i if
If i
Tl j
! i
i j
i i
i T~^ i i T^ 1
CDr^COCno'r-'CMCO
ws ws
ws
ws
ws
ws
WSWS
WS
WS WS
Tigure F.28 Boxplot Displays for HF TLC 50MHz Measurements (<1B) at Pre-test by Surface Finish
F-57
-------
APPENDIX F
Post 85/85
HF TLC 50MHz
t
Q-
P
SiteFlux
HASL
Boxplots of DPHF TL by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5 —
4 —
3 —
2 —
1 _
Q ^
-1 -
-2
-3 —
-4 —
r
h
i X !
5 I
i !
f f
* i i
1 . * i
I ' ' ' A I '
i R I *8 ^ A A Pi & $ 5 A
1 Ea? ^ Pi ^ T ra ^ Eltpa
J ™ | ®S "|lp
i • * i
i :
: i
* * * * ' • • i
1 * »
i !
i
I
3
£
f i
i i
i !
1 f 1
I *i 1
1 5 1
1 i ||
1 1 i |
n t @ ^1 1 i ^
a ! pi T i |j ! fb |3
i | 1 i w
I 1 i
I 1 J
l| . i
if * 5
i 1 *l
f J *
l
CO
\
CM
ws ws ws ws
ws
ws wsws ws ws
ws
Figure F.29 Boxplot Displays for HF TLC 50MHz Post 85/85 - Pre-test Measurements ((IB) by Surf. Finish
(Acceptance Criterion = ±5 dB of Pre-test)
Post Thermal Shock
TLC 50MHz
HASL
Boxplots of DTHF TL by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn . ImmAg
Ni/Au Ni/Au/Pd
DTHFTL50
5 —
4-
3-
2-
1 -
0-
-1 —
-2-
•3-
-4-
-5 —
SiteRux
< 1 I J «
x i i i i
i f ! !
I I i i i
* i S 1 . *i 1
i * 51 t i
I 1 M I f ^
5 * 1 . I 1 if
1 I i I * !
1 ' K I-* i 1 1
1 J ! f 1
i 5 I | 5 *
? 5 r • j
i i i i i i i i i i i i i i i i i i i i i i i
ws ws ws ws ws ws wsws ws ws ws
Figure F.30 Boxplot Displays for HF TLC 50MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = ±5 dB of Pre-test)
F-58
-------
APPENDIX F
Pre-Test
HF TLC 500MHz
HASL
Boxplots of HF TL500 by SiteFlux
(means are indicated bysolid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
-15-
-16-
0 -17 r-
o
LO
_J
1-
LL. 1R
•j- -18 —
-19-
-20 —
SiteFlux
t"
5
i
i
i '• 'A' A
i p] ia I ^ i' il T Pf T
: |j "P I H 13
; T ' I T 1 i
i «l h
J
I
I
I
i
1
i i i n i i i r~^
CM CO T lO CD h- CO ' O> C
ws . ws ws ws
^
ra
i ra
-
1 1
T- cv
ws
I. ,
j
•
tf
)
•q
w
— ! 1
1 I
i I
-.1 1
\{ ' }
ill , 1 1
MI hi
* I m @ ra i |
} T p ra | l|
I T i
i 1 !
I !
E j {
i i
i 1
t i
^n i TI —
in CO !*•» 0
'S WS WS • WS WS WS
i
i
* i
* i
*!
J
, j
1 1 B i|
i i
j
" *I
* I
\ I
•
*
i
i
II 1
3 O5 O T-
ws ws
K
i ®
1 1
CM CO
CM CM
WS
Figure F.32 Boxplot Displays for HF TLC 500MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = ±5 dB of Pre-test)
F-59
-------
APPENDIX F
Post Thermal Shock D . . , „-,-,,,- -...,-, «., ,_,
HF TLC 500MHz Boxplots °f DTHF TL5 by SlteFlux
(means are indicated bysolid circles)
HASL OSP ImmSn ImmAg
2 ^
1 —
o o —
1
i -1-
Q
-2 —
•3 —
i i n
J * I
i * I 1
I * ' 1 1
1 * [:•: *
Fj 1 { • • 1 i
y i i \ l i
B • i s T T I \ v
H i
ll ! *
ll i
if !
I I
l l
\ * !
i i i i i i i i i i 1 1 I 1 l
SiteFlux ••-cscOTiocor^toooT-cMco-rincor-
T
}
I
!
!
1
i
I
i
i
I
1
{
(
WS WS WS WS WS WS WSWS
Ni/Au Ni/Au/Pd
1
I
* f
i
"l *
i
1 • !
1 | | ||i i
i
i
i
i
* \
j
:
\
i i i ii i
CO O) O T— CN CO
WS WS WS
Figure F.33 Boxplot Displays for HF TLC 500MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = ±5 dB of Pre-test)
PrA-"Tc*
-------
APPENDIX F
Post 85/85
HF TLC RNR
HASL
10-
5-
"3
or
U-
X 0-
D_
Q
-5 —
f
i
$
1
f
if
* !
ig
TST _ a. ji
88 !
I
* '
'
1
}
* 1
i
1 1 i 1 1
SiteFIux *- w to T in
WS WS
Boxplots of DPHFTLRN by SiteFIux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
i . j — .
* l
i
\
i
* i
f
* f
I
* B ® S *
W^™"«&®'F1 — @^I«s>fSi
s? s ta i *w
i
i
* $
* !
» i
S
|
i i i i~i i — i — i — r^\ — i —
COI*^CO O>'OT— OfCO^lOCD
I—
1
i
i
I
i
* 1
1
^ J
' i
1
{
I
i
I f
1
1- 1
Ni/Au Ni/Au/Pd
I
$
j
!
* *i
{
I 1
ra & ^
L •» ta l 11
* I
I
I
i
; 1
r i
;
1 — T H — i 1 —
1^ CO O) O -r- OJ (O
WS WS WS WS WSWS
WS WS WS
Figure F.35 Boxplot Displays for HF TLC RNR at Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = <10 dB increase over Pre-test)
Post Thermal Shock
HF TLC RNR
HASL
10-
^~5
~^r
£ °-
1 1
x
I—
Q
-10 —
i
I
I
I
[
XX |
,
a- «• ^ •§: 1 =3
1
* * * 1
I
i
i
i
i
!
I I 1 1 T~
SiteFIux T- CM 00 •» 10
WS WS
Boxplots of DTHFTLRN by SiteFIux
(means are indicated bysolid circles)
OSP > ImmSn ImmAg
i j — :
•*!
* !
i
1
* {
* i
^®@®«.®*S-i_*
S • * * 1 &
\
i
{
x !
i
i
1
j
1 1 1 1~T 1 1 1 l~l 1
!
1
i
i
I
1
1 1
*
i{
if
ll
ij
1 I
j
i
;
j
* 1
1 1
- ' 03
WS WS WS WS WSWS
Ni/Au Ni/Au/Pd
1
i
1
1
* *i
X
& «.! i
! 8
* 1
1
*!
j
1
j
j j£
1
1 1 T^ 1 —
O> O *— CM CO
WS WS WS
Figure F.36 Boxplot Displays for HF TLC RNR Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = <10 dB increase over Pre-test)
F-61
-------
APPENDIX F
Post Thermal Shock
1.0-Mil Pads
HASL
Boxplots of DTPads by SiteFlux
(means are indicated bysolid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
15-
14-
o
T- -^ r- T- r- T- CM
*JPI i
! I if
{ 1 T
\
i
i
i
j
i
i
j
}
!
i
i i
T- CM CO
CM CM £M
WS WS WS WS
WS
WS WSWS WS WS WS
Figure F.37 Boxplot Displays for 10-Mil Pad Post TS - Pre-test Measurements (logio ohms) by Surf. Finish
(Acceptance Criterion = Resistance > 7.7 logie ohms)
Post Mechanical Shock
10-Mil Pads
HASL
Boxplots of DMPads by SiteFlux
(means are indicated bysolid circles)
OSP Imm Sn Imm Ag
Ni/Au Ni/Au/Pd
Q
15 —
14 —
13-
12 —
8-
-
&
*
- m
lf
J E
* 1
{
1
!
J
f
I
I
i
i
I
?
— —
X
s
ji;i
|
1
I
|
i
I
*
1
«
I
—
'
•
* 1 I
i £ :-: si i * & i
* g ffi ;.; i S ••: I
i ? 1 1 i 1 ! i
! 1 I lii .1 i
ii j hy . i .j
i : i
i i i
i : i
i i i
i i i
f : i
J i i . * i
ii \
i. i i
i i i
! i i
SiteFlux
T- CM CO
I I
o r-
in to
WS WS WS WS
• WS
WS WSWS WS WS
WS
Figure F.38 Boxplot Displays for 10-Mil Pad Post MS - Pre-test Measurements (log™ ohms) by Surf. Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
F-62
-------
APPENDIX F
Post 85/85
PGA-A
HASL
13 —
12-
CL
& "-I
10 —
SiteFlux
Boxplots of DPPGA A by SiteFlux
(means are indicated by solid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
ws
ws ws
T-
c
i
3j Ka KJ «f
a
i-ij !
1 !
i
* !
i
i
1
J
!
1 I 1 1
•^ CO ^ If
1 'J
j
r
(C
u 1 0
: K-l EH F"|
4 ** tf
i
X
x
1 1 j-J-
1^ CO O> C
.
s
V
I
3
A ^ 5
: tji] 1 I§|
( 1
* i
1 i
I '\
I
I
I
I
I
I
j
1 1 1 1
CM co ^- in <
1
jj:
i-i
i
1
j
1
o
[^
111
i J
: '
i l !
! I
: *
II L
: i
i
1
j
1
1
I
i
i
1
j
i
CO O
I I
: Op &J |
; S
: I
1 i
i
I
*;
i
1
i
I
i
i
I
i
1 1
) O T- C
1 1,
i |
1 i
5 •:•:
i T
1
M CO
M tNJ
ws
ws
ws wsws
ws
ws
ws
Figure F.39 Boxplot Displays for PGA-A Post 85/85 - Pre-test Measurements (tog10 ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post Thermal Shock
PGA-A
HASL
Boxplots of DTPGA A by SiteFlux
(means are indicated bysolid circles)
OSP , ImmSn ImmAg
15 —
14 —
< -13 —
C5
CL
Q 12-
11 -
10 —
SiteFlux
l
V
« ".' 1 : 1 1 J 1
• 1 . . * * i I I
I i i I
S3
; ll " " I i *\* * If
i « I.I f { * l i '
!•..•«! T « . i i " '
J ll I J I
i T i i i
i ' i i i
i iii
I i ! i
I i i 1
i * ill
f i r i
i i i i i i i i i i i i i i i i i i i i i
ws ws ws ws ws ws wsws ws ws ws
Figure F.40 Boxplot Displays for PGA-A Post TS - Pre-test Measurements Q.ogw ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 log! o ohms)
F-63
-------
APPENDIX F
Post Mechanical Shock D ,, r™/m^AAu ^ m
PGA A Boxplots of DM PGA A by SiteFlux
(means are indicated bysolid circles)
HASL OSP
14-
13 —
<
1 12~
"^>
O
11 -
10 —
B $ ca
* * 8
i B
i
i
I
IT
i i
i
1
1
I
i
{
1
1 I I 1 1 I I 1 I
SiteFlUX r-NCOT10CDI-.COO)C
Imm Sn Imm Ag
i
f T [I
i i '
i $
1 !
i
i
i
i
i
i
i
i
\
i i i i i i
•<- N m *t u> a> t
s
5
;;;
:|:
I
|
S»
Ni/Au Ni/Au/Pd
T 1
I • , ll
| IT
! i
i i
! !
1 i
! i
1 !
i I
: i
i i
i :
i i
i :
t i
i i i i i
^
1
s
ws ws ws ws
ws
ws wsws ws ws
ws
Figure F.41 Boxplot Displays for PGA-A Post MS - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post 85/85
PGA-B
Boxplots of DPPGA B by SiteFlux
(means are indicated bysolid circles)
HASL
13 —
12-
m
g
0.
CL
Q n_
10 —
1 A
m £3
*
X
i
H i
T 1
I
X
1 1 1
| |
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
\ f, i
i
i
i i
i ®
i:
i
i
X
ill pi | A i i 1 i i | 1 i
^1 ^ H ta X T p ta ^
i 1 % T Q i p
i 1 1 1 -1 x 1 f
i 1
! • "
i
;
i
i
i *
\
i
i
I
:
i
i i i i i i . i i i i i i i i i i
SiteFlUX r-<\jco^rmcot^cocnOT-cMco- 7.7 logic ohms)
F-64
-------
APPENDIX F
Post Thermal Shock
PGA-B
HASL
m
g
a.
Q
15-
14 —
13 —
12-
11
SiteFlux
Boxplots of DTPGA B by SiteFlux
(means are indicated bysolid circles)
OSP I mm Sn I mm Ag
Ni/Au Ni/Au/Pd
* IS
•PI *
B i
« ^ ! *
i m
* \
i • !
i i
i i
i !
1 I
1
1
* }
i i 1 1 1 1
r- CM CO T If) CO 1-
\NS \NS \NS
• t
•
i
a
V\,
1 =
3 •
e
1 CT
'S
— I 1 ! ! —
I *! *! 1
I* 1 $
II ! 1
1 Hf- S1 « {
•'• j ||j j g j t
' in ' 1
: i . i f i k
J I 1 * 1
i i :
i i i
*•
1 1 f
I i i
i ! i
1 J $
i i i
i j {
iiiiilllll.il
WS WS WSWS WS WS
»
1
j
•*
1
a co
WS '
Figure F.43 Boxplot Displays for PGA-A Post TS - Pre-test Measurements (log™ ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post Mechanical Shock D .. fr^..n^K ^ u. ^-* ^,
ppA R Boxplots of DM PGA B by SiteFlux
(means are indicated bysolid circles)
HASL OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
15-
14 —
m
§ 13-
2
Q
12 —
11 —
* ! 1 *l
! i f
! i i
! J „ i
I! pa '0
* ij § ^
I '
i *
\
\
\
\
\
x- '
_ ^ ^»
» iSjT |B" J ™ |
5 | MIT i ^
II \ • i i I
I * ' i i
1 1 !
1 i 1
1 I 1
T3 - - 1 1
i !
• * 1 I
i !
? I !
i i i i i r i i i i i i i i i i i i
i
I
|
i
1
1 i i1 PI- *
i § 81 1]
i T $ 1
* M
i '
i
i
i
i
i
i
i
tiii
SiteFlUX i-CMCO-srmCDI^-COOTO^-CMCOTlOCOI^'COCn.OT-CMCO
WS WS .WS WS WS WS WSWS WS WS WS
Figure F.44 Boxplot Displays for PGA-A Post MS - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logio ohms)
F-65
-------
APPENDIX F
GuHWinq Boxplots of DPGulIWi by SiteFiux
(means are indicated bysolid circles)
HASL OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
14 —
13 —
12 —
I 11~
§ 10-
DL
Q
g —
8 —
•
7 —
,j|. , * ,
1
1
i ? i % i a
; fil i
j |T II
r * 1
i
i
!j * 1
s 1 |
! 1 I
i 1 '
I T
f ' •
I
liS
" f
S ' '
1
I
i
rj
i :
i i
! * !
i i
i i
i i
i j —
: I 1 | 1
8 1 1 i
if 1
! 1 i
! i
i i i
* i i
< i
3 i i
i . * ?
i !
f 1
\ i i j
i i i T~T i i i i~i i i i r~^ — i — r^ — i — i — r^ — i —
SiteFiux •r-cMcOTincots.corooT-cMcOTiocDts.cocnoi-cMco
WS WS WS WS WS WS WSWS WS WS WS
Figure FAS Boxplot Displays for the Gull Wing Post 85/85 - Pre-test Measuremts. (logio ohms) by Suirf. Fin.
(Acceptance Criterion = Resistance > 7.7 logio ohms)
GufwnT31 Sh°Ck Boxplots of DTGulIWi by SiteFiux
(means are indicated bysolid circles)
HASL OSP ' ImmSn ImmAg Ni/Au Ni/Au/Pd
15-
14-
13-
^
§ 12 —
g
<•£ 11-
Q
10 —
9 —
8 —
* I
i
„ _
T !;
i
1
« _, j _ „_ _, _
H 18 B
1 j|j * 8
3t Ki | Hr
{
1
1
1
{
i i i r 1
di A iM •*-' J. "i *
fa P ~ la
¥ | B
1 11
|!j | )(
|
I
1
4
i , in i i
1 1 ^
:
i ;
?
^ i
s i
5 !
1 I
j ?
I I L
„ \ I
- i 5
1 * I
; j
i i i i i i i 1
||| nH sain ea
* } 1 @ Y ¥{1
! S if
|T | • '
i
i I
I i
i i
"1 1
I i
i I
\ \
I !
i i
i i 1 I II I
SiteFiux t-'cMcoTincois.cocooT-cMcO'a-incois.cocnor-cMco
Figure F.46 Boxplot Displays for the Gull Wing Post TS - Pre-test Measurements (Iog10 ohms) by Surf. Fin.
(Acceptance Criterion = Resistance > 7.7 logio ohms)
F-66
-------
AfPEffDIXF
Post Mechanical Shock ' . . rr,..~ ..,..., ' _
Gull Wina Boxplots of DMGulIWi by SiteFIux
(means are indicated bysolid circles)
HASL OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
14-
13-
o, 12~
§
= 11-
5
Q 10-
9 —
8 —
1 * | A si | §
li x 1 I ? ll
1 i | J
T t * y * * -
I I' *
' i
i
i
i
i
i
i
i
i
« t
• S Si Ifj " I B
i 8 r^ 1 1 'III
; * * i Ml
* U i I i »• 1 1
.: | m Lj :j j |
1 IT I*
1 T | 1 i
j Tj
i
I
i
j
I
I
i i
i j
r~ i i i T i i i i i r I i i i
, j_
II
i
III* i
1 li l
i H i
i. h i
.-. "i •
l i i
* i \
1 ' 1 '5
i I 1
i i
i 1 I
f $
{ J
i i
T T^T r*—
S
•|:
8
:::
;i;
f
r
SiteFIux ---CMCOTrincoi^cocnot-cM'cOTiocoi^cocnoT-cM
WS WS WS WS WS WS WSWS WS WS
I
PI
|xj
— r —
CO
CM
WS
Figure F.47 Boxplot Displays for the Gull Wing Post MS - Pre-test Measurements Q.og10 ohms) by Surf. Fin.
(Acceptance Criterion ~ Resistance > 7.7 log™ ohms)
Prp "Tp^t
Stranded Wire 1 Boxplots of StWire 1 by SiteFlux
(means are indicated bysolid circles)
HASL OSP . ImmSn . Imrr
i Ag Ni/Au Ni/Au/Pd
20—I
15-
10 —
SiteFIux
. , , — :
1 ,
i
ll
;
A :
;; :
I ^
>: '
'fl
1
¥
1
*
5
0
K
f
.
i
h « r"
: ;
! i-
'• \m
l id :
! i PI
• *
M
L
i Jl [
1 1 ^
i n n n |
s ^ i 1
* s
Jy 1 T i
L
1
:
i
1
:
I
I
I
I | I |
1 ^ 1
1 1
1 ! — : 5
i 1 I
i I
U
!
•
||
I'l
1 ;
* :
« :
! 1
f :
f
Ji
j
:
.
s
i! , J
{ 1
s | j. r
J
JI
IM
!
!
1 t
L
)
1
H
1
i
I'l
j &
i •:•
. ;J
! « S
i i
i I
i i
I
L
•
:
i
'•
T
I I 1 f
i i i
1 1 !
1 1 1
1 1
l.
T-CMCO^TtOCOf^COO>OT-CMCO^TIf}COf-.COO>O*— CMCO
WS WS WS WS WS
WS WSWS WS WS WS
Figure F.48 Boxplot Displays for the Stranded Wire 1 Measurements (volts) at Pre-test by Surface Finish
F-67
-------
APPENDIX F
Pre-Test
Stranded Wire 2
w
28 —
27-
26 —
25 —
24-
23 —
22-
21-
20 —
19 —
SiteFlux
HASL
Boxplots of StWire2 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
A
1
S
f
i
{
i
f
f
I
!
i
i
i
\
Is
•S
T
1
1 J
: •
;
L
i
i
j
*
i
*
i
1
5
T
L
1
t
I
i
i
l
I!
!
i
|
i
l
i
j
l
I
l
I
l
1
$
J
i
i
I
If
i i
i ?
i \
L
•
1
1
i
1
f
i
i
1
I
I
i
I
I
i
i
t
i i
S
1
1
I I I
T- CM CO
I I
in CD
i i i i i r
in o
WS WS WS WS
WS
WS WSWS WS WS WS
Figure F.49 Boxplot Displays for the Stranded Wire 2 Measurements (volts) at Pre-test by Surface Finish
F-68
-------
APPENDIX F
F. 10 Design and CCAMTF Baseline Testing of the Test PWA
F.10.1 Test PWA
As mentioned in Chapter 4, the primary test vehicle used in both the DfE project and in the
CCAMTF evaluation of low-residue technology was an electrically functional PWA. This assembly
was designed at Sandia National Laboratories in Albuquerque based on input from LRSTF members
and from military and industry participants during open review meetings held by the task force. The
PWA measures 6.05" x 5.8" x 0.062" and is divided into six sections, each containing one of the
following types of electronic circuits:
• High current low voltage (HCLV)
• High voltage low current (HVLC)
• High speed digital (HSD)
High frequency (HF)
Other networks (ON)
Stranded wire (SW)
The layout of the functional assembly is shown in Figure F. 50. The components in the HCLV,
HVLC, HSD, and HF circuits represent two principal types of soldering technology:
• Plated through hole (PTH)—leaded components are soldered through vias in the circuit board
by means of a wave soldering operation
• Surface mount technology (SMT)—leadless components are soldered to pads on the circuit
board by passing the circuit board through a reflow oven.
The other networks (ON) are used for current leakage measurements: 10-mil pads, a socket for a
PGA, and a gull wing. The two stranded wires (SW) are hand soldered.
The subsections for PTH and SMT components form separate electrical circuits. The PWA
includes a large common ground plane, components with heat sinks, and mounted hardware.
Each subsection shown in Figure F.50 contains both functional and nonfunctional components
(added to increase component density). A 29-pin PTH edge connector is used for circuit testing. High
frequency connectors are used to ensure proper impedance matching and test signal fidelity as
required. Board fabrication drawings, schematics, and a complete listing of all components are
available by contacting the authors of this report. A discussion of each of the sections of the test PWA
is now given. This discussion is supplemented with baseline test results for each of the 23 electrical
responses listed in Table 4.1.
F. 10,2 High Current Low Voltage
The HCLV section of the board is in the upper left-hand corner of PWA (see Figure F.50). The
upper left-hand portion of this quadrant contains PTH components with SMT components immediately
beneath.
Purpose of the HCLV Experiment
Performance of high-current circuits is affected by series resistance. Resistance of a conductor
(including solder joints) is determined by the following equation:
F-69
-------
APPENDIX F
pL
R = f-—ohms(£l}
•A...
(F.7)
where p = resistivity, the proportionality constant
L = length of the conductor
AC = cross-sectional area of the conductor (solder joints)
Resistance is most likely to change due to cracking or corrosion of the solder joint that may be
related to the soldering process. These conditions decrease the cross-sectional area of the solder joints,
thus increasing resistance as shown in Equation F.7. Use of high current to test solder joint resistance
makes detection of a change in resistance easier. A 5 Amperes (A) current was selected as a value that
would cover most military applications. A change of resistance is most conveniently determined by
measuring the steady state performance of the circuit, which will now be discussed.
-^ u.uj ^
High Current Low Voltage (HCLV)
PTH
High Current Low Voltage (HCLV)
SMT
High Speed Digital (HSD)
PTH
High Voltage Low Current (HVLC)
PTH
High Voltage Low Current (HVLC)
SMT
OTL2
Hiqh Frequency (HF) 1
| TL4
"ransmission Lir
P
!
n
c
High Speed °
Digital (HSD) "
SMT °
t
o
r
r^
"'C:
TL3 1
Figure F.50 Layout of the PWA Illustrating the Four Major Sections and Subsections
F-70
-------
APPENDIX F
Steady State Circuit Performance
Overall circuit resistance, Rtotai, is the parallel combination of the seven resistors, R1? R2,..., R7,
(all resistors = 1OQ) used in the HCLV circuit:
7
ion
(F.8)
R —
•"• ~
ion
(F.9)
Since a current (I) of 5 A will be applied to the circuit, the resulting voltage (V), according to
Ohm's Law, is
(FAQ)
Changes in resistance are thus detected by changes in voltage. However, a pulse width had to be
chosen that would not Overstress the circuit components. With current equally divided among the
seven parallel resistors, the power (P) dissipated in each resistor, according to Joule's Law, is:
5A
' = (—j xlOCl=5Watts(W)
(F.11)
Since the power rating for the PTH wire-wound resistor is 3W, the rating is exceeded by a factor
of 1.7 for steady state (5.1 / 3). Design curves from the resistor manufacturer indicate the PTH wire-
wound resistors could tolerate the excess power for about WQms. The SMT resistors are rated at 1W,
so the steady state rating is exceeded by a factor of five. With the manufacturer unable to provide the
pulse current capability of the SMT resistors, a pulse derating factor could not be determined. A pulse
width of lOOpus was selected, which is three orders of magnitude less than the capability of the wire-
wound resistors. This width is also sufficiently long for the circuit to achieve steady state before the
measurement is taken. .
Circuit Board Design
Traces carrying the 5A current were placed on an inner layer of the circuit board because: (1) the
primary concern was the possible degradation of the solder connections as discussed above and (2) the
bulk electrical characteristics (resistivity) of the traces should not be affected by flux residues. High-
current trace widths were-designed to be 250 mils whenever possible (following MIL-STD-275). This
width with a 5 A current should cause no more than a 30°C temperature rise under steady-state
conditions.
The resistor and capacitor values were selected to be readily available. If other values are used,
care should be taken to not over-stress the parts, as discussed above.
Baseline Testing Results for HCLV V
A gauge repeatability and reproducibility (GR&R) study (Iman et al, 1998) was conducted for the
CCAMTF ATS as part of the CCAMTF program. The LRSTF PWA was utilized in this study. In
particular, 120 LRSTF PWAs were tested for each of the following four surface finishes: OSP,
F-71
-------
APPENDIX F
immersion Ag, immersion Au/Pd and HASL with solder mask. Half the PWAs in each surface finish
group were processed with low-residue (LR) flux and the other half with water soluble (WS) flux.
Data modeling showed that surface finish and flux type did not significantly affect the voltage
measurements for HCLVPTH and HCLV SMT. Figures F.51 and F.52 provide dotplot displays of 4
x 120 = 480 voltage measurements for HCLV PTH and 480 voltage measurements for HCLV SMT,
respectively. The summary statistics HCLV PTH and HCLV SMT voltages are given in Table F.31.
+ + "-"-- + + + + VoltS
6.60 " 6.72 6.84 6.96 7.08 7.20
Figure F.51. Dotplot for 480 HCLV PTH Voltage Measurements
(each dot represents up to 10 points)
•
6.90
7.00
•
' 7110 7.2(
.
) 7.30
7.40
Volts
Figure F.52. Dotplot for 480 HCLV SMT Voltage Measurements
(each dot represents up to 16 points)
Table F.31. Summary Statistics for HCLV Circuitry Test Measurements
Circuitry Mean Median St Dev. Min Max
HCLVPTH
HCLV SMT
6.88V
7.20V
6.96
7.20
0.163
0.106
6.60
6.88
. 7.20
7.44
F.10.3 High Voltage Low Current
The HVLC circuitry is immediately below the HCLV circuitry and above the high frequency
transmission lines in Figure F.50. The PTH circuitry is in the upper part of this subsection and the
SMT circuitry is hi the lower part.
Purpose of the HVLC Experiment
Flux residues could decrease the insulation resistance between conductors. The impact of this
decrease could be significant in circuits with a high voltage gradient across the insulating region.
Decreased resistance can be detected by an increase in current when a high voltage is applied to the
F-72 ™~~""
-------
APPENDIX F
circuit. A voltage of 250V was selected as the high potential for this test. The change in leakage
current is determined by measuring the steady-state performance of the circuit, which will now be
discussed.
Steady State Circuit Performance
Steady-state operation of the HVLC circuit can be determined by considering only the resistors. The
total resistance of the series combination is the sum of the resistances.
Rtota!=Rl+R2+R3+R4=R5 = 50MO
(F.12)
since all resistors are 10MH each. From Ohm's law, the current flowing into the circuit with 250V
applied is '
R 50MH
(F.13)
Care was taken to not overstress the individual components in the circuits. The voltage stress across
each resistor-capacitor pair is one-fifth of the applied 250V, or 50V. The voltage ratings are 250V for
the PTH resistors, 200V for the SMT resistors, and 250V for all the capacitors. Power rating is not a
concern due to the low current.
Circuit Board Design
High voltage traces were placed next to ground potential traces by design. The spacings between
the high voltage and intermediate traces were selected using MIL-STD-275.
Voltage Spacing Between Traces frails')
0-100
101-300
301-500
5
15
30 •
These guidelines were followed except the 5-mil spacing, where 10 mils was used to facilitate board
fabrication. Table F.32 lists the voltage on various board circuit traces and the spacing to the adjacent
ground trace.
Resistors and capacitors were selected to have readily available values—different values could have
been used to achieve particular experimental goals. For instance, higher resistance values could be
used with lower value capacitors. Reverse biased, low-leakage diodes could also be used for higher
sensitivity to parasitic leakage resistance.
Baseline Testing Results for HVLC
Data modeling showed that surface finish and flux type had very little effect on the voltage
measurements for HVLC PTH and HVLC SMT. Figures F.53 and F.54 provide dotplot displays of
480 voltage measurements for HVLC PTH and HVLC SMT, respectively. The summary statistics for
HVLC PTH and HVLC SMT voltages are given in Table F.33. Note that two slight outliers for HVLC
PTH are identified in Table F.33, but are not included in Figure F.53.
F-73
-------
APPENDIX F
Table F.32 HVLC Circuit Board Trace Potentials
Technology Trace Connected to:
Resistor Capacitor
PTH R15
R16
R17
R18
R19
SMT R20
R21
R22
R23
R24
C21
C22
C23
C24
C25
C26
C27
C28
C29
C30
Potential (V)
250
200
200
150
150
100
100
50
50
250
200
200
150
150
100
100
50
50
Trace Length at
Potential (in)
0.8
0.4
0.4
NA
NA
0.4
0.4
NA
NA
5.0
1.0
1.0
NA
NA
0.9
0.9
NA
NA
Spacing
(mills)
30 ,
15
15
10
10
30
15
15
10
10
NA = not applicable since no 50V or 150V traces were adjacent to ground potential
Table F.33 Summary Statistics for HVLC Circuitry Test Measurements (sans outliers)
Circuitry Mean Median St Dev. Min Max Outliers
HVLC PTH
HVLC SMT
5.04nA
4.95nA
5.04
4.95
0.024
0.011
4.972
4.914
5.148
4.976
5.203 5.232
4.970 5.005 5.040 5.075 5.110 5.145
Figure F.53 Dotplot of 478 Voltage Measurements for HVLC PTH
(each dot represents up to 2 points)
-uA
F-74
-------
APPENDIX F
4.920
+-
4.932
4.944
4.956
4.968
lift.
4.980
Figure F.54 Dotplot of 480 Voltage Measurements for HVLC SMT
(each dot represents up to 2 points)
F.10.4 High Speed Digital
The HSD circuitry is in the upper right-hand corner of the LRSTF PWA shown in Figure F.50.
This subsection contains the PTH circuitry and consists of two 14-pin Dual In-line Package (DIP)
integrated circuits (ICs). The SMT subsection 1C is a single 20-pin leadless chip carrier (LCC)
package. Each of these ICs is a "Fast" bi-polar digital "QUAD-DUAL-INPUT-NAND-GATE." Both
subsections contain two ceramic capacitors that bypass spurious noise on the powerinput line (VCC)
to the ICs and an output high-frequency connector. Inputs to both subsections are applied through the
edge-connector on the right side of the board. Figure F.55 shows a simplified schematic of the ICs.
2.5V
Pulse
5V
1.
VCC
Quad-Dual-lnput-NAND-Gate 1C
V,
out
Ground
Figure F.55 Simplified Schematic of the ICs in the HSD Subsection
F-75
-------
APPENDIX F
Purpose of the HSD Experiment
The output signal of each gate in Figure F.55 is opposite in polarity to the input signal. If the
traces of these two signals are in close proximity on the printed circuit board (capacitively coupled),
the gate switching speed might be affected by the presence of flux residues. A 5VDC bias is applied to
the VCC inputs during environmental testing to accelerate aging. One PTH 1C (U02) is hand soldered
during assembly to introduce hand solder flux residue in the experiment.
Circuit Description
The schematic in Figure F.55 represents the ICs in the PTH and SMT subsections. The ICs are
random logic circuits that are NAJND (Not AND) gates. An AND gate's output is high only when all
inputs are high. The logic of a NAND gate is opposite the logic of an AND gate. Therefore, the
output of a NAND gate is low only when all inputs are high, otherwise the output is high. With the
two connected inputs, the output of each gate is opposite the input. Since the four gates are connected
in series, the output of the last gate is the same logic level (high or low) as the input, with a slight lag.
The output pulse does not change logic levels instantaneously, but the switching times from low to
high (rise tune) and from high to low (fall time) should be less than Ins. ICs should perform within
these criteria if the VCC input is 5±0.5V DC, the output load does not exceed specifications, and the
circuit has a proper ground plane as shown in Figure F.55. The HSD circuits also provide an
intermediate test for high frequencies, with switching tune dictating a high frequency spectrum. The
frequency spectrum of switching circuits can be expressed in terms of bandwidth (BW). For a
switching circuit, the respective BWs (in Hertz) for rise (tr) and fall (tf) tunes are:
035
BW, = -—Hz and
f
(F.14)
Bipolar technology was used rather than a complementary metal oxide semiconductor (CMOS)
since it is not as vulnerable to electrostatic discharge (BSD) damage. Available military bipolar
technologies have the following typical switching speeds and bandwidths:
Technology
5404 TIL
54LS04Low
Power Schottky
54S04 Schottky
54F04 Advanced
Schottky (Fast)
Typical trorf(ns)
12
9
3
2.5
Bandwidth (MHz)
29
39
117
140
The Fast technology was selected since it had the shortest switching time and largest bandwidth,
which provides the widest frequency spectrum for this test.
Circuit Board Design
Ground planes were provided for proper circuit operation of the ICs. The PTH subcircuit utilized
the large common ground plane on layer 3 since most of the input and output traces are on layer 4.
Since the SMT circuit traces are on the top layer, a smaller ground plane was added on layer 2, The
"QUAD-DUAL-INPUT-NAND-GATE" was selected since other solder studies of national attention
have used that particular type of 1C, which makes direct comparisons with these studies possible.
F-76 "
-------
APPENDIX F
Baseline Testing Results for HSD
Data modeling showed that surface finish and flux type had very little effect on the total
propagation delay measurements (msec) for HSD PTH and HSD SMT. Figures F.56 and F.57 provide
dotplot displays of 480 voltage measurements for HSD PTH and HSD SMT, respectively. The
summary statistics HSD PTH and HSD SMT total propagation delay are given in Table F.34 (Note one
slight outlier for HSD PTH).
•*. + + + + +- u sec
12.64 12.80 12.96 13.12 13.28- 13.44
Figure F.56 Dotplot of 480 Measurements of Total Propagation Delay for HSD PTH
(each dot represents up to 2 points)
• * • *••*•!•!
***• •••*•*••
• • »**••••••*•*«*•
+—• + + +- + +_u sec
4.80 4.92 5.04 5.16 5.28 5.40
Figure F.57 Dotplot of 480 Measurements of Total Propagation Delay for HSD SMT
(each dot represents up to 2 points)
Circuitry
Table F.34 Summary Statistics for HSD Circuitry Total Propagation Delay (jasec)
Test Measurements (sans outliers)
Mean Median St. Dev. Min Max Outliers
HSD PTH
HSD SMT
13.04|asec
5.02)4. sec
13.04
5.02
0.124
0.086
12.56
4.75
13.44
5.39
14.40
4.20 4.29
F-77
-------
APPENDIX F
F.10.5 High Frequency
The HF section shown in the lower right-hand comer of Figure F.50 contains two major
subsections, the low-pass filters (LPF) and the transmission line coupler (TLC). The TLC traces on
layer 4 of the board are on the backside of the board. The LPF/PTH subsection is above the LPF/SMT
subsection. Each of these subsections has discrete ceramic capacitors and three inductor-capacitor
(LC) filters, with the inductor printed on the circuit board in a spiral pattern. The HF circuits allow
evaluation of circuit performance up to 1 GHz (1000MHz).
Purpose of the High Frequency Experiment
Flux residues may affect the performance of LPF printed circuit inductors and transmission lines
due to parasitic resistances and parasitic capacitances. Since the transmission lines are separated by
only 10 mils, flux residues between the lines may affect their performance.
LPF Circuit Description
An inductor-capacitor (LC) LPF consists of a series inductor followed by a shunt capacitor. A
low-frequency signal passes through the LPF without any loss since the inductor acts as a short circuit
and the capacitor acts as an open circuit for such signals. Conversely, "a high-frequency signal is
blocked by the LPF since the inductor acts as an open circuit and the capacitor acts as a short circuit
for such signals.
When a sine wave test signal is passed through an LPF, its amplitude is attenuated as a function of
frequency. The relationship between the output and input voltage amplitudes can be expressed as a
transfer function. The transfer function, Vout / Vin, was measured to determine any effects of the low-
residue fluxes.
The transfer function is measured in decibels (dB) as a function of frequency. A decibel can be
expressed in terms of voltage as follows:
dB = 20 log
(F.15)
The PTH transfer function differs from the SMT transfer function due to the self inductance of the
capacitor through-hole leads.
LPF Circuit Board Design
The three LC LPFs for each of the SMT and PTH circuits were designed to have the following
cutoff frequencies: 800, 400, and 200 MHz. Cutoff frequency is that frequency for which the transfer
function is -3 dB. The respective component values chosen for the LC filters are 16 nH (nano-Henries)
and 6.4jpF(pico-Farads), 32 nH and 13 pF, and 65 nH and 24 pF. Most LPF circuitry was placed on
Layer 1, with Layer 2 used as a ground plane. Crossovers needed to connect the LPF circuits are on
Layer 4.
The LPF circuits were designed to operate with a 50O test system, so all interconnect traces
longer than 0.10 in were designed as 50fl transmission lines to avoid signal distortion. The LPF
circuits were predicted to have less than 2 dB loss below 150 MHz, approximately 6 dB loss near 235
-------
APPENDIX F
MHz, and greater than 40 dB loss at 550 MHz and beyond. The measured response of the LPF/SMT
circuit is close to that predicted except that the transfer function decreases more rapidly than predicted
above 350 MHz. As stated previously, the PTH circuit transfer function did not perform similarly to
the SMT, particularly at frequencies above 150 MHz.
-0.325 -0.300 -0.275 -0.250 -0.225 -0.200
Figure F.58 Dotptot of 473 Measurements of the Response for HF PTH at 50 MHz
(each dot represents up to 2 points)
240.0 244.0 248.0 252.0 256.0 260.0
Figure F.59 Dotplot of 472 Measurements of the Frequency for HF PTH at -3dB
(each dot represents up to 2 points)
• * *•*•••*
424.0 432.0 440.0 448.0 456.0 464.0
Figure F.60 Dotplot of 474 Measurements of the Frequency for HF PTH at -40dB
(each dot represents up to 2 points)
F-79
-------
APPENDIX F
Baseline Testing Results for HF LPF
Data modeling showed that surface finish and flux type had slights effects on the HF LPF
frequencies and responses for HF PTH 50 MHz, HF PTH f(-3dB), HF PTH f(-40dB), HF SMT 50
MHz, and HF SMT f(-3dB). The response, HF SMT f(-40dB), was 5 to 12 MHz lower for PWA with
OSP, immersion Ag, or immersion Au/Pd surface finishes. However, the range of frequencies for this
response was only from 630.7 MHz to 680.60 MHz, so the changes in frequency are relatively small.
Figures F.58 to F.59 provide dotplot displays of 480 measurements for the six HF LPF responses. The
summary statistics for these responses are given hi Table F.35 (Note there are several outliers
identified in this table).
-0.315
.-0.280
-0.245
-0.210
-0.175
+-dB
-0.140
Figure F.61 Dotplot of 473 Measurements of the Response for HF SMT at 50 MHz
(each dot represents up to 2 points)
273.6
275.2
276.8
278.4
280.0
.281.6
Figure F.62 Dotplot of 469 Measurements of the Frequency for HF SMT at -3dB
(each dot represents up to 7 points)
630
640
+-
650
-MHz
660
670
680
Figure F.63 Dotplot of 469 Measurements of the Frequency for HF SMT at -40dB
(each dot represents up to 2 points)
F-80
-------
. APPENDIX F -
The distribution in Figure F.59 is different from the other 22 electrical responses in that it displays
a bimodal distribution for HF PTH f(-3dB) with one group of frequencies centered at approximately
245MHz and the other group at 256MHz. Data modeling showed that the differences between these
two groups were not related to any of the experimental parameters (surface finish or flux) nor were
they related to fixture or time of test. A possible explanation for the bimodal distribution is differences
hi date lots for the components. However, date lot information were not recorded prior to processing
and thus, the date lot hypothesis cannot be confirmed. Since the JTP acceptance criterion is based on
change after exposure to environmental conditions, the bimodal distribution could potentially be
important if the measurements were not repeatable. Twenty board serial numbers were randomly
selected for retest to see if the measurements were repeatable with 10 boards from the distribution
centered at 245MHz and 10 boards from the distribution centered at 256MHz. These two groups of 10
were equally split between fixtures A and B on the CCAMTF ATS. Table F.36 gives the differences
between the initial baseline measurements and those from the repeat test. The differences hi this table
are all quite small. The correlation of the measurements on fixture A is 0.995 and on fixture B it is
0.982, which indicates excellent repeatability. Thus, other than being a curiosity, the bimodal
distribution for HF PTH f(-3 dB) will have no practical effect on the test results.
Table F.35 Summary Statistics for 393
Circuitry Mean Median
Test Measurements for Response (dB) or Frequency (MHz)
for HF LPF (sans outliers)
StDev. Min Max Outliers
HF PTH 50 MHz -0.254 dB -0.252 0.022
HFPTH-3dB 250.6MHz 250.7 5.65
»
HFPTH-40dB 440.7MHz 440.1 6.01
*
HFSMTSOMHz -0.242 dB -0.242 0.023
HFSMT-3dB 278.3MHz 278.6 1.20
HFSMT-40dB 660.2MHz 661.0 7.66
-0.319 -0.194 -0.351
-0.148
-0.130
-0.096
240.0 260.8 227.4
305.3
307.1
308.3
425.3 464.4 506.6
507.8
513.7
-0.329 -0.144 -0.447
-0.066
-0.061
273.8 282.2 225.2
299.4
302.9
355.2
383.1
389.6
630.7 680.6 694.8
708.5
721.5
862.8
877.7
-0.150
-0:138
-0.107
230.5
306.5
307.7
308.9
507.2
513.1
514.3
-0.074
-0.062
295.8
301.8
302.9
381.9
384.3
701.9
719.8
758.3
872.3
890.2
924.6
F-81
-------
APPENDIX F
Table F.36 Results from Repeat Testing of the HF PTH f(-3dB) Circuit
Fixture A Fixture B
Test Baseline Repeat Difference | Baseline Repeat Difference
1
2
3
4
5
6
7
8
9
10
244.2
245.3
246.5
247.1
253.1
255.4
256.0
257.2
259.0
259.6
243.0
244.8
246.5
247.1
254.3
255.4
256.0
257.8
259.0
259.0
1.23
0.55
-0.03
-0.03
-1.15
-0.04
-0.03
-0.61
0.00
0.60
242.4
244.2
245.3
246.5
248.9
253.7
254.8
256.0
257.8
259.0
243.0
245.3
245.9
244.2
250.1
255.4
255.4
258.4
258.4
259.0
-0.57
-1.14
-0.64
2.34
-1.19
-1.74
-0.64
-2.41
-0.61
0.00
TLC Circuit Description
Figure F.64 shows a diagram of the TLC subsection. The LPFs described above are lumped
element circuits since the capacitors are discrete components. The TLC lines are distributed element
circuits with the resistors, inductors, and capacitors distributed along the lines. A circuit model for the
lines is shown in Figure F.65.
J9
\
J10 '\ J8
Figure F.64 Diagram of the HF/TLC Subsection
V(n • >ieairage •
• '
•wv • • • — -
retrace LL
v CL Rleakage \
: ?
Mrace
LL
'leakage
Vout
CL
Figure F.65 HF/TLC Distributed Element Model
The inductance and capacitance for a transmission line with a ground plane are, respectively:
LL =0.085^ Js~rnHlin (F.16)
85
(F.17)
F-82
-------
APPENDIX F
where Ro = characteristic resistance and sr = dielectric constant of the board material.
The TLC Ro was designed to be 50O for operation with a 50Q test system. For FR-4 epoxy
(board substrate material), LL is about 9.6 nHlin and CL is about 3.8 pFlin.
The TLC was tested with a sine wave signal similar to the one used in testing the LPFs. The
source resistance was 50Q. and the three output terminals were connected to 50Q. loads.
TLC Circuit Board Design
The transmission line coupler (TLC) circuit has a pair of coupled 50H transmission lines with
required measurable performance frequencies less than 1000 MHz. Layer 4 of the printed wiring board
(PWB) was used to route the TLC circuit, with Layer 3 used as the ground plane. The TLC circuit is a
5 in long pair of 0.034 in wide 5QQ. transmission lines spaced 0.010 in apart. The circuit design
incorporated the board dielectric constant of about 3.8 and the .020 in spacing between copper layers.
A computer-aided circuit design tool (Libra) was used to model the TLC circuit. Performance
measured on a test PWB agreed very closely with the forward and reverse coupling predictions
between 45 MHz and 1000 MHz. .
Baseline Testing Results for HF TLC
Data modeling showed that surface finish and flux type had very slight effect on the HF TLC
frequencies and responses for HF TLC 50 MHz, HF TLC 500 MHz, HF TLC 1000 MHz, HF TLC
Reverse Null Frequency, and HF TLC Reverse Null Response. Figures F.66 to F.70 provide dotplot
displays of 480 measurements for the five HF TLC responses. Summary statistics for these responses
are given in Table F.37 (Note the outliers identified in this table).
-42.0 -40.0 -38.0 -36.0 -34.0 -32.0
Figure F.66 Dotplot of 479 Measurements of the Response for HF TLC at 50 MHz
(each dot represents up to 4 points)
-18.90 -18.20 ' -17.50 -16.80 -16.10 -15.40
Figure F.67 Dotplot of 479 Measurements of the Response for HF TLC at 500 MHz
(each dot represents up to 3 points)
F-83
-------
APPENDIX F
-13.20 -12.80 -12.40 -12.00 -11.60 -11.20
Figure F.68 Dotplot of 478 Measurements of the Response for HF TLC at 1000 MHz
(each dot represents up to 2 points)
636.0
642.0
648.0
654'.0
660.0
666.0
Figure F.69 Dotplot of 479 Measurements of the HF TLC Reverse Null Frequency
(each dot represents up to 2 points)
-66.0 -60.0 -54.0 -48.0 -42.0 -36.0
Figure F.70 Dotplot of 479 Measurements of the HF TLC Reverse Null Response
(each dot represents up to 2 points)
Table F.37 Summary Statistics for 480 Test Measurements for Response (dB) or Frequency (MHz) for HF TLC
(sans outliers)
Circuitry Mean Median St Dev. Min Max Outliers
HF TLC 50 MHz
HF TLC 500 MHz
HF TLC 1000 MHz
HFTLCRNF
HFTLCRNR
-37.57 dB
-18.34 dB
-12.56 dB
649.6 MHz
-44.82 dB
-37.34
-18.43
-12.60
649.1
-44.01
0.974
0.403
0.258
4.77
5.25
-42.74
-19.29
-13.15
636.6
-64.89
-33.05
-15.57
-11.07
665.1
-34.12
-6.13
-6.90
-7.05
935.3
-9.67
-8.94
F-84
-------
APPENDIX F
F.10.6 Other Networks (Leakage Currents)
The test PWA also contains three test patterns to provide tests for current leakage: (1) the pin grid
array (PGA), (2) the gull wing (GW), and (3) 10-mil spaced pads. A 100V source was used to
generate leakage currents. '
Purpose of the Experiments
The PGA, GW, and 10-mil pads allow leakage currents to be measured on test patterns that are
typical in circuit board layouts. These patterns contain several possible leakage paths and the leakage
could increase with the presence of flux residues and environmental exposure. In addition, solder
mask was applied to portions of the PGA and GW patterns to evaluate its effect on leakage currents
and the formation of solder balls.
Pin Grid Array
The PGA hole pattern has four concentric squares that are electrically connected by traces on the
top layer of the board as shown in Figure F.71. The pattern also has four vias just inside the corners of
the innermost square that are connected to that square. Four vias were placed inside the innermost
square to trap flux residues. Two leakage current measurements were made: (1) between the two inner
squares (PGA-A) and (2) between the two outer squares (PGA-B), as shown in Figure F. 71. Solder
mask covers the holes of the two outer squares on the bottom layer, allowing a direct comparison of
similar patterns with and without solder mask.
Rather than an actual PGA device, a socket was used since it provided the same soldering '
connections as a PGA device. Also, obtaining leakage measurements on an actual PGA is nearly
impossible due to complexity of its internal semiconductor circuits.
Gull Wing
The upper half of the topmost GW lands and the lower half of the bottom most GW lands were
covered with solder mask to create a region that is susceptible to the formation of solder balls. The
lands were visually inspected to detect the presence of solder balls. A nonfunctional GW device is
installed with every other lead connected to a circuit board trace forming two parallel paths around the
device. Total leakage current measurements were made on adjacent lands of the GW device
10-mil Pads
The 10-mil pads were laid out in two rows of five pads each. The pads within each row were
connected on the bottom layer of the board and leakage between the rows was measured.
Baseline Testing Results for Leakage Currents
The leakage currents are converted to resistance (ohms) through the basic equation R = V/I. Since
the applied voltage is 100 V and the current is measured in nanoamps, this equation can be expressed
as logio R = 11 - logic I. ' . . % -
F-85
-------
APPENDIX F
PGA-B*
PGA-A m
Solder
Mask
Figure F.71 PGA Hole Pattern with Solder Mask
Table F.38 Significant Coefficients for the GLM Analyses of Leakage Currents
Experimental Variables • 10-MUPad PGA A PGAB Gull Wing
Constant
OSP
Immersion Ag
Immersion,Au/Pd
Flux
OSP*Flux
Ag*Flux
Au/Pd*Flux
Model R2
Standard Deviation
11.43
0.68
0.59
• 0.28
1.61
-0.33
-0.37
;,
60.99
0.606
10.63
0.92
0.84
0.49
1.77
-0.26
-^ W>
74.52
0.542
9.88
1.22
1.22
1.52
2.74
-0.60
-0.90
-0.90
88.12
0.432
11.57
0.61
0.67
0.40
0.89
?!' ;«»*"<, xV*
-*• ^'- '- V' ^ ,
-0.31
35.04
.681
General linear modeling (GLM) results for logio R are given in Table F.38. The GLM results
show that surface finish and flux type strongly affect leakage currents. To illustrate these effects,
dotplot displays of 480 measurements for the four leakage responses are given by surface finish and
flux in Figures F.72 to F075 and by flux hi Figure F.76. The summary statistics for these responses are
given in Tables F.3 9 and F.40.
F-86
-------
APPENDIXF
-OSP LR
-OSP WS
-Ag LR
-Ag WS
+ + + + Au/Pd LR
-Au/Pd WS
-HASL LR
+ + + + HASL WS
10 . 0 11.0 12.0 13.0 14.0
Figure F.72 Dotplots for 480 Measurements of Leakage on 10-MU Pads by Surface Finish and Flux
F-87
-------
APPENDIX F
----- + --------- + --------- + --------- + ------- OSP LR
ws
----- + --------- + --------- + --------- + ------- Au/Pd LR
----- * --------- 4. --------- + ......... + ....... Au/Pd WS
----- + --------- + --------- + --------- + ------- HASL WS
10.0 11.0' 12.0 13.0 14.0
Figure F.73 Dotplots for 480 Measurements of Leakage on PGA A by Surface Finish and Flux
F-88
-------
+ + + _ + osp
+ -__ + + •__ + osp
_ + + + + Au/Pd LR
+ ; + +--.—. + Au/Pd WS
+ + + _ + HASL LR
APPENDIX F
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.74 Dotplots for 480 Measurements of Leakage on PGA B by Surface Finish and Flux
F-89
-------
APPENDIX F
LR
-OSP WS
-Ag LR
WS
_ + + + + Au/Pd LR
+ + + + Au/Pd WS
-HASL LR
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.75 Dotplots for 480 Measurements of Leakage on the Gull Wing by Surface Finish and Flux
F-90
-------
APPENDIX F
Table F.39 Summary Statistics for Leakage Currents
and Flux
Circuitry Surface Finish Flux Mean
10-Mil Pads OSP
Immersion Ag
Immersion Au/Pd
HASL
PGAA OSP
Immersion Ag
Immersion Au/Pd
HASL
PGAB OSP
Immersion Ag
Imm'ersion Au/Pd
HASL
Gull Wing OSP
Immersion Ag
Immersion Au/Pd
HASL
LR
WS
LR
WS
LR
WS
LR
WS '
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
12.11
13.39
12.02
13.26
11.81
13.22
11.29
13.15
11.59
13.28
11.47
12.98
11.23
12.78
10.45
12.56
11.10
13.23
11,10
12.94
11.47
13.16
9.74
12.70
12.15
13.10
, 12.23
13.14
11.99
12.53 '
11.57
12.44
Test Measurements
Median St. Dev.
11.94
13.52
11.90
13.30
. 11.73
13.22
11.29
13.40
11.62
13.30
11.39
12.94
11.20
12.80
10.46
12.66
11.11
13.30
11.12
13.00
11.44
13.10
9.75
12.70
12.40
13.22
12.32
13.46
12.02
12.66
11.52
12.70
0.77
0.55
0.76
0.38
0.54
0.60
0.33
0.67
0.67
0.26
0.66
0.33
0.56
• 0.62
0.28
0.58
0.43
0.25
0.47
0.27
0.50
0.39
0.29
0.35
0.90
0.65
0.60
0.70
0.57
0.64
0.39
0.86
by Surface Finish
Min Max
10.91
11.12
10.73
12.48
10.47
11.91
10.34
11.57
10.38
12.12
10.16
12.18
10.18
11.67
9.94
11.29
9.91
11.85
10.13
12.19
10.09
12.51
9.11
11.65
9.01
11.44
10.66
10.91
10.35
10.69
10.26
9.48
15.00
14.00
15.00
14.00
14.00
15.00
12.30
15.00
13.15
13.70
13.22
14.00
13.15
15.00
11.10
13.40
12.09
13.52
12.40
13.30
13.15
15.00
10.35
13.40
13.52
16.00
13:52
14.00
13.22
14.00
12.62
13.52
F-91
-------
APPENDIX F
__+ lOmilPad LR
+ + + + lOmilPad WS
*• **•!•»»•» ••
+ + , + + PGA A LR
Each, dot represents up to 2 points :: .
+ + + + PGA A WS
10.0 11.0 12.0 13.0 14.0
Figure F.76 Dotplots for 480 Leakage Measurements by Flux
F-92
-------
APPENDIX F
Each dot represents up to 3 points ::
•. *•••*•
**•••**•••»
+ _ + + + ---PGA B WS-
*
. :
::::::::::
. ::::::::::.;..
*!!**!!!!£*!£ £*!!!
• • £*!!!*•*!!?£!!!!!£
• •••••••••••••••••••••S5JSJS** ••
+ + __ + +_ GullWing LR
+ __ + + + GullWing WS
10.0 11.0 - 12.0 13.0 14.0
Figure F.76 Continued
F-93
-------
APPENDIX F
F.10.7 Stranded Wires
Two 22-gauge stranded wires were hand soldered just to the left of the edge connector. One wire
was soldered directly into the board through holes and the other were soldered to two terminals, E17
and E18. Each wire is 1.5 in long, is silver coated, and has white PTFE insulation. All wires were
stripped, tinned, and cleaned in preparation for the soldering process.
Purpose of the Stranded Wire Experiment
Stranded wires were used to evaluate flux residues and subsequent corrosion.
Table F.40 Summary Statistics for Leakage Currents Test Measurements by Flux
Circuitry
Flux
Mean Median St Dev.
Min
Max
10-MiIPads
PGA A
PGAB
Gull Wing
LR
WS
LR
WS
LR
WS
LR
WS
11.80
13.25
11.18
12.90
10.85
13.01
11.99
12.80
11.68
13.30
11.10
13.00
11.00
13.07
12.02
12.94
0.70
0.56
0.72
0.54
0.79
0.38
0.68
0.78
10.34
11.12
9.94
11.29
9.11
11.65
9.01 '
9.48
15.00
. 15.00
13.22
15.00
13.15 ,
15.00
13.52
16.00
Circuit Description
The 5A 100|j.s pulse used to test the HCLV circuit was injected into each of the stranded wires for
electrical test. A separate PWB trace was connected to each end of the stranded wire. Test wires were
connected to the separate traces allowing to provide the means to measure the voltage drop across the
stranded wires. In this manner, the voltage drop was measured independently from any voltage drop in
the test wires conducting the 5 A pulse to the stranded wires.
Baseline Testing Results for Stranded Wires
Surface finish and flux type had very little effect on the HF TLC frequencies and responses for HF
TLC 50 MHz, HF TLC 500 MHz, HF TLC 1000 MHz, HF TLC Reverse Null Frequency, and HF
TLC Reverse Null Response. Figures F.77 and F.78 provide dotplot displays of 480 measurements for
the two stranded wire voltages. The summary statistics for these responses are given in Table F.41.
8.0
: : :
10.0 12. t
:
) 14. (
) 16.0
18.0
mV
Figure F.77 Dotplots for 480 Voltage Measurements for Stranded Wire 1
(each dot represents up to 11 points)
F-94
-------
APPENDIX F
Circuitry
20.0 22.0 24.0 26.0 28.0 3J}.0
Figure F.80 Dotplots for 476 Voltage Measurements for Stranded Wire 2
(each dot represents 8 points)
Table F.41 Summary Statistics for Stranded Wires Voltage Test Measurements
Mean Median St. Per. Min Max Outliers
Stranded Wire 1
Stranded Wire 2
11.75mV
24.82mV
12.00
25.00
1.60
2.41
8.00
19.00
'18.00
30.00
42,43, 45,45
F. 10.8 Summary Statistics for AH Baseline Measurements
For ease of reference, Table F.42 gives the summary statistics for all 23 electrical responses from
the test PWA. ,
F.10.9 Listing of Components
All functional component types conformed to commercial specifications and were ordered pre-
tinned (to the extent possible). Components were not pre-cleaned before use. A listing of all
components is given in the Table F.43.
F-95
-------
APPENDIX F
Table F.42 Summary Statistics for All Baseline 480 Measurements (sans outliers)
Circuitry Mean Median St. Dev. Min Max
Outliers
:M<%rc<^l
-------
APPENDIX F
Table F.43 Listing of Components for the Test PWA
MFGP/N Description ^'"""l,1
Assembl
ACC916228-2
350-60-2
402-632-38-0110
231-632-A-2
RWR89N10ROFR
M55342M09B10MOM
RLR07C1005FR
. M55342M09B10POM
2309-2-00-44-00-07-0
KA29/127BPMCTH
C1825N474K5XSCxxxx
. C0627104K1X5CS7506
C1825N104K1XRC
C062T105K5X5CSxxxx
C052G130J2G5CR
CDR31BP130BJWR
C052G240J2G5CRXXXX
C0805N240J1GRC37317537
C0805N629B1GSC373 17535
C052G629D2G5CR7535
JM38510/33001B2A
JM38510/33001BCA
QFP80T25
CS1
CKR06
SC1210E7Axxxx
D034
RN65
RN55(sub for CS1, Qty 800)
SR1210E7A
T05
TO220M-3
5162-5013-09
... 131-3701-201
PGA Socket, 18X18 (223 PINS)
6 Split washer
6-32 UNC Mach Screw
6-32 UNC Mach Screw Nut
Resistor, 10 Ohm, Axial
Resistor, 10 Ohm, Surface Mnt
Resistor, lOMeg Axial
Resistor, lOMeg Surface Mount
Swage pin
29 Pin Connector,Pretin
CAP, .47 UF, Surf Mnt
CAP, 0.1 UF, Radial
CAP, 0.1 UF, Surf Mnt
CAP, 1 UF, Radial
CAP, 13 PF, Radial
CAP, 13 PF, Surf Mnt
CAP, 24 PF, Radial
CAP, 24 PF, Surf Mnt
CAP, 6.2 PF ±0.5%, Surf Mnt
CAP, 6.2 PF, ±0.5%, Radial
20PinLCC
14 Pin Dual-In-Line
80 Pin SQ Flat Pack
Cap
Cap
Cap
Diode
Resistor
Resistor
Resistor
Transistor
Transistor
Connector, RF, OMNI Spec
Sub for 5 162-50 13-09
1
3
3
3
7
7
5
5
17
1
7
. 7.
7
• 7
1
1
1
1
1
1
1
2
1
1
2
13
13
1
5
18
4 •
3
10
10
yCF Supplier
AMP
Barnhill Bolt
BarnhillBolt
Barnnili Bolt
Dale
Dale
Dale
Dale
Harrison HEC
Hypertonics
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
- Kemet
TI (808810.1001)
TI (808810.1)
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
TTI
Penstock
F-97
-------
APPENDIX F
F.ll Design for the Environment Printed Wiring Board Project Performance
Demonstration Methodology for Alternative Surface Finishes
Note: This methodology is based on input from members of a Performance Demonstration Technical
Workgroup, which includes representatives of the printed wiring board (PWB) industry manufacturers,
assemblers, and designers; industry suppliers; public interest group; Environmental Protection Agency
(EPA); the University of Tennessee Center for Clean Products and Clean Technologies; and other
stakeholders. As the testing continues, there may be slight modifications to this methodology.
I. OVERVIEW
A. Goals
The U.S. Environmental Protection Agency's (EPA's) Design for the Environment (DfE) Printed
Wiring Board (PWB) Project is a cooperative partnership among EPA, the PWB industry, public
interest groups, and other stakeholders. The project encourages businesses to incorporate
environmental concerns into their decision-making processes, along with the traditional parameters of
cost and performance, when choosing which technologies and processes to implement. To accomplish
this goal, the DfE PWB Project collects detailed data on the performance, cost, and risk aspects of one
"use cluster" or manufacturing operation, and makes it available to all interested parties. This use
cluster focuses on surface finishes used in PWB manufacturing. Analyses on the performance, cost,
and risk of several alternative surface finishes will be conducted throughout this project, and the results
will be documented in the final project report, titled the Cleaner Technologies Substitutes Assessment
or CTSA. This methodology provides the general protocol for the performance demonstration portion
of the DfE PWB Project. The CTSA is intended to provide manufacturers and designers with detailed
information so that they can make informed decisions, taking environmental and health risks into
consideration, on what process is best suited for their own facility.
Surface finishes are applied to PWBs to prevent oxidation of exposed copper on the board, thus
ensuring a solderable surface when components are added at a later processing stage. Specifically, the
goals of the DfE PWB Surface Finishes Project are:
1) to standardize existing information about surface finish technologies;
2) to present information about surface finish technologies not in widespread use, so PWB
manufacturers and designers can evaluate the environmental and health risks, along with the cost and
performance characteristics, among different technologies; and
3) to encourage PWB manufacturers and designers to follow the example of this project and evaluate
systematically other technologies, practices, and procedures in their operations that affect the
environment.
B. General Performance Demonstration Plan
The most widely used process for applying surface finishes in commercial PWB shops is hot air solder
leveling (HASL). In this process, tin-lead is fused onto exposed copper surfaces. This process was
selected as the focus of the Design for the Environment Project because HASL is a source of lead
waste in the environment and because there are several alternative surface finishes available on the
market. A comprehensive evaluation of these technologies, including performance, cost, and risk,
however, has not been conducted. In addition, a major technical concern is that the HASL process
does not provide a level soldering surface for components.
F-98
-------
- ' APPENDIX F
The general plan for the performance demonstration portion of the Project is to collect data on
alternative surface finish processes during actual production runs at sites where the processes are
already in use. Demonstration facilities will be nominated by suppliers. These sites may be customer
production facilities, customer testing facilities (beta sites), or supplier testing facilities, in that order of
preference. Each demonstration site will receive standardized test boards which they will run through
their surface finish operation during their normal production operation.
The test vehicle design will be tested on the test board designed by the Sandia National Laboratory
Low-Residue Soldering Task Force (LRSTF).. The same test vehicle was used by the Circuit Card
Assembly and Materials Task Force (CCAMTF). CCAMTF is a joint industry and military program
evaluating several alternative technologies including Organic Solderability Preservative (OSP),
Immersion Silver, Electroplated Palladium/Immersion Gold, Electroless Nickel/Immersion Gold, and
Electroplated Palladium. CCAMTF conducted initial screening tests on coupons for each of these
surface finishes, however, they will conduct functionality tests only for the OSP (thick), Electroplated
Palladium/Immersion Gold, and Immersion Silver technologies.
H. PERFORMANCE DEMONSTRATION PROTOCOL
A. Technologies to be Tested
The technologies that the DfE Project plans to test include:
1. HASL (baseline)
2. OSP-Thick
3. Immersion Tin
4. Immersion Silver ' . • . . .
5. Electroless Nickel/Immersion Gold
6. Nickel/Palladium/Gold "
B. Step One: Identify Suppliers and Test Sites/Facilities
Performance Demonstration Technical Workgroup members identified suppliers of the above product
lines. Any supplier of these technologies who wanted to participate was eligible to submit its product
line, provided that it agreed to comply with the testing methodology and submit the requested
information, including chemical formulation data. All proprietary information submitted is bring
handled as Confidential Business Information. For each product line submitted, the supplier
completed a Supplier Data Sheet detailing information on the chemicals used, equipment requirements,
waste treatment recommendations, any limitations of the technology, and other information on the
product line. .
Performance demonstration sites were nominated by suppliers. They identified sites that are currently
using their alternative surface finish product line in the following order of preference: '
customer production facilities (first preference)
beta sites — customer testing facilities (second preference)
supplier testing facilities (third preference)
The final number of product lines evaluated for each type of alternative surface finish was determined
based on the number of suppliers interested in participating and on the resources available. Each
accepted product line was tested at one or two sites. If a supplier has more than one substantially
F-99
-------
APPENDIX F
different product line within a technology, the supplier was allowed to submit names of test facilities
for each of the products.
C. Step Two: Fabricate Test Vehicles
Test board were fabricated based on the Sandia National Laboratory Low-Residue Soldering Task
Force (LRSTF) test board design. This general design was also used in the CCAMTF testing. For the
DfE Project, uncoated test boards with comb pattern spacing of 8 mil, 12 mil, 16 mil, and 20 mil will
be used.
All test boards are of the same design, and were fabricated at a single shop to minimize the variables
associated with board production. All manufacturing steps, up to but not including the soldennask
application, were completed by the test board fabricator. For each supplier's product line, 24 boards
were shipped to the demonstration site where the alternative surface finish was applied, beginning with
the soldermask application step.
The design of the LRSTF PWB was based on input from a large segment of the manufacturing
community, and-thus reflects the multiple requirements of the commercial sector. Each quadrant of the
LRSTF PWA contain one of the following types of circuity: .
High-current low-voltage (HCLV)
High-voltage low current (HVLC)
High speed digital (BSD)
High frequency (HF)
The components in each quadrant represent two principal types of soldering technology:
Plated through hole (PTH) - leaded components are soldered through vias in the circuit board by
means of a wave soldering operation.
Surface mount technology (SMT) - components manufactured with solder tips on two of their
opposite ends are temporarily attached to the substrate with an adhesive and then they are soldered to
pads on the circuit board by passing the circuit board through a reflow oven to reflow the solder tips.
The LRSTF PWA also, has two stranded wires (SW) that are secured to the circuit board with hand
soldering, such as used hi repair operations. This assembly also contains other networks that are used
to monitor current leakage.
D. Step Three: Collect Background Information
After the suppliers identified appropriate test facilities and completed a supplier data sheet, an
independent observer contacted the designated facilities. The observer scheduled a date for the on-site
performance demonstration. A questionnaire was sent to each facility prior to the site visit to collect
information on the surface finish technology used and background information on the facility, such as
the size and type of product produced. On the day of the performance demonstration, the observer
reviewed the background questionnaire and discussed any ambiguities with the facility contect.
F-100
-------
APPENDIX F
E. Step Four: Conduct the Surface Finish Performance Demonstration
After test boards were distributed to the demonstration sites, the surface finish performance
demonstrations were conducted. The surface finish was applied to the test boards as part of the normal
production run at the facility. The test boards were placed in the middle of the run to reflect actual
production conditions. The facility applied the solder mask it normally uses in production. The usual
process operator operated the line to minimize error due to unfamiliarity with the technology. All test
boards were processed in the same production ran.
On the day of the performance demonstration, the observer collected data on the surface finish process.
During the demonstration, the observer recorded information on surface finish technology
performance., including information on chemicals, equipment, and waste treatment methods used. In
addition, other information needed for the performance, cost, or risk analyses, as described below, was
collected.
1. Product Cost: A cost per square foot of panel processed will be calculated, this number will be
based on information provided by product suppliers, such as purchase price, recommended bath
life and treatment/disposal methods, and estimated chemical and equipment costs per square foot
panel per day. Any "real world" information from PWB manufacturers, such as actual dumping
frequencies, treatment/disposal methods, labor requirements, and chemical and equipment costs,
will be collected during performance demonstrations, as required for use in the cost analysis. The
product cost may differ for difference shop throughput categories.
2. Product Constraints: Information on any incompatibilities such as soldermask, flux, substrate
type, or assembly process will be included. This information will be submitted by the suppliers
and may also be identified as a result of the performance demonstrations.
3. Special storage, safety, and disposal requirements: Information on flammability or special
storage requirements of the chemicals used in the process will be requested from the suppliers.
Suppliers will provide recommendations on disposal or treatment of wastes associated with the use
of their product lines. Information on these issues was also collected from participating facilities
during the performance demonstrations. The storage and disposal costs will be a factor in
determining the adjusted cost of the product. This project does not entail a life cycle analysis for
disposal of the boards.
4. Ease of use: During the performance demonstration, the physical effort required to use the various
surface finishes effectively will be qualitatively assessed based on the judgement of the operator in
comparison to the baseline technology, HASL. Specific questions such as the following will be
asked: What process operating parameters are needed to ensure good performance? What are the
ranges of those parameters, and is there much flexibility in the process steps? How many hours of
training are required to use this type of surface finish?
5. Duration of Production Cycle: The measured time of the surface finish application process and
the number of operators required will be recorded during the performance demonstration. This
information will be used to measure the labor costs associated with the use of the product line.
Labor costs will be based on the operator time required to run the process using an industry
standard worker wage. The process cycle has been defined as the activities following soldermask
application up to, but not including, gold tab plating. The facilities participating in the
performance demonstration will use the same soldermask they typically use in production
conditions. The observer recorded the type of soldermask used, and information on the facilities'
experiences with other soldermasks to determine if any known incompatibilities exist.
6. Effectiveness of Technology, Product Quality: The performance characteristics of the
assembled boards will be tested after all demonstrations are complete and the boards are assembled
F-101
-------
APPENDIX F
with the functional components. Circuit electrical Performance will be tested to assess the circuit
performance of the functional test vehicle under applicable environmental stress. Circuit
Reliability Testing (functional tests) conditions will include Thermal Shock and Mechanical
Shock. These tests are described in greater detail in Step 5. Qualitative information on shelf life
considerations were collected through the performance demonstrations, where applicable.
7. Energy and Natural Resource Data: Information will be collected from the suppliers and during
the performance demonstrations to evaluate the variability of energy consumption for the use of
different surface finishes. The analysis will also address material use rates and how the rates vary
with the different surface finishes.
8. Exposure Data: Exposure data will be used to characterize chemical exposures associated with
the technologies. Exposure information collected during the performance demonstration may be
supplemented with data from other sources, where available.
F. Step Five: Assemble and Test the Boards
After the surface finish was applied to the test boards at each demonstration facility, the facility sent
the processed boards to one site for assembly. Two different assembly processes were used: a halide-
free, low-residue flux and a halide-containing, water-soluble flux. Table 1 shows the different
assembly methods, and number of test vehicles used for each method. The boards were not assembled
as originally planned, resulting in the .uneven distribution of assembly methods.
Table 1: Test Vehicle Distribution by Site and Flux
Site#
1
2
6
3
13
16
4
5
10
11
8
9
7
12
14
Surface Finishes*
HASL
HASL
HASL
HASL Totals
OSP-Thick
(DSP-Thick
OSP-Thick
OSP Totals
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin Totals
Immersion Silver
Immersion Silver
Immersion Silver Totals
Electroless Ni/Immersion Au
Electroless Ni/Immersion Au
Electroless Ni/Immersion Au
NI/Au Totals
Subtotals
# of Boards
Assembled with Low
Residue Flux
8
0
8
16
4
8
8
20
0
4 •
8
8
20
0
8
8
0
8
4
12
84
# of Boards
Assembled with
Water Soluble Flux
8
8
0
16
8
8
0
16
8
8
0
0
16
8
4
12
8
0
8
16
80
Total test boards: 164
Total Boards by
Site and by Surface
Finish
16
8
8
32
12
16 -
-8
36
8
12
8
8
36
8
12
20
8
8
12
28
* Corresponding board identification numbers are listed in Appendix A.
Following assembly, the performance characteristics of the assembled boards will be tested. Testing
will include Circuit Electrical Performance testing and Circuit Reliability Testing.
F-102
-------
APPENDIX F
Circuit Electrical Performance
TMs test assesses the circuit performance of a functional test vehicle under applicable environmental
stress. The assembled test vehicles will be exposed to 8.5 ° C at 85% relative humidity for 3 weeks.
The assemblies will be tested prior to exposure, and at the end of three weeks of exposure. Good
experimental design practices will be followed to control extraneous sources of variation. For
example, the assemblies will be placed randomly in the test chamber. If all assemblies cannot be
accommodated in the test chamber at the same tune, they will be randomized to maintain balance
among the experimental factors at each test time. A staggered ramp will be used to prevent
condensation (during ramp-up, the temperature will be raised to test level before the humidity is raised
and the procedure will be reversed during ramp-down). The pre-tests and post-tests will be identical.
Circuit Reliability Testing
The same test vehicles used to test circuit electrical performance will be used for the circuit reliability
tests, which include:
Thermal Shock
Mechanical Shock
The electrical functionality of the LRSTF PWA will be evaluated through 23 electrical responses, as
follows:
HCLV PTH voltage
HCLV SMT voltage
Stranded wire 1 voltage
Stranded wire 2 voltage
HVLC PTH current
HVLC SMT current
10-mil spaced pads current leakage
PGA A current leakage
PGA B current leakage
Gull wing current leakage
HSD PTH total propagation delay
HSD SMT total propagation delay
HF LPF PTH 50 MHz response
HF LPF PTH frequency response at -3 dB
HF LPF PTH frequency response at -40 dB
HP LPF SMT 50 MHz response
HF LPF SMT frequency response at -3 -dB
HF LPF SMT frequency response at -40 dB
HF TLC 50 MHz forward response
HF TLC 500 MHz forward response
HF TLC 1000 MHz forward response
HF TLC reverse null frequency
HF TLC reverse null response
Table 2 shows the total number of electrical responses that will be measured.
Table 2. Number of Tests to be Conducted
Test Environment
85/85
Thermal Shock
Mechanical Shock
Totals
Number of
PWBs
164
164
Number of Test
Times
2
1
1
4
Number of
Tests
164x2 = 328
164 x 1 = 164
164 x 1 = 164
656
Number of Electrical
Responses Measured
164x2x23-7,544
164x1x23 = 3,772
164 x 1 x 23 = 3,722
15,088
G. Analyze Data and Present Results
The details of the data analysis and results are presented in the "Technical Proposal for this project, in
Appendix B.
F-103
-------
APPENDIX F
PERFORMANCE DEMONSTRATION PARTICIPANT REQUIREMENTS
A. From the Facilities/Process Operators:
1. Participating facilities were contacted by the project observer to arrange a convenient data for the
performance demonstration. The observer sent a fact sheet describing the facility's role in the
project.
2. Each facility was asked to complete a background questionnaire prior to the scheduled date of the
performance demonstration and return it to the observer.
3. Each facility was asked to make its process line/process operators available to run the 24 test
boards on the agreed upon date.
4. The process operator met with the independent observer before running the test boards through the
line to explain the unique aspects of the line to the observer. The process operator was asked to be
available to assist the independent observer in collecting information about the line.
B. From the Suppliers of the Process Line Alternatives:
1. Suppliers were asked to submit product data sheets, on which they provided information on
product formulations, product constraints, recommended disposal/treatment etc. The information,
including chemical formulation information, was requested prior to testing. Any proprietary
information was submitted to the University of Tennessee as Confidential Business Information.
2. Suppliers were asked to identify and contact the demonstration sites.
3. Suppliers were asked to attend the on-site performance demonstration if they wishes to do so, but
they were not required to attend.
Attachment A to this Methodology lists "Identification Numbers for Assembled Boards." To
conserve space this information as not been reprinted as part of the CTSA.
Attachment B to this Methodology is the "Technical/Management Proposal for Validation of
Alternatives to Lead Containing Surface Finishes." This Attachment contains the testing and analysis
methodology submitted by Dr. Ronald L. Inman, President, Southwest Technology Consultants in
Albuquerque, MN. Dr. Inman's methodology and results are presented in Chapter 6 of the CTSA and
in Appendix F, and therefore, Attachment B of the Methodology is not repeated here.
F-104
-------
Appendix G
Supplemental Cost Analysis Information
-------
-------
APPENDIX G
G-1 Example Graphic Representation of Cost Simulation Model
G-2 Bath Replacement Criteria for Surface Finishing Processes
G-3 Bills of Activities for Surface Finishing Processes
G-4 Simulation Model Outputs for Surface Finishing Processes
G-5 Chemical Costs by Bath for Individual Surface Finishing Processes
G-6 Total Materials Cost for Surface Finishing Processes
G-l
-------
APPENDIX G
G-l. Example Graphic Representation of Cost Simulation Model
G-2
-------
APPENDIX G
G-2 Bath Replacement Criteria for Surface Finishing Processes
Process: HASL
Chemical Bath
Cleaner
Microetch
Flux
Solder
Bath Replacement Criteria*
(ssf/gal)
750
570
NAb
NAb
a Values were selected by averaging the replacement criteria for similar bath types from other alternatives.
b This bath is refilled or continuously maintained through chemical additions rather than replaced; The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
Process: Electroless Nickel/Immersion Gold
Chemical Batb
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Bath Replacement Criteria3
(sstfgal) i
750
570
830
1,500
130
890
a Values were determined from data provided by two electroless nickel/immersion gold suppliers. To convert to units of
• racks per bath replacement for non-conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack.
Process: Electroless Nickel/Electroless Palladium/Immersion Gold
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator -
Electroless Palladium
Immersion Gold
Bath Replacement
(ssffgal)
Criteria3
750
570
830
1,500
130
1,200
150
890
Values were determined from data provided by two electroless nickel/immersion gold suppliers and one electroless
nickel/palladiuni/immersion gold supplier. To convert to units of racks per bath replacement for non-conveyorized
processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack.
G-3
-------
APPENDIX G
Process: OSP
' Chemical Bath [
Cleaner
Microetch
OSP
Bath Replacement Criteria3
(ssffgal)
750
570
NAb
* Values were determined from data provided by two OSP suppliers. To convert to units of racks per bath replacement
for non-conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack. To convert to units of panels per
bath replacement for conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation. •
Process: Immersion Silver
Chemical Bath
Cleaner
Microetch
Predip
Immersion Silver
Bath Replacement Criteria3
(ss#gal)
'750
570
1,000
NAb
Values were determined from data provided by two OSP suppliers. To convert to units of panels per bath replacement
for conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the Initial bath make-up for the purposes of the computer simulation.
Process: Immersion Tin
Chemical Bath %
Cleaner
Microetch
Predip
Immersion Tin
Bath Replacement Criteria3 •
(ssffgal)
750
570
1,250
NAb
Values were determined from data provided by two OSP suppliers. To convert to units of racks per bath replacement
for non-conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack. To convert to units of panels per
bath replacement for conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
G-4
-------
APPENDIX G
G-3 Bills of Activities for Surface Finishing Processes
Activities Associated with the Bath Setup
-r"" '.I..- " .j^S^i^et^^* " "'" " V
Wear masks, goggles, rubber gloves, and suitable clothing
Go to storage area
Locate protective equipment
Put on protective equipment
Return to tank
Put in base liquid (usually water)
Open water valve
Wait for measured amount
Close water valve
Document water amount/level
Mix the bath solution
Open the chemical containers
Add the chemicals to the bath
Turn on the agitator
Wait for mixing
Turn off the agitator
Titrate sample
Document
Repeat as necessary
Flush containers
Turn on water valve
Spray containers
Turn off water valve
Place empty container in storage area
Take container to storage
Documentation
Return to tank
Total =
! -C^Jhtfssr " -
$/bath setup
labor
labor
labor
protective equipment
labor
$/bath setup
labor
labor
labor •
labor
S/bath setup
labor
labor
labor
labor
labor
labor
labor
labor
S/bath setup
labor
labor
labor
S/bath setup
labor
labor
labor
Sper testing
1 C0s#Ae*l«#
$2.50
$2.60
$5.00
$3.00
$2.00
$15.10
G-5
-------
APPENDIX G
Activities Associated with the Tank Cleanup
- ^&X$yi$$$&$rfa$®£ "*-t "/ ' ,",%, ',
Rinse with water
Obtain spray/rinse equipment
Turn water on
Spray equipment
Turn water off
Obtain scrubbing and cleaning tools
Go to storage area
Find necessary tools
Return to tank
Hand scrub tank
Put on gloves, choose tool
Scrub tank
Return cleaning tools
Go* to the storage area
Place tools in correct place
Return to tank
Spray according to schedule
Wait for time to elapse before spraying
Obtain spray equipment
Turn spray on
Spray all cleaning solution from tank
Turn spray off
Operator opens control valve
Find correct control valve
Open valve
Water goes to treatment facility
Wait for water to drain
Operator closes control valve
Locate correct control valve
Close valve
Total =
£a&$titir&
S/cleanup
labor
labor
labor
labor
S/cleanup
labor
labor
labor
$/cleanup
labor
labor
cleaning supplies
$/cIeanup
labor
labor
labor
S/cleanup
labor
labor
labor
labor
labor
S/cleanup
labor
labor
S/cleanup
labor
S/cleanup
labor
labor
Sper testing
C&rtflfc&rtfrJ
$25.00
$1.00
$30.00
$1.25
$5.00
$1.00
$2.75
$1.00
$67.00
G-6
-------
APPENDIX G
Activities Associated with Sampling and Testing
" %% " ," ", , MfMt^JDe^rJptioji " „ ' , ,
Get sample
Go to the line
Titrate small sample into flask
Transfer to lab
Test sample
Request testing chemicals
Document request
Locate chemicals
Add chemicals to sample
Mix
Document the results
Return testing chemicals
Relay information to line operator
Return to line
Inform operator of results
Document
Total =.
! C&$**&i¥*r
S/testing
labor
labor
materials
labor
$/testing
labor
labor
labor
labor
materials
labor
labor
labor
S/testing
labor .
labor
labor
Sper testing
r
-------
APPENDIX G
Activities Associated with Filter Replacement
^tt^^&J&ldptfe* ? '','"'
Check old filter
Pull canister from process
Inspect filter
Decide if replacement is necessary
Get new filer
Go to storage area
Locate new filters
Fill out paper work
Return to tank
Change filter
Pull old filter from canister
Replace with new filter
Replace canister
Fill out paper work
Dispose of old filter
Take old filter to disposal bin/area
Dispose of filter
Return to tank
Fill out paper work
Total =
Ofci&rwer -
S/repIacement
labor
labor
labor
S/replacement
labor
labor
labor
labor
$/replacement
labor
labor
filter
labor
labor
S/replacement
labor
labor
labor
labor
Sper replacement
I €wt&#frfferj
$1.50
$1.75
$12.25
$2.00
$17.50
G-8
-------
APPENDIX G
Activities Associated with Transportation
AeSyiijr Description
Paperwork and maintenance
Request for chemicals
Updating inventory logs
Safety and environmental record keeping
Move forklift to chemical storage area
Move to forklift parking area
Prepare forklift to move chemicals
Move to line container storage area
Prepare forklift to move line container .
Move forklift to chemical storage area
Locate chemicals in storage area
Move forklift to appropriate areas
Move chemical containers from storage to staging
Move containers from staging to storage
Preparation of chemicals for transfer
Open chemical container(s)
Utilize correct tools to obtain chemicals
Place obtained chemicals in line containers)
Close chemical containers)
Place line containers) on forklift
Transport chemicals to line
Move forklift to line
Unload line containers) at line
Move forklift to parking area
Transport chemicals from line to bath
Move line container(s) to bath
Clean line container(s)
Store line container(s) in appropriate area
Total =
Cost j&rfy«r
S/transportation
labor
labor
labor
S/transportation
labor
labor
labor
labor
labor
S/transportation
labor
labor
filter
S/transportation
labor
labor
labor
labor
labor
S/transportation
labor
labor
labor
S/transportation
labor
labor
labor
Sper testing
CsstTAetivify
$1.10
$3.22
$1.15
$1.78
$1.15
$.88
$9.28
G-9
-------
APPENDIX G
G-4 Simulation Model Outputs for Surface Finishing Processes
NAME:
Throughput:
HASL, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 17831.4 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
TaktTime
Time in system
Counters
5.7866
19.957
(Corr)
4.8613
1.4700
7.9560
141.10
168.71
3080
3081
Identifier Count Limit
Parts Done 3081 ' Infinite
Frequencies
Identifier
STATE (Microetch3_R)
STATE (CIeaner3_R)
STATE (flux3_R)
STATE (solder3_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3075
3075
9
2251
2250
7
3081
3082
1
3081
3082
1
AvgTime
1.4728
3.9279
136.00
4.7494
2.7503
136.00
.18000
5.5615
136.00
.12600
5.6155
136.00
Percent
25.40
67.74
6.86
59.96
34.70
5.34
3.11
96.13
0.76
2.18
97.06
0.76
Percent
25.40
67.74
6.86
59.96
34.70
5.34
3.11
96.13
0.76
2.18
97.06
0.76
G-10
-------
APPENDIX G
NAME:
Throughput:
HASL, non-conveyorized
60,OOOKssf
ARENA Simulation Results -
Replication ended at time: 2876.64 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in system
Counters
3.8531
89.058
.69813
(Corr)
3.4700
7.9560
139.47
279.95
710
711
Identifier Count Limit .
Parts Done 711 Infinite
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (flux3_R)
STATE (solder3_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
577
575
3
3
1
2 .
711
712
1
711
712
1
AvgTime
1.8113
2.4756
136.00
822.39
137.47
136.00
.18000
3.6694
136.00
.12600
3.7233
136.00
Percent
36.33
49.48
14.18
85.77
4.78
9.46
4.45
90.82
4.73
3.11
92.16
4.73
Percent
36.33
49.48
14.18
85.77
4.78
9.46
4.45
90.82
4.73
3.11
92.16
4.73
G-ll
-------
APPENDIX G
NAME:
Throughput:
HASL, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 2348.82 min.
Tally Variables
Identifier Average Half Width
Minimum
Maximum
Observations
Takttime
Time in system
Counters
.19281
19.009
.02704
(Corr)
.16654
4.9888
136.00
140.82
10600
10601
Identifier Count Limit
Depart 33_C . 10601 Infinite
Frequencies
Identifier
STATE (CleanerJR)
STATE (so!der_R)
STATE (flux_R)
STATE (Microetch_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
9825
9823
2
10601
10601
1
10601
10601
1
10601
• 10601
1
AvgTime
.00539
.17549
136.00
.00500
.17544
136.00
.00500
.17544
136.00
.00500
.17544
136.00
Percent
2.59
84.14
13.28
2.59
90.77
6.64
2.59
90.77
6.64
2.59
90.77
6.64
Percent
2.59
84.14
13.28
2.59
90.77
6.64
2.59
90.77
6.64
2.59
90.77
6.64
G-12
-------
APPENDIX G
NAME:
Throughput:
HASL, conveyorized
260,OOOKssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
Identifier Average
8908.24 min.
Half Width Minimum
Maximum
Observations
Time in system
Takttime
21.188
.18000
10.277
(Corr)
4.9888
.16654
140.91
136.00
45936
45935
Counters
Identifier Count Limit
Depart.33_C 45936 Infinite
Frequencies
Identifier
STATE (CleanerJR.)
STATE (solder_R)
STATE (Microetch_R)
STATE (flux_R)
Category .
BUSY
IDLE
• FAILED
. BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
42056
42051
6
45936
45936
1
45936
45932
6
45936
45937 '
1
AvgTime
.00546
.17506
136.00
.00500
.17506
136.00
.00500
.16027
, 136.00
.00500
.17506
136.00
Percent
2.73
87.56
9.71
2.73
95.65
1.62
2.73
87.56
9.71
2.73
95.65
1.62
Percent
2.73
87.56
9.71
2.73
95.65
1.62
2.73
87.56
9.71
2.73
95.65
1.62
G-13
-------
APPENDIX G
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
260,OOOKssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
Identifier Average
114576.0 min.
Half Width Minimum
Maximum
Observations
Time in system
TaktTime
Counters
116.79
38.848
1.0484
(Corr)
106.86
17.830
278.21
131.33
308
3080
Identifier Count
Parts Done 3081
Frequencies
Identifier
STATE (Acid Dip_R)
STATE (Catalyst_R)
STATE (CleanerJR.)
STATE (Electroless Palla
STATE (Immersion Gold_R
STATE (PreinitiatorJR.)
STATE (Electroless Nicke
•
STATE (Microetch_R)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3073
3070
4
3075
3070
6
3069
3062
7
3008
2975
34
2803
2798
6
3081
3082
5
2872
2833
40
3064
3056
9
AvgTime
1.6342 •
37.226
113.00
3.7372
35:045
113.00
3.4835
35.362
113.00
4.7321
34.179
113.00
19.598
22.926
113.00
2.3000
36.375
113.00
1'9.663
20.743
113.00
1.4781
37.373
113.00
Percent
4.19
95.43
0.38
9.60
89,84
0.57
8.93
90.41
0.66
11.89
84.91
3.21
45.87
53.56
0.57
5.92
93.61
0.47
47.16
49.07
3.77
3.78
95.37
0.85
Percent
4.19
95.43
0.38
9.60
89.84
0.57
8.93
90.41
0.66
11.89
84.91
3.21
45.87
53.56
0.57
5.92
93.61
0.47
47.16
49.07
3.77
3.78
95.37
0.85
G-14
-------
APPENDIX G
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 25807.8 min.
Tally Variables
Identifier Average
Half Width Minimum
Maximum
Observations
Time in system
Takt Time
Counters
115.87
38.929
1.7495
(Corr)
106.86
17.830
199.39
131.33
711
710
Identifier Count
Parts Done. 711
Frequencies
Identifier
STATE (Acid Dip_R)
STATE (Cleaner_R)
STATE (Catalyst_R)
STATE (Electroless Palla
STATE (Immersion Gold_R
STATE (Preinitiator_R)
STATE (Electroless Nicke
STATE (Microetch_R)
Limit
Infinite
Category
BUSY
IDLE
FAILED .
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
711
712
1
709
707
2
707
706
2
695
688
8
652
651
2
711
711
1
670
663
9
707
706
3
AvgTime
1.6300
37.269
113.00
3.4797
35.522
113.00
3.7511
35.311
113.00
4.7263
34.329
113.00
19.443
22.895
113.00
2.3000
, 36.651
113.00
19.451
20.751
113.00'
1.4783
37.427
113.00
Percent
4.17
95.43
0.41
8.87
90.32
0.81
9.54
89.65
0.81
11.81
84.94
3.25
45.59
53.60
0.81
5.88
93.71
0.41
46.87
49.48
3.66
3,76
95.02
1.22
Percent -
4.17
95.43
. 0.41
8.87
90.32
0.81
9.54
89.65
0.81
11.81
84.94
3.25
45.59
53.60
0.81
5.88
93.71
0.41
46.87
49.48
3.66
3.76
95.02
1.22
G-15
-------
APPENDIX G
NAME:
Throughput:
Nickel/Gold, non-conveyorized
260,OOOKssf
ARENA Simulation Results
Replication ended at time: 86437.5 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
TaktTime
Time in system
Counters
27.062
98.948
1.2220E-14
2.0602
17.830
86.100
134.33
286.16
3080
3081
Identifier Count
Parts Done 3081
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Acid Dip2_R)
STATE (Electroless Nickel)
STATE (CIeaner2_R)
STATE (Catalyst2_R)
STATE (Immersion Gold2_
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3056
3048
9
3068
3065
4
2448
2409
40
3063
3056
7
3067
3062
6
2966
2961
6
AvgTime
1.4820
25.546
116.00
1.6369
25.432
116.00
23.069
9.2664
116.00
3.4903
23.538
116.00
3.7470
23.268
116.00
18.521
9.3911
116.00
Percent
5.43
93.32
1.25
6.02
93.42
0.56
67.69
26.75
5.56
12.81
86.21
0.97
13.77
85.39
0.83
65.84.
33.33
0.83
Percent
5.43
93.32
1.25
6.02
93.42
0.56
67.69
26.75
5.56
12.81
86.21
0.97
13.77 •
- 85.39
0.83
65.84
33.33
0.83
G-16
-------
APPENDIX G
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
19427.7mm.
Identifier
Average
Half Width Minimum
Maximum
Observations
Takt Time
Time in system
Counters
27.150
95.321
(Corr)
4.1505
17.830
86.100
134.33
193.43
710
711
Identifier . Count
Parts Done 71 1
Frequencies
Identifier
STATE (Electroless Nicke
STATE (Acid Dip2_R)
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (Catalyst2_R)
- *
STATE (Immersion Gold2_
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
605
597
9
711
712
1
705
704
3
708
706
2
711
710
2
684
683
2
AvgTime
21.541
8.9632
116.00 "
1.6300
25.495
11 6:00
1.4825
25.617
116.00
3.4847
23.694
116.00
3.7300
23.300
116.00
18.533
9.5440
116.00
Percent
67.08
27.54
5.37
5.97
93.44
0.60
5.38
92.83
1.79
12.70
86.11
. 1.19
13.65
85.16
1.19
65.25
33.55
1.19
Percent
67.08
27.54
5.37
5.97
93.44
0.60
5.38
92.83
1.79
12.70
86.11
1.19
13.65
85.16
1.19
65.25
33.55
1.19
G-17
-------
APPENDIX G
NAME:
Throughput:
OSP, non-conveyorized
260,OOOKssf
ARENA Simulation Results
Replication ended at time: 14371.9 min.
Tally Variables
Identifier Average Half Width
Minimum
Maximum
Observations
TaktTime
Time in System
Counters
4.7599
399.53
.59985
(Corr)
4.6200
21.330
150.67
513.90
3080
3081
Identifier , Count Limit
Depart 7_C . 3081 Infinite
Frequencies
Identifier
STATE (Cleaner_R)
STATE (Osp_R)
STATE (Microetch_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
2301
2294
7
3081 .
3081
1
2711
2703
9
AvgTime
4.6462
1.2850
149.00
1.6700
3.0469
. 149.00
1.6706
3.2600
149.00
Percent
72.82
20.08
7.10 ,
35.04
63.94
1.01
30.85
60.02
9.13
Percent
72.82
20.08
' 7.10
35.04 _
63.94
1.01
30.85
60.02
9.13
G-18
-------
APPENDIX G
NAME:
Throughput:
OSP, non-conveyorized
60,OOOKssf
ARENA Simulation Results
Replication ended at time: 3731.92 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
5.0236
172.58
.57885
(Corr)
4.6200
21.330
150.47
322.15
710
711
Identifier Count Limit
Depart 7_C 711 Infinite
Frequencies
Identifier
STATE (Cleaner_R)
STATE (Osp_R)
STATE (Microetch_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
581
579
2
711
. 711
1
619
618
3
AvgTime
4.2464
1.6696
149.00
L6700
3.3692
149.00 '
1.6884
, 3.6241
149.00
Percent
66.11
25.90
7.99
31.82
64.19
3.99
28.01
60.02
11.98
Percent
66.11
25.90
7.99
31.82
64.19
3.99
28.01
60.02
11.98
G-19
-------
APPENDIX G
NAME:
Throughput:
OSP, conveyorized
260,OOOKssf
ARENA Simulation Results
Replication ended at time: 6568.83 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takttime
Time in system
Counters
Identifier
Depart 22_C
Frequencies
Identifier
.14724
30.442
Count
45937
.01562
14.465
Limit
Infinite
.13961
5.1777
149.00
154.12
45936
45937
Category
Number
AvgTime
Percent
Percent
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (osp_R)
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
45937
45932
6
40587
40582
6
45937
45937
1
.00500
.12290
149.00
.00566
.13910
149.00
.00500
.13911
149.00
3.39
83.40
13.21
3.39
83.40
13.21
3.39
94.41
2.20
3.39
83.40
13.21
3.39
83.40
13.21
3.39
94.41
2.20
G-20
-------
APPENDIX G
NAME:
Throughput:
OSP, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 2002.0 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
.15805
. 27.077
.03019
(Corr)
.1356
5.1777
149.00
154.07
1060
10600
Identifier Count Limit '
Depart 22_C 10601 Infinite '
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (OSP_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
10601
10601
1
9531
9530
2
10601
10601
1
AvgTime
.00500 •
..16979
149.00
'.00556
.17324
149.00
.00500
.16979
149.00
Percent
2.65
89.91
7.44
2.65
82.47
14.89
2.65
89.91
7.44
Percent
2.65
89.91
7.44
2.65
82.47
14.89
2.65
89.91
7.44
G-21
-------
APPENDIX G
NAME:
Throughput:
Immersion Silver, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
5425.08 min.
Identifier
Average
Half Width Minimum
Maximum
Observations
Time in System
Takttime
Counters
14.998
.51074
5.9815
(Corr)
11.189
.48953
125.07
113.99
10601
10600
Identifier Count
depart 44_C . 10601
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (Immersion Silver)
STATE (prodip_R)
Limit
Infinite
Category
BUSY
IDLE'
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY.
IDLE
FAILED
Number
10601
10601
1
10372
10370
. 2
10601
10601
1
10601
10600
2
AvgTime
.00500
.49600
114.00
.00511
.49605
114.00
.00500
.49600
114.00
.00500
.48529
114.00
Percent
0.98
96.92
2.10
0.98
94.82
4.20
0.98
96.92
2.10
0.98
94.82,
4.20
Percent
0.98
96.92
2.10
0.98
94.82
4.20
0.98
96.92
2.10
0.98
94.82
4.20
G-22
-------
APPENDIX G
NAME:
Throughput:
Immersion Silver, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 26206.7 min.
Tally Variables
Identifier • Average Half Width Minimum
Maximum
Observations
Time in System
Takt Time
Counters
18.921
.50495
4.1632
(Corr)
11.189
.48995
238.69
114.03
45937
45936
Identifier Count Limit .
depart 44_C 45937 Infinite ' . ' • • •
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
-
STATE (Immersion Silver)
STATE (prodip_R)
Category
BUSY
IDLE
- FAILED
BUSY
IDLE
FAILED,
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
45937
45932
6
44792
44786
6
45937
45937
1
45021
45017
5
AvgTime
.00500
.48535
114.00
.00513
.49777
1 14.00
.00500
.49770
114.00
.00510
,49775
114.00
Percent
0.99
96.06
2.95
0.99
96.06
2.95
0.99
98.52
0.49
0.99
96.55
2.46
Percent
0.99
96.06
2.95
0.99
96.06
2.95
0.99
98.52
0.49
0.99
96.55
2.46
G-23
-------
APPENDIX G
NAME:
Throughput:
Immersion Tin, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 30669.2 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
TaktTime
Time in System
Counters
9.8516
40.215
(Corr)
4.5278
8.5500
26.010
93.550
185.18
3080
3081
Identifier Count Limit
Depart 7_C 3081 -Infinite
Frequencies
Identifier
STATE (Cleaner_R)
STATE (predip_R)
STATE (Immersion Tin_R)
STATE (Microetch_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3009
.3002
7
3049
3045
5
2003
2003
1
3008
3000
9
AvgTime
3.5530 '
6.3568
85.000
1.1822
8.6500
85.000
13.151
1.9678
85.000
1.5056
8.3583
85.000
Percent
35.20
62.84
1.96
11.87
86.73
1.40
86.74
12.98
0.28
14.91
82.57
2.52
Percent
35.20
62.84
1.96
11.87
86.73
1.40
86.74
12.98
0.28
14.91
82.57
2.52
G-24
-------
APPENDIX G
NAME:
Throughput:
Immersion Tin, nonTConveyorized
60,OOOKssf
ARENA Simulation Results
Replication ended at time: ~ 7144.18mm.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
9.9108
36.380
.36935
7.8297
8.5500
26.010
88.470
104.68
710
711
Identifier Count Limit
Depart 7_C 711 Infinite
Frequencies
Identifier
STATE (Cleaner_R)
STATE (Predip_R)
STATE (Immersion Tin_R)
STATE (MicroetchJR.)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED,
BUSY
IDLE
FAILED
BUSY'
IDLE
FAILED
Number
699
697
2
711
712
1
527
527
1
693
692
3
AvgTime
3.5295
6.4663
85.000
1.1700
8.7462
85.000
11.535
1.8598
85.000
. 1.5081
8.4451
85.000
Percent
. 34.53
63.09
2.38
11.64
87.17
1.19
85.09
13.72
1.19
. 14.63
81.80
3.57
Percent
34.53
63.09
2.38
11.64
87.17
1.19
85.09
13.72
1.19
14.63
81.80
3.57
G-25
-------
APPENDIX G
NAME:
Throughput:
Immersion Tin, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 43501.6mm.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
TaktTime
Time in System
Counters
.95367
21.375
(Corr)
(Corr)
.93728
12.350
85.005
160.23
45936
45937
Identifier Count Limit
Depart 22_C 45937 Infinite
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (Predip_R)
STATE (Immersion TinJR.)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
• FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
45936
45931
6
45487
' 45481
6
45576
45572
5
45937
45937
1
AvgTime
.00500
.91794
85.000
.00505
.92702
85.000
.00504
.92704
85.000
.00500
.92707
85.000
Percent
0.54
98.28
1.19
0.54
98.28
1-19 '
0.54
98.47
0.99
0.54
99.27
0.20
Percent
0.54
98.28
1.19
0.54
98.28
1.19
0.54
98.47
0.99
0.54
99.27
0.20
G-26
-------
APPENDIX G
NAME:
Throughput:
Immersion Tin, conveyorized (Tin h 60)
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
Identifier Average
10029.78 min.
Half Width Minimum
Maximum
Observations
Takt Time
Time in System
.95796
23.910
(Corr)
(Corr)
.93728
12.364
85.260
• 110.71
10600
10601
Counters
Identifier Count Limit
Depart 22_C 10601 Infinite
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R) -
STATE (Predip_R)
STATE (Immersion Tin_R)
Category
BUSY
IDLE
• FAILED
BUSY
IDLE
FAILED.
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
10601
10601
1
10476
10475
2
10601
10600
2
10601
10601
1
AvgTime
.26000
.67102
85.000
.26310
.67098
85.000
.26000
.66307
85.000
.26000
.67102
85.000
Percent
27.69
71.46
0.85
27.69
70.60
1.71
27.69
70.60
1.71
27.69
71.46
0.85
Percent
27.69
71.46
0.85
27.69
70.60.
1.71
27.69
70.60
1.71
27.69
71.46
0.85
G-27
-------
APPENDIX G
G-5 Chemical Costs by Bath for Individual Surface Finish Processes
G-28
-------
APPENDIX G
M
OJ
2
"3
02
ffi
2
CL,
•
O
r-
O\
r— 1
•6(9
VO
00
oo
So
.
CQ
oo
C3
oo
^
oo
VO
oo
cs
00
cs
cs
CO
"8
S3.
W
8
•3
o
u
c
"S
§-
"8
•s
l|
o CS .
G-29
-------
APPENDIXG
s.
O
a
.2
S2
S ~4
en S
f!
ii.
53
C3
3
"o
o ^
e S
•
-a .!s
•a S.
II
81 1
^_* ^ pi
o ^
c ^»
.52 tS
«.S
CO -O
•5s
> s
« g
2 "3
H S
o f
G-30
-------
APPENDIX G
a
H
a
0>
P* CC
" s «
3*
1
2»
i
e-
«_
S
9
as
O
VO
CO
(N
S
o
CN
r-H
oe-
in
vd
vo
0
00
•69
0
OX)
o
vq
vd
oo
69
W
c
VO
vo
69
1
VO
vo
CN
O
00
a
w-
O
O
o
3
•u
«> "S
^ .2
"
2
s'S
(-« C
±1 _O
•d oo
ii
ii
-° 2
« s
£ o
eg c
£ .2
£.
*_»
-a
*• ea
^ R
G-31
-------
APPENDIX G
ja
a
o
"H
O
gl
il
s- a
« S SS~
U(§.|
S «w ^,
«£*£
a*i
"5 S -8
-5 i*3 v»
fit
.S
3*11
I
«*
>
*o ^^^
5>
ja
>Iume in Ba
(in gallons)
Horizontal
£
P
-
CO
s
»— 4
o
1— <
1/-J
es
<
_
M
«r>
in
VO
0
s
u
O
CO
uS
VO
*— 4
t^
o
VO
«
_4
r-4
^
VO
VO
CO
-n
— M
0
«
1
ffl •
04
CM
o
ca
ca
o
0
cT
(D ,
CJ
3
•o . . .
2
0.
1
cr
2
S ' .
43
C
ll .
•s'i-
ta H
° a
±-.2
II
ca o
4-* ca
CO ™
Jl! ^3
§*s
O CO
. *c3
"o ^
"s*°
il
CO C3
-o .2
** 3i
g "o
•a ^
-C to
c 'o
'3 e
S .2
JJ S •
CQ OJ
i ^ • .
Is
•a" n.
| Ǥ ' '
.S3 tS
J3 2 •
|1.
l|
"3
a U
G-32
-------
APPENDIX G
2
e
o
>2
o»
i
1
32
M
2
•^•k
Process: Eleci
Supplier #4
o- « 'S
** ^5 >?
,g. *g ^
^ "^.
ill
c& ^s *S
^ **-4 *2
°~
DC
"SL **
JM O
C^ *tT?
Ifi
*3
in
$ * -
O
*S "£
e bx 3
j^.
S3 ^» ^"
31 ^
&"
JS
"S
M x~s p-«
A | J
| OD'C
-M
^
i
-
&3-
1
•o
O
"
•a
o
JQ
6^
>n
-S
-a
o
1
0
o
€/*
•o
O
"
,0
•0
1
CO
CO
ffl
1 — <
>0
Rt
'3
•a
o
Microetch
3
2
1
•o
^
_H
o
Electroless Nil
"j3
^
o
ee
T— t
ffi
"c3-
-S?
C3
en
&*
i-r,
|
°s
«e
«j
•a
O '
"
•*-»
"S
^
•n-
m
09
*c3 c^
OJO CQ
~t2 -2?
f— i
S^
So,
^
•
tn
1
•a
^
2
Immersion Go
^
CO
€«
1
N
O
CO
hJ
^
s
_c
i
T3
O
s.
c
!>
CJ
^2
S
e co
M 8
3 O
2 2
•S o.
"O (^
c •*->
o tJS
8 §
CM
•~ T3
f^
^ Q)
CU rf
en ^5
•3 o
Q> «
•U 60
^ 2
11
ll
C/5 4— »
^ W
II5
•S 1
S3 •£
€ ~a
c «
01 , u
13 ^
C3 3
S xi
13 -s
"8 S
||
"73 iu
0 |j
2 "^
> 0
CJ
« o
8 S
||
-------
APPENDIX G
t
13
*0
a
e
o
2
CJ
s Nickel/Imm
tn
C}
"o
M
CJ
Q
il
u
« f-s
.S «2 .2
S *§)^
»j3 C^
_ _^^ '^
ti^"
43
"S >*»V «M«
Volume in Bs
(in gallons
Horizqpta
«
00
cs
r^r
1
•O
^2;
"3
-S?
0
S
0
,-*
* — 1
»n
13
-a
,
in
1>«
Q
*«
-S?
o
cs
CO
W
0
CO
cS^
"
I
•o
o
"E
f«^.
cs
0
CO
*
>— >
T— 4
aS
tS
O
CO
S
1
O
W
C^
O
I1
<5
m
CS
a
CO
'""'
a
•v
o
g
*""*
o\
CS
€«•
«
H- (
^1
t— H
in
"oJ
-o
O
O.
Q
•a
"o
o
o\
co^
cs"
f%
1
•o
J2;
. »
1
t— C
S
vq
i-<
in
No data
J"j
c5
g
Electroless
. _
S
OS
O
CO
>n
1—4
- __
S
Tj"
CO
VO
VO'
1-J
o
O
CO •&
s *o
o w
^ s
Cfl P
ther bath;
d then av
O CM
ra u
j= &
C! 3
S 13
£t2 ^<
=3-8
"8 S
M S
3 s-i
s •§
CO "3
rrt 3
acement cost w;
!d bath was calc
& o
V 00
"- a)
•o j=
a Immersion
overall cost f
G-34
-------
APPENDIX G
2g»
OH CC
«
I*
*
"I
a is
u
1
I
U
w a «
,5 .3 J
Is*
H 0
~ sS o
» «* -.H
"«
-
O
5
VO
v
m
1
OS
O
a
cs
W
vq
en
o\
rn
ON
PH
0)
O
PH
00
O
fofPWB
§
o"
\o
•£,
2
« t,
o- §
If
I'l
3
G-35
-------
APPENDIX G
'
CO
»• L«
S s
0 SI
O Cl
u, a
fW CO
•*»,$".^s
O C3 C£
r\ pa c^
or -^ if
*•* $5 • >?i
?2 'o S'
Total Cost
of the Bath
(Horizontal)
; to ' V
' .5 ' '• ' ;
il-2
*** CS
S
f"O
O u
**"* *es
s «
U J3
o
. .
^ .2 ; '••
. • QJ " s - -
tlfi *^B _.
ta "S .2
C> K CD
^J C> ^^
A^
s « :'•
II
u
*
•»•*
o 5 §
4> *3 .."S
i rt !>
§'"
5
CD 52 c*
*^^ ' S3 • ^^
""* ^ S
3 S3 £
•o ^2<"
^*
JS
*#•*
CJ
.
^
*o
€^
oq
Si
0
O9
O
j—t
<
1— 1
*— »
v>
in
vc>
^^
c
CO
O
O
M
IO
•€^
S
09
^_^
CB
vq
in
CQ
^
i— <
«n
vq
OO
o
o
o
S
^_^
ra
•5?
cs
o\
09
c-
o
.
•
.
^
09
•
OO
Q
^
^1
t
09
a
"cB
00
-S?
r— 1
09
O
O
«
.
^
|
p t
oo
O
'i
ffl
^
PU
CM
O
CM
en
0
o
s" • . •
cs
I • .. •
•a
§,
o
<4^
T3
.!3
1
fe
0^3 • ' . .
OJ P
st -
o 2
<-• *•
II
C3 .^2
'E "o
jjj tn
1 j ' '
g °
DC ,
" S
•E.-2
"T3 "5
S -a
§ J
^ > 3 -- r .
'w "°
e3 w
•a -S
>
.52 o
<*\ "O
eu "
CO ^
0| •
£ ^
a -0
G-36
-------
APPENDIX G
2
"o
O
a
•i
g
s
s
_g
•3
™5
£
c«
cu
"3
1
Sjj
a
CJ
jg^
CO
0
•Q
^^
CJ
"°" oo
W %
In cu
S S
-«- •« x-v
111
W ^S 4J
£i£>
ipfaoziafi
;:;xM:¥:t=SSS:S5g;tsJ
** ° 0
0
11
*3" &«
s
*!
*» *?!'
§1
0 JS
IP
ts ^
A*
«.
1 «
^y-« ^^
V
jH
33 ^2^.
-j^ tft ,
4) W t-
fl &£ $)
*O ^^
5
0$ X"*!, MM
.s J 1
CU OS ^
Sa 5MD *£N
5 g j9
5 '••"'
^*
JS
cs
oo
CN
ol
C3
ts
•o
,
o
o
<
1—H
to
1
-o
0
z
L-i
1
G
£
09
1
1— <
13
00
VO
vq
>n
en
CQ
,
en
en
O
!
bo
en
i
Q.
t
e>)
in
oo
«
T— (
to
1
•a
0
iZ
•g
2
_o
0
oo
cf
O9
1
•o
t
s>
t-
o
en
fe
i— H
"a
0
0^
o
O
_
1)
0
5
CN
HI
"
*— i
T— <
in
03
T3
O
|2
13
ts.
en
IT
CB
•a
o
_
oo
ON
CN
J
HH
1— H
m
1
•a
o
^
ex
5
"3
'o
0
OS
en
of
1
•o
VO
3
S
vd
I— s
VO
3)
ON
O
en
O9
£
&4
>n
1§)
00
04
VO
VO
_,
*~i
in
cj
•s
•o
o
z
"g
o
52
CO
1
w
0
en
of
ca
t
S
g
1
01
09
O
i — i
g
1— <
0
oo'
O9
T— 1
O
"~!
m
1
-o
0
z
J5
*-»
]H
(£
0
oo
en"
09
Rj
ts
•o
en
-^
en
09
oi
PH
r— 4
1
00
en
09
CS
cx
01
13
bO
04
OO
oi
t*
en
-^
en
en
09
<=j
oo
T— (
«n
•o •
o
)7|
S
3
'S •
_2
13
OH
CO
OJ
1
"o
5
1
fC
09
|
"3
-S?
1 — I
0
H
en
CUD
o
o
O9
en
D
< •
^
•o
o
T3
"o
O
o
Immersi
1
CO
(U
3
•a
S
EX,
C
to
I-
3
-G ^
3 S ,
0 0
« t!
^2 ^
CO .^j
S |
Ij
•3 a>
"O 60
"& 2
j= c3
tt; •s
o a
o o
en -t*
3 -^
0 §>
-o S
ft C
o3 *^-»
f i
c "
C
"*-* .«
^ 0
S -i
S§ o.
*~J ^— t
^O *^
T3 ^
o o>
!Ǥ
^ 13
03 a>
^
is
« g
CO ^*
O co
O 03
•*-» ^
(D -=
£ e3
S ^
"S."o
ca M
2 •£
§) g
If. '
g 8 .
11 '
« o
G-37
-------
APPENDIX G
G-6 Total Materials Cost for Surface Finishing Processes
Process: HASL, non-conveyorized
Throughput: 260Kssf ofPWB
Bath
Cleaner
Microetch
Flux
Solder
Chemical CostflBath
', Replacement" -
$188
$197
$16,250 c
$55,460 d
Number of Batti i
Replacements h
7
9
1
1
Total Materials Cost
Total Chemical Cost
$1,320
$1,770
$16,250
$55,460
$74,800
* Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
e Flux bath is not replaced, but rather refilled'as flux is consumed. Cost of flux was calculated at $12.50/gal and is
consumed at 200 ssf/gal.
d Solder is not replaced, but rather refilled as solder is consumed. Cost of solder was calculated using a solder cost of
52.57/lb and an average solder consumption rate, including solder wastage, of 0.083 Ib/ssf which was obtained from
three PWB facilities.
Process: HASL, conveyorized
Throughput: 260KssfofPWB
. Bath
Cleaner
Microetch
Flux
Solder
^Chemical Cost(Bat&
>1 Replacement3
$244
$344
$16,250 c
$55,460 d
Number of Bath
Replacements b
6
6
1
1
Total Materials Cost
Total Chemical Cost
$1,460
$2,060
$16,250
$55,460
$75,200
* Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
e Flux bath is not replaced, but rather refilled as flux is consumed. Cost of flux was calculated at $12.50/gal and is
consumed at 200 ssf/gal.
d Solder is not replaced, but rather refilled as solder is consumed. Cost of solder was calculated using a solder cost of
$2.57/lb and an average solder consumption rate, including solder wastage, of 0.083 Ib/ssf which was obtained from
three PWB facilities.
G-38
-------
APPENDIX G
Process: Electroless Nickel/Immersion Gold, non-conveyorized
Throughput; 260K ssf of PWB
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemical Cost/fBata
Replacement a
$92.8
$386
$1,640
$315
$890
NAC
Number of Bath
Replacements *
7
9
6
4
40
6
Total Materials Cost
Total Caemieal Cost
$649
$3,470
$9,830
$1,260
$35,500
$57,900
$108,600
a I jj VV...VKA. * «.*VVw MAW U.T w^u^^s wMu* wwoi, wi an piwt-coowo suuililtlCU. 1U1 CVcUimUOIl 111
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
c Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and
throughput between product lines. The overall cost for the gold bath was calculated for each product line and then
averaged together to give the gold cost for the process.
Process: Electroless Nickel/EIectroless Palladium/Immersion Gold, non-conveyorized
Throughput: 260KssfofPWB
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Chemical CostfBath
Replacement *
$128
$266
$2,810
$11.3
$2,390
$2,430
$3,980
NAC
Number of Bath
' Replacements6
7
9
6
4
40
5'
34
6
Total Materials Cost
Total Chemical Cost
$900
$2,390
$16,860
$45
$95,600
$12,150
$135,300
$57,900
$321,000
a Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
0 Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and
throughput between product lines. The overall cost for the gold bath was calculated for each product line and then
averaged together to give the gold cost for the process.
G-39
-------
APPENDIX G
Process: OSP, non-conveyorized
Throaghput; 260K ssf of PWB
Bath ;
Cleaner
Microetch
OSP
Chemical Cost/Bath
Replacement *
$51.6
$157
$16,750 c
Number of Bath
Replacements b
7
9
1
Total Materials Cost
Total Chemical Cost
$361
$1,420
$16,750
$18,500
* Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
c OSP bath is not replaced, but rather refilled as the OSP is consumed. Cost of OSP was calculated at $105/gal and is
consumed at 1,630 ssf/gal.
Process: OSP, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
OSP
Chemical Cost/Bath
Replacement a
$67.2
$271
$16,750 c
Number of Bath
Replacements b
6
6
1
Total Materials Cost
Total Chemical Cost i
$403
$1,630
$16,800
$18,800
* Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
c OSP bath is not replaced, but rather refilled as the OSP is consumed. Cost of OSP was calculated at $105/ga.l and is
consumed at 1,630 ssf7gal.
G-40
-------
APPENDIX G
Process: Immersion Silver, conveyorized
Throughput; 260KssfofPWB
Bath
Cleaner
Microetch
Predip
Immersion Silver
Chemical Cost/Bath
Replacement a
$958
$124
$1,200
$40,170°
Number of Bath
Replacements b
6
6
5
1
Total Materials Cost
Total Ciiemical Cost
$5,750
$744
$6,000
$40,200
$52,700
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
c Silver bath is not replaced, but rather maintained as the silver bath is depleted. The cost of the silver bath was
calculated at $30.9/gal and is consumed at 200 ssf/gal.
Process: Immersion Tin, non-conveyorized
Throughput; 260K ssf of PWB
Bath
Cleaner
Microetch
Predip
Immersion Tin
Chemical Cost/Bath
Replacement8
$174
$74
$659
$23,850 ° ,
Number of Bath
Replacements b
7
9
5
1
Total Materials Cost
Total Chemical Cost
$1,220
$665
$3,300
$23,850 '
$29,000
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
0 Tin bath is not replaced, but rather maintained as the tin bath is depleted. The cost of the tin bath was calculated at
$133/gal and is consumed at 1,450 ssf/gal.
Process: Immersion Tin, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner .
Microetch
Predip
Immersion Tin
Chemical Cost/Bath
Replacement *
$226
$125
$597
$23,850 c
Number of Bath
Replacements b
6
6
5
• 1
Total Materials Cost
Total Chemical Cost
$1,350
$752
$2.990
$23,850
$28,900
a Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in
this surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
0 Tin bath is not replaced, but rather maintained as the tin bath is depleted. The cost of the tin bath was calculated at
$133/gal and is consumed at 1,450 ssf/gal.
G-41
-------
-------
Appendix H
Environmental Hazard Assessment and
Ecological Risk Assessment Methodology
-------
-------
APPENDIX H
H-l. HAZARD PROFILE
The environmental hazard assessment of chemicals consists of the identification of the
effects that a chemical may have on organisms in the environment. An overview of this
assessment process has been reported by, for example, Smrchek and Zeeman (1998) and by
Zeeman and Gilford (1993). The effects are expressed in terms of the acute and chronic toxicity
of a chemical on the exposed organisms. These are generally given as either the lethal
concentration (LC) or as the effective concentration (EC) that describe the type and seriousness
of the effect for a known concentration .of a chemical.. When the effective concentrations for a
range of species for a chemical are tabulated, the tabulation is called a hazard profile or toxicity
profile. A more detailed discussion of a comprehensive hazard profile has been presented by
Nabholz (1991). The most frequently used hazard profile for the aquatic environment consists of
a set of six effective concentrations as reported by Nabholz et al. (1993a). These are:
• Fish acute value (usually a fish 96-hour LC50 value) .
• Aquatic invertebrate acute value (usually a daphnid 48-hour LC50 value)
• Green algal toxicity value (usually an algal 96-hour EC50 value)
• Fish chronic value (usually a fish 28-day chronic value [ChV])
• Aquatic invertebrate chronic value (usually a daphnid 21 -day ChV)
• Algal chronic value (usually an algal 96-hour NEC or GMATC value for biomass)
For the acute values, the LC50 (lethality or mortality) (EC50) (non-lethal/lethal effects)
refers to the concentration that results in 50 percent of the test organisms affected at the end of
the specified exposure period in a toxicity test. The chronic values represent the concentration of
the chemical that results hi no statistically significant sublethal effects on the test organism
following an extended or chronic exposure.
The hazard profile can be constructed using effective concentrations based on toxicity test
data (with measured test chemical concentrations) or estimated toxicity values based on structure
activity relationships (SARs). The measured-values are preferred because they are based on
actual test data, but in the absence of test data SAR estimates, if available for the chemical class,
can be used. Thus the hazard profile may consist of only measured data, only predicted values, or
a combination of both. Also, the amount of data in the hazard profile may range from a minimum
of one acute or chronic value to the full compliment of three acute values and three chronic
values. ,
In the absence of measured toxicity values, estimates of these values can be made using
SARs. SAR methods include quantitative structure activity relationships (QSARs), qualitative
SARs, or use of the chemical analogs. The use of SARs by OPPT has been described (Clements,
1988; Clements, 1994). The use and application of QSARs specifically for the hazard assessment
of TSCA new chemicals has been presented (Clements et al., 1993 a). The development, .
validation, and application of SARs in OPPT have been presented by OPPT staff (Zeeman et al.,
1993b; Boethiing, 1993; Clements et al., 1993b; Nabholz et al., 1993b; Newsome et al., 1993 and
Lipnick, 1993).
H-l
-------
APPENDIX H
The predictive equations (QSARs) are used in lieu of actual test data to estimate a toxicity
value for aquatic organisms within a specific chemical class. A total of 140 have been listed ,
(Clement et al., 1995; Smrchek and Zeeman, 1998). Although the equations are derived from
correlation and linear regression analysis based on measured data, the confidence intervals
associated with the equation are not used to provide a range of toxicity values. Even with
measured test data, the use of the confidence limits to determine the range of values is not used.
H-2. DETERMINATION OF CONCERN CONCENTRATION
Upon completion of a hazard profile, a concern concentration (CC) is determined. A
concern concentration is that concentration of a chemical in the aquatic environment, which, if
exceeded, may cause a significant risk to aquatic organisms. Conversely, if the CC is not
exceeded, the assumption is made that probability of a significant risk occurring is low and no
regulatory action is required. The CC for each chemical is determined by applying assessment
factors (AsF) (U-.S. EPA, 1984) or uncertainty factors (UF) (Smrchek et aL, 1993) to the effect
concentrations in the hazard profile.
These factors incorporate the concept of the uncertainty associated with: 1) toxicity data,
laboratory tests versus field tests, and measured versus estimated data; and 2) species sensitivity.
For example, if only a single LC50 value for a single species is available, there are several
uncertainties to consider. First, how reliable is the value itself? If the test were to be done again
by the same laboratory or a different laboratory, would the value differ and, if so, by how much?
Second, there are differences in sensitivity (toxicity) among and between species that have to be
considered. If the species tested the most or the least sensitive? In general, if only a single
toxicity value is available, there is a large uncertainty about the applicability of this value to other
organisms in the environment and a large assessment factor, i.e., 1000, is applied to cover the
breadth of sensitivity known to exist among and between organisms in the environment.
Conversely, the more information that is available results in more, certainty concerning the toxicity
values and requires the use of smaller factors. For example, if toxicity values are derived from
field tests, then an assessment factor of 1 is used because tests measure chemical effects on field
organisms.
Four factors are used by OPPT to set a CC for chronic risk: 1,10, 100, and 1000. The
factor used is dependent on the amount and type of toxicity data contained in the hazard profile
and reflects the amount of uncertainty about the potential effects associated with a toxicity value.
In general, the more complete the hazard profile and the higher the quality of the generated
toxicity-data, the smaller a factor that is used. The following discussion describes the use and
application of uncertainty or assessment factors.
1. If the hazard profile only contains one or two acute toxicity values, the concern
concentration is set at 1/1000 of the acute value.
2. If the hazard profile contains three acute values (called the base set), the concern
concentration is set at 1/100 of the lowest acute value.
H-2
-------
APPENDIX H
3. If the hazard profile contains one chronic value, the concern concentration is set at 1/10 of
the chronic value if the value is for the most sensitive species. Otherwise, it is 1/100 of the
acute value for the most sensitive species.
4. If the hazard profile contains three chronic values, the concern concentration is set at 1/10
of the lowest chronic value.
5. If the hazard profile contains a measured chronic value from a field study, then an
assessment factor of 1 is used.
H3. HAZARD RANKING
Chemicals can be also ranked by their hazard concern levels for the aquatic environment.
This ranking can be based upon the acute toxicity values expressed in milligrams per liter (mg/L).
The generally accepted scoring used by OPPT is as follows (Smrchek et aL, 1993; Wagner et al.,
1995):
High Concern (H)
Moderate (or Medium) Concern (M)
Low Concern (L)
> 1 and < 100
>100
This ranking can also be expressed in terms of chronic values as follows:
High Concern (H)
Moderate (or Medium) .Concern (M)
Low Concern (L)
>0.1 and < 10.0
> 10.0
Chronic toxicity ranking takes precedent over the acute ranking.
H-2
-------
APPENDIX H
REFERENCES
t
Boethling, R.S. 1993. Structure Activity Relationships for Evaluation of Biodegradability in the
EPA's Office of Pollution Prevention and Toxics. Environmental Toxicology and Risk
Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch, F. James Dwyer, Christopher G.
Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing and Materials,
Philadelphia, pp. 540-554.
Clements, R.G. (Ed.) 1988. Estimating Toxicity of Industrial Chemicals to Aquatic Organisms
Using Structure-Activity Relationships. U.S. Environmental Protection Agency, Office of
Pollution Prevention and Toxics, Health and Environmental Review Division. Environmental
Effects Branch. Washington, DC. EPA-560/6-88/001,NTIS#PB89-117592.
Clements, R.G. (Ed.) 1994. "Estimating Toxicity of Industrial Chemicals to Aquatic Organisms
Using Structure Activity Relationships," 2nd Ed. EPA-748-R-93-001. Environmental Effects
Branch, Health and Environmental Review Division (7403). Office of Pollution Prevention and
Toxics, U.S. EPA, Washington, DC. PB94-108206, National Technical Information Services
(NTIS), U.S. Department of Commerce, Springfield, VA 22161.
Clements, R.G., J.V. Nabholz, D.W. Johnson, and M. Zeeman. 1993a. "The Use and "
Application of QS AR's in the Office of Toxic Substances for Ecological Hazard Assessment of
New Chemicals." Environmental Toxicology and Risk Assessment. ASTM STP 1179. Wayne
G. Landis, Jane S. Hughes, and Michael A. Lewis (Eds.). American Society for Testing and
Materials, Philadelphia, pp. 56-64.
Clements, R.G., J.V. Nabholz, D.W. Johnson, and M. Zeeman. 1993b. "The Use of Quantitative
Structure-Activity Relationships (QSARs) as Screening Tools in Environmental Assessment."
Environmental Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,
F. James Dwyer, Christopher G. Ingersoll, and Thomas W. La Point (Eds.). American Society for
Testing and Materials, Philadelphia, pp. 555-570.
Clements, R.G., J.V. Nabholz, M.G. Zeeman, and C. Auer. 1995. "The Relationship of
Structure-Activity Relationships (SARs) in the Aquatic Toxicity Evaluation of Discrete Organic
Chemicals. SAR and QSAR in Environmental Research 3:203-215.
Lipnick, R.L. 1993. "Baseline Toxicity QSAR Models; A Means to Assess Mechanism of
Toxicity for Aquatic Organism and Mammals." Environmental Toxicology and Risk Assessment,
2nd Volume. ASTM STP 1216. J.W. Gorsuch, F. James Dwyer, Christopher G. Ingersoll, and
Thomas W. La Point (Eds.). American Society for Testing and Materials, Philadelphia, pp. 610-
619.
Nabholz, J.V. 1991. "Environmental Hazard and Risk Assessment Under the United States
Toxic Substances Control Act." The Science of the Total Environment 109/110:649-665.
H-4
-------
APPENDIX H
Nabholz, J.V., P. Miller, and M. Zeeman. 1993a. "Environmental Risk Assessment of New
Chemicals Under the Toxic Substances Control Act (TSCA) Section Five." Environmental
Toxicology and Risk Assessment, ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and
Michael A. Lewis (Eds.). American Society for Testing and Materials, Philadelphia, pp. 40-55.
Nabholz, J.V., R.G. Clements, M.G. Zeeman, K.C. Osborn, and R. Wedge. 1993b. "Validation
of Structure Activity Relationships Used by the U.S. EPA's Office,of Pollution Prevention and
Toxics for the Environmental Hazard Assessment of Industrial Chemicals." Environmental
Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.S. Gorsuch, F. James
Dwyer, Christopher G. Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing
and Materials, Philadelphia, pp. 571-590.
Newsome, L.D., D.E. Johnson, and J.V. Nabholz. 1993. "Quantitative Structure-Activity
Predictions for Amine Toxicity to Algae and Daphnids." Environmental Toxicology and Risk
Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,. F. James Dwyer, Christopher G.
Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing .and Materials,
Philadelphia, pp. 591-609. . .
Smrchek, J.C. and M.G. Zeeman. 1998. "Assessing Risks to Ecological Systems from
Chemicals." Handbook of Environmental Risk Assessment and Management. P. Calow (Ed.),
Blackwell Science Ltd, Oxford, UK, pp. 24-90.
Smrchek, J.C., R. Clements, R. Morcock, and W. Robert. 1993. "Assessing Ecological Hazards
under TSCA: Methods and Evaluation of Data." Environmental Toxicology and Risk
Assessment. ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and Michael A. Landis
(Eds.). American Society for Testing and Materials, Philadelphia, pp 22-39.
U.S. EPA (Environmental Protection Agency). 1984. "Estimating Concern, Levels for •
Concentrations of Chemical Substances in the Environment." Environmental Effects Branch,
Health and Environmental Review Division (7403), Office of Pollution Prevention and Toxics,
U.S. EPA, Washington, DC. . .
Wagner, P.M., J.V. Nabholz, and R.J. Kent. 1995. "The New Chemicals Process at the
Environmental Protection Agency (EPA): Structure-Activity Relationships for Hazard
Identification and Risk Assessment." Toxicology Letters 79:67-73.
Zeeman, M.G. and James Gilford. 1993. "Ecological Hazard Evaluation and Risk Assessment
Under EPA's Toxic Substances Control Act (TSCA): An Introduction." Environmental
Toxicology and Risk Assessment. ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and
Michael A. Lewis (Eds.). American Society for Testing and Materials, Philadelphia, pp. 7-21.
H-5
-------
APPENDIX H
Zeeman, M.G., J.V. Nabholz, and R.G. Clements. 1993. "The Development of SAR/QSAR for
Use Under EPA's Toxic Substances Control Act (TSCA): An Introduction." Environmental
Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,. F. James
Dwyer, Christopher G. Intersoll, and Thomas W. La Point (Eds.). American Society for Testing
and Materials, Philadelphia, pp. 523-539.
H-6
-------