m*m I * '*.;!;. /jft^^^rfj^'^^iiL^^iLiI' r'wv.;'-- ------- ------- Table of Contents Appendix A Data Collection Sheets . A-l Appendix B Bath Chemistry Data B-l Appendix C Chemical Properties Data C-l Appendix D Supplemental Exposure Assessment Information D-l Appendix E Drag-Out Model E-l Appendix F Supplemental Performance Demonstration Information F-l Appendix G Supplemental Cost Analysis Information G-l Appendix H , Environmental Hazard Assessment and Ecological Risk Assessment Methodology H-l ------- ------- Appendix A Data Collection Sheets ------- ------- Contents Workplace Practices Questionnaire ...... A-l Facility Background Information Sheet A-58 Observer Data Sheet A-67 .Supplier Data Sheet -....' - A-74 ------- ------- APPENDIX A Workplace Practices Questionnaire IPC Design for the Environment Printed Wiring Board Project Workplace Practices Questionnaire Please complete this questionnaire, make a copy for your records, and send the original to: JackGeibig UT Center for Clean Products 311 Conference Center Building Knoxville TN 37996 Phone:(423)974-6513 Fax:(423)974-1838 FACILITY AND CONTACT INFORMATION Facility Identification Company Name: Site Name: Street Address: City: -....-•-••... •-• ..--.• •. - . ..-.-.. • . - • . .. 1 - . .•'-'•. ' • ' - , .••-... .;.-••- | State: | Zip: Contact Identification Enter the names of the persons who can be contacted regarding this survey Name: Title: Phone: Fax: E-Mail: " ' A-l ------- APPENDIX A —INSTRUCTION SHEET— FOR THE DESIGN FOR THE ENVIRONMENT (DFE) ALTERNATIVE SURFACE FINISHES (ASF) PROJECT WORKPLACE PRACTICES QUESTIONNAIRE INTRODUCTION This questionnaire was prepared by the University of Tennessee Center for Clean Products and Clean Technologies in partnership with The US EPA Design for the Environment (DfE) Printed Wiring Board (PWB) Program, IPC, and other members of the DfE PWB Industry Project Work Groups. The purpose of this questionnaire is to collect data that will be used in preparation of a Design for the Environment (DfE) Alterative Surface Technologies report. This report will present an analysis and evaluation of the risk.. performance, and costs associated with operating each of the alternative surface finish processes. Much of this report will be based on data submitted by PWB manufacturing facilities. You can obtain more information about this project and other DfE PWB projects from the US EPA's website at http://www.epa.gov/opptintr/dfe/pwb/pwb.html). CONFTOENTIAIJTY All information and data that is entered into this questionnaire is confidential. The sources of responses are only known to the IPC and have been coded by the IPC for industry research purposes. Any use or publication of the data will mot identify the names or locations of the respondent companies or the individuals completing the forms. INSTRUCTIONS w Respondents must complete Sections 1 (Facility Characterization) and Section 2 (HASL Process) of this questionnaire. Section 3 is divided into five processes (3 A through 3E) as shown below: 3A. Organic Solder Preservative (OSP) Process 3B. Immersion Silver Process 3C. Immersion Tin Process 3D. Electroless Nickel/Immersion Gold Process 3E. Electroless Nickel/Electroless Palladium/Immersion Gold Process *• Of these five subsections, 3A-3E, please fill out only the top two alternative processes, based on PWB through-put, that are currently being implemented at your facility. If your responses do not fit in the spaces provided, please photocopy the section to provide more space or use ordinary paper and mark the response with the section number to which it applies. *• Please make a copy of the completed sections and retain them for your records. If you have questions regarding the survey, please contact Jack Geibig of the University of Tennessee Center for Clean Products and Clean Technologies at (telephone 423/975-6513; fax 423/974-1838; emailjgeibig@utk.edu) or Star Summerfield at IPC (telephone 847/790-5347; fax 847/509-9798; email summst@ipc.org). Please return the completed questionnaire by January 8,1999 to: Star Summerfield, IPC 2215 Sanders Road, Northbrook, IL 60062-6135 Phone: 847/790-5347, FAX 847/509-9798, email summst@ipc.org A RETURN LABEL TO IPC IS ENCLOSED FOR YOUR CONVENIENCE. A-2 ------- APPENDIX A Section 1. Facility Characterization This section focuses on general information specific to the facility. This information is not process-specific. Please estimate manufacturing data for the previous 12 month period, or other convenient time period of 12 consecutive months (e.g., F Y97). 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: sq.ft. Overall amount of PWB produced in surface square feet (ssf): ; . . ssfiyr 1.2 Process Type Estimate the percentage of PWBs manufactured at your facility using the following methods for surface finishing (SF). Specify "other" entry. • Surface Finish Process HASL OSP-Thick OSP-Thin , (benzotriazole-based) Immersion Tin Electroless Palladium Immersion Silver Percent of Total : • '." •..-..'••-. ••'••••%"' % % % '• •:••;- ' % % Surface Finish Process Electroless Nickel/ Immersion Gold Electroless Nickel/Electroless Palladium/ Immersion Gold Other: Other: Other: Total Percent of Total % % %• % % 100% 1.3 Wastewater Discharge and Sludge Data Wastewater discharge method (circle one): Direct Indirect Zero (tosfream) (toPOTW) Throughput of facility wastewater treatment system: Annual weight of sludge generated: Is sludge dewatered prior to disposal (circle one)? Water content prior to dewatering: Water content after dewatering: gals/day ' ' -- Ibs Yes No % ' • % A-3 ------- APPENDIX A | d> SB i ?« ij Il$9fci£ SJ?|Afi£t- ft.flr'^osg I I a fr t A-4 ------- APPENDIX A 2.2 General Data-HASL Number of days HASL line is in operation: Estimated scrap rate (% of defective product) for HASL process: ;,;> .r.-: days/yr •: '"'-• ' „- •" " -,,..'.. •/-<_-- •.--. .vr ",f-,''C^\ Number of hours per day the HASL line is in operation: Total of PWB surface square feet processed by HASL line per year: hrs/day : , .ssffyr 2.3 Process Area Employees--HASL Complete the following table by indicating the number of employees of each type that perform work duties in the same process room as the HASL 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 HASL Area Worker Line Operators Lab Technicians. Maintenance Workers Wastewater Treatment Operators Supervisory Personnel Other (specify): Number of Employees in HASL Process Area •- ... • • • - --1 • • ," " ' "• • '-"-"' • - ' Average Hours per Week per Employee in HASL Process Area ••- V ' " .•- • :'•'•:< hrs ; . '.'-'>' .• -.'•-• . '-'. '. hrs - .•••': •.-•'••'.' '. '-'•,..-• , . '• hrS - hrs hrs hrs 2.4 Physical Settings-HASL Size of the room containing the HASL process: Are the overall process areas/rooms ventilated (circle one)? Do you have local vents (circle one)? sq.ft. Yes . No ' Yes "•-•. No :,; Height of room: Air flow rate: Local vent air flow rate: , ... •.' --:.''- '... fL cu. ft./min. cu..ft./min. Overall surface finishing process line dimensions Length (ft.): Width (ft.): Height (ft.): 2.5 Rack Dimensions—HASL Average number of panels per rack: Average size of panel in rack: Length (in.): Average space between panels in rack: in. Width (in.): Do you purposely slow the withdraw rate of your panels from process baths to reduce drag-out? (Circle one) Yes No A-5 ------- APPENDIX A 2.6 Rinse Bath Water Usage-HASL Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse baths present in your HASL 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 water flow rate of one bath in the cascade. I Total volume of water used by the HASL line when operating: gat/day | Process Step Number • Example: 8 Flow Control b R Daily Water Flow Rate0 2,400 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- k Flow control - Consult key at right and enter the letter for the flow control method used for that specific rinse bath. * 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. Cascade Water Process Steps a '.:". •.-.'• .;'.-. 8->6 . •' ': -. • • " . -• • , - '-.'•'• : • .' . ' . ' Flow Control Methods Kev [C] - Conductivity Meter [P] - pH Meter [V] - Operator Control Valve [R] - Flow Restricter [N] - None (continuous flow) [O] - Other (explain) 2.7 Filter RepIacement-HASL Not Applicable n Bath(s) filtered (enter process step ft from flow diagram in 2. 1) Frequency of replacement: Duration of replacement process: Personal protective equipment (see key): Persona! Protective Equipment Kev: [FJ- Eye Protection [G]- Gloves [L] - Lab coat/Sleeved garment [A] -Apron [R] - Respiratory Protection [B] - Boots [Z] - All except Respiratory Protection pSTj-None 2.8 Rack or Conveyor Cleaning-HASL A-6 ------- APPENDIX A 2.8 Rack or Conveyor Cleaning-HASL Not Applicable Q 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: Rack Cleaning Method: [C]-Chemical bath on SF process line pJJ-Chemical bam on another line Pi-Temporary chemical bath [S]-ManuaI scrubbing with chemical [MJ-Non-chemicaI cleaning [N]-None [O]-Continuous cleaning Conveyor Cleaning Method: £C]-Chemical rinsing or soaking [S]-Manual scrubbing with chemical • M-Non-chemical cleaning fN]-None [O]-Continnous cleaning 'Personal Protective Equipment: [E]-Eye Protection [G]-Gloves " [L]-Lab coat/Sleeved garment [A]-Apron [K]-Respiratory Protection ,[B]-Boots [O]-Continuous Cleaning {NJ-None [Z]-A11 except Respiratory Protection 2.9 Solder Unit Maintenance and Waste disposal Complete the following maintenance and waste disposal questions for only the unit of the process that performs the hot air solder leveling Frequency of maintenance: Duration of maintenance : Personal protective equipment (see key): Number of personnel involved: mfn. ' " Personal Protective Equipment - Enter the letters of aH the protective equipment used by the workers who physically replace the spent bath. [E] - Eye protection [B] - Boots [A] - Apron [G] - Gloves [L] - Lab coat/Sleeved garment [R]- Respiratory protection HZ] - All except Respiratory Protection [N]-None Method of dross removal: Frequency of dross removal: Quantity of solder waste disposed (per day): Method of solder waste disposal . " • Method Of Solder Waste Disposal - Indicate method of solder waste disposal from key below: [M] - Metals reclaimed off-site p.]- Recycled on-sfte [RO] - Recycled off-site [D] - Drummed and treated as hazardous waste [O]- Other (specify) A-7 ------- APPENDIX A o. 1 a co ,5 . 5 ng Conditions S-* 03 CU O « Q u , o PH •2 Cl Q 1 m >> fS PL, a ">i o 5r U o o jg ex ^ I « cu 1 ° £M >"""x li" , H | Q ^ § «> "5 'o> "0 *-i « a a •§ S ^^ 1 ^ S _C3 S £?* S!3 C3 *Q ^ J2J >• ^ ^^ 5 J s § ^^ *** N*-/ .a^ •^ a> MJS a o CU {j ""' ° 1 ^ -8" CO c> ID en Cct to .a .a s u g^ : t o 8- •3 oo .a .a I 2 w is ,g. o -s o "flj •g g s •s •4 o .a 3 S i s mersion Ti HH A •« ."§ "2 '=- S S |3 .3 ts o B § ^ g -S o 8 a -S. 8 g 2 S g.^, •S3 >> "S ^ J3 i- ta !g ® S 2 2 5,^ii^o O STO U 57 rA « E > > E o U_l t__J l_l U_J I__J |___] & ' -1 1. ^ B a ^ S> o « Ofl s ^ 1 1 ii ^ ^ • *^ ~"H ^ 1 l«gl o H "3 eg I III A-8 ------- APPENDIX A a* T3 CO P .-a- •*•» i m ^ **. a a a a •^1 . S *\ = 2 -S e W- C 11 IB M 3 ^ a 3 3 LI 3 S 1> U s j •"* b. a 1 U e>) " M- " 1-1 i ^ 3 ; ^ ••i "* 1—1 I s en •* ^ J ) ) CM en T* '-' a M ^ > w 01 en "•• ^ •% [ a > J o 04 en •* .S S ' .2? g ID to" & 2 "o > O •a 1 CO 's en 73 o 1 •s J3 "i 1 a. CM 0 A •+•* a> j3 tj_i ni *a tg» P i _>> ™ "-C -g CQ ^ •a -3 O-g C3 ^ '§ *s ^1 0 . &, A-9 ------- APPENDIX A <«-) O t-i 73 *M co ^3 •0 0 CO 'fS to « c l« «.§ O CB &s va o •S « g-s ^•2 a> ** 0 0 cal balan dditions t E "* «> "ca "§•1 Jl e o *C3 H-H f ^ S 0> 1 CO « & <£« CO 'S •a J 4_* 3 •1J » -S "2 >° cS a> o u ill •sis * § es JsS| 13 rsona tective ipment1" Pers rot P Equi 1 = 'S 5 .w a / 3 .2 .2 3 P3 U O PH ^ O i * 2S A-10 ------- APPENDIX A cal bat che 0° 8 .§• -3 1 ••s 1 G. u I * 3 £ SSs 38-S « ca S S&2 «** 03 60 H.S a 2 0 S3 —- &1 oc 5 Method Bath R • a *••*" a oo -• « l"i — S S1® W,o "S-e u S So ££ US a> at ^5 is w s e. §* r O «s £ Cleaner • g d, - * •& § "9 -S • 2 " ^ 13 ca 1^^ no too! or fy u O O a " fe 111! II ------- APPENDIX A 2.14 Chemical Bath SampIing-HASL Bath Type Example: Cleaner Microetch Other (specify): Type of Sampling a A. . Frequency b 3 per day : - '""'•- ;;;:- '•-. :' ' ' . Duration of Sampling0 •-;V-5riiiit :;;•'• : '. Protective Equipment d E,G,A • . Method of. Sampling e •• :-:-^3r.-v, ••• ''' . • • - - '•,-•' ',-.',. *Tvne of Sampling c Duration of Sampling: Enter the 'Method of Sampling; [A] -Automated average time required to manually take a [D] -Drain or spigot [M]- Manual sample from the tank. rp] - Pipette [N]-None • [L]- Ladle d Protective Equipment; Consult the key [O] - Other (specify) * Frequency: Enter the average time for the above table and enter the letters for elapsed or number of panel sq. ft. all protective equipment used by the processed between samples. Clearly person performing the chemical sampling. specify units (e.g., hours, sq.ft.) 2.15 Process Waste Disposal — HASL Bath Type Cleaner Microetch Flux Solder Post-Clean Other (specify): Annual Volume Treated or Disposed a • •". • ;•"- "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 Disposal11 RCRA Waste Code (if applicable) B Methods of Treatment or Disposal - [P] - 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 disposal [RN] - Recycled on-site [RF] - Recycled off-site [O]- 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 [T]- Chemical tote [O]- Other (specify) A-12 ------- APPENDIX A an £ I i a a ! u 0) •N H § % . - fi^,i «|gg,o| "al-Stl 4) •w A 1 f S* c 1 3 5 « •<* it a T d in the •o AS5 I« .2- 4> O "S « J> .£ il 1 2 41 ai * Oj I ,5 /-\ (i) « 0 4> I I •••4 2 o •« o •a I 1 § '§ c o *d 1 CO 05 8 2- ^ 4) p. € 0 ^3 ";!3 O *3 £M 8 ^ 1 1 •3 O S1 I U 1 •o 2> s a . o O s**\ s 'S 4> 1; 1 O O 0 •g 3 o <-> 3 1 -3 3 S s- A-13 ------- APPENDIX A 3.2 General Data-Nickel/Gold Number of days the nickel/gold line is in operation: Estimated scrap rate (% of defective product) for the nickel/gold process; days/yr ! % Number of hours per day the nickel/gold line is in operation: Total of PWB surface square feet processed by the nickel/gold line per year: his/day ssf7yr 3.3 Process Area Employees—Nickel/Gold Complete the following table by indicating the number of employees of each type that perform work duties in the same process room as the nickel/gold 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 Line Operators Lab Technicians Maintenance Workers Wastewater Treatment Operators Supervisory Personnel Other (specify): Number of Employees in Surface Finish Process Area is fi t * : 5 5 i"c r»* Average Hours per Week per Employee in Surface Finish Process Area hrs hrs hrs hrs hrs hrs 3.4 Physical Settings-Nickel/Gold Size of the room containing the surface finish process: Are the overall process areas/rooms ventilated (circle one)? Do you have local vents (circle one)? sq. ft. Yes No Yes Mo Height of room: Air flow rate: Local vent air flow rate: ::::::::::::::::::::::::::: ;::::::::::::::i:::/v::: !!iiniilii;!!;!IH!i!!!!!l!!ll!l!Hnii;l;i!St!l ;;;;:-::::;-::;;F;i-:-!;;:€iJE;:ft/riBn;:; !;;;;i;;;;;:;;;:Hn;;:ncU:;ftWuh'::: Overall surface finishing process line dimensions Length (ft.): Width (ft.): Height (ft.): 3.5 Rack Dimensions—Nickel/Gold Average number of panels per rack: Average size of panel hi rack: Average space between panels in rack: in. Length (in.): Width (in.): Do you purposely slow the withdraw rate of your panels from process baths to reduce drag-out? (Circle one) . Yes .. , No A-14 ------- APPENDIX A 3.6 Rinse Bath Water Usage-Nickel/Gold Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse baths present in your nickel/gold 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 water flow rate of one bath in the cascade. I Total volume of water used fay the surface finish line when operating: gal/day | Process Step Number* ;T? : KJ ia. "$? ::;:i: •"•:::::: llllll!!!!!!!!!!!!!*:! iliiiiiliHIH!!!!!;!!:!; Flow Control" ':'•'•'• iiii Hill Daily Water Flow Rate' ilili 4ii )( & ::::: 18 !i i i i i i & i H 1 1 n i 4 4 4 4 4 Hi ap ail! a m all ail! * Process step number - Consult the process schematic in question 2. 1 and enter the process step number of the specific water rinse tank b 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. d Cascade water process steps - Use the step numbers for rinses that are cascaded together. Cascade Water Process Steps' Ii lilll Flow Control Methods Key [C] - Conductivity Meter |P]-pH Meter [V] - Operator Control Valve [R] - Flow Restricter [N] - None (continuous flow) [O] - Other (explain) :l :: 3.7 Filter Replacement-Nickel/Gold 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: •liHii;!; [E]-Eye Protection [G]-Gloves [Z]-A11 except Respiratory Protection [L]-Lab coal/Sleeved garment [A]-Apron [N]-None [R]-Respiratory Protection [B]-Boots A-15 ------- APPENDIX A 3.8 . Rack or Conveyor Oeaning-Nickel/Gold Not Applicable EH 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: [Q-Chernical bath on SF process line (Pi-Chemical bath on another line m-Temporary chemical bath [S]-ManuaI scrubbing with chemical [Ml-Non-chemical cleaning [N]-None [O]-Contiinious cleaning Conveyor Cleaning Method: [C]-Chemical rinsing or soaking [S]-Manual scrubbing with chemical [M]-Non-chemical cleaning [N]-None [O]-Continuous cleaning Personal Protective Equipment: (EJ-Eye Protection [GJ-Gloves [L]-Lab coat/Sleeved garment [A]-Apron [RJ-Respiratory Protection |B]-Boots [O]-Continuous Cleaning [M]-None (ZJ-A11 except Respiratory Protection 3.9 Chemical Bath Sampling-Nickel/Gold Bath Type Example: Cleaner/ Conditioner Microetch Catalyst Acid Dip Activator Electroless Nickel Immersion Gold Other (specify): 'Type of Sampling [A]-Automated [M]-Manual [N]-None bFreauencv: Enter t time elapsed or numb ft processed between Clearly specify units I sq.ft.) Type of Sampling1 • A 3 • *" . t Frequency6 3 perday m it : • Duration of Sampling0 5 nun ' 3 : Protective Equipment* E,G?A Method of P - c Duration of Sampling: Enter the • Method of Sampling; average time required to manually [DJ-Drain or spigot take a sample from the tank [P]-Pipette [Li-Ladle d Protective Equipment: Consult [O]-Other (specify) he average the key for the above table and er of panel sq. enter the letters for all protective samples. equipment used by the person e.g., hours, performing the chemical sampling. A-16 ------- APPENDIX A •o ^^ a 09 O 1 U BO 'i 1 08 Q 01 V) 2 5 o •35 CM §.•§ JsS 1? If IP V /Sf. S -3 S a S (5 '§ I ill % >; »a J— .2 0 ^ •a a s « j c :=- A "es P3 llii ;;^;;;;;;:; iii^ili!!!!; igflllill!! :••••;::•:••••. •:-^^T:::".:T 1 Cleaner/ | Conditioner : ;!!!;;!! m ;!!& •••W: ill : -:«-r. : Microetch I iSHii :::::<:: ;;;&; =iiii •*•* i" 0 m ••••^ !;!^i li^ii :«.: 5 •e 3 IliH! ;;;^-; |:;»fc; :::«i:: !!!!! :::;;:: •?§!; 1 Activator . [: iH! •••ft! Ui,: iliii '::::;:: ::«;; 1 Electroless Nickel [ ii^l! :;;^; L^J ill r:\lv: ;:«:•; Immersion Gold lifii :-;^i ::^-: ilid: ::!i:jii ••^!: [ Other (specify): |i i!l!::'r: "e I | > as •B fi S co 1 1 f-4 ^ . S s 5 2 b Drip Time- Enter the the specific process bath to allow d A-17 ------- APPENDIX A S3 a . "3> S.s 5 OS ^ O en ^ 0. •§ 1 '•g c 3 •; "S s S < t- /— . H ^ & Ctt S£ ^ e i j I f, "S •a 2 "S .a JS U .e "S HHf: ,'IIIU *— ( Cleaner 1! iilili ;;:::: ii!|! CM In It 11 :!;::, :;;::; co ^ 1 Microetch :::'! :: rh CN co :::;: r-1 | "« M :Hj: CN CO -1 a. •3 •a li' y!!!! :i!!.:!i i! cs 1 CO 1 "-1 I Activator \:\[': 1 IP CS Li::!: ^i CO -1 "a3 ^ o S 1 Electroless iiiiiii «s CO -1 •s 1 1 Immersion c-l CO iiii: lil :::f; ::::::: :::::!: :r:i::!; iiiiii: IHl ^ S 1 45 o :::::: :::!:: :::::; 11 ;:;;:: •::'::' ------- APPENDIX A H* 2 "* . o, g -2 o co a d i i > . 55 *s "Bb 5 •» ss n.-2 "3 to es ca *tf o is o *-> «.&• ST§ I jf * w *•••§«> to « »a |«S •g M £3 "§ JJ 1 ^ CC sf g • •g g fe c ro j^ fl.iS' 13 SS 5 S '* S ^ w « ^ 2 « -S *- 3J2£ Sa cs o u, S * S «S ^ "1 « g f^i T& QQ* P i "3 ^ ^ g O O C i Jfr'S s ^ 60 g te |i|i "S S S ~ ® *° "S •* "J? « ^ "a ^*^ ,£* CJ J3 1 S) g g . M -| H S *3 o a a js ~ o ^ J Cj-| '•§ S S « •§ « 0 5 § jj> C "S a '- g § S s S 1 g "O j 5 ^2 j Q | ||| S u «2 la •2 s |i u •"• •o •a •o •M ^ •Q 2 • PH u 1 5 4> - "3 > = C '•" 1 -S 1 o) o ^ a, £• 5 5 «» <§ " 5 gi o S a •Sac 85 fc. .5 0- >-> 1 i — s ffll fa 0 S HI r_ ,^ . t^ts S M -y w co w S o "K o W -g.g oTS o v ^i^ ^rf^T1 -^ S 'is- £5^-g 2 5-8*8. I ^ . |.lt fit "^ 4-^ >^ u o P-I 18.? M iy . 2^| 1 Itf PLi-Nl"^ ^ CO 03 g ,^g B.gs-||g. * ° "§ ^ §*^§ d ? !§ « §5 J3 i * i t i » ^ "S »c; ^ ptj • *^ -2 5^~ o «3 *i °"S "u » 0"° _u S " T3 "S, g ^^ » 8 §" •g-S |JS o-| &J - a i^ g^wO S«fe if "ISo1 ^s "„ 5| J-|i|i •g g 2s §^^S § >-^ 3 os *§ o °s ^l*g-a "j^-S "*'&» §• g's.llf 'o-aP-il '5*«S0 JB o " ft S > 2 -^ S'gS',!, |'|f .2 J ^^ 11 ^-S -a J*o ™ t-j K ^ Is g^ Ers'-S'oo *j^ 'j-^ d QJ gj ^ ^^ ^^ CS *T« I3'""' A-19 ------- APPENDIX A C3.S 5 °«s S2-S.S « S S *^^ fi o o § « •III is rs Q Method of Spe Bath Removal - o 1^1 O 1) I tch I ' Activator 1 Electroles o O ! S =2? 1 c. I 3 §!§• H-i'Sl Q w 5 .— S^ It !£§ S'.e ill Hi l|t -l S «>(_)>';3HV-' as g g ~ eo 2ST8 tl'SO -"8 iiti5 t^P 05— S — •rlijil "*"r^2,f|as n to^S w r* o^ Ist a^^ A-20 ------- APPENDIX A 3.14 Process Waste Disposal- Nickel/Gold Bath Type Cleaner/ Conditioner Microetch Catalyst Acid Dip Activator Electroless Nickel Immersion Gold Other (specify): Annual Volume Treated or Disposed3 ;:i ii ;; !! a 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 i bath change outs and bailout \ before entering the totaL < . ' i Ii •:! Method of Treatment or Disposal b : :?:: :::' ii:::: i^ii RCRA Waste < Code (if applicable) iiiiiiiHiHiiiiiiiiiillii Container Type 3 Methods of Treatment or Disposal- Container Type - indicate P] - Precipitation pretreatment on-site the type of container used N| - pH neutralization pretreatment on-site for disposal of bath wastes rS] - Disposed directly to sewer with no [OH]- Open-head drum reatment [CH]- Closed-head drum D] - Drummed for off-site treatment or [T]- Chemical tote lisposal [O]- Other (specify) UN] - Recycled on-site 1RF] - Recycled off-site O]- Other (specify) A-21 ------- APPENDIX A CO a o o Jl«« 2:3 «a« ra.a» 6 3«Q O,« '3 o»*3 Of w «" «n a a 03 § 6 41 «ri a 0 o •-B 2 4) T ca u 4) U I e 4) a A I 8- oe 8 2 "3 1 s •PN •8 es « .2 a. 2 a a | « I. 'I US ^52 -S p 4> LI i» -"f' •g ' O A-22 ------- APPENDIX A 4.2 General Data-Nickel/Palladium/Gold Number of days the nickel/palladium/gold line is in operation: Estimated scrap rate (% of defective product) for the nickel/palladium/gold process: ;;;!;;!!!!;!;day.S/p;;: Number of hours per day the nickel/palladium/gold line is in operation: Total of PWB surface square feet processed by the nickel/palladium/gold line per year: !!!!;!!!!!!;!!!!!lKtisf(liayi; ;;;;;;;;;;:;;;;;;;;;;sgjyp: 4.3 Process Area Employees—Nickel/Palladium/Gold Complete the following table by indicating the number of employees of each type that perform work duties in the same process room as the mckel/palladium/gold 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 Line Operators Lab Technicians Maintenance Workers Wastewater Treatment Operators Supervisory Personnel Other (specify): Number of Employees in Surface Finish Process Area :::::::::::if::::T::::::::::::::;:::::::::::::::;1:::i:!iiii!"--"^ = ^^^"^ :::::::::::::::::::::::: ::::::::;:!::::: :::;:;::::::::: \\\\ \ I ::::::•::::::::::: i Average Hours per Week per Employee in Surface Finish Process Area ;;i;:;i:;l:i!!il!i;!!ill!!n;;!![;!;!!;!!!;!!!J!!!!!;!!!i!l!ii!!!::i!S; PiyyiHn!ii!iilil!:li!!!niiili&! ;;ii;-H!l!iil;!iii;!iii!!i!!!j!:!j::^n:::::H::ii^i^;^;:::n;m%: 'iiiill^ilHiiiiilllfiiiiiiillHiiihiliiiliiiiliHiliiiiiliiiiJiiij^i •i;!ii;;;:nM!- = i:::; = : = :::::;M=M;;:;;|!!;i;!i;;:;::::|:!::|:::ai!ihit|: •;;:;~::;--:=;i;;!;i!;i!i!!!Hl!!!;!iiii!!!!!!!!i!!!!!!H^!!!!!!!;nife; 4.4 Physical Settings-Nickel/Palladium/Gold Size of the room containing the surface finish process: Are the overall process areas/rooms ventilated (circle one)? Do you have local vents (circle one)? ;;;;::;;;;;;;;;:;;;;;;i;:;;;;s(j!;ff::; !!!!!iXlea!!!i!!!!!!!!i!iSS!J!i!!J ::::-.\r>iQ:::::-;:Tr:;::rM>.:;:::;:: : : : : : ;i S2» :::;::::::;::: :J;:fU- •;•••- Height of room: Air flow rate: Local vent air flow rate: ::i;;;:;:;;-:;:;;:::::::;::;;;;;;;;;;;;;;;:;;«f:;; :::..::::::::::::::::::. :::::::::::::::::::::1!U::I •-•••.•.•.•.•.::::::::y.. ::::::::•{* f.i'ff.jfftfa: • iiiiillliiiiiiiiilll^l^iitti/ffiSfii;! Overall surface finishing process line dimensions Length (ft): Width (ft): Height (ft): 4.5 Rack Dimensions—Nickel/Palladium/Gold Average number of panels per rack: Average size of panel in rack: Average space between panels in rack: in. Length (in.): Width (in.): Do you purposely slow the withdraw rate of your panels from process baths to reduce drag-out? (Circle one) Yes No A-23 ------- APPENDIX A 4.6 Rinse Bath Water Usage-Nickel/Palladium/Gold Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse baths present in your nickel/palladium/gold 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 water flow rate of one bath in the cascade. | Total volume of water used by the surface finish line when operating: gal/day} Process Step Number3 Flow Control" Daily Water Flow Ratec Cascade Water Process Steps'1 Example: ! R. a Process step number - Consult the process schematic in question 2.1 and enter the process step number of the specific water rinse tank. b Flow control - Consult key at right and enter the letter for the flow control method used for that specific rinse bath. e Daily water flow rate - Enter the average daily flow rate for the specific water rinse tank. d Cascade water process steps - Use the step numbers for rinses that are cascaded together. Flow Control Methods Key [C] - Conductivity Meter [P]-pH Meter [V] - Operator Control Valve [R] - Flow Restricter [N] - None (continuous flow) [O] - Other (explain) 4.7 Filter Replacement-Nickel/Palladium/Gold Not Applicable |~~] Bath(s) filtered (cater process step # fiom flow diagram in 2.1) Frequency of replacement: Duration of replacement process: Personal protective equipment (see key): Personal Protective Equipment Kev: - • [E]-Eye Protection [G]-GIoves [Zj-AII except Respiratory Protection [L]-Lab coat/Sleeved garment [A]-Apron [N]-None [R]-Rcspiratoiy Protection pB]-Boots A-24 ------- APPENDIX A 4.8 Rack or Conveyor Cleaning-Nickel/Palladium/Gold . 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: :.: •:::•:::::!! Rack Cleaning Method: [C]-Chemical bath on SF process line [D]-Chemical bath on another line m-Temporary chemical bath - [S]-Manual scrubbing with chemical (Ml-Non-chemical cleaning {N>None [O]-Continuous cleaning Conveyor Cleaning Method: [C]-Chemical rinsing or soaking [S]-Manual scrubbing with chemical [M]-Non-chemical cleaning [N]-None [O]-Continuous cleaning Personal Protective Equipment: El-Eye Protection [G]-Gloves BJ-iab coat/Sleeved garment [A]-Apron [R]-Respiratory Protection [B]-Boots [Pi-Continuous Cleaning [N]-None EZ3-A11 except Respiratory Protection 4.9 Chemical Bath Sampling-Nickel/Palladium/Gold Bath Type Example: Cleaner/ Conditioner Catalyst Acid Dip Activator Electroless Nickel Electroless Paladium Other (specify): a Tvoe of Sampling jAj-Automated [M]-Manual [N]-None * Freauencv: Enter t time elapsed or numb ft. processed between Clearly specify units sq.ft.) Type of Sampling* A :::: :: ie average er of panel sq. samples. Ie.g., hours, Frequency" !;i;l 1 3 jpriday- c Duration of Si average time req take a sample frc d Protective Eqi the key for meal enter die letters f equipment used! performing the d Duration of j 1 •: !5;Hiiii:;;:j;: !!l!!H!i;!l impling: Enter the uired to manually an me tank. lipment: Consult K>ve table and. or all protective >y the person lemical sampling. Protective ::::*ji:::jpc ',::•.£;; ::::;:: :::Jfc»!fer^^k; ;:::;:; lilllN Method of iiiiiPJi e Method of Sampling; [D]-Drain or spigot {P]-Pipette [L]-Ladle [OJ-Other (specify) A-25 ------- APPENDIX A .5 .S3 >> G S II If a> •?; ^ o O f», tl o eo o .5 eo c> I| o _g S 'S '3 en Average eye * O D. CO ,0 ^ (3 Dfi 1 Q« O OS ts en CO O > O 43 « "Sb^ S e3 « c H^ ?^< 1 _g, g § OQ To en •T-t g- i. Cleaner/ Conditione £•• 8 CO 8 CO "ra eo .g .s- j Microetch 6f* i i "i _d _c •«^ 1 ^ i § i .s .g s. S 2 IF- ji i "I! .2 -S; | Activator gn j g i 1 u •S -S i [ Electroless Nickel <-• & o i „ & i _«? ~ .s. Electroless Palladium g, O o it .g _a •a "3 O 1 Immersion i .. ^ CfM ' i &, cu § • X o 1 g 2 C 5 &. a 1 CU en •a 1 | '"S •"5 1 i i W3 1 •a s 2 -sr s § i.I 1 8 73 o 8 -- la, ^s4! i-tj W PH ^ f5n /—J iliiiig B- Q S /2* *_J^« ^** ^ II 1* 03 d CO g^ § s ™ s •? 9 '"? o ^"57^ A EI^^U-J ep 2 •! CO 2 03 . g 0. § •g .«• 1 2 CS ------- APPENDIX A I — t 00 2 -S S « as as -2 " 3 O JS 1! T3 rS p Ifl I'M M § C 1 ?* \ o :::: CM CO f 0 5 s 0) I _S P Clj 0 1 ' 1 S 1 "o e "e3 to » .S "eb "S •a re .§ CO i CO » CC 1 *o 5 "3 o •5-; a • 05 3 ^ 3 ^ 2 -^ f3 s ^ fp^ 1 "o -3 ------- APPENDIX A U at CJ P ^-7 •S 3 5 o ,-v"5 w»,« C5 < Manufacturer (if annlicable) Product Name "« .0 f> •4-» C3 M i— i Cleaner :i:H • 1 ' < ( i • • C ; H CS s 1 3 i •t > ^ C^3 i— H 1 Other (specify) ::~ «S ii" CO t u 43 S S i "3 stead of v ,r weight (i.e., crystalline chemicals) in cular chemical used is measured ty i PI ce 2¥ C ^ §.s ci § o S2 fc) cC? •as H 8 Annual Quantity Us« in pounds and clearly s A-28 ------- APPENDIX A §•51 Jg -35 s * M 03 f 4».g •* C^ § g> = = i-s s •g M . a. « a i-( o S o U 1 o» ;;;s IIP ills iiiic;;!;; iiiiii H o «f|| - tc g «-• O o s>"" ^|,fil *^ *-* • S Si2 We S3 Bo ^§g Sill85 goj fe^J^as l-s/g- =1-1 ll -rllJl^ J>53<£opO "*" ' ' r^-,J-,J-, A-29 ------- APPENDIX A 4.14 Process Waste Disposal- Nickel/Palladium/Gold Bath Type Cleaner/ Conditioner Microetch Catalyst Acid Dip Activator Electroless Nickel Electroless Palladium Immersion Gold* Other (specify): Annual Volume Treated or Disposed2 .. • • * Annual Volume Treated or Disposed - Enter the yearly amount of the specific bath treated or disposed. Be sure to consider the volume treatedfrom both bath change outs and bailout before entering the total Method of Treatment or Disposal b • RCRA Waste Code (if applicable) - - Container Type B Methods of Treatment or Disposal- Container Type - [P] - Precipitation pretreatment on-site indicate the type of [N] - pH neutralization pretreatment on-site container used for [S] - Disposed directly to sewer with no disposal of bath wastes treatment [OH]- Open-head drum [D] - Drummed for off-site treatment or [CHJ- Closed-head disposal drum [RN] - Recycled on-site [T]- Chemical tote fRF] - Recycled off-site [Ol- Other (specify) [O]- Other (specify) A-30 ------- APPENDIX A o>*2 O «• 83 03 4) O 2 o l-jjl! ll| a ------- APPENDIX A 5.2 General Data-OSP Number of days the OSP line is in operation: Estimated scrap rate (% of defective product) for OSP process: days/yr % Number of hours per day the OSP line is in operation: Total of PWB surfece square feet processed by the OSP line per year: ;Ji;H!!!;!ii!!J!:I^Jap •••••••••••••••-••••issfi^: 5.3 Process Area Employees—OSP Complete the following table by indicating the number of employees of each type that perform work duties in the same process room as the OSP 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 Line Operators Lab Technicians Maintenance Workers Wastewater Treatment Operators Supervisory Personnel Other (specify): Number of Employees in Surface Finish Process Area •£- - : : * ' Average Hours per Week per Employee in Surface Finish Process Area hrs hrs hrs hrs hrs hrs 5.4 Physical Settings-OSP Size of the room containing the surfece finish process: Are the overall process areas/rooms ventilated (circle one)? Do you have local vents (circle one)? Overall surface finishing process line di Length (ft): sq.ft. Yes No Yes No mensions Width (ft.): Height of room: Air flow rate: Local vent air flow rate: Height (i !!!!!!!!!ii!!!lii!l!!!liii!!!!!!li=ill|i!==illff:!! ••:ii = ?:H;;::::::;::::;i;iea.:LSi/mih;:: milimiiliililili^&fBmii: I): 5.5 Rack Dimensions—OSP Average number of panels per rack: Average size of panel in rack: Average space between panels in rack: in. ::LengtK;(;in.): Width (in.): Do you purposely slow the withdraw rate of your panels from process baths to reduce drag-out? (Circle one) : ''•' !Y<» Ti:;;W A-32 ------- APPENDIX A 5.6 Rinse Bath Water Usage-OSP Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse baths present in your OSP 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 water flow rate of one bath in the cascade. I Total volume of water used by the surface finish line when operating: galVday | Process Step Number3 ill am] ieJp • !;; I::::::::::::;::::::::: I::::::::::;:::;;:::::; ii \ '.' Flow Control" liliilli I i!::!iii!!!!y!!!H •• lljljlllllliiiili j \ Daily Water FlowRatec :' :::::: '.'.'.^'.: J1J4 : I:::::::::::::::: OgaL/daj ! ; iiiliiigl '•• ;;;;;;;i;!gai ii iilii ii iiiiiiiii^fl ii iiiiiiiiiis^l #?il?!i nm Wx& WM #$&& J$is:£ " Process step number - Consult the process schematic in question 2. 1 and -enter the process step number of the specific water rinse tank b 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. . d Cascade water process steps - Use the step numbers for rinses that are cascaded together. Cascade Water Process Steps'1 :-: illllffliil ;!H :::: ;:r;:;;p;; Flow Control Methods Key [C] - Conductivity Meter [P] - pH Meter [V] - Operator Control Valve pi] - Flow Restricter [N] - None (continuous flow) [O] - Other (explain) 5.7 Filter Replacement-OSP Not Applicable • . n 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 [Z]-A11 except Respiratory Protection (L]-Lab coat/Sleeved garment [AJ-Apron [N]-None [R]-Respiratory Protection [B]-Boots A-33 ------- APPENDIX A 5.8 Rack or Conveyor Cleaning-DSP 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: mm. Rack Cleaning Method: [C]-Chemical bath on SF process line [D]-Chemical bath on another line m-Tempoiary chemical bath [S]-Manual scrubbing with chemical [M]-Non-cbemical cleaning [N]-None [O]-Continuous cleaning Conveyor Cleaning Method: [C]-Cbemical rinsing or soaking [S]-Manual scrubbing with chemical [MJ-Non-chemical cleaning . [N]-None [O]-Continuous cleaning Personal Protective Equipment: [EJ-Eye Protection [G]-Gloves |LJ-Lab coat/Sleeved garment [A]-Apion [R]-Respiiatory Protection [B]-Boots [O]-Cbntinuous Cleaning [N]-None El-All except Respiratory Protection 5.9 Chemical Bath Sampling-OSP Bath Type Example: Cleaner Microetch Other (specify): "Tvoe of Sampling [A]-Automated [M]-Manual pSTJ-None b Frequency: Enter t time elapsed or numb ft. processed between Clearly specify units sq.ft.) Type of Sampling3 A • •t Frequency1* 3 perday Duration of Sampling0 Smin Protective Equipment* E,G»A - Method of Sampling" P - c Duration of Sampling: Enter the * Method of Sampling: average time required to manually [D]-Dram or spigot take a sample from the tank. [P]-Pipette [L]-Ladle d Protective Equipment: Consult [O]-Omer (specify) he average the key for the above table and ter of panel sq. enter the letters for all protective L samples. equipment used by the person (e.g., hours, performing the chemical sampling A-34 ------- APPENDIX A o 8 a o I I g o s 1 "o 'o o a . ir S'S .fa> "•S "C ^ es 53 ^j O C ^ a. «> « S S -*• •a jj <8 >>-S « o g « 1-< ,S. •*? 3 I" I ol c e CO CO CO C ge cycle t s (include a _« T t <3f> ^C ^C o; a C O cd Q CO s 1 5 Q "« "S £ a* *• "3 ' II j v— ' it it. Oi H ^ "a -s? S -ss H | '§*<§, a 1 « f « s § S H 3 S <£- *"* I "o >» g \»5 "i o -* /»5V £ « 3 •§ > .5 ••j •*SJV •B,g S ^ ,_5 .S .s n - •.jf-f- . it* 1 1;^: iisHi iijliii l;^! ii^ «..A. : is' s. Si 1 S: : :r4.: iiS* 1 iiiiSl I! ;|S :::::: Hi a*; . 1 u - f' :::: iii: •iii :: — ::bT II ;;;y; || IIP ::1~* :l::; "•!:•: C" S p ex* 1 iilijj ill •i"«; : OJt f = i" ":•?:! «*•• ^ S-: :xf if* 11 ijyj; :::«: JS| ::?* -..*. S; C" : I 1: l^lill lill r.!..,^ II iiliH! iiiii ce 0 f • •e.'l'g - ^S w -*2 ri *' « jj o s S .s •s -I? H ? ° =3 1 2, O G O o a ^ -S o a> •*-* -*s ^ < O **| ' I |-"'"-| r ' i ^ > fadl E3. fe, ^ ^ Si 2 J §• ^^ C! 5 WQ o 5 C?* *o "S ** -5 o s g iff cs 5 "S\ -2 *>o PI ^ co co c3 ^ glltj. "1 A, ^ ri, ^ ISSSs V O 1 -I 4 ------- APPENDIX A .2 P ill •3 "S t? S 2 *5 m s j3 fl a 7 o-J "3 a 3 Sf H Cw ^1 ^ i c C3 _~ § Jc -*-« s™^> i 1 CO CO I CO i CO !3 *c3 o 1 o ,2 "3 o '•5 . co CO ^ ^ 1 s O ^"" JS CO CO 3 M ^5 •8 'o o> o> »S Q, 3 co •S a §•§ _. C3 2 CO II A-36 ------- APPENDIX A >-> III ,0, § i o * < j S rg "bb p C3 C Q M1 "* as {* o i H 8 «-«S 3 a as "s :i H a cs ^ lis S i 8 r 121 .III c sf 2 1 g -* i .S <§ ^ CO C3 »rt E **"* <1> all' 3jr C3 *Q •*-* do si w II li s <5 .y t; O "O C js !§ -e s .g j- *§ • O •a S « a 2 a S o "s ^ "o'^ .a ^ o ~ ^^ "*"* S o •23 <4-( '^ N l*^' ^ 5«§ ^ 1 - O « '•"' •S "5 1 2 S -i » < J 8 _o -^ CM 0 3Ss o<1 •S -s£ ^|2 6 Criteria for Addition" •s iiiiiill !!i!l!!i;!!!yi; !!;|;i;!;nrg; .: ::;;::::::;-..; g 3 i. ^ \ 5 CO II . "s'l ^3 ^ S^S mg [^ j>» . O^ (- i .Q rS j§ O .2 •** o *S | i u § § *J 1 r^K1 -2 a, § Q, • o ^-t Hi !1I ig-tg III o "0 >> 8 53 |^ _ /i ^F*.*^J— IP5.N^ii a S«o o ^^-5 « 1 a -| t1™ "S o ^ «fl 1- 11 « P-l»g 5 ^3 ^ ^1 Iff I H'H.^ "J-s'if fill . ilfl •O^OOT'O' 3*3 °"o li Si li1?! l|.il •*•» ?- t" Q^ ^*3 « q^*g r— i i .3 j-* ^SU O 5J (U P^ I— i Q (JJ*^-^H JU o ^ &*2 !| £32 .§ i *^ •> oo *jj *w 55 J- 2 2 oj ."ti! i i i i » "C *^ ixi (X( O £"* O A-37 ------- APPENDIX A C o •3J! els .2 5-5 •5 cj o> €5 S3 c> IP Method of Spe Bath Remova |§ « o1 u I Bath Type Cle croetch 2 "o en pecify her s v cr^S ? 513 ?.§•§ § 22 2 «J.a il|I i «-sfs^ « •s s.o.gS 2' III S »&2.So a'S'S-g S a S S •§§-g,gS W^^«fi .2-^, 1(2300 ^O^ffiO ]Tgg d> i i * ' 5 • -1 • * M OT. o • § a" g 3 e|5 -•s Hi! ! Hi « ,« * • | •S • • ' ' ' 2 Ist •§ j sr A-38 ------- APPENDIX A 5.14 Process Waste Disposal—OSP Bath Type Cleaner Micretch Flux Solder Post-Clean Other (specify): Annual Volume Treated or Disposed' Annual Volume Treated or Disposed - Enter the yearly amount of the specific bath treated or disposed. Be sure to consider the volume treated from birth 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- Pi - 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 disposal {RNj - Recycled on-site fRF] - Recycled off-site [O]-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 [O]-Other (specify) A-39 ------- APPENDIX A o6.§ ass. t{ «a u ^^Ift; |J»i (u w O<^3 < *« mi a. §• CO O .m o o •«• •H a a £ e T v> u a i A A fr ! GB « o £ ^ ersion Silve: £ 1 >• Pre-Cleane 1 ^ fc 41 e Microetch Pre-Condit 1 ' ^•ly e •t- 1 ! 1 « •w 1 -s CO 4) *2 1 1 • 1 Si'r^E^ 2 A4 CM O O a A-40 ------- APPENDIX A 6.2 General Data—Immersion Silver Number of days immersion silver line is in operation: Estimated scrap rate (% of defective product) for immersion silver process: daysfrr % '.- Number of hours per day the immersion silver line is in operation: Total of PWB surface square feet processed by immersion silver line per year* ;;;!!H;!!;;i;ll;irs!diai^; J!i!!!!lnil!Jii!!iJl!SSE$£.: 6.3 Process Area Employees—Immersion SUver 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 silver 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 Immersion Silver Area Worker Line Operators Lab Technicians Maintenance Workers Wastewater Treatment Operators Supervisory Personnel Other (specify): Number of Employees in Immersion Silver Process Area ;;;:;:;:;;;;;;;;;;:;:-[[[ ::::::::-.:::;:::::::::-::::::::::::::;:::;;;::;;:;::::""" •""*:"":::::::::::: : ':j:::"""::.:::::::::::;::: I:::::::;:::::::::-.;:::::::::;:::: ::::-::!::::::•'::: ::jiijl:!illl!!!!iJ!ii!!!ii!?!i!!n!;!!;l;!:!!!!i!iH!ii!i!ljJ!iiJ!iiiliiiiii!i Average Hours per Week per Employee in Immersion Silver Process Area :::;:!:;:;;::;;:::::; ;:;:::p;;;;;n;;;;:i;;;;i;;;:;;uj;i;;;;;;;i;iirs;: ;;;i;i;;:;:;;;;;;;;:;;;;;;;;;;;;;;;:;;;:;:;i;;;;;;;;;;;;;?;;;;;;;;::;iirs;i !:i;!;!i;;|:|!;::::::;:^::::::::;;;;;:::::::::K:r;^;:«:-:;::;: = ;JirSc: !!!!!l!H!!!i!!!;!lli!!!j!l!i!IJ!j!?!!l!p;:^~:^:^^:^:;:::lnre; 6.4 Physical Settings—Immersion Silver Size of the room containing the surface finish process: Are the overall process areas/rooms ventilated (circle one)? Do yon have local vents (circle one)? sq.ft. Yes No Yes No Height of room: Air flow rate: Local vent air flow rate: ^^mzmx&Mtfffiax Mm^^^WMZMm Overall surface finishing process line dimensions Length (ft): Width (ft): Height (ft): Rack Dimensions—Immersion Silver Average number of panels per rack: Average size of panel in rack: 1 verage space between panels in rack: .:1::::::^:::::::::::::::::-::::::::: i, :::;--:-:::::; = ":-V::"-lIfc;; Lengthen.): . ~Width(inu): v :J!)!ij!l!!!lJ!;i!nnji!n!J!J!i! Do you purposely slow the withdraw rate of your panels from process baths to reduce drag-out? (Circle one) ------- APPENDIX. A 6.6 Rinse Bath Water Usage—Immersion Silver Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse bath? present in your immersion silver 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 water flow rate of one bath in the cascade. I Total volume of water used by the surface finish line when operating: gat/day I Process Step Number1 Example: ' 8 • • • . *» • : . Flow Control" R 1 1 ! i f _ - ' Daily Water Flow Ratee i 2,400 gaL/day gal/day gaL/day gat/day gaL/day gaL/day gaL/day a Process step number - Consult the process schematic in question 2. 1 and enter the process step number of the specific water rinse tank. b Flow control - Consult key at right and enter the letter for the flow control method used for that specific rinse bath. e 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. Cascade Water Process Steps'1 8-*« \ j Flow Control Methods Key [C] - Conductivity Meter [P]-pH Meter [V] - Operator Control Valve [R] - Flow Restricter [N] - None (continuous flow) [O] - Other (explain) 6.7 Filter Replacement—Immersion Silver 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 Kev: Not Appli cable D (E]-Eyc Protection [G]-Gloves [Z]-A11 except Respiratoiy Protection [L]-Lab coat/Sleeved garment [A]-Apron [N]-None [R]-Rcspiratory Protection [B]-Boots A-42 ------- APPENDIX A 6.8 Rack or Conveyor Cleaning-Immersion Silver Not Applicable Q 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: •Hi;! : Rack Cleaning Method: [C]-Chemical bath on SE process line [DJ-Chemical bath on another line JTJ-Teinporary diemical bath [S]-Manual scrubbing with chemical [VQ-Non-chemical cleaning DN]-None [O]-Cpntinuous cleaning Conveyor Cleaning Method: [C]-Chemical rinsing or soaking [S]-Manual scrubbing with chemical [M]-Non-chemical cleaning Bfl-None [O]-Continuous cleaning Personal Protective Equipment: [E]-Eye Protection [G]-Gloves M-Lab coat/Sleeved garment [A]-Apron [R]-Respiratory Protection [B]-Boots [O]-Continuous Cleaning [N3-None [Zl-AH except Respiratory Protection 6.9 Chemical Bath Sampling—Immersion silver Bath Type Example: Pre-Cleaner Type of Sampling3 \M Frequency6 Duration of Samplii Protective Equipment'' Method of Sampling' III Pi! Microetch Pre-Conditioner Immersion Silver Other (specify): 'Type of Sampling [A]-Automated [MJ-Manual [N]-None b Frequency: Enter the average time elapsed or number of panel sq. i. processed between samples. Nearly specify units (e.g., hours, sq.ft.) c Duration of Sampling: Enter the average time required to manually take a sample from the tank d Protective Equipment: Consult the key for the above table and enter the letters for all protective equipment used by the person performing the chemical sampling. " Method of Sampling;: [D]-Drain or spigot [P]-Pipette IL]-Ladle [O]-Other (specify) A-43 ------- APPENDIX A i-( o o O W5 Operating Conditio OS •8 Q en en if & £5 ^ If •£* Cli g /-> w __ en *S T3 ™ g a a .§ ° a JJH o> S >£• * — f I s=> II* ^ ^3 W ts *S S (3 •o pfi 'S? B1! •-I 0 f P9 B* tt tti ox. , . 1 OS* X R? I 0S 1 Pk fe 0 : ox' i= i a ff • J.' •2 .1 § 64 ~ § 03 A 1 "i .g' .s" i- V B .2. ^i» •3 | £ fe 0 1 1 §> .a .s" b ------- APPENDIX A A-45 ------- APPENDIX A •2 "a§ •«^ f^ CO S S S S bo-M ^ .§ •5b^ SI o -2 Hi •S « H 2 a 0< o" cr w W a SOW 0 S II « g S "S P Conditioner Immerson Silver a. il-! pfl if P*j— 12 8 t-' § S ** •a ------- APPENDIX A £1 1 I CS •s Of H a" H «§ IS II «§ li i ,J5 I I = ' S3 es S II ,o a, £ JS I Slf »«>.a « „ |-Sg, -S 8 *• O.S Ci TS § " S 5 B lll------- 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*NfflM6B> •*••"* 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. | «§ ' ' .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> !«§ ^ 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 -------