UNCLASSIFIED

AD NUMBER

AD018113

NEW LIMITATION CHANGE

TOApproved for public release, distribution

unlimited

FROMDistribution authorized to U.S. Gov't.

agencies and their contractors; Otherrequests shall be referred to AFWAL,Wright-Patterson AF, OH 45433

AUTHORITY

AFWAL Ltr. 14 Jan 88

THIS PAGE IS UNCLASSIFIED

Reprodued by

ervIces Tec a rmaton gency" SRVICE CENTER

KNOTT BUILDING, DAYTON, 2, OHIO

.1

~1

-- D

IN r*orIL r

" Al- rE&IIN1(-AL I{i.olr[ 5692

AN INVESTIGATION OF ELECTRODEPOSITED ALLOYSFOR PROTECTION OF STEEL AIRCRAFT PARTS

A I 11 I. IRI'I.' R. JR.

(,1. I:% i:' . L: RaR. C II A'. 1:N L.. F.:iVjS

It. I11:1 1 F. 4ll. r ) f.k I. I1 i 1 VLT

P 1.) Vi5,A

RIIl1lT Alit DEVELOPMENT CENTEU

AF TECHNICAL REPORT 5692SUPPLEMENT 4

AN INVESTIGATION OF ELECTRODEPOSITED ALLOYSFOR PROTECTION OF STEEL AIRCRAFT PARTS

Arch B. Tripler. Jr.

Glen Fuller

Dr. Charles L. Faust

Battelle Memorial Institute

July 1953

Materials Laboratory

Contract No. AF 33fO38)-8750RDO No. 611-11

Vright Air Development Center

Air Research and Development CommandUnited States Air Force

Wright-l'atterson Air Force Base, Ohio

POR E"VIO) O)

Thi-~- rn nt 2 .anareA it +1-1 n 1 4-+a I I hin-n IN* 1i.l lusttlt

Thrp r r-- J -- - I a i"=derUSAF Contract No. AF 33(03)0750. The contract was initiated under Re=rearoh ar.: Development Order No. 611-11, "Electrodepositon and Electro-chemical Treatments" and was administered under the direction of theMaterials Laboratory, Directorate of Research, Wright Air DevelopmentCentor, u3 th Major L. E. Michael acting as project engineer. This is theFinal Report on the phase of the work related to outdoor exposure testingof certain experimental coatings on steel.

The work was conducted by A. B. Tripler, Jr., J. Edwin Bride, GlennSohaer, 07.en Fuller, and C. L. Faust, all of Battelle Memorial Institute,Columbus, Ohio.,

A nQ Wrq a-' 'p

kure-manganese, mnganese-zinc-alloy, and tin-tinllloy coatingnon eteel wmere exposed for 21 months at the Patteile North Florida ResearchStation& Panels coated with zinc, cadmium, and chromated zinc wore expoaedconcurrently as standards.

Th6 pure-manganese and mnganee-zinc coatings were inferior to thezinc and cadmium utandards.

The zinc-tin-alloy coatings of 0,3-mil thickness "d greater pro-tected the underlying iteel as well as the zinc rnd cadi-nu standards.

PUBLICATION REVIEW

This report has been reviewed and is approved.

FOR THE COMMAND-ER

* .J SORTEColonel, USAF

- 4/ Ahief, Material LaboratoryDirectoratte of -a -at h

SiT D) Dvri TG0 Awl'

SUMMARY . . , , * . * , * . . . . . . . * * * 9 9 1 * 2

DISCUSSION OF ESSENTLkLDATA. , . . . . . . . . . . . 3

Introduction . , . . , . . , . . , a . a a a 3Zinc-Tin-Alloy Coatings. , . . . . . . . . . . . 5 * 5Pure-Manganese Cuatings . , . . . . . . . . . . * . 6Manganese-Zinc-Alloy Coatings . . . . . . . . 6 . * . 6

EXIEPLMENTALWORK . . . . . . . . . . , . . . . , . 7

Preparation of Exposure Panels *. . . , . a . . a . 7General. *. . . 0 6 a a a a a a .a a 7Zinc Plating. . . . . . , * . * . . * . .* a a 9Zoadmium Plating * * . . . . . , . . , . . . . 9Manganese Plating. . . . . 9 . . a a . a a a a a 9Zinc-Tin-Alloy Plating.. . . . . . . a .a .• 9Manga ese-Zinc-Alloy Diffusion Process.. . . . . IiChromnate Conversion Coatings . , . . . . . . . 11

he Outdoor Exposure Site . ....... ... * . .

APPENDIX A. CUMULATIVE RESULTS OF OUTDOOR-EXPOSURETESTS FOR EACH TWO-MONTH PERIOD . . . . . . . . . . . 12

APPENDIX B. DETAILS OF PLATING AND CLEANING PROCEDURESUSED IN PREPARING PANELS FOR THE OUTDOOP.-EXPOSURETEST * * . • . . . . . . . . . 4 * * * . . . ... . 52

I ,, I

1,TST OF R;JY.S

Table PP

A-1 C U TLATWVE(P5 1SLTS OF OUT DOOR-EKXPOSU"RE TESTS A~h 6o) DAYS15Exp%0U£R0 0 a a c C, a Sa a c a a a0 6 1

iL2 CUMULATTVE R]'PS TILTS OF OUTDO)OREXPOSURE TESTS AFTER 127 DAYS'NJ1Z'pSIRE(). 4;z 7z:z=21- - - - a a a a a a 16

A-3 CUMULATIVE RESULTS OF OUTDOORs.EXPOS UE TESTS AFTER 18 7 DAYS'(AvPRoxnakTELY 6 MONTHS' ) KXFOSURE-21) ***.. ....

A-4 CUMULATIVE RE-SULTS OF QU3 DCOR-EXEOSUIE TESTS AFTER 245 DAYS'(8 MONTHS' ) zXrOsunEU) . * * * . **. ***a * ,,...6 2

A-5 CUIITLATIVE RESULTS OF 0V EXPOMSURE TESTS AFTER 302 DAYS'(10oMONTIs')EXPOSURSU ***.... o0 &a !.. .,, 26

A-6 CUMULATIVE RESLTS ,OF OUTTDOOR- EXOSURE TESTS AFTER ONE YEAR'S 3*KOUEW 0 5 * 0 5 0 & 0 0 0 0 S 0 0 0 0 S 0 3

A-7 CUMLATVEdi hTS OFOUTDOOR-EXFOSUPE TESTS AFTER 4l MOThS1

EXPOSURE ~ 0 0 0 0 a a a 40 * a S 0 a a 3

A-S CWUIE(WSULTS OF OUTDOOR-EXPOSURE TESTS AFTER 16 MONTHS'.~ ~ ~ EXOUES 6 * 0 5 0 0 0 * 0 9 C a 0 0 0 * 0 0 39

A-9 CUMULATIVE (R5 ULTS *OF OUTDOOR-EXPOSURE TES TS AFTER 18 MO-:T HS'IXPOSURE~* . 0 0 0 0 0 0 0 4 0 S a B 0 6 -

A-10 ,WTAT .FUL17So O0UTDOO' CR-EXPOSURE TESTS AFTER 20 MONTHS'- a a a a * . * a a * a a * q * 0

In a temperate climate, electrodeposited zino or cadlmium keepsstoel from. rus-tin for tea years or more. In the tropics, zinc and cadmiumco~ttrLng; dctcrlor . "rli"ly nd tir mtanl thwn nfirrniu 2 7n many 0aB8C.

this occurs in lest thn a year. Duwin, World War II, this condition becamea menace to aircraft which were operating in the tropics. In addition, thestorage of steel aircraft parts could not be accomplished with safety.

In 19h6, the Air Force contracted with Battelle Memorial Institute toconduct an exploratory research for substitutes for zinc and cadmium coatingsfor steel parts of aircraft. In planning this work, two restrictions were es-tablished for the substitute coatings: (1) the coatings must provide sacriicial (cathodic) protection for SAE hl130 steel, (2) the coatings must be capableof being electrodeposited.

The praoticrl galvanic series of metals and alloys shows that, of themetals which can be electrodeposited from aqueous solutions, there are onlythree expected to be more active than iron. These three metals are zinc,cadmium, and manganese.

This limited the search to alloys of cadmium or zinc, and pure manganeseand its alloys. In order to keep the problem from becoming too complex, thestudy was further liuitod to binary alloys.

Although climatic conditions vary in the "tropicsw, the common denomi-nator wab the daily cycle of moisture condensation and drying. In acceleratedlaboratory tests made to simulate those conditions, pure nanganese coatings,zinc-tin,.alloy (20-80) coatings, and manganese-zinc-alloy (50-50) coatingsshowed promise of bsing superior to zinc and possibly also to cadmium.

This report tells what happened when panels platud with these experilien-tRd ccatings were exposed outdoors at the inland site of the Battelle NcrthFlorida Research Station, Daytona Beach, Florida.

SUJb RY

Of the thrse exporimnnt.l coatings tested at the Bs.ttlI. e North9lorida Research Statimoi only the zinoctiv-alloy ooatings of 0, 3 - aindO.5=ndl thicknesses offered proteotion equal to the zino -.nd oudmiu isteadard coatings. Because the test ive discontinued after ppro.imtely21 months' exposure, no conolusions ean be drawn as to whohetr &ho zinc-tin is superior or inferior to zinc and oadmium at this exposure site,

Zinc-tin-alloy ooatings of Ol-nil thiokness were infei-or to bothzino and cadmium coatin&a of like thickness.

Pure-manganese and mnganese-zino-alloy ooatings failed rapidly.A 0.5-mil coating of either one was somewhat inferior to 0.1 ml of zinc,and was not nearly so good as 0.1 nil of cadmium.

The preparation of the teat panels is described in detail In thereport.

.~ p. ' ' *:.p} I,

VJV3GCUSSIOil2 ( " .,.J.NCI, L)P0A .

int r oduction

Three thicktc's1- ° -e -.tecd for ear' expev int.! coating: 0. 1 mU,0.3 itail; and 0. 5 mil. There were foul 4 x 6-inch ;oua.tls for each thick-ness. For the three experimental coatings, this aoded up to 36 panels.

As standards of comparison, panels coated with pure zinc, chromatedzinc, and cadmium were exposed simultaneously. Tivey wzi,- alzo pre.-aredin the three thicknesses, and there were four panels for each thickness.Thus, there were 36 standard panels, making a grand total of 72 panels.

Each of the 72 panels comprised a "four-in-one" test; that is, eachpanel had four significant areas which were observed separately during thetest period.

The surface facing the sky (the panels were mounted on ASTM racks

at an angle of 30' from the horizontal) is referred to as the top, and the

surface facing the ground is referred to as the bottom. Each of these twosurfaces had a 4 x 4-inch area which was unmarked, and a Z x 4-inch area

which had two intersecting, diagonal scratches (see Figure 3 in Experimental

Work section). The coatings were scratched in order to determine the degreeof sacrificial protection given to the underlying steel. The scratches were

milled to a width of 0. 006 inch and a depth sufficient to expose the underlying

steel.

In the tabulated results, found in Appendix A, four symbols are usedto designate the four areas. T refers to the unmarked portion of the panel

that faced the sky. TX refers to the scratched por.xoa of the panel whichfaced the sky. B refers to the unmarked portion of the panel which facedthe ground. BX refers to the scratched portion of the panel that faced theground.

During the first four months, the panels were examined semimonthly,

after that they were examined monthly.

The arrangement on the exposure rack ia shown in Figure 1.

Each panel was notched according to a code illustrated in Figure 3

in the Vxperimental Work section.

)-", ' U l~ 3

I - -..--

Li - C

__ ___ 4

~ j £~

- fl N! I4 4 4

84 ~

IB

4

Sb

-~ ~.,45

4N

Ma Ma

-____ w

____ ____

-1-------I ~4

-I-----4

_______ I_____il-i c~) FN! _ -

C -

-_NJ ____ 4 C'

(5 ,~ -

LI hiN

-1C

:2 Li -Li - U-

K- IC

4 - i 0

~ LK- I-. a

APT 1? 5692 $uppl 1~ 1-s

In addition to the regular Laboratory Record Book entries, KodachrOTmcslides were made of the panels during exposure. These were made after 6mor hs, one year, and 21 months. Two sets were made each time. Oneset was retained by Battelle, the other was sent to the project enginleer atWright Field. The slides were marked according to the numerical arrange-ment given in Figure 1.

The teats commenced May 17, 1951, and were discontinued on Feb-ruary 7, 1953.

The sole criterion for judging the protective value of the experimental

coatings lies in how they compare with the zinc or cadmium coatings.

Zinc-Tin-Alloy Coatings

For the twenty-one months that the panels were tested, the 0.3 mil.and 0.5 mil zinc-tin-alloy coatings protected the underlying steel as wellas the zinc and cadmium standard coatings of like thicknesses.

The 0. 1-mnil zinc-tin-alloy coatings were much inferior to the zincand cadmium coatings of this thickness. The bottoms showed considerablerusting after two months (see Table A-l, Appendix A). Table A-2, AppendixA, shows that, after four months, rusting was severe.

The observation here, that thin coatings of zinc-tin alloy fail rapidly,confirms the findings of other observers, The thin coatings are probablymore parous than thicker coatings.

Zinc-tin-alloy coatings confer about the same amount of sacrificialprotection on steel that cadmium coatings do. This was shown by the earlyappearance of rust in the scratches. Rust was present to approximatelythe same extent in the scratches of the cadmium coated panels. Corrosion-current-density measurements were made several years ago when thelaboratory phase of this work was in its early stages. Zinc-tin (20-80) hadabout the same static potential as pure zinc. Yet, when coupled with baresteel, a smaller current flowed in the cell containing the zinc-tin alloy.This is because the zinc-tin alloy polarizes to a greater extent than purezinc, thus accounting for the lower degree of cathodic protection.

It is not recommended that zinc-tin coatings as light as 0.1 mil beused. Since the testing of the 0. 3-mil and 0. 5-mil coatings was discontinuedbefore they had failed, no conclusions can be drawn concerning their relativeeffectiveness as compared to zinc or cadmium.

. 6 3692 unol l 5

Pur eC-Mangba'.ij r G o stiu

Pure-manganese coatings were inferior to zinc or cadmium coatings.Reference to the cumulative data tabulated in Appendix A shows that 0. 1 milof either zinc or cadmium was better than 0.5 ntl of mangancr .

The manganese coatings failed largely due to undercutting (this was alsotrue of manganese-zinc-alloy coatings). The point of weakness was the intei -

face between the coating and the steel. The rust spread laterally beneath thecoating, prying it loose. After the coating flaked off, rusting proceededrapidly.

The degree of sacrificial protection given by manganese coatings isdifficult to determine. Manganese corrosion products formed in the scratches,but, being brown, they masked any iron corrosion products which formed.Any sacrificial protection offered by manganese was not of long duration, be-cause of the rapid oxidation of the manganese. The corrosion products werecathodic to steel and this may account for the undercutting which wasobserved.

Mananese-Zinc-Alloy Coatings

The manganese-zinc-alloy coatings were formed on the steel by platingmanganese on the steel, plating zinc over the manganese, and then heatingso as to cause diffusion. Details are found in the Experimental Work section.The composition of the coating approximated 50Mn-5OZn (weight per cent).

Manganese-zinc showed the same failing at the basis metal-coatinginterface that pure manganese did.

The manganese-zinc-alloy coatings were also inferior to zinc orcadmium coatings. Zinc or cadmium coatings of 0. 1 nil were superior tothe 0. 5-nil manganese-zinc coatings.

SP5692 SuppI h 6

EXPERIMENTAL WOM{K

Preparation of £Mposure Panels

Gene ral

The SAE 4130 steel on which the experimental and standard coatingswere plated measured 4 inches x 18 inches x 1/32 inch as received, Be-cause the larger size facilitated polishing and buffingp these operations wertcompletea before cutting .L... t 6- ch *w.a

The large panels were cleaned in a hot, alkaline, soak cleaner. Theywere then polished on 240-grit, substantially new, emery belts, The finalfinishing was done with a sisal buff (Tampico wheel), which produced a finishabout equivalent to a 300 grit. The panels were then cut to the 4 x 6-inchsize, care being taken not to mar the surfaces. The cut panels were storedunder kerosene until ready to plate.

A "robber" type rack was used for plating the panels (see Figure 2).It was formed by bending 1/8 x 1-inch hot-rolled, plain-carbon steel intoa rectangular frame measuring 6-1/2 inches x 4-1/2 inches, inside dimen-sions. The butting ends of the frame were welded together. As shown inFigure 2, the steel panel was supported within this frame by three contactpoints. Two of them were located on the insice of the lower 4-1/2-inch side.The third was a spring clip located on the upper 4-1/2-inch side. A 1/4-inchrod was brazed, end on, to the outside of the upper 4-1/2-inch side. The rackwas suspended in the plating bath by this rod, which in turn was fastened toa reciprocating or to a stationary work rod.

The dimeniions of the rack were determined experimentally as thosegiving unifor-n plate distribution. A slight modification of the rack wasnecessary in o-der to get aniform distribution when plating manganese. Theone-inch strip of ;!e "robber" was narrowed to three-quarters inch, and four,1/4-inch holes were drilled in he top strip to allow gas to escape. Srrzalldefects in the plates occurred at the three points of contact with the rack,They were lacquered to eliminate them as foci of corrosion.

Three coating thicknesses, 0. 1 mil, 0.3 mil, and 0.5 m-il, were pre-pared ior each type of coatings and there were four pr-nele for e ch thickness.The MagneeGage** was used for determining coating thicknesses. A toleranceof = 10%, relative to the nominal thicknesses given above, was allowed. Amagnet was especially calibrated for measurement of manganese coatingthicneesa. The method is described later in this section under Manganese

a g L-s y pew;i Wo A N. nl1, 9p 64 ". and No. 6M. pp I-8.

'in~~~r Ani 9'.,n

FIGUW 2 'gB E TYPE PLATING RACKr&, ,DOR UI p PA N OF

' n FOUR -INC H By siX -INCB OUT O RE PSU EPN L

.\FTR 5692 3uppT 1

Identification of the type of coating and thickness was made on each

panel by a series of V notches cut in the edge of the panel prior to plating.

The code used is explained in Figure 3. The observer is looking at the top(side which faced the sky during the test) when the coating code notches are

at the upper left and the thickness code notches are at the upper right.

Figure 3 also shows the position of the scratches or scribe marks. The

latter were made after the panels were plated. They were cut accurately

on a milling machine to a width of 0. 006 inch and a depth sufficient to expose

the underlying steel. The milling was facilitated by use of a specially pre-

pared jig. The diagonal cuts were made on each side of each panel.

Zinc Plating

The zinc plating on those panels having zinc alone was done from a

cyanide-type bath. The composition of the solution and the conditions for

plating are given in Appendix B.

Cadmium Plating-

The cadmium plating was done from a proprietary cyanide-type bath.The composition of the solution and the conditions for plating are given in

Appendix B.

Manganese PlatinL

Sound, lustrous-gray deposits of manganese were obtained from asulfate-type bath containing large amounts of ammonium ion and a very

small amount of sulfite. The bath composition and plating conditions aregiven in Appendix B.

A calibration curve for the measurement of manganese plate thickness

was not available, A calibration was made by first testing the thicknessmagnetically, and then, using exactly the same spot, measuring the true

thickness of a microsection with a microscope. The magnetic values wereplotted against the true values to obtain the curve.

Zinc-Tin-Alloy Plating

The zinc-tin alloy was deposited from a cyanide-stannate bath developedby the Tin Research Institute in England*. The alloy had a nominal composi-

tion of 80% tin and 20% zinc.

;. Elec roclem Soc., 94, 73 (1948)

A TR 5692 Suppi 4 9

I 2 3

C

A

I KI

~lI

A:Z~cCoding

A= Zinc C = Manganese 1 O.I milA+ C Zinc + chromate A+B2 Zinc-tin 2w 03 mil

B Cadmium +C a Manganese-zinc 3-0.5 mil

FIGURE 3. SCALE DRAWING OF TEST PANEL SHOWING CODING SCHEME FORIDENTIFICATION AND LOCATION OF DIAGONAL SCRATCHES

Afieq 5()2 Sippl 4 X0

Both composition and plating conditions are given in Xppendix T.

There is also a note regarding the importance of using high-purity nodiuim

stannate.

Manganese -Zinc-Alloy Diffusion Process

The manganese-zinc-alloy coatings were prepar"d by heat treating

manganese-zinc duplex coating. so am to cause interdiffuuion of the two

elements.

Manganese was plated directly on the steel from the regular manganese

bath. A very dilute sinc zyanide bath was then used to place a zinc strike

over the manganese plate. The balance of the zinc was then dcpeQe! d over

the strike from a special acid zinc-plating solution.

After plating1 the panels were placed in a cold blower-type furnace.The temperature rose to 600 F in one hour and twenty minutes. The fur-

nace was held at this temperature for six hours. The panels were furnace

cooled to room temperature in 6-1/2 hours.

Chromate Conversion Coatings

The chromate conversion coatings were formed by the Cronak* pro-

cess on 12 of the zinc-coated panels. Details of its use are given inAppendix B.

The Outdoor Exposure Site

The panels were mounted on standard ASTM racks at the inland site

of the Battelle North Florida Research Station, Daytona Beach, Florida.This site is approximately two miles from the Atlantic Ocean. Salt spray

is essentially nil and a heavy dew occurs nine out of ten nights.

The racks faced south and were elevated 30* from the horizontal.

* New krsey Zinc Co. . U. S. Patent 2,035,380.

AFTi 5692 Suppl L

APPENDIX A. CUMULATIVE RESULTS OF OUTDOOR-EXPOSURETESTS FOR EACH TWO-MONTH PERIOD

AFTh 56)2 Ju- - 12

I'ABLE A -. (UMtULATIVE RESULTS OF OUTEXXOR-EXPOSURlt TESTS

AFTER 60 DAYS' EXPOSURF (M'

Coatug( 2 Stcflon

_Type Thickness of Pamel(3 Description of Corosion

Mn 0o1 T Scttard pin-point nstilng throughout; very numerous pin poinni of

-iti on vertical edges

TX Nu.erous pin points of rustB Small area of rust ,.. U_ -.c left corners of all 4 specimens

RX Numerous phi points of ust on 1 specimen; f few pin points on 3upceiancm

Zn 0.1 T No rtTX Very slight uace of rut in scratches; scratches mostly filled with

white corrosion productsB No rustBX Same = TX

Mn-Zn 0.1 T No rut on 2 panels: trace on 2 panels

TX A fw Ist pin poinn on 1 panel; rust traces in scratches on 3; 3 panelsshow small tumed area

B One pin point and a few small rutd areasDX Same u B

Cd 0.1 T No rust

TX Slight scattered traces of rust In scratchesB No rusteX Same as TX

Zn-Sn 0.1 T Several pin points of rust

TX Panel 1 shows 1% rust; panel 2 shows 2% rust; panels 3 and 4 show 4%rust

B Pin-point rust coven 80M, 40A. 20%, and 3%, respectivelyBX Rust spreading from scribes

Zn(Cr) 0.1 T One small area of rust on each of 2 specimens

TX Same as T& No rustBX Traces of rut in scratches

M11 0.3 T No rust

TX One pin point of rmn visible zftt- 1S day4; no longer visible after 60days

UNo rustKo ruzt

0.3 T No rut

TX No rwt; white coi|o(i .ductu in scribesSiHo r ta.3x No ratl; white cojrrfpc- prodtct' in scribeg

AYrR 569.2'UP)h

TABLE A-1. (COniinucd)

... cint 2 ) SectionfTv - Thickness of Panel (3 ) DescrLfption of Corrion

Mi-Z 0.3 T No rustTX Very slight trace of rust in scratches after 18 days; no longer visible

after 60 daysB fo ruatBX No rust

Cd 0.3 T No rustTX Slight trace of rust in scratchesa No roteX Scattered traces in scratches

Zn-Sn 0.3 T No rutTX Traces of rwLt in scratchesB No rutDX Traces of rust in scratches

Zn(Cr) 0.3 T No rust

TX No rutB No rustBX Few taces of rust in scratches; white corrosion products in scratches

Mn 0.5 T No rustTX No rust3 No rstDX No rust

Zn 0.5 T No rustTX Possible tace of rust in scratches; white corrosion products in scratchesa No rstOX No rut; white ccrrmlon products in cratches

Mn-Zn 0.5 T No rustTX Posible slight traces of rust in scratches3 No rustBX No rust

Cd 0.6 T No rut; 1 pin-point blister on each of 2 specimesTX Numerous traces of rust in scratches; several tiny blisters on I specimenB No rust• , l st L ;crzach^ for &about Wo loength

in-Sn 0.5 T No rustTX Traces of rust in acratches for about 75. of LengthB No int

BX Slight cr..zes of rust in wcratches for about 2$.* of length

'Jik 5) up it ]It

TABLi A-I. (Continiucd)

Coatin&2 ) SectionT*y Thickneo of PAnMD(3 scription of Corroion

Zn(Cr) 0.5 T No rustTX No rustB No rustaX No rust, trace of whit corrotion poducu in scratches

(1) Test uuted May 17, 1951.(2) Tii thlckneen v-." (0.1. 0.3. 0. 5) ae in terms of mil. One mil 0.001 inch.(3) T = Unmarked portion of panel which faces sky.

TX = Scratched portion of panel which faces sky.8 = Unmarked portion of panel which faces ground.OX = Scratched portion of pkAnl which faces ground.

*'.Ftrj 56r)2 6uppl J 15

I0 ibal

VV

VV

0. Ch I.,t

fria 88o2" 2

I '~ fit j ttlfr t%

Bo Ieq

0 im

4 4- - 4r

AFDR 5692 Suppi. 4

4

fl

a

\~:x ~:~x1~

$ ~4

2 I',

Co C

Ill C

0 0 cvC'

- 4

a

C'

1 ,~4? c;111-jVt h

~1 -4

C

A .go 1. ~o

A-.

B4

4 -Aa Ia a -'-3 ~ - 0

a ?~.' S aU

U C.a ~ - ~j~0aD

a ~

swK ~ A A~A 4uU4.

4

4.

~- ~-00 0

~- a C 'C

sfl 0 13r~ ~

a -. .4

aI A-

32

3'

'2

i ;-,

~ ~ ii.~

tjt

C6C

CD CC

44L4 -0

o o eq

InI

006a

a

O..

S co.

tA~

131.4

Po

C3-

000

r r

CAA

'A 1

CL c

ILQ

unc .2 C 4

S 06

LUU

a p - C

0 g

igd-

JUTT8

26'.

c 2L

04 4

L L~

R 0

k-x F. I-

II

I

I

'11-'3

i 4'3

IixI '% I RIx \tI U I

Sg ~0

0

U o

~ *

% 4 ji~5 8- -~

~

~

S 6 C

C

C.

Ph

r1 (p?

1

I I

.1

it I.Ii j A

Is I a

a5,-. a

2 1S

c~

CI ~.1I ~ .~1iii

ii. hI I, I i

I I ~ 4 ~

,, .f4

I 'JI

-: IIII-, it

'~I ~i ~-

I..- J~

U ~

1~~o

ID.z.

.2'U'-. C, o 0

w ~ ~o ~t3II I'IIii 1 j1

I.1 111 Ii

'a ~* ~ ii *I*I1r~ 1. *4 U~ 1ii I A 4'

!~ Ui *A .4*1 ~

'-

4

.~''~4 I4 .~. M

4

IU- I0I-

LI.) h I-' ~

I-'~~~

01

'-I .

~ p4C

.. ',

U~.

Li,

ILrR 5~2 .Lp1A. h

I

a

d

~pI~ ~i a

~ * ~

In a

U*0

N

A? 2 2 ~ ~

CL

at

I

I'lllit I

ii Iii AIj 4

m 0

I- 8I C

HI- mt ~;1Lc&o

I~ ~ ~

I t-~ U.- s-. N M.~

0 ~

K II3

I -y

-W u

cr a78 72

96 3

AD -

co

Ld c

I

- - .43

4 .- -

010

-C 4iC

IL

R.nQ 7Z

-c m5C a~ ccc mi

4 U

c~ r

IL

E4 f

z '.7l

-tx-V- F

- .4

61 x

c #, -

32a

0

00 00

C CQ

0x

H F- m

j ~ 2

d 00 P & . C-

to i

- I -F)

-C "

~~r X>0~

-. z z z /

x o x x

All. ~ ~

5692C -

0 -r,

%.4 t

-e.d

4d

U -U u~

i f4 "0.- 1.

1" -4 4.e4 - - t o .

,0 - 14 u

U

-4 14 L) $4u l 4

k4 ;a v 0 1

o I -otZ. Z

J-4~

C-2- 0 3 ~ .

'.' -35

a. 04 t

44A

00 0~ toFA 0 to0

4' V 0a.4. u0~0 )

oUU

to 0 W U 4" 0)'S: 0. iv) 0 ~ 2~

.4'.v 4.0- 0

0) U 0 u 005401.0 0is

00

0

.-. r~0 L

93

u 93

Ai~f R 5692 upp i 4 36

0

00 0

4 0s C3

k0 - n0o fd 0 A)

~~~~w 0 0' ' a~~~ 'a Utv..~oo T ks 0uf

54 k4 93~ 0.1 0gAo .1 - I

wm u A 0 0s4 Of .00 v .4 atos

A in QaAQ) N

a m ~ 5 4 o

S ~ .. s 'I isdkt -' c 0k

.0t 0 0 I t~ 4 #A 0u OA k' a 0

0 U a. 0 0 Nd 0. 2 s0 0 0

at o Ut Ut x

Un inLuL

o U a... 5 54tos

a 94ht 3 0i

a56)2e 5'

Footnotes for Table A-7.

(1) Test started May 17, 1951.

(2) The coating-thickness values (0. 1, 0. 3, and 0. 5) are in terms ofmils. One mil a 0.001 inch.

(3) T a Unmarked portion of panel which faces the sky.TX. Scratched portion of panel which faces the sky.

B a Unmarked portion of panel which faces the ground.BXu Scratched portion of panel which faces the ground.

M

0

-1 w. L. ) c40 - 0 0 M

II0 I D - . 4, - -jm

M I I 'fl -.i - .

tt~u 0 0--

II at _ u ja Uk 4. ,. ,

0/

II ~ ~~ IL M (9 :. ..

0) = .4)Az

ILI 00'.oU

0" L.U04 0

A. 4o

bCL

a I..

0 Z'.4 0~

a.~ 44 0U''4U ,4

0 0 ~ 4 , ~ ' , 0 -

nC 0 >o"' >

0~ -

01 c 3u1,

E~ C' CuM

'-oS 0,,

1.. ..0. 4V1 4. Cu Cu

46 C

00 at'. $4 - , ~ u

4. 0 W )

(Af-

PT-.-

to II14- %.4 -l

0 " 0 :3 :5

v4 go 0. Cl

14 0 0 0u

u v 0 0 g

1. 14 L

-40 '4~0~ 14 Q

z ~ U 0 u0 lu0 *0 0 40

"~ 0~4 cr Ina . . .

LI) 0% U00*0

0 in

t 1~~o x ~

0 41

-0 00~ k~ d% t t0 co

4 -0 * N 0 t- 0cc

0 bD bo t00z 00 Ln 84 o 4'a a0 o .4 .4 .4 '-

~~~ *z zt4.~

o5a4 J0U ., X:

C;P ..

.41 ~0 0C'

t4U ' av 44 *10 a0 ' a'

u 4~a ~ ) 'N '-

0 ~ t 0 14~ 4 1 4 O ~V

5692 Zz- u0.4 01o

4) C

(4 o

00 0 C.-

O fu( %. 4U4

0

034 0 (4 x 4

0 00 $4

.4 0 C

.5 4 -0o

6) 4 00

lu 0

004) 00'

ot

01 (At) 0.co 4) t- 4

Ui 00

C '44

-4)

alA'4 2 supp 4

Cl U

IVj 00.

A 0 dP.CA

iv "W 0a.

4 -PC -d

o.0 * C i

In 4

* 6 *

S.- *6'a

0 ~ 0

- U*l

A..II 555 ' ML 4

00

14 0

oc

00

F4tt H If i

!ABLE A-9. CUMU.kLAIVE ;E-LTS OF OUT DOOR-EXPOSURE TESTS AFTER 'EMONT!S ' EXPOSURE()

Thickneu, Section

Type mll(2 ) of panel(3 ) Description of CorToion

Zn 0.1 All panels have a uniform gray-white fim of corosion products withinumerable pinpoint to pinhead spou of heavier white corrosion

producu, One panel has no rust, two panels each have a trace of,c-ttered pinpoint to light ru"n at top cdgC, the fouath panel lhs a1% area At top edge stowing scattered pLnpoint to light run

TX Three panels have 2%, 8%, and 12% r.pectivel), of their surfacesshowing sca tteed pinpoint to moderate rut; general appearancesame as T

B 1%, 5%, 50h and 35%. respectively of the surfaces of four panelsshowing moderate to heavy rust

RX 16%, 20%, 26% and 30D%, respectively, of the surfaces of four panelsshowing mroera&t to heavy rust

Cd 0,. T No rust

TX glighz incane'd tness of Iuoi in scratches

3 So ruit

IX Sliht scatured trac4s of rust in scratches

Zn (Cr) 0.1 7 No rUt all panels have arying amount. of pinpoint to pinhead spotsof whim cotroion product

TX 2% 10, 0%, and 501% respecivuly. of the surfaces of four panelshav moderate to heavy rust

5 No rust

BX 3%. 12%, 351, and 350, respectively, of the surfaces of four panelshave mo.:erate to heavy rust

Zn 0.3 T Small area of pinpoint rust on cne pane; all panels have uniformgray-whitue corrotion produL.u over entire surface, with innumerablemicroscopic to pin.ead spots of white corrosion product

TX Faint brown. stains Al ng lower edges of All 'Cur panels; gray-white.rd white corrosion products; same at

B No rut

BX No rust; white co::osion products in. scrazcsYcs

n-Zn 0. .9 T 5&, 75&/Q, 885v, D9,., respect Ively, of suraces cf four panelscovered with thickly cblaered pLapoint to 2-milhmeter russ spou

TARS e A - U ((M.,'2[hi: 1 J)

C oa tin gIKne$. Section

Type mil( 2) of Pael(3) Description of Corrosion

TX i0:YS N I OSrfa-,f ail fou: paVelS rust!'ng

A80 of un panel, and i00"/' of three remaining panels have rLstbreaking through coating or undercuttig coating which as lostauhes'.on

LX 96, of one panel, and 1001 of three remaining paneLs have rust

breaking through coating or undercutting coating wich has lost

adhesion.

Cd 0. 3 T No roat

TX Slight uace of rust in scratches

B No rust

BX Slight uace of rust La scratches

Zn-Sa 0.3 T No rast

TX Traces of rust in jcratches

B No rust

BX Slight uiace of rLt in scratches

ZXn(Cr) 0. 3 T No rust; scattered pinpoint specks of white corrosion products alocg

TX lovir edges of al panels; a Lew scattered large white pots else-where on all pan'els

B No rust

RX Few traces of rust in scratches; white co;rosior, products in scratches

NT 0.5 T IA 65%, 950,, and n%, resctively, of coatings on four panelshave flaked off and the exposed steel is rusting

T....X Vt. I 0 and 0%, £tipcLtctiiy, Uf coatigs 01 four pan.eishave flaked off and the exposed steel is rusting

B 31, 8%, 12%. and 40%, respectively, of four panels have rust

(mostly along edge,) undercutting the coating or where coating hasalready flaked off

BX 10%, 12%, 15%, and 15%, respectively, of fou panels have rust

(mostly along edges) undercutting the coating o. wixre coating hasAlread) flaked off

... .. . . , .-.

rABLE A-9, (Cont-ued)

Coa tniThickneu, Section

Type raIU() of P&.AIP Deicriptom of Corrosion

Zn 0.6 T No rust

TX Ponibille taces of rust In scfrathes; whit corroion pfo4=ts in scratchips

a No rust

RX No uti Whigs cnosim pzo ucu in scratchas

?&-Zn 0.1 T 15%, 15%, 88%, and 100% xeusctivuiy, of the ccadns of four paneLimaLr flaked off or biLng undercut by u iz

TX 855% 90% 100M and IDA respectively. of the coatLgs of four p-acels

either flaked off o being undercut by rust

90A. 95%, W%, and U16. rripecti-lely. of the coatings of four panelseithr flaked off ir beng undercut 'by rust

BX 100% of the coatings of all four pan lh either flAked off or beingundercut by rust

Cd 0.5 T No rust; one pinpoint blIsne on .each of two p4nels

TX Nurnrous traces of rust in scratches; several tiny blirtrs on ome panel

3 No rust

BX Rust in scratches for about 50r. of lengths

Zn- S 0.5 T No ruAt

TX T'rices of rust Ln scratches for about 50% of lengths

B No rust

ax Slight iaces of rut In scratches for about 2&4 of lengths

.n(Lr) UJ. b T No rust

TX No rut

B No rust

BX No rust; trace of white corroiio:- productx in -cratches

(1) Ten sstarr ed on MVay 1 Qc!,

(2) The coat tg hic kness yW (0, 2, O. 3. and , 5) Are n terrrs of is. One m... 0. C01 inch.(3) T m Unmarked portion of painel Which faces tne sky.

TX P Scratchedi portion of par.el whict faices te skyv.

B r Un.rnarked port,;n of pa'lel w11ich faces Liw gruUni.

-,* I cat~cw~ tor ot parcel whlcn faces the ground.

TABLE A-10. CUMULATIVE RESLILTS Of OUTDOOR-EXPOSURE T2S3S AFTER 20MONTHS' FXF(usLIK;:( 1 )

CoatirspThickness, Section

Type mil 2 ) of Panel 3 Description of Corrosion

Zn 0. 1 T All paneis have a uniform gray-white film of corrosion products with in-numerable pinpoint to pinhead spots of heavier white corrosion products.One pauel has no rust, two panels each have a trace of light to moderaterust zt top edges. .nd thc fourth pan.l has, 1'. zer.a at top edge showinglight to moderate rust

TX Two panels have no rust, two panels have 8% and 12%, respectively, oftheir surfaces showing light to moderate rust; general appearance tameas T. See text for explanation of apparent discrepancy between data ofthis report and those of Tenth 3imonthly Progtts Report

8 3%, W, 10%, 40%, respectively, of the surfaces of four panels showinglight to moderate rust

BX 16%. 25%, 30%. 40%. respectively, of the surfaces of four panels showinglight to moderate rust

Cd 0.1 T No rust

TX Slight scattered trace: of rust in scratches

B No rust

eX Slight scattered traces of rust in scratches

Zn(Cr) 0. 1 T No rust; all panels htve varying amounts of pinpoint to pinhead spots ofwhite corrosion products

TX 3A. 12%. 50, and 50%, respectively, of the surfaces of four panels havemoderate to heavy rust

B No rust

BX 5%, 12%. 3A. and 35%. :especuivly. of the surfaces of four panels havemoderate to heavy rust

Zn 0.3 T Small area of pirEniut rust f-, one panel; all panels have uniform gray-white corrosion products r er entire sutface. with innumerable microscopicto pinhead spots of whi, torrosion products

TX Faint biown stains along lower edges of all four panels; gray-white andwhite corrosion products; general appearance same as T

B No rust

BX No rust; white corrosion products In scratches

'P'R 569? _upp1l

47

TABLE A-IC. (Continued)

CoatinRrMdc.knt es, Section

Type rotl (2) of p.nel(3) Detcription of Corrosion

M.-Zn 0.3 T vf^. 9A, , and 10. respectively, of the surfaces of four panelscovered with thickly clustered 1- to 3-mm rust spots

TX 100% of surfaces of %ll four panels coveted with thickly clustered 1- w3-mm rust spots

B 100% of surfaces of all four panels have moderately heavy ruit breakingthrough coating or undercutting coating

BX 100" of surfaces of all four panels have moderately heavy rust breakingthough coating or undercutting coating.

NOTM. The 0.3-ml rn aganewa-zinc-coated panels were removed

from the test rack on January A, 1963

Cd 0.3 T No rust

TX SUght trace of rust in scatchns

a No rut

BX Slight trace of rust in scratches

Zn'Sn 0.3 T No rust

TX Traces of rust in scratches

B No rust

8X Slght trace of rust In scratches

Zn(Cr) 0. 3 T No rust; s-atr 'd pinpoint specks of white corrosion products along lowerTXJ edges of al nDs; a few scattered large white spot: elsewhere on all

panels

B No rust

BX A few traces of rust in scratches; white corrosion products in scratches

Mn 0.5 T 1A%, 75%, 9't, and 9'., respectively, of coatings on four panels haveflaked off and the exposed steel is rusting

TX 2,., S , 95%, and 95%, respectively, of coatings on four panels haveflaked off and the exposed steel is rusting

S 1A . 10%. 15%. and 4$, respectively, of four panels havc ru!n (mostlyalong edges) undercutting tha coating or where coating has already

flaked off

A T? 5&92 -UPP! i

I A 3:ui E; '

Coatina

Thicknes, SectionType mil 2 ) of Panel ( - ) Deicription of Corro.ion

Mn 9X 10, 2A. 25-, and 30%, respectiveiy1 of four panels have rust (rnost:yalong edges) undercutting the coating or where rust hat already flaked off

Zn 0.5 T No rust

TX Possible traces of rust in scratches; white corroslon products in scratches

B No rir

BX No ru!t; white corrosion products in scratches

Mn Zn 0,5 T 16%, 25%, 85%, and 1O00, respectively, of the coatings of four panelseither have flaked off or are being undercut by rust

TX 8,. 90%. 100, and 100%, rerpcctively, of the coatings of four panelseither have flaked off or are being undercut by rust

B M%. 96 , 10C%. and 100%, respectively, of the coatings of four panelseither have flaked off or are being undercut by rust

3X 100D of the coatings of all four panels either have flaked off or are beingundercut by rust

Cd 0. 8 T No rust; one 3uinpoint blister on each of two panels

TX Numerous uaces of rust in scratches; several tiny blisters on one panel

B No rust

BX ust In scratches for about 6Y% of lengths

Znln 0.5 T No rust

TX 'L.aces of rust in scratches for about 5 of lengths

B No rust

BX Slight traces of rust in scratches for about 126' of lengths

Zn(Cr) 0.5 T No rust

TX No rust

8 No rust

BX N1.1o rust, traces of whue corrosion prod'uc's in scratches

Fooir ts ppear on the following page.

Foctnotes or Table A-IC

(1) Test stated on May 11, 19,51.(2) The coating thickness values (0. 1. 0.3, and 0. 5 are In terms of mili. 1 mil 0. 001 inch.

%3) T - Unmarked ortion of panel w lch faces the sky,TX a Scratched portion of panel which fas the sky,B - Unmarked portion of panel which faces the ground.IX - Scratched portion of panel which faces the pround.

APPENDIX B. DETAILS OF PLATL4G AND CLEANINGPROCEDURES USED IN PREPARING PANELS FOR

TINE OUTDOOR-EXPOSURE TEST

Y-PTR 5642 nirm,>_ ,: Wi

APPENDIX B

Cleanin&

After rernovirg the pane'.s from the kerosene in which they were

stored, they were degreased ir hot trichioroethylene vapors. Followingdegreasing, they were cleaned cathodically in Anodex* for 3 minute. at 50asf and 200 F, brushed with hot Anodex solution, and then given an additionalminute of cathodic cleaning. After rinsing, they were dipped in 6N HCIrolution (70 F) for 30 seconds.

Zinc Plating

The zinc-coated panels and the zinc-plus-chrornate coated pane:swere plated in the following solution:

Zr(CN) Z 90 g/I

NaCN 37.5 g/l

NaOH 9j g/l

Temperature: 100 F

Current: 10 arnperes**

Anodes: Horse Head Special Zinc enclosed in cutton bags

Plating Tixhes:

0. I rr il 5 minutes

0.3 ril 17. 5 minutes

0.5 nil 28 mninutes

Cadmium Plating

Cadolyte Single Salt*** 120 g/l

Temperature: 90 F

Current: i0 amperesAnodes: Steel

Bath was continuously filtered.

,cCDermI'W Inc., Watrb.ly. Cm eclc .61n thii Lid all odef plamWlg, the cuiz, t is that per 4;:x6 ' p AI plus iu "ob~ r-, ' ,

A"Ths UdyR m Couporltiop . , Detroit 11.

52,tAy -R up]

Platitg Times:

0. 1 mil 4. 7 minutes

0.3 rcil 16 minutes

0. 5 uni1 25 minutes

IManganese Plating

MnSO 4 . HzO 40 g/l

(NH4 2 SO 4 135 g/1

Na 2SQ 3 7HZO 0. 5 g/l

Temperature: 100 F

Current: 30 amperes

Anodes: Carbon rods in porous Alundurn cups

Anolyte: (NH 4 ) 2 S0 4 135 g/l

pH: 7.5

Plating Times:

0.1 mil 3.5 minutes

0. 3 .il 11 minutes

0. 5 mil 19 minutes

Manganese-Zinc Duplex Plating

The steel panels were first plated with manganese, using themanganese bath which has becn described.

Zinc could not be plated directly on electrodeprsitea manganese fromthe ordinary zinc baths, so a special strike solution was devised. Itscomposition is as follows:

Zn(CN)- 5 to 7.5 g/i

NaCN 5 g/1

NaOH 5 g//1

Temperature: 80 F

Current: 30 amperes

Anodes: Stainlrss steel*rr'' 69wL

In using the strike solution, the manganese-coated panie mtust bernmersed wt'.th the current on,

After the strike coating, the balance uf the required zinc wasdeposited fro n a special acid-type zinc solution developed here for otherapplications. Since the manganese-zinc cjting did not perf. .'n well inexposure t&;ts, release of information on the zinc dtposition is not nteded.

Plating Times:

Time, minute.Tota Zinc Zinc

Th ickne ss _Mangane se Strike Plate

0. 1 il 1.75 1.0 1.5

0.3 nil 5,5 1.0 4.7

0. 5 rnil 9. 5 1.0 8.3

Zinc-Tin-Alloy PLatin&

The bath used for plating the zinc-tin alloy has been described byCuthbertson*. Its composition is as follows:

Tin (as sodium stannate)** 30 g/l

Zinc (as zinc cyanide) 2. 5 g/l

NaOH (free) 4 to 6 g/l

Total cyanide (as NaCN) 25 to 48 g/l

Free NaCN 17, 5 g/l

Temperature: 140 F

Anodes: Cast 20% zinc-80% tin alloy

Current: 15 amperes

0. Elecuochem. Soc. 94, 13 (14a).'High p'-'1Iy stairna is esunald ?.J u6al com rn c rna riALI was rot puae erl'h. Material of sluble

quA:4ty was obtatrxd bom . N tli Ar.d Tne:.n; Cc:p., Rahway. New jcity, and . T. Ba Ke C rnical Co .Ph..lipsburg, New Jersey.

k tP 5.2 1uppl L

The bath should be made up as folows,: fill the container to two-thirds of its volume with water, preferably distilled or deicnized softenei,

and heat to 140 F, D:sso've the NaCN, NaOH, and /r(CN) 2 , in that Qr.er.Finally, add the sodium stannaic. Analyze the so.ution a-d adjust thecomponents. It Is well to hold the batlh at temperature ior 2 to 3 clays hc,;re

plating.

Chromate Ccatn, on Z:nc Plate

The Cronak* proces was used to produce a chromate converuancoating on 12 of the zinc-coated pane .-,

H 50 4 (Cone,) 30 mn!l

Na 2 Cr2 C 7 200 g/1

Tenperature: 70 F

The zinc-coated panels were rnrnersed wet for 10 to 15 second.with slight agitationj removed, drained for 15 to 20 seconds, rinsed in

70 F water, then in 150 F water, an finally dried,

ABT: JEBG5:GF:CLF/rnas/pjs/lhMarch 18, 1953

0New Jc cy Zti Co, U. S. Paer.t 2,0.35,3 0.

.Lf'l ?V)C $upPI 55

Reproduced by

Urmed ervices Technical Inforn l g rDOCUMENT SERVICE CE :. ik1

mtoKNOT-of BUILDINGR, DAYTON, 2, oi;0

AD

P i 'El

-- - - 0---------- "q-I ..... . . ._