Office of Research and
                       Development
                       Washington, DC 20460
                        EPA/600/R-92/114
                        July 1992
&EPA
Modifications to Reduce
Drag Out at a Printed
Circuit Board
Manufacturer

-------
                                   EPA/600/R-92/114
                                  'July 1992
    MODIFICATIONS TO REDUCE DRAG OUT
AT A PRINTED CIRCUIT BOARD MANUFACTURER
                   by
               Paul  Pagel
 Minnesota Technical  Assistance Program
     Minneapolis,  Minnesota   55455

                  and
             Teresa Harten
             Project Officer
  Pollution Prevention Research Branch
 Risk Reduction Engineering  Laboratory
        Cincinnati, Ohio  45268
             CR-815821-01-0
 RISK REDUCTION  ENGINEERING  LABORATORY
   OFFICE OF RESEARCH AND DEVELOPMENT
  U.S.  ENVIRONMENTAL PROTECTION AGENCY
         CINCINNATI,  OHIO   45268
                                  Printed on Recycled Paper

-------
                                  DISCLAIMER
        The  information  in this document has been funded wholly or  in  part  by
the United States Environmental Protection Agency under Contract No. CR-
815821-01-0 to the University of Minnesota, Minnesota Technical Assistance
Program.  It has been subjected to the Agency's peer and administrative
review, and it has been approved for publication as  an EPA document.  Mention
of trade names or commercial  products does not constitute endorsement or
recommendation for use.

-------
                                   FOREWORD

        Today's  rapidly  developing technologies and industrial products  and
practices frequently carry with them the increased generation of materials
that, if improperly dealt with, can threaten both public health and the
environment.  The U.S. Environmental  Protection Agency is charged by Congress
with protecting the Nation's land, air, and water resources.  Under a mandate
of national environmental laws, the Agency strives to formulate and implement
actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life.  These laws direct the
EPA to perform research to define our environmental problems, measure the
impacts, and search for solutions.

        The  Risk Reduction  Engineering  Laboratory  is  responsible for
planning,  implementing, and managing research, development, and demonstration
programs to provide an authoritative, defensible engineering basis in support
of the policies, programs, and regulations of the EPA with respect to drinking
water, wastewater, pesticides, toxic substances, solid and hazardous wastes,
and Superfund-related activities.  This publication is one of the products of
that research and provides a vital communication link between the researcher
and the user community.

        This  report,  "Modifications  to  Reduce  Drag Out  at  a  Printed Circuit
Board Manufacturer," discusses evaluation of two low cost, low technology
modifications to rinsing operations at a printed circuit board manufacturer.
Both modifications reduce waste by reducing drag out of process chemicals from
the line.  The information contained in this report should be helpful  to
operators and designers of circuit board lines, and other closely related
metal finishing  processes, in identifying and implementing waste reduction
technologies within their operations.


                            E. Timothy Oppelt, Director
                            Risk Reduction Engineering Laboratory
                                      iii

-------
                                    ABSTRACT

         The MnTAP/EPA WRITE project at Mi com, Inc. demonstrated the waste
 reducing capability of two simple rinsing modifications  on  two processes that
 are typically found within electronic  circuit board  manufacturing  operations.
 The simple,  low (or no)  cost,  low technology changes that were made were
 1) slowing the withdrawal  rate of the  racks containing the  printed circuit
 boards as they are pulled  from concentrated process  tanks and 2) combining an
 intermediate withdrawal  rate with a longer drain  time over  the process tanks
 before transfer to the rinse tanks. As  compared  to  baseline  sampling, both
 modifications significantly reduced drag out of concentrated  copper containing
 bath solutions into the rinse water systems that  followed the bath tanks.

         The two processes tested were: an etchant bath  and  the countercurrent
 rinsing system following it and an electroless copper plating bath and the
 countercurrent rinsing system following  it.   The  reduction  in drag out for the
 micro-etch bath was 45% as a result of the first  modification and  41% as a
 result of the second modification.  For  the electroless  bath, drag out was
 reduced by 50% after the first modification and 52%  after the second
 modification.

         By reducing drag out in these amounts, 203 and  189  grams of copper
 per day were prevented from being discharged as waste in the  rinse water waste
 stream,  for  modifications  1  and 2 respectively.   Because copper concentration
 in rinse water was reduced,  the potential  for conserving rinse water flows was
 also shown,  although this  was  not directly tested.   Rinse water flows could be
 turned  down  proportionate  to the reduction in drag out and  still maintain  the
 same rinsing  efficiencies.

         The  economic savings due to these reductions were calculated by
 considering  avoided  cost of  treatment  of the rinse water and  avoided charges
 for  water  and  sewer  service.   If implemented,  the first modification would
 save  the company $3350  $2640  savings in  treatment  costs and $710 in avoided
 water and  sewer costs.   The  same figures  for implementing the second
 modification would be  $3120  -  $2460 in treatment  costs and  $660 in  avoided
 water and  sewer charges.   Since no capital  costs  were incurred  in  making the
 changes, payback would be  immediate.

         This report is submitted in fulfillment  of EPA  Cooperative Agreement
 No. CR-815821-01-0 by  Minnesota Technical  Assistance  Program, University of
Minnesota, under the sponsorship of the  U.  S.  Environmental  Protection Agency.
The report covers  a  period from June 26,  1989  to  December 30, 1990  and work
was completed  as of  December 30,  1990.
                                      IV

-------
                                   CONTENTS

DISCLAIMER	ii
FOREWORD	iii
ABSTRACT	iv
CONTENTS  	  v
INTRODUCTION   	  1

         Program Objectives    	  1
         Rinsing Operations  at Plating  Facilities    	  1
         WRITE Company Selection   	'.  .  2
         WRITE Company Description   	  2

TECHNOLOGY DESCRIPTION:  DRAG OUT REDUCTION   	  5

         Benefits of Reducing Drag Out   	5
         Modifications for Micom   	  5

SAMPLING AND  ANALYSIS    	  8

         Baseline Sampling   	  9
         First Modification  Sampling    	  9
         Second Modification Sampling    	  9

RESULTS   	11

         Differences in Drag Out  Due to Rinsing  Modifications    	14
         Quality Assurance Project Plan Results    	 15


DISCUSSION	18

         Technical  Evaluation     	 18
         Economic Analysis   	 18
         Implementation   	 19
         Future Modifications     	 19
         Transferring Project Results    	 20

RECOMMENDATIONS    	 21

BIBLIOGRAPHY   	 22

APPENDIX - Raw Data  and  Calculated Drag Out Values	23

-------
                                 INTRODUCTION

         The Waste Reduction Innovative  Technology  Evaluation  (WRITE)  Program
is a national research demonstration program designed to evaluate the use of
innovative  engineering and  scientific technologies to reduce the volume and/or
toxicity of wastes produced from the manufacture, processing, and use of toxic
materials.  The U.S. EPA Office of Research and Development, Risk Reduction
Engineering Laboratory, and the Minnesota Technical Assistance Program (MnTAP)
entered  into a cooperative  agreement as part of the "WRITE Pilot Program with
State and Local Governments" in July, 1989.  Funding of the WRITE program is
provided jointly  by the EPA and the state and local governments.  The joint
approach was chosen because state and local government officials are familiar
with local  industry practices and regional manufacturing interests, and these
factors  can affect the potential success and widespread applicability of
proposed pollution prevention technologies.

PROGRAM  OBJECTIVES

         The Minnesota/EPA WRITE program is  one  of  seven  such  programs
nationwide.  The  program  in Minnesota targets the metal finishing industry,
specifically rinsing processes within metal finishing operations, as the focus
of waste reduction evaluations.  The two most effective methods for reducing
wastes from these rinsing processes are 1) reducing drag out, the carryover of
concentrated solutions from plating baths, and 2) practicing water
conservation.

         The present  report  discusses  the  results of the  first  project
performed under the Minnesota/EPA WRITE program.  The project evaluated
modifications which reduce  drag out at a single plating line within a printed
circuit  board manufacturing facility.  It is hoped that by demonstrating the
success  of  the modification in a fully operational  setting, the technology
will be  transferred to other plating/rinsing systems within the company as
well as  to  other  facilities within the metal finishing industry.

RINSING OPERATIONS AT PLATING FACILITIES

         The basic plating operation  involves  submerging  parts  in  a  process
solution, then rinsing off  the excess film of plating chemicals known as drag
out.  This  rinsing process can waste several pounds per day of expensive
plating chemicals which must be removed from the rinse water before rinse
waters can  be discharged to the sewer to comply with effluent limit
requirements.  More efficient rinsing operations could reduce the loss of
plating chemicals, saving on the material cost and reducing the amount of
waste to be managed (Cushnie, 1985).

-------
         Rinsing is essentially an operation of dilution;  its objective is to
 dilute the dissolved  chemicals on the surface  of the  work to the point where
 they are insignificant,  not only  in  their effect on the  quality of the work
 being processed,  but  also with respect to plating solution contamination in
 the operation of  a plating line over a long period of time.   Efficient rinsing
 is obtained when  this objective is accomplished  with  the minimum use of water
 (Durney, 1984).  Reducing drag out should reduce treatment needs and allow for
 reductions in the flow of water required for rinsing.

 WRITE COMPANY SELECTION

         Solicitation of metal  plating companies for participation in the
 WRITE Program began in early 1989 through newsletter  articles,  metal finishing
 association presentations, and direct mailings.   Twenty-five companies
 responded with interest in evaluating source reduction opportunities via
 rinsing modifications.  Work began in mid-August 1989 when site visits were
 conducted to assess each company's interest and  applicability to the project.
 Companies were evaluated for inclusion in the  WRITE Program on  the basis of a
 number of criteria which included potential for  pollution prevention,
 willingness to make modifications, willingness to share  information,
 production variability,  and measurable quantity  and concentration of
 contaminant to be reduced.

 WRITE COMPANY DESCRIPTION

         Micom  Inc. was selected  largely because of its willingness  to make
 changes and because of its relatively stable production.  Micom is a medium-
 sized job  shop in New Brighton, Minnesota,  employing  approximately 240 people.
 The company manufactures  printed  circuit boards  under a  number  of military and
 commercial  contracts.   Annual  revenues for 1989  were  $17 million.  An  average
 production day consists  of two eight-hour shifts with a  work load of 1,000-
 1,200 square feet  (ft  ) of  panels.

         Micom manufactures double-sided and multilayered printed circuit
 boards.  The average  circuit board size  is  18  inches  by  21 inches with 8000
 holes per  panel.   Printed  circuit boards are produced by depositing  and
 etching  metal  from a  fiberglass sheet (board).   The steps in this process
 include:   cutting  the  boards to size,  coating  the boards with a photosensitive
 material (resist),  developing  the resist,  drilling holes,  deburring, cleaning,
 etching, plating the  inside of the holes (using  an electroless  plating
 process),  and plating  the  boards;  inspections  are performed  at  many  points
 along the way.  Water  rinses follow  many of these steps.  All printed  circuit
 boards at Micom pass through the  sensitize  line  (Figure  1).   This line is  used
 to deposit copper  onto the  inside of the circuit board holes and consists  of
micro-etch,  activator, accelerator,  electroless  copper and rinse tanks.

         The  WRITE  project tested  changes to the  operation of the micro-etch
bath and the two countercurrent rinses which follow it,  and  the  electroless
copper bath  and the two countercurrent rinses  which follow it.   Micom was
interested in evaluating modifications to the  micro-etch and  electroless

-------
\^^ Soft
Water Line *2


J*
Water Line •
1 '



PTH 160
Cleaner

>

Counter
Current
Rinse
f
)

26
r >

Counler
Current
Rinse
\
f

IOX
Sulfurlc
Acid
Pre-Dlp

Micro-etcn

Counler
Current
Rinse

-
gpm
L_

Counter
Current
Rinse
1
Copper Ion Exchange Line * 1

Activator
Pre-dlp
A



Activator


Counter
Current
Rinse '
r
>

•

>
*
Water line *3


Counter
Current
Rinse
"\
r

Accelerator
Tank
                                                                  Sewer Line « I To pH Adjust
                                              Soft Water Line *l
CO
                work flow
Water Line "3
>



Accelerator


<

Flowing
Rinse

>



Dead
Rinse
1
Electroless
Copper
1
|
fj*\ (Water Line "31
3.3 gpm
V >

Counter
Current
Rinse
N >
y Copper Ion Exchange Line • 1

m

Counter
Current
Rinse

Dead
Rinse

IOX
Suifuric
Add
f

Flow ing
Rinse
f
\

«>cetlc Acid
f
                                                  Sewer Line * I To pH Adjust





                     ^^^ Represents primary valve on each water line.




                      A    Represents a sampling point.
                                                           Figure  1.   Micom Sensitize  Line

-------
copper plating and rinsing processes because these processes are significant
contributors of copper to the ion exchange system used to treat metal bearing
wastewaters.  Softened water at a restricted flow of about 3 gallons per
minute makes up the influent to the countercurrent rinses following the micro-
etch bath and the electroless copper bath.  The copper bearing waste stream
from the rinse tanks is piped to an ion exchange system for copper removal and
then sewered.  The ion exchange canisters are regenerated off site.

         The printed circuit  boards  enter  the sensitive line  in  racks which
hold 24 boards each.  The boards vary in size with the largest being 18 inches
by 24 inches.  The racks are 34 inches by 19.5 inches wide by 13 inches deep
and are transported by a hoist from tank to tank.  The operator controls the
hoist and allows the rack to drain 3-5 seconds before proceeding to the next
tank.  The  approximate residence time is 75 seconds in the micro-etch bath, 30
minutes  in  the electroless copper bath, and 2 to 3 minutes in each rinse bath.

         The electroless  copper  plating  process  is the rate determining  step
for production throughput.  The plating tank holds two racks of printed
circuit  boards which remain in the bath for 30 minutes each.  In order to
maximize production, each pair of racks are started in the line at
approximately 30 minute intervals.  This allows the electroless copper bath to
remain in use and minimizes the delay between plating each rack.

-------
                  TECHNOLOGY DESCRIPTION:   DRAG OUT REDUCTION


BENEFITS OF REDUCING DRAG OUT

         The micro-etch process  prepares the  printed circuit  board  for  plating
by removing oxidized copper from its surface.  Reducing the drag out from the
micro-etch process solution will also reduce the amount of water required for
rinsing and reduce the operating costs of the  ion exchange wastewater
treatment system.  However, the micro-etch process  solution may require more
frequent replacement due to the additional buildup  of copper which was
previously removed by drag out.

         Reducing the  copper drag out  from the  electroless  copper plating
solution results  in three possible benefits.   If copper in the rinse water is
reduced by capturing the process solution before rinsing, raw material  costs
may be reduced,  since the need for chemical additions could be reduced by
returning process solution to the bath.  Treatment  costs may also be reduced,
especially where  relatively expensive treatment is  used to remove copper from
the rinse waters.  In addition, reducing copper discharge by capturing process
solution can reduce the amount of water needed for  rinsing, since less process
solution requiring dilution (rinsing) will be  present.

         Disposal of  spent  micro-etch  and  electroless  baths included  copper
recovery in both  cases and regeneration of etchant  in the case of the micro-
etch solution.   When the micro-etch bath had to be  removed from the line and
replaced with fresh bath it was company practice to reclaim copper from the
bath as copper sulfate and reuse it in another copper plating solution on
site.  The etchant was regenerated and also used on site for less critical
operations such  as stripping copper from carrier racks.  The company also
recovered copper  from the spent electroless copper  bath, although the copper
was sent off site for reuse.  Thus, copper-containing bath solution retained
in the process baths due to implementing drag  out reduction changes would
ultimately be subjected to the company's recovery operations.

MODIFICATIONS FOR MICOM

         At Micom, the  criteria  for  selecting implementable rinsing process
modifications included:  availability of equipment, impact of a modification
on process solutions, maintaining product quality,  and effect on production
throughput.  Production objectives (quality and throughput) were paramount, as
might be expected.

-------
         Before  the WRITE Program's involvement with Micom, the company had
already made modifications which reduced rinse water requirements  including
the installation of countercurrent rinses,  water flow restrictors,  softened
water, air agitation and mechanical agitation.  Countercurrent rinses had been
installed following the micro-etch and electroless copper process  baths,  and
water flow restrictors on rinse tank inlets maintained an approximate flow of
3 gallons per minute (gpm).   Softened water was used in the copper bearing
rinses to improve rinsing and the efficiency of the ion exchange system.   The
filled racks were mechanically agitated in each tank to force solution through
the circuit board holes, and some rinses were air agitated to increase the
dilution of the drag out by mixing the rinse.

         The  first  modification  that  was  evaluated  at  Micom was  slowing the
rate of withdrawal of the racks from the process solution tanks.  This was
accomplished by lowering the speed of the motor on the mechanical  hoist used
for raising, lowering and transporting racks of circuit boards.  The slower
rate of withdrawal allows the process solution viscosity to aid solution
removal by pulling the solution from boards like a squeegee.  The total drain
time is also increased as the boards are being withdrawn more slowly, thereby
allowing the part to drain longer.

         The  effects  of reducing the  drag out on process  solutions  is an
important consideration.  The micro-etch bath removes copper  from the printed
circuit boards  and continually  builds up copper  in the solution until it no
longer is able  to remove copper.   At this  point, the  bath is  replaced with new
solution.  For  an etchant bath, retaining  solution which was  formerly lost to
drag out could  mean the etchant bath must  be  replaced more often due to the
increased build-up of copper.   However,  managing the  concentrated waste stream
which results from bath replacement  is  preferable  to  managing diluted rinse
waters because  opportunities for  metals  recovery are  improved if the
wastestream is  concentrated.  As  previously  discussed, at Micom, spent
concentrated micro-etch and electroless  bath  solutions were subjected to on-
site copper recovery processes, and  for  the  micro-etch bath,  an etchant
recovery process.  Another advantage of  reducing drag out to  rinse water is
that rinse water can be reduced proportionate to drag out reduction, thereby
requiring less  treatment chemicals and  reducing  sludge generation at on-site
wastewater treatment systems.

         Retaining  electroless  plating  solution  which was formerly  lost to
drag out could  increase the life  of  the  plating bath.  Since  fewer plating
chemicals are lost to drag out  of solution,  fewer chemical additions will be
required to maintain the chemical  composition of the  bath.  However, a
filtration step may have to be  added to  remove  impurities which were
previously removed by drag out.   Drag out  reduction from preceding tanks could
also lower the  amount of impurities carried  into the  plating  bath.

         The  second modification evaluated  was increasing  the  drain time  over
the process bath before transferring the racks into the rinse tanks.   This was
accomplished by having the line operator wait longer  (10 seconds for this
evaluation) before moving the draining rack to the rinse.  The baseline

-------
withdrawal rate was not reproduced, for the second modification due to the
installation of a new air hoist, which could not be adjusted to exactly the
same withdrawal rate as the original hoist.  This resulted in a withdrawal
rate that was slower than the baseline and faster than the first modification
This rate will be referred to as the "intermediate" withdrawal rate.  The
combination of increased drain times and the intermediate withdrawal rate
allow solutions to drip back into the process tank, thus reducing the drag
out.

-------
                             SAMPLING AND ANALYSIS

         To evaluate the  modifications of the rinsing process, MnTAP
established a baseline for process solution drag out,  rinse water use, rate of
rack withdrawal, and drain time for the Micom sensitize line.  To measure the
drag out of solution from the process tank, the incoming softened water flow
to the rinse tanks was temporarily shut off.  To account for copper already
present  in the rinse tanks (Figure 1),  samples were taken before and after a
known quantity of printed circuit boards were rinsed (one or two racks of 24
boards).  The total copper concentration in the samples was analytically
determined by atomic absorption.  The change in copper concentration was
calculated by subtracting the initial (before rinsing) concentration of copper
in each  rinse from the final (after rinsing) concentration.  The process tank
was sampled immediately  after a rack of boards was removed.  The surface area
of the printed circuit boards rinsed during the sample period was calculated
to enable a comparison based on square footage of production.  The additional
mass of copper in the rinses is equal to the change in copper concentration in
the rinses multiplied by the volume of each rinse.  The drag out was
calculated, as shown below, by dividing the total amount of copper in the
rinses by the concentration of the process solution and then normalizing the
result by dividing by the surface area of the boards processed.  (Quality
Assurance Project Plan,  Section III, page 7, 1990)


               (change in copper
               concentration in rinses) X (volume of rinses)
Drag out  =    	mq/L	'	m]	
(ml/ftf)
               (copper concentration         (surface area
                in process solution)    X       of boards)
                          mg/L                     ft2


The tank following the countercurrent rinses was sampled to check the amount
of copper build-up.  This was monitored during the baseline period to indicate
the effectiveness of these rinses.  This measurement provides a standard to
use while comparing rinse water reduction modifications.

         The water  flow rate  was calculated  by measuring  the  time  it  took to
fill a five gallon container from the rinse tank discharge.  The rinse water
flow rate is expressed in gallons per minute.  The withdrawal time was
determined by measuring  the time it took to lift a rack of boards from the
process bath to the height needed to clear the wall of the tank (A to B in
Figure 2).   The withdrawal height was 34 inches.  The average withdrawal rate
was determined by dividing the height by the average withdrawal  time.


                                      8

-------
                        withdrawal height
Withdrawal Rate  =  	
  (ft/min)
                     average withdrawal time


The drain time was determined by measuring the time after the rack was
withdrawn until the rack was halfway over the next tank (B to C in Figure 3).
The total time is the sum of the withdrawal time and the drain time (A to C in
Figures 2 & 3).

BASELINE SAMPLING

         The  baseline  samples were collected  from March  12-23,  1990.  The
micro-etch bath and rinses were sampled before and after a single rack of
printed circuit boards were rinsed.  The electroless copper plating bath and
rinses were sampled before and after each pair of racks were rinsed.    One
hundred thirty-six samples were analyzed to determine the volume of drag out
from the micro-etch and the electroless copper baths for twelve pairs of racks
over a two week period.  The withdrawal time and total drain time were
measured for each rack transfer.  The withdrawal rate was calculated  from the
average withdrawal time.  The flow rates of the rinses following the  micro-
etch and electroless copper baths were monitored twelve times over two days.
The surface area of the boards in each rack was calculated from measurements
of the boards.

FIRST MODIFICATION SAMPLING

         The  first modification,  slowing the  rate of withdrawal, was tested  on
November 15, 1990.  One hundred thirty-six (136) samples were analyzed in
order to determine the volume of drag out from the micro-etch and the
electroless copper baths for twelve pairs of racks over a one day period.  The
withdrawal time and total drain time were measured for each rack transfer.
The flow rates of the rinses were not sampled due to the water being  turned
off to determine the drag out.  The surface area of the boards in each rack
was calculated from measurements of the boards.

SECOND MODIFICATION SAMPLING

         The  second modification,  increasing  the drain time with an
intermediate withdrawal rate, was tested on December 10 and 11, 1990.  One
hundred and nine samples (109) were analyzed in order to calculate the volume
of drag out from the micro-etch and the electroless copper baths for nine
pairs of racks over a two day period.  The withdrawal time and total  drain
time were measured for each rack transfer.  During chemical sampling of the
rinse tanks, rinse water was turned off; flow rates were measured after the
rinse waters had been turned back on.  The surface area of the boards in each
rack was calculated from measurements of the boards.

-------
               Rinse
               Tank
Process
 Tank
Rl nse
Tank
Figure 2.  Rack Positions Used in Determining
           Withdrawal Rate







i6
-^-J-Li-



* i
6 $
-x — "-^_
FV-ocess
Tank
B

Rinse
Tank







6
Process
Tank


C
'd
Rinse
Tank





Figure  3.   Rack Positions Used  in Determining
            Drain Time
                          10

-------
                                    RESULTS


        The calculated  amount  of  drag  out during  the  baseline evaluation was
12.1 ml/ft  from the micro-etch bath and 6.0 ml/ft  from the  electroless
copper bath.  Racks of printed circuit boards were withdrawn from the tanks
very quickly,  usually taking less than two seconds.  The average drain time
for racks being removed from the electroless copper bath was longer than those
removed from the micro-etch bath because of an additional  distance some racks
travel while still draining over the wider electroless copper tank.  The
results of the baseline evaluation are summarized in Tables 1 & 2.

            TABLE 1.  RESULTS OF BASELINE MICRO-ETCH EVALUATION
Average
Range
n Std.
Dev.
                                           Low   High

 Drag out (ml/ft2)               12.1      9.3   14.5    12      1.5

 Withdrawal Time  (seconds)       1.7       1.0   3.2     12      0.8

 Withdrawal Rate  (ft/min)        100       -              -        -

 Drain Time (seconds)            3.4       1.2   7.6     12      2.1

 Total Time (seconds)            5.1       2.3   8.9     12      2.3

 Surface Area/Rack  (ft2)          88        48    126     12      24

 Water Flow Rate  (GPM)           2.6       2.5   2.7      6    0.04
                                      11

-------
         TABLE  2.  RESULTS OF BASELINE ELECTROLESS COPPER EVALUATION
Average

Drag out (ml/ft2) 6.0
Withdrawal Time (seconds) 1.8
Range
Low
4.7
1.0
High
7.6
3.2
n

12
23
Std.
Dev.

1.1
1.6
 Withdrawal Rate (ft/min)        94         -             -       '

 Drain Time (seconds)            5.2        0.8   10.2   23     2.6

 Total Time (seconds)            7.0        1.3   11.4   23     3.1

 Surface Area/Rack (ft2)         169        124   234    12      33

 Water Flow Rate (GPM)           3.3        3.2   3.3     6    0.05

        The calculated  amount  of drag  out  during the  evaluation of  the  first
modification was 6.7 ml/ft2 from the  micro-etch bath and 3.0 ml/ft   from the
electroless copper bath.  Racks  of printed circuit boards were withdrawn from
the tanks slowly, taking from thirteen to  sixteen seconds.  The results of the
first modification evaluation are summarized  in Tables 3 & 4.
          TABLE 3.   RESULTS OF MODIFICATION 1 MICRO-ETCH EVALUATION -
                             SLOWED RATE OF  WITHDRAWAL
Average Range

Drag out (ml/ft2)
Withdrawal Time (seconds)
Withdrawal Rate (ft/min)
Drain Time (seconds)
Total Time (seconds)
Surface Area/Rack (ft2)

6.7
14.9
11
2.5
17.4
83
Low
5.2
13.6

1.4
15.5
56
High
9.7
16.0

4.6
19.4
114
n

12
11
-
11
11
12
Std.
Dev.

1.3
0.7
-
0.9
1.1
14
                                      12

-------
      TABLE  4.   RESULTS  OF  MODIFICATION  1  ELECTROLESS  COPPER  EVALUATION
                            SLOWED RATE OF WITHDRAWAL
                                 Average      Range      n      Std.
                                                               Dev.

                                           Low   High

 Drag out (ml/ft2)                3.0       2.1     3.4    12      0.4

 Withdrawal  Time (seconds)       13.9      13.1  15.4    23      0.7

 Withdrawal  Rate (ft/min)        12        ...

 Drain Time (seconds)            3.2       1.1   8.7     23      2.1

 Total Time (seconds)            17.1      14.4  23.0    23      2.3

 Surface Area/Rack (ft2)         162       104   220     12      32


         The calculated  amount  of drag  out  during the evaluation of the second
modification was 7.1 ml/ft  from the micro etch bath and 2.9 ml/ft  from the
electroless copper bath.  Racks of printed circuit boards were withdrawn from
the tanks slightly faster than the baseline but much slower than the first
modification,  taking from two to five seconds.  The results of the second
modification evaluation are summarized  in Tables 5 & 6.
          TABLE 5.   RESULTS OF MODIFICATION 2  MICRO-ETCH  EVALUATION
                LONGER DRAIN TIME WITH INTERMEDIATE WITHDRAWAL RATE


Drag out (ml/ft2)
Withdrawal Time (seconds)
Withdrawal Rate (ft/min)
Drain Time (seconds)
Total Time (seconds)
Surface Area/Rack (ft2)
Average

7.1
4.3
40
12.1
16.4
92
Range
Low
5.5
2.1
-
11.2
13.7
60
High
8.2
5.6
-
13.7
18.0
119
n

12
10
-
10
10
10
Std.
Dev.

0.9
0.9

0.9
1.3
17
                                      13

-------
       TABLE  6    RESULTS OF MODIFICATION 2 ELECTROLESS COPPER EVALUATION
                  LONGER DRAIN TIME WITH INTERMEDIATE WITHDRAWAL RATE
                                  Average       Range       n       Std.
                                                                   Dev.

                                             Low   High

 Drag out (ml/ft2)                2.9          1.9   3.7     12        0.5

 Withdrawal Time (seconds)       4.3          3.5   4.9     19        0.4

 Withdrawal Rate (ft/min)        40           -

 Drain Time (seconds)            11.9         11.1  16.4    19        1.2

 Total Time (seconds)            16.3         15.0  20.5    19        1.2

 Surface Area/Rack (ft2)         175          60    228     10        49


DIFFERENCES IN DRAG OUT DUE TO RINSING MODIFICATIONS

         Slowing  the  rate  of withdrawal  (modification  1)  lowered  the  drag  out
from the micro-etch solution from the baseline drag out of 12.1 ml/ft  to
6.7 ml/ft ,  while extending the drain time combined with an intermediate
withdrawal rate  (modification 2) yielded a similar drag out of 7.1 ml/ft .
The drag out from the electro!ess copper plating bath was reduced from 6.0
ml/ft  to 3.0 ml/ft2  when  the withdrawal  rate was slowed and  reduced  to 2.9
ml/ft2 after the drain time was lengthened using an intermediate withdrawal
rate.

         These  lower  drag  out volumes  represent a 45  percent  reduction  for
modification 1 and 41 percent reduction for modification 2 from the micro-
etch; for the elecroless  copper, modification  1  resulted in a 50 percent
reduction, while modification 2 resulted in a  52 percent reduction in drag
out.  The differences in  drag out are summarized in Table 7 and Table 8.

         Comparisons  of  the baseline  to  each  modification  using  statistical
analysis confirm significant reduction at an a error level less than 0.005.
This yields a confidence  level greater than 99.5% when each modification is
compared to the  baseline.  The evaluation indicates a significant reduction in
drag out when the withdrawal rate is slowed or the drain time is increased
with an  intermediate withdrawal rate.  A statistical  analysis to determine the
significance of  the difference between modifications 1 and 2 was not
performed.

         The  reduction  in  drag  out from  the first modification was equivalent
to preventing 194 grams of copper from the micro-etch bath and 9 grams of
copper from the  electroless bath per day from entering the rinse water waste
stream,  for a total of 203 grams/day.  The figures for the second modification


                                      14

-------
                   TABLE 7.  SUMMARY OF MICRO-ETCH RESULTS

BASELINE
MODIFICATION 1
slower rate of
withdrawal
MODIFICATION 2
longer drain time
with intermediate
withdrawal rate
Withdrawal
Rate
(ft/min)
100
11
40
TABLE 8. SUMMARY

BASELINE
MODIFICATION 1
Withdrawal
Rate
(ft/min)
94
12
Time
Withdrawal
(seconds)
1.7
14.9
4.3
Drain
Time
(seconds)
3.4
2.5
12.1
OF ELECTROLESS COPPER
Time
Withdrawal
(seconds)
1.8
13.9
Drain
Time
(seconds)
5.2
3.2
Total
Time
(seconds)
5.1
17.4
16.4
RESULTS
Total
Time
(seconds)
7.0
17.1
Dragout
(ml/ft2)
12.1
6.7
7.1

Dragout
(ml/ft2)
6.0
3.0
slower rate of
withdrawal

MODIFICATION 2
longer drain time
with intermediate
withdrawal rate
40
4.3
11.9
16.3
2.9
                                      15

-------
are 180 grams prevented from leaving the micro-etch bath and 9 grams,  for a
total of 189 grams/day, prevented from leaving the electroless bath.   These
figures assumed a copper concentration of 30 grams/liter in the micro-etch
bath and 2.4 grams/liter in the electroless bath and a production level  of
1200 ft  of printed circuit board per day.

         Because  copper concentration  in rinse water was reduced, the
potential for conserving rinse water flows was also shown,  although this was
not directly tested.  Rinse water flows could be turned down proportionate to
the reduction in drag out and still  maintain the same rinsing efficiencies.

QUALITY ASSURANCE PROJECT PLAN RESULTS

         Sampling procedures were followed  according to  the  Quality Assurance
Project Plan (QAPjP).  The baseline sampling was performed over a two week
period, the first modification was tested and sampled over a two shift period,
and the second modification was  sampled over a two day period.  A wide variety
of printed circuit boards were being manufactured during each of the sampling
periods.  Although the boards were not identical in each case, they were
representative of the work at Micom.

         To  assure  compliance  with the QAPjP,  field  audits  were  conducted by
Barb Loida, MnTAP engineer.  The  auditor reviewed the field activities at
Micom such as sample collection,  chain of custody, and sample information.

         Analytical  methods were  followed  according  to the  approved QAPjP by
PACE Laboratories, Inc., with the exception of the electroless copper plating
bath samples.  These samples, when preserved with nitric acid according to
instructions from the laboratory, precipitated out the copper as the solution
cooled.  The analytical procedure was modified to include preservation with a
hydrochloric/nitric acid mixture, digestion of the sample, and analysis for
total copper.

          The analytical  results for the  electroless  copper bath samples for
the baseline evaluation averaged  1600 mg/1 and one sample was lost due to a
laboratory accident during the digestion process.  Despite additional
validation of the procedure, the  results from the samples taken during the
first modification period averaged only 1800 mg/1 with one sample as low as
900 mg/1.  The samples during the second modification averaged 2200 mg/1 after
the laboratory recalculated the  results to account for the 30 ml of acid that
had been added to each sample per the modified analytical procedure.

         These results  do  not  compare well  with  the  electroless  copper bath
operating concentrations according to Micom's laboratory analysis and line
operating procedures.  It should  also be pointed out that an electroless
copper plating solution will not  plate at such low copper concentrations.
This led to the decision to accept an average bath copper concentration of
2400 mg/L as determined by Micom  process control charts and results from the
Micom laboratory.  The control charts show an operating range of ± 100 mg/L of
copper; additions are automatically pumped into the plating bath as required.
                                      16

-------
This average of 2400 mg/L was used in place of analytical results for the
electroless copper solution only, and was used for baseline, modification 1,
and modification 2 evaluation calculations.

        Although the  reason  for  the  poor  analytical  results  for  the
electroless copper samples is not clear, several possibilities include:
improper acidification, plate out of sample on container walls, sample not
maintained at bath temperature of 103 F, or the results may not have been
adjusted to account for the addition of acid.

        Quality control  samples  were analyzed  for  copper.   Field blank,
precision - relative percent difference (RPD), and percent recovery - matrix
spike/matrix spike duplicates (MS/MSDs) were analyzed.  Copper was not
detectable at the method detection limit in the field blanks.  All  of the RPDs
between duplicates were less than the 20 percent QC target range.  All of the
percent recoveries were within the 75 to 125 percent recovery target range.
                                       17

-------
                                  DISCUSSION
TECHNICAL  EVALUATION
         Although possible,  the  ability  to  slow  the  mechanical  hoist's  rate  of
withdrawal  is complicated by a few items.  The mechanical hoist at Mi com is
capable  of  a slower vertical rate, but this not only slows the rate of
withdrawal  but also slows the rate of insertion.  The slower overall rate
therefore demands more of the hoist's operating time, making it difficult to
move the hoist to the opposite end of the line in time to transfer another
rack, thus  slowing production.  The maximum horizontal speed of 40 feet per
minute is not fast enough to allow travel from one end to the other and still
maintain current production levels.   The mechanical hoist was also prone to
breakdown,  requiring repair.

         The extended drain  time was  tested  by manually signaling  the rack
transfer after a ten second drain time over the copper bearing process
solutions.  The  line operator could implement this by counting to ten slowly
before transferring the rack.

ECONOMIC ANALYSIS

         An  economic evaluation  showed that  the  company could save $2640 per
year in  rinse water treatment and disposal costs by implementing modification
1 or $2460  per year by implementing modification 2.  Rinse water was treated
using on-site ion exchange cannisters which had a capacity of 46 pounds of
copper before requiring off-site regeneration at a cost of $1096 per
cannister.  Savings were calculated using the following formula:


            (amount of copper prevented)  (cost of canister)
Savings  =  	
                    capacity of ion exchange canister


         The savings do not  take into account the  time required  to change the
peed of the hoist or train the operators to extend the drain time.  The
additional  time required to withdraw or drain the racks could be offset by
shortening  some rinsing and/or holding times or by reducing the spacing
between production runs.   One production run consists of a pair of racks which
enter the electroless copper bath at approximately the same time.   Filled
racks are often held in rinse tanks or placed at the loading area before
entering the sensitize line.  The electroless copper plating process is the
                                      18

-------
rate determining step and a production run remains in this bath for 30 minutes
before being rinsed.

IMPLEMENTATION

         For  a  number of site  specific reasons  the  company decided  to
implement a longer drain time to reduce drag out instead  of altering the
withdrawal rate.  As discussed above, the programmable mechanical hoist used
for modification 1 at Micom was unreliable and often broke down.  Another
problem with the mechanical hoist at  Micom was its inflexibility in
programming.  To slow the withdrawal  rate for modification 1 the vertical
speed had to be adjusted, thus slowing not only the withdrawal rates but also
the insertion rates; these rates were slowed for insertion and withdrawal into
all tanks on the sensitize line, not  just the micro-etch  and the electroless
tanks.  In order to maintain production rates at previous levels, operators
supplemented the mechanical hoist with an air-assisted hoist.

         As  a result  of  recurrent  break downs of the mechanical  hoist and the
inability to specifically target and  program  slower withdrawal rates for the
two bath tanks, the company took the  mechanical hoist out of service.   Air-
assisted hoists were used as replacements.  With this type operation it made
more sense for the company to implement a longer drain time to achieve drag
out reduction.   The company believed  that it would be easier to train
operators to increase the drain time  over the two tanks rather than have them
slow withdrawal rates.

         For  modification 2  the  additional time added to the sensitize line to
allow the intermediate withdrawal rate and longer drain time as compared to
baseline was 21 seconds.  This amount was negligible when compared to the 60
minute total production  time through  the sensitize line.   Minor modifications
to the operation of the  line could offset the added time  so that the baseline
production rate could be maintained.  At Micom, filled racks were often held
beyond the necessary times in rinse tanks while processing in the sensitize
line or placed at the loading area before entering the line.   Changes such  as
shortening the timing between rack starts and/or reducing holding times in
rinse tanks that were known to be more than adequate could makeup for the
added 21 seconds.

FUTURE MODIFICATIONS

         Once drag  out reduction  has  been  optimized, less  water would be
required to maintain the cleanliness  of the rinses.  Using less water will
decrease water and sewer charges, which is especially important at Micom where
incoming water must be softened before use on the sensitize line.   Decreased
rinse water volumes may  reduce treatment costs by making the ion exchange
system at Micom more effective at removing copper from rinse water and by
reducing the amount of water which must pass through the resin beds.
                                      19

-------
        Water use can be reduced proportionately with contaminant reduction.
An additional savings of $710 per year for modification  1,  or $660 for
modification 2,  could be realized in avoided water and sewer charges if the
company reduced rinse water flow rates proportional  to the  reduced copper
contamination of the rinse water that would result from  implementing the drag
out reducing modifications tested.   Less water would also need to be softened,
but lower costs would not be evident as water softening  is  billed as a monthly
fee independent of water use.  This fee may also be reduced through
negotiations with the water softening company.

        MnTAP intended  to maintain  or  improve the current rinsing
effectiveness.  This was monitored during the baseline by checking the amount
of copper build-up in the non-flowing tank which follows the countercurrent
rinses.  Copper buildup in tank 4,  Figure 1, ranged between zero and 12 mg/1
over eight hours while copper buildup in tank 8 ranged between 0.01 and 0.08
mg/1.  This monitoring was not continued during the modifications as the rinse
water was turned off for a considerable amount of time during the drag out
evaluations.  These rinses should be monitored when the  rinse water flow rates
are reduced.  The buildup of copper in these tanks can be a good monitor of
the quality of the rinsing when it is compared to a baseline.

TRANSFERRING PROJECT RESULTS

        For  the Micom project,  not  only  is  the work centered  on  one process
line, but also further narrowed to concentrate on two process steps within
that line, the micro-etch and the electroless copper plating solution.  The
reason for the narrow focus was first a matter of available budget for
performing the evaluation, since a limited amount of time and money is
available to perform the necessary sampling, analysis and data reduction for
such an evaluation.   Another reason is that these two processes are the most
concentrated copper sources on the process line.  The focus is also part of a
larger philosophy, one which MnTAP calls "planting seeds."   The hope is that
if rinsing process modifications are demonstrated to be successful in one
location in a plating operation, plant management will be more likely to
attempt similar modifications on other process lines.   Rather than attempting
plant-wide modifications, or changes applied to large, aggregated waste
streams, a staged approach is not only more manageable,  but may improve
accuracy of cost and product quality assessments.
                                      20

-------
                                RECOMMENDATIONS
        Besides  the  two  objectives  of reducing  copper  discharge  and water
use, in the future MnTAP plans to develop and test a procedure for evaluating
and modifying rinsing processes which will be transferable to and useful to
other operations.  This procedure would address such imperatives as protecting
product quality while modifying the production process, evaluating the impact
of modifications on production throughput, and providing a "rule-of-thumb" for
such decisions as number of samples required for evaluation.  This procedure
will be an important product of the project.  By evaluating various rinsing
modifications at additional companies and additional process lines, MnTAP
intends to collect the information necessary for companies to attempt a
modification to their rinsing practices.
                                      21

-------
                                 BIBLIOGRAPHY


Control and Treatment Technology for the Metal Finishing Industry - In-Plant
Changes.  EPA 625/8-82-08.

Cushnie, George C., Jr.   Electroplating Wastewater Pollution Control
Technology.  Noyes Publications, Park Ridge, NJ, 1985.

Durney, Lawrence Jr., ed.  Electroplating Engineering Handbook (4th Edition),
Van Nostrand Reinhold, NY, 1984.

Environmental Pollution Control Alternatives -- Reducing Water Pollution
Control Costs in the Electroplating Industry.  EPA 625/5-85-016.

Foecke, Terence.  Personal interview with Terence Foecke, Director, WRITAR,
Minneapolis, Minnesota, Feb. 13, 1991.

In-Process Pollution Assessment:  Upgrading Metal-Finishing Facilities to
Reduce Pollution. EPA 625/3-73-546.

Kushner, Joseph B.  Hater and Waste Control for the Plating Shop.  Gardner
Publications, Cincinnati, OH,  1981.

Lowenheim, Frederick, Modern Electroplating.  Wiley Interscience.  Meltzer,
Michael, Ph.D., "Reducing Environmental Risk:  Source Reduction for the
Electroplating Industry."  Dissertation, UCLA.

Minnesota Technical Assistance Program (MnTAP).  Quality Assurance Pro.iect
Plan for the WRITE Program With States: Evaluation of Modified Rinsing
Technologies at Micom, Inc., New Brighton, Minnesota.  March 1, 1990.

Nunno, T., et. al.  Toxic Waste Minimization in the Printed Circuit Board
Industry, Noyes Data Corporation, Park Ridge, NJ, 1988.
The Minnesota Technical Assistance Program (MnTAP) was created in 1984 with
support from the Minnesota Office of Waste Management (OWM).  MnTAP is located
at the University of Minnesota and is a non-regulatory program designed to
help Minnesota business and industry prevent pollution at its source and
properly manage industrial wastes.  These services include:  telephone
assistance, on-site visits, student interns, and an information clearinghouse.
                                      22

-------
               APPENDIX
Raw Data and Calculated Drag Out Values
                   23

-------
Laboratory Data and Calculations for Baseline - March 12-23, 1990
number board area
date time rack # of boards In rack
In rack oqft

03/12/90 14:00 9 24 56.0
10 21 71.2
9&10 45 127.2





03/13/90 13:00 1 24 48.0
2 24 76.0
1&2 48 124





03/13/90 16:30 8 24 108.5
9 20 32.4
8&9 44 140.9





03/14/90 16:00 11 24 84.0
10 24 89.0
10&11 48 173




tank #
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Cu cone. Cu cone change In dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/eqft mg/sqtt mg/sqft

160
4.8
38

3
0.07
0.05

0.63
0.03


2.2
0.02


9
0.07
54

0.17
0.07
0.1

240
2.5


8
0.26

33800
250
7
44
240
9.1
0.14
0.11
33700
68
0.56

250
8.4
0.1

31400
205
1.7
62
380
11
0.16
0.12
30600
370
4.55

2200
18
0.36


90 21483 895 384
2.2
6

6.1 1450 32 11
0.07
0.06

67.37 15821 659 330
0.53


6.2 1476 31 12
0.08


196 46048 1919 424
1.63
8

10.83 2566 58 18
0.09
0.02

130 30768 1282 366
2.05


10 2373 49 14
0.1

dragout
volume
ml/sqft

11.3



4.7



9.8



5.0



13.5



7.6



12.0



5.7



-------
                                                                         Baseline
IX)
en
number board area
date time rack # of boards in rack tank #
in rack sqft
1
03/14/90 16:30 12 24 88.5 2
13 24 84.0 3
12&13 48 172.5 4
5
6
7
8
1
03/15/90 13:00 6 24 86.0 2
5 24 96.0 3
5&6 48 182 4
5
6
7
8
1
03/16/90 16:00 12 24 126.0 2
13 25 107.5 3
12&13 49 233.5 4
5
6
7
8
1
03/19/90 16:15 7 24 84.0 2
6 24 84.0 3
6&7 48 168 4
5
6
7
8
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/sqft mg/sqft

220
2.7
70

8.3
0.19
0.08

440
7.4
86

7.8
0.06
0.07

1
0.02
100

0.17
0.05
0.15

140
1.3
88

1
0.07
0.16
29000
355
4.8
79
2400
17
0.37
0.15
38500
570
9.4
88
2100
17
0.21
0.14
31400
200
1.5
110
1800
18
0.26
0.22
31600
290
3.2
100
1300
11
0.19
0.17

135 31944 1331 361
2.1
9

8.7 2086 43 12
0.18
0.07

130 30756 1282 358
2
2

9.2 2197 46 12
0.15
0.07

199 46712 1946 371
1.48
10

17.83 4239 87 18
0.21
0.07

150 35393 1475 421
1.9
12

10 2378 50 14
0.12
0.01
dragout
volume
ml/sqft

12.4



5.0



9.3



5.0



11.8



7.6



13.3



5.9



-------
Baseline
number boa'd area
date time rack # of boards in rack tank #
in rack sqft
1
03/2O/90 12:45 1 24 64.0 2
2 24 64.0 3
1&2 48 128 4
5
6
7
8
1
03/20/90 14:30 8 24 84.0 2
7 24 107.8 3
7&8 48 191.8 4
5
6
7
8
1
03/22/90 14:15 3 24 105.0 2
4 23 74.5 3
3&4 47 179.5 4
5
6
7
8
1
03/23/90 14:00 7 24 115.5 2
8 24 85.8 3
7&8 48 201.25 4
5
6
7
8
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/scrft mg/sqft

0.03
0.37


1.1
0


340
4.4
120

3.7
0.06
0.09

91
2.5
140

2.4
0.02
0.06

300
7.9
160

4.1
0.05

31900
110
1.6

NA
9
0.1

25400
470
7
120
2400
18
0.28
0.12
34000
290
4.7
160
2200
11
0.16
0.09
38200
510
14
160
2300
18
0.25
0.1

109.97 25910 1080 405
1.23


7.9 1880 39 15
0.1


130 30896 1287 368
2.6
0

14.3 3412 71 18
0.22
0.03

199 46880 1953 446
2.2
20

8.6 2053 44 11
0.14
0.03

210 50351 2098 436
6.1
0

13.9 3313 69 16
0.2
0.1
dragout
volume
ml/sqft

12.7



6.1



14.5



7.4



13.1



4.8



11.4



6.9



-------
                                             Laboratory Data and Calculations for Modification 1  •  November 15, 1990
ro
number board area
date time rack # of boards In rack tank #
In rack sqft
1
11/15/90 08:30 2 24 80.0 2
1 20 140.0 3
1 &2 44 220 4
5
6
7
8
1
11/15/90 10:15 3 19 66.5 2
4 24 84.0 3
3&4 43 150.5 4
5
6
7
8
1
11/15/90 11:00 5 24 84.0 2
6 24 84.0 3
5&6 48 168 4
5
6
7
8
1
11/15/90 11:45 7 24 84.0 2
8 24 104.0 3
8 24 104 4
5
6
7
8
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/sqft mg/sqft

91
0.78
190

0
0
0.15

40
0.55


0.45
0


55
0.93


2
0.24


110
1.1


4
0.09

22000
150
2.5
210
2100
4.7
0.09
0.215
32400
100
2.45

2100
4.8
0.09

35700
140
1.7

1900
6.8
0.14

29700
180
2.5

900
7.4
0.13


59 14148 589 177
1.72
20

4.7 1125 26 5
0.09
0.065

60 14423 759 217
1.9


4.35 1043 24 7
0.09


85 19984 833 238
0.77


4.8 1105 23 7
-0.1


70 16636 693 198
1.4


3.4 808 34 8
0.04

dragout
volume
ml/sqft

8.0



2.1



6.7



2.9



6.7



2.7



6.7



3.2



-------
                                                                            Modification 1
ro
CO
number board area
date time rack # of boards in rack tank #
in rack sqft
1
11/15/90 13:15 10 24 84.0 2
9 24 94.0 3
9&10 48 178 4
5
6
7
8
1
11/15/90 14:00 12 24 84.0 2
11 13 45.5 3
11&12 37 129.5 4
5
6
7
8
1
11/15/90 14:45 13 24 84.0 2
14 24 84.0 3
13&14 48 168 4
5
6
7
8
1
11/15/90 15:15 15 24 84.0 2
16 24 100.0 3
15&16 48 184 4
5
6
7
8
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/sqft mg/sqft

170
2


2.8
0


200
2.8


4.55
0.09


190
3.9


5.7
0.2


320
9.2


7.2
0.33

32400
280
5.1

1600
8.8
0.18

37600
270
5.2

2100
8.1
0.18

39900
275
5.5

1500
11
0.32

36100
400
9.7

2000
12
0.41


110 26352 1098 314
3.1


6 1452 30 8
0.18


70 16869 703 201
2.4


3.55 855 23 7
0.09


85 20178 841 240
1.6


5.3 1273 27 8
0.12


80 18757 782 223
0.5


4.8 1147 24 6
0.08

dragout
volume
ml/sqft

9.7



3.4



5.3



2.8



6.0



3.2



6.2



2.6



-------
Modification 1
date time

11/15/90 16:00







11/15/90 16:45







11/15/90 17:30







11/15/90 18:00







11/15/90 13:10

rack*

17
16
17&18





20
19
19&20





21
22
21 8.22





23
24
23&24





9

number
of boards
In rack

24
24
46





25
24
49





26
24
50





24
17
41





24

board area Cu cone. Cu cone change in dragout/
In rack tank # before rinse after rinse Cu cone rack
sqtt mg/l mg/l mg/l mg/sqtt
1
64.0 2
84.0 3
168 4
5
6
7
8
1
55.5 2
84.0 3
139.5 4
5
6
7
8
1
114.0 2
84.0 3
198 4
5
6
7
8
1
84.0 2 X
51.6 3 X
135.6 4
5
6
7
8
1
94.0 2
3
27800
290 350 60 14155
5.8 6.55 0.75

1400
4.9 9.2 4.3 1133
0.14 0.67 0.53

41400
290 340 50 11860
4.1 5 0.9

2300
4.5 8.7 4.2 1003
0.15 0.22 0.07

39500
170 300 130 30593
2.5 3.8 1.3

2800
4.1 8.7 4.6 1107
0.11 0.22 0.11

40600
430 430 101355
5 5

1500
5.7 9 3.3 794
0.14 0.22 0.08

32400
72 170 98 23300
022
dragout/ dragout/ dragout
board area volume
mg/sqft mg/Eqft ml/eqft

590 169 6.1



24 7 2.8



474 214 5.2



20 7 3.0



1177 268 6.8



22 6 2.3



4223 1207 |X



19 6 2.4



971 248 7.7


-------
                                             Laboratory Data and Calculations for Modification 2  -  December 10 & 11, 1990
00
o
number
date time rack # of boards
in rack

12/10/90 11:00 5 24
6 24
58,6 48

1 1 :30 7 24
8 24
7&8 48

12/10/90 12:00 9 21
10 24
9&10 45





12/10/90 13:15 11 24
12 24
11 & 12 48





12/10/90 13:20 12 24

11&12 48



board area
in rack
sqft

108.0
108.0
216

108.0
108.0
216

94.5
108.8
203.25





87.5
93.5
181





93.5

181



tank*
5
6
7

5
6
7

1
2
3
4
5
6
7
B
1
2
3
4
5
6
7
8
1
2
3

1
2
3
Cu cone. Cu cone change in dragout/ dragout/
before rinse after rinse Cu cone rack board
mg/l mg/l mg/l mg/sqft mg/sqft

3.8
0


7
0.2


250
8


4.5
0.08


140
3.3


1.8
0


240
5.2


140
3.3
2300
9.3
0.17

2100
14
0.32

36200
340
11

2000
10.4
0.22

33200
240
5.2

2300
6.4
0.06

33200
330
8.1

33200
330
8.1

5.5 1332 28
0.17


7 1673 35
0.12


90 21669 1032
3


5.9 1419 32
0.14


100 23743 989
1.9


4.6 1095 23
0.06


90 21646 902
2.9


190 45764 953
4.8
dragout/ dragout
area volume
mg/sqft ml/sqft

6 2.6



B 3.2



229 6.3



7 2.9



271 8.2



6 2.5



232 7.0



253 7.6


-------
Modification 2
number board area
date time rac'< # °' boards in rack tank #
in rack sqft
1
12/10/90 14:45 13 24 108.5 2
14 24 119.0 3
13&14 48 227.5 4
5
6
7
8
1
12/10/90 14:50 14 24 119.0 2
3
13&14 48 227.5
1
2
3
1
12/11/90 09:30 4 21 59.5 2
3
4 21 59.5 4
5
6
7
8
1
12711/90 10:15 5 24 108.0 2
6 11 49.5 3
5&6 35 157.5 4
5
6
7
8
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/sqft mg/sqft

54
1.4


0.95
0


170
3.5


54
1.4

160
4.7


7.9
0


120
2.9


6.5
0.04

31100
170
3.5

2000
7.95
0.08

31100
290
7.4

31100
290
7.4
32950
220
4.7

2200
10
0.13

32900
200
7.2

2300
9.5
0.14


116 27517 1147 254
2.1


7 1664 35 7
0.08


120 28869 1203 243
3.9


236 56852 1184 250
6

60 13980 666 235
0


2.1 524 25 9
0.13


80 19642 818 182
4.3


3 728 21 5
0.1

dragout
volume
ml/sqft

8.2



3.0



7.8



8.0


7.1



3.7



5.5


1.9



-------
                                                                     Modification 2
Co
no
number board area
date time rack # of boards in rack
In rack sqft

12/11/90 11:00 7 24 84.0
8 24 84.0
7&8 48 168





12/11/90 11:45 9 24 80.0
10 24 64.0
98,10 48 144





12/11/90 12:15 11 25 87.5
12 26 91.0
11&12 51 178.5





12/11/90 12:15 11 25 87.5

12 26 91.0



tank*
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
5
6
7

5
6
7
Cu cone. Cu cone change in dragout/ dragout/ dragout/
before rinse after rinse Cu cone rack board area
mg/l mg/l mg/l mg/sqft mg/sqft mg/sqft

135
5.4


4.2
0


190
6.4


9.5
0.12


240
9.3


4.1
0.12


4.1
0.12


6.7
0.19
36000
230
7

2300
8
0.09

33300
255
8.7

2200
13
0.31

35700
320
11

2500
9.9
0.32

2500
6.7
0,19

2500
9.9
0.32

95 22508 938 268
1.6


3.8 914 19 5
0.09


65 15681 653 196
2.3


3.5 867 18 6
0.19


80 19036 761 218
1.7


5.8 1409 28 8
0.2


2.6 622 25 7
0.07


3.2 782 30 9
0.13
dragout
volume
ml/sqft

7.4



2.3



5.9



2.5



6.1



3.3



3.0



3.6

        notes:

          03/20/90


          11/15/90
12:45    The sample from tank #5 was destroyed in a laboratory accident.


18:00    Tank #2 & tank #3 were not sampled before the racks were rinsed - Sampling error.

         Dragout volumes could not be calculated.

-------
CO
CO
                                       Dragout and Withdrawal/Drain Time
                                          Micro Etch Baseline
                                                 04/23/91
dragout withdrawal
DATE

03/12/90
03/13/90
03/13/90
03/14/90
03/14/90
03/15/90
03/16/90
03/19/90
03/20/90
03/20/90
03/22/90
03/23/90


standard

TIME

14:00
13:00
16:30
16:00
16:30
13:00
16:00
16:15
12:45
14:30
14:15
14:00


RACK#

9
1
8
11
12
6
12
7
1
8
3
7
total
average
deviation(n-l)
range
9.3
volume
ml/sqft
11.3
9.8
13.5
12.0
12.4
9.3
11.8
13.3
12.7
14.5
13.1
11.4
145.2
12.1
1.5
14.5
time
(s)
1
1.5
2.9
1.3
1.5
3.2
2.7
1.8
1.2
1
1.7
1
20.8
1.7
0.8

drain
time
(s)
3.4
5.5
2.7
7.6
6.2
3.2
4.3
1.2
1.5
1.3
1.5
2.3
40.7
3.4
2.1

total
time
(s)
4.4
7
5.6
8.9
7.7
6.4
7
3
2.7
2.3
3.2
3.3
61.5
5.1
2.3


-------
Dragout and Withdrawal/Drain Time
  Electroless Copper Baseline
           04/23/91

DATE TIME RACK #

03/12/90 14:00 9 & 10
03/13/90 13:00 1 & 2
03/13/90 16:30 8 & 9
03/14/90 16:00 10&11
03/14/90 16:30 12 & 13
03/15/90 13:00 5 & 6
03/16/90 16:00 12 & 13
03/19/90 16:15 6 & 7
03/20/90 12:45 1 & 2
03/20/90 14:30 7 & 8
03/22/90 14:15 3 & 4
03/23/90 14:00 7 & 8
Total
Average
Standard deviation (n-1)
Range 4.7


dragout
volume
ml/sqft
4.7
5.0
7.6
5.7
5.0
5.0
7.6
5.9
6.1
7.4
4.8
6.9
71.7
6.0
1.1
7.6


Withdrawal Time
(seconds)
first second
1
1.6 2.2
8.7 0.5
1 1
1.3 0.8
1.9 1.6
3.7 1.1
1.6 1.2
0.7 1.3
1.7 1.4
2.2 1.5
1.8 1.3
26.2 14.9
2.4 1.2
2.2 0.4
0.5 8.7
ave. 1.8
stds. 1.6
Drain Time
(seconds)
first second
9.7
4.7 8.3
6.3 0.8
2.7 9.8
5.2 7.5
4.1 5.5
1.4 2.9
4.4 10.2
7.6 4
3.3 3.8
4.8 6.5
2.9 4
47.4 73.0
4.3 6.1
1.7 3.1
0.8 10.2
5.2
2.6
Total Time
(seconds)
first second
10.7
6.3 10.5
15 1.3
3.7 10.8
6.5 8.3
6 7.1
5.1 4
6 11.4
8.3 5.3
5 5.2
7 8
4.7 5.3
73.6 87.9
6.7 7.3
3.0 3.2
1.3 11.4
7.0
3.1

-------
CO
en
                                 Dragout and Withdrawal/Drain Time
                                       Micro Etch - Modification 1
                                                 04/24/91
dragout withdrawal
DATE

11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90


Standard

TIME

08:30
10:15
11:00
11:45
13:10
13:15
14:00
14:45
15:15
16:00
16:45
17:30

Average
Deviation
Range
RACK#

2
3
5
7
9
10
12
13
15
17
20
21
Total

(n-1)
5.2
volume
ml/sqft
8.0
6.7
6.7
6.7
7.7
9.7
5.3
6.0
6.2
6.1
5.2
6.8
81.0
6.7
1.2
9.7
time
(s)
13.6
14.6
14.8
14.5

14.8
14.8
14.4
16.0
15.0
15.6
15.6
163.7
14.9
0.7

drain
time
(s)
1.9
3.0
2.2
1.6

4.6
3.0
2.4
1.8
3.0
1.4
2.4
27.3
2.5
0.9

total
time
(s)
15.5
17.6
17.0
16.1

19.4
17.8
16.8
17.8
18.0
17.0
18.0
191.0
17.4
1.1


-------
                                               Dragout and Withdrawal/Drain Time
                                               Electroless Copper - Modification 1
                                                          04/23/91

DATE

11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90
11/15/90


Standard




TIME

09:30
11:15
12:00
13:00
14:15
15:00
15:30
16:15
17:00
17:42
18:30
19:00

Average

RACK#

1&2
3&4
5&6
8
9&10
11&12
13&14
15&16
17&18
19&20
21&22
23&24
Total

Deviation(n-l)
Range


2.1


dragout
volume
ml/sqft
2.1
2.9
2.7
3.2
3.4
2.8
3.2
2.6
2.8
3.0
2.3
2.4
33.49
3.04
0.4
3.4


Withdrawal
(seconds)
first
14.9
13.8
13.5

13.4
13.1
13.2
14.9
13.8
14.1
14.0
14.1
152.8
13.9
0.6
13.1
ave.
stds.
Time

second
14.3
13.5
14.8
13.2
13.1
13.4
13.4
14.2
13.8
15.4
13.5
13.2
165.8
13.8
0.7
15.4
13.9
0.7
Drain
Time
(seconds)
first
3.7
2.3
1.4

2.0
4.3
2.2
1.1
1.6
1.1
2.6
2.1
24.4
2.2
1.0
1.1


second
8.7
2.2
4.8
6.7
1.3
2.3
3.0
4.5
6.6
5.9
2.4
1.9
50.3
4.2
2.4
8.7
3.2
2.1
Total
Time
(seconds)
first
18.6
16.1
14.9

15.4
17.4
15.4
16.0
15.4
15.2
16.6
16.2
177.2
16.1
1.1
14.4


second
23.0
15.7
19.6
19.9
14.4
15.7
16.4
18.7
20.4
21.3
15.9
15.1
216.1
18.0
2.8
23.0
17.1
2.3
CO
CTi

-------
GO
                                        Dragout and Withdrawal/Drain Time
                                         Micro Etch - Modification 2
                                                 04/24/91
dragout withdrawal drain
DATE

12/10/90
12/10/90
12/10/90
12/10/90
12/10/90
12/10/90
12/10/90
12/11/90
12/11/90
12/11/90
12/11/90
12/11/90


Standard

TIME

12:00
13:15
13:15
13:20
14:45
14:45
14:50
09:30
10:15
11:00
11:45
12:15

Average
RACK#

9
11
11&12
12
13
13&14
14
4
5
7
9
11
total

deviation(n-l)
Range
5.5
volume
ml/sqft
6.3
8.2
7.6
7.0
8.2
7.8
8.0
7.1
5.5
7.4
5.9
6.1
85.17
7.1
0.9
8.2
time
(s)
4.5
4.5

4.3
2.1

5.6
4.3
4.4
4.7
4.2
4.1
42.70
4.3
0.9

time
(s)
13.5
11.4

11.1
11.6

11.8
11.8
11.8
12.2
13.7
12
120.90
12.1
0.9

total
time
(s)
18
15.9

15.4
13.7

17.4
16.1
16.2
16.9
17.9
16.1
163.60
16.4
1.3


-------
        GO
        oo
JO
z

n
z
H
2


Z
f>

O
                                                       Dragout and Withdrawal/Drain Time

                                                       Electroless Copper - Modification 2

                                                                  04/24/91
DATE
12/10/90
12/10/90
12/10/90
12/10/90
12/10/90
12/11/90
12/11/90
12/11/90
12/11/90
12/11/90
12/11/90
12/11/90
Standard

TIME
11:00
11:30
12:00
13:15
14:45
09:30
10:15
11:00
11:45
12:15
12:20
12:25
RACK#
5&6
7&8
9&10
11&12
13&14
4
5&6
7&8
9&10
11
12
11&12
total
Average
deviation(n-l)
Range 1.9


dragout
volume
ml/sqft
2.6
3.2
2.9
2.5
3.0
3.7
1.9
2.3
2.5
3.0
3.6
3.3
34.47
2.9
0.5
3.7

Withdrawal
(seconds)
first
4.9
4.0
3.5
3.9
4.2

4.3
4.4
4.9


3.5
37.6
4.2
0.5
3.5
ave.
stds.
Time
second
4.4
4.8
4.2
4.1
4.9
4.2
4.0
4.9
4.4


4.4
44.3
4.4
0.3
4.9
4.3
0.4
Drain Time
(seconds)
first second
11.2
11.4
11.6
12.1
11.8

12.4
11.9
11.1


11.5
105.0
11.7
0.4
11.1

11.2
11.8
12.8
16.4
11.8
11.3
11.9
11.4
11.1


11.6
121.3
12.1
1.6
16.4
11.9
1.2
Total Time
(seconds)
first second
16.1
15.4
15.1
16.0
16.0

16.7
16.3
16.0


15.0
142.6
15.8
0.6
15.0

15.6
16.6
17.0
20.5
16.7
15.5
15.9
16.3
15.5


16.0
165.6
16.6
1.5
20.5
16.2
1.2

-------