THIS REPORT HAS BEEN DELIMITED

AND CLEARED FOR PUBLIC RELEASE

UNDER DOD DIRECTIVE 5200.20 ANDNO RESTRICTIONS ARE IMPOSED UPON

ITS USE AND DISCLOSURE,

DISTRIBUTION STATEMENT A

APPROVED FOR PUBLIC RELEASEj

DISTRIBUTION UNLIMITED,

i~med ervices echnlcai nformation gency

NO0TICE: WHEN GOVERNMENT OR OTHER DRAWING-5, SPECIFICAT-IONS" OR OTHER DATAARE YD FOR ANY PURPOSE OTHER THAN IN- -ONNECTION WITH A DEFINITELY RELATEDGOVERNMENT PROCUREMENT OPERATION., THE U. S. GOVERNMENT THEREBY INCURSNO0 RESPONSIBILIT, NOR ANY OBLIGATION WHATSOEVER; AND THE FACT THAT THEGOVERNMENT MAY RAVE FORMULATED, FURNISHED, OR IN ANY WAY SUPPLIED THESAM DRAWINGS, SPECIFICATIONS, Olt OTHER DATA IS NOT TO BE REGARDED BYIMPLICATION OR OTHERWISE AS IN ANY MANNER LICENSING THE HOLDER OR ANY OTHERPERSON OR CORPORATION, OR CONVEYING ANY RIGHTS OR PERMISSION TO MANUFACTURE,USEC OR BE~LL AN~Y PATENTED INVENTION TRAT MAY IN ANY WAY BE RELATED THERETO.__

Reproduced 6DOCUMENT SERVI#E CENTER

KNOTT BUILDING, DAYTON, 2, OHIO

N

-

FE

I S iEC

Twhnkil D'

to fhecs ()fck o;. Naval Reumarch

'AIA11OK OFM METALS AT.CONTROLJLED'POTENTI~rAL

by

~Pb ~*~I 4.b~uie'i~, A.'Pirrv. and,

Report No. 12

't'-.chmical Report

tothe

I. Office of Naval Research

Report No 112.

Project fl-05l-258

ANODIC CXIDATIL75 OF METALS AT CO"ROLLED POTETIAL -

EIfPRIITAL METHODS AID CASE (#F MRON

by

•.•-lhaw, .- Tnc4 - .T•n1la0n, .Tnhn A. Parr•,.• a.d George L. Stiehi- ----- a rg eS ih

March 1954o hearDepartment of Chemitry

Louieana S . UniversityI Baton Rouge 3D Louisiara

II i

-. a - - w -

IDISUIBUTION LIST FOR TBECHIICL RBDRTS

No. of Copies Addressee

1 Commanding OfficerOffice of kaval oesearch Branch Office150 Causeway StreetBoston, Massachusetts

2 Commanding OfficerOffice of Naval Research Branch OfficeThe John Crerar Library Building86 X. PAndolph StreetTenth FloorChicago 1, Illinois

1 Commanding OfficerOffice of Naval Research Branch Office346 BroadwayNow lork 13, New York

1 Commanding OfficerOffice of Naval Research Branch Office1000 Geary StreetSan Francisco 9. California

1 Commanding OfficerOffice of Naval Researckr Branch Offies1030 N. Green StreetPasadena 1, California

2 Officer-in-ChargeOffice of Naval Research Branch OfficeNavy *amber 100Fleet Poest OfficeNow Yorks Now lork

6 Director, Naval Research LaboratoryWashington 25, D. C.Attention: Technical Information Officer

2 Chic.P of Naval ResearchOffice of Naval ResearchWashington 25, D. C.

Attantion: Chemistry Branch

S5 AST], Document Service CenterKnott Building,zyton 2, Ohio

1 Dr. Ralph G. H. Siu, Research DirectorGeneral Laboratories, QV Depct2800 S. 2O0h StreetPhiladelphia 45, Poonsyl.vania

L i --. -.",' -I,,,,,_

DISTRIJTION LIST OR ECHNICAL RE2TS Page 71

No). of _o2ies Adresses

1. Dr. Warriu Stubblebine, Research DirectorChemical and Plastics Section, RDB-KPDQuartermaster General's OfficeWashington 25, D. C.

1 Dr. A. Stuart Hunter, Tech. DirectorResea'ch and DeTelopment Pranch )EPDQuartermaster General's OfficeWashington 25, D. C.

1. Office of Technical ServicesDepartment of ComeoceWashington 25: D. C.

1 Dr. A. WeisalerDepartment of the ArmyOffice of the Chief of OrdnanceWashington 25, D. C.

Attn: O1 DTB-PS

Research and Development GroupLogistics Division, General StaffDepartment of the ArmyWashingoon 25, D. C.Attn: Dr. W. T. Read, Scientific Adviser

2 Director, Naval Research LaboratoryWashington 25, D. C.

Attention: Chemistry Divisiou

Navy Department

Washington 25, D. C.Attention: Code 340

2 Chief of the Bureau of AeronauticsNavy DepartmentWashington 25, D. C.Attention: Code TD-4

* 2 Chief of the Bureau of OrdnanceNavy DepartmentWasshingto-n 2%, D. C.Attention: Code Rexd

I. Office of the Secretary of DefensePentagon Room 3D1041Waahington 25, D. C.Attention: Library Branch (R+D)

1 Dr. A. G. HcrneyOffice Scientific ResearohR and D Consead USAFBox 1395Baltimore, Karyland

1' ,4 - -- - --. ~ - .

D18T1-3•UON LIST FOR TSCHNCAL RSPFRTS Page III

No of enin Afddressee

Dr. . k. Zahl, Tech. DireoctorSignal Corps Ingineering LaboratoriesFort Konsoutha, Now Jersey

U. S. Naval Radiological Defense 1!aborstarySan Ferancisco 24, CaliforniaAttention: Technial Library

! Naval Ordnance Test Statiom iayoker'o)China Lake, CaliforniaAttentionc Head, Chemistry Division

1 ~Office of Ordnance RLesearch

2127 Myrtle DriveDurham, North Carolina

1 Technical CounandCh.emical Corps

Chemical Center, Maryland

U. S. Atomic Anergy CommissionResearch DivisionWashington 25; D. C.

1 U. So Atomic Xnergy CommissionChemistry DivisionBrookhaven National LaBoratoryUpton, New York

1 U. So Atomic Snergy CommissionLibrary Branch, Tech. Information OREP. O BoxAOak Ridge, Tennessee

Dr. Ao Bý RemickDepartment of ChemistryWayne UniversityDetroit I, Michigan

1 Dri Frank RavorIaDepartment of h 3emistryWestern Reserve UniversityClevelmid 6, Ohio

S1 Dr. David C. GrahamaDepartment of ChemistryAmherst College

amherstp aassachuoetts

.' I7 .- , . . .. . • " -

DISTRIBUTION LIST FOR TECHNICAL REPORTS Page IV

No. of vopies Addressee

1. Dr. P. J. oivingDepartmen6 of ChemistryUniversity of MichiganAnn Arbor, Michigan

Dr. N. H. FurmanDepartment of Chemistry

Princeton UniversityPrinceton, New Jersey

1 Dr. I M. KoltboffDepartment of ChemistryUniversity of MinnesotaMinneapolis 14, Minnesota

1 Dr. H. A. LaitinenDepartment of ChesistryUniversity of IllinoisUrbana, Illinois

1 Dr. louis MeitesDepartment of ChemistryYale UniversityNev a.yen, Connecticut

1 Dr. 0. H. MullerDepartment of PhysiologySyracuse UniversitySyracuse, New York

1 Dr. Pierre Van RysselbergheDepartment of ChemistryUniversity of OregonEgene, Oregon

1 Dr. Stanley Waw--onekDepartment of ChemistryState University of IowaIowa City, Iowa

S1 Dr. J. J. Lingane

Department of ChemistryHarvard University ICambridge 38, Massachusetts

1 Dr. Leon 0. MorganDep-atment of ChemistryUniversity of TexasAustin, Texas

'ii

-- - ....- -. in .-

o__ __ _- •" -- * -EITI

ABSTRACT

Rates of anodic oxidation of iron are measured at c.-Atant potential

in dilut perchlorie acid, acetate and borate buffers, and sodium hbyroxide

solutions. A treatwment for the, kinetics of anodie oxi4dation of metals at

constant potential is presented for processee in which a soluble species is

formed and processes yielding insoluble products are also interpreted.

axperimental methods are described.

The kinetics of the anodic oxidation of metals in involved especially

when insolulia products are formed. These processes do not lend themselves to

mathematical analysis, and an essentially experimental approach must be

follmed. Such an approach is adopted here, an4 results are reported for iron.

~PDEllI4AL

POTEMTIOSTAT,

The potentiostat was a modified version of the instrument described

Iby ]Taphere and Rogers . So many potentiostats have been described in

(1) R. W. Lamphere and L. B. Rogers, nme. Chek., 2, 4,63 (1190).

the literature that there is no need for a detailed description herej

(2) J. J. Lingane, "Blectroanalytical Chedistry", interscience, New Yorks

N. Y-9 1953s pp. 202-244.

2

The diagram of Fig. 1 in sufficiently explicit. The potentiostat had an

output voltage of approximately 160 voltal the imxim current was 2.4 amperes.

E9tOLTTIC CMLL.

The cell was designed to achieve uniform currenL density over the whole

area of the electrode, The arrangement is quite apparent from Figs 2. The

electrode of the metal being studied was a disk placed at tbe bottom of the

Lucite cell. A disk was pressed against the electrode by means of three screws.

Connection (ki) was obtained by pressing a metallic point on one face of the

electrode. A thin (0.2 nu or so) plastic washer inserted between the metal and

the wall of the cell ensured tightnesso The diameter of the working circmler

area of the electrode was 2.5 -,m, The auxiliary electrode was a platinum disk

(E2) whose diameter was only slightly smaller than that of the effective circular

area of the electrode. Electrode E2 was inserted in a separate compartment

whose bottom was a fritted glass disk. An agar-agar plug (not represented in

Fig* 2) covered this disk and decreased the rate of diffusion of electrolyte

f~ro- one compartment to the other. The assembly of electrode E2 was held by a

rubber ring R,

Unifrm stirring was obtained by circulating the electrolyte at the rate

of 200 il. min. by means of a pump P (F.ig. 2 right) made of uncorrodable3

material * Oxygeu was removed by bubbling nitrogen in cell C2 . This cell was

maintained at 35 4 0.,1. The total volume -, solution was 420 m.i

(3) Vanton Corporation: New !ork, N. Yo

The potential of the working electrode (BI) ims measured against a

saturated calomel electrode which was connected to the main crll by a kalt

bridge (filled -with the same electrolyte as the main cell). The usual Haber-

Lo-ain tip for reference electrodes was not utilized, because the resulting

shielding of the working electrode causes a local decrease in current density

in the area of the electrode where tho potential is being measured@ This!4

effect -ms studied extensively by Piontefli and coworkers

(4i) -R Alettif U. Bertocci, 0. Bianchis C. Guercis It. Piontelli, G. Polis

and G. Serravalle, int. Corn. Electrochem. Therm. Kin., Proceedings (f 3rd

Maeen 2 mfred6I Milan9 1953, pp. 30-42.

"who recommended a cylindric tip pressed against the electrode. This procedure

was used in this studye The end of the tip was made of a Lucite piece which

was cemented to an ordinary glass tube. There was virtuafly no electrolysis

at ths point of contact, but the current density was otherwise uniform.

Connection between electrode. Ex and E2 wIs provided by a small canal 0

of tho -.•rking electrode.

tRFATMENT OF WORKING ELOCTR!ME.

The worldng electrode was made of *pure" iron o The metal was firet

eee 1

V 4

(5) Obtaiued -,a- D. H. Maejay, 198 Broadway, New York, N. Y. The emission

spectrum Indicated that the metal was at least as pure as "spectroscopically

pure ironm.

cleaned with acetone and water. It was then iwersed for 15-30 sec. in 6 N

hydrochloric acid and rinsed with distilled water. The electrode was dried

and finally weighed. At the end of a!eotrelysis the electrode was rapidly

removed from the cell by unfastening the three screws at the bettom of the

cell,. tbe electrolysis circuit remlning closed. The electrode wax washed

ini 6 X. '41C1 for 30 Lqqc. t.0 ramWU +.!-A vwtvliy't! +8- a-~no 1 'idaion; this was

follred by waahig in water and drying under vacuum. Losses of weight an

1w as 092- 0.3 mg. cm. hr. could be m4asured.

DESCRIPTION AND DISCUSSIOW OF RESULTS

OXIDATION IN 0.1 N PERCHLORIC ACID.

Rates of oxidation of iron -In 0.1 H perc'---,rie acid are plotted

against potential in Fig. 3. The rate of oxidation decreases as the potential

become more cathodic. These results can be explained as follows.

The current for the anodic oxidation of a matal with the formation of

a soluble species Is

-. = - "- " -) - --

where the k's are rate constants for the anodic and cathodic processes res-

pectivelys femn the activity coefficient of ion fen, (C~n)x . the con-

centration of ion 16n at the electrode surfaces and A the area of the electrode.

Equation (1) i written on the assusption that the activity of the metallic

electrode in unity and that the anodic process is controlled by a single rate

determining step. Furthermore it will be assw d that A is constant during

electrolysis, i.e. that there is no preferential attack.

Tha current can also be written in terms of the flux of ion MHn at

the electrode g'iriace. Thus

(2)rIiwhere Mi. is the diffusion coefficl.ent of ion Mn The calculation of the

derivative !n (2) is exceedingly difficult in the case of a stirred solution sven

under the case of luminar floui6 and an approximate solution based on

(6) See for exaala B. Levich; Acta Pn tcochim. U. R S. S., 110, 257 (5042).

Nernst's concept of diffusion layer will be given here. Thus

0 (3)

where C•~n is the concentration of ion Men in the bulk of solution, and

th: tcness of the diffusion layer. Since J is very small in coLparison

vlth the dimansions of th -vessels Com*n at time t is

0 r L 1% 4 )rL41 ,o '-, V.

where V i8 -the volume of solution.

I,• By combining equations (1) to (4) one obtains a differential jquation

whose solution in

This solution is written £Zw C* - 0 at t 0 0. The average rate of

oxidation W for a duration cis obtained from (5) by application of Faraday's

law. If ona assumes that is independent of time, one obtaine

where V is the atomic weight of metal M and AJ i defined by

A -1)

IV

The function of the dimensionleas gAoup I" C inL equation (7) is equal

to unity for )A =1; this function, which decreases as A t Increasesp

apprnaheA ACero for sufficiently large values of X * Two important con-

clusiams -c-n be drawn from equation (7).

(1) When the potential is sufficiently anodic, Kc and cornsequently A

- is very sn1ll equation (7) can then be written under The form W - WICka. Uince

K& is an exponentLr-l function of the electrode potential, a plot of log W versus

potential should yield a straight line. This in essentially the case in Fig 2

n-. p vn g Q *- +and -- A - 2 uwnl + -tA +caa thodnicjt pr-s Me morer

anodic the influence of the cathodic process becomes more prcnocuiced, iLe.

7

kf increases. The rate W is thus smaller than one would expect from pro-

portionality between W and WM ka, and there is a departure from linearity in

the diagram of Fig. 3. iu Irecise calculation of the dependence of ka on

* potentirAL can be deduced from Fig. 2 because of the following cause of errors:

the value of the potential of the working electrode with the cell of Fig. I

includes the ohmic drop between the end of canal 0 and the surface of the

electrode. It would be difficult to evaluate this error, although it could be

minimized by decreasing the diameter of canal 0. However, thir solution la ,ot

practical because the canal 0 is then too easily plugged when a precipitate (see

below) is formed at the surface of the working electrode.

(2) The averave value of W at a given potenti!9! depands on the d uration

of electrolyie. Thus, A is c•rnsant at a Piven potential: hi A. i

increases with the duration of electrolysis, and the function of C in (6)decreases. This effect is apparent from the data of Fig. 4. There is some

discrepancy between theory (curve) and experiment for the longer durations of

electrolysis. This 7ay rescult •."a ;.-ar-iations in the ealectar.; ar-a while it

was asealw4 in the above analysis that this area is constant, In quoting rates

of anodic oxidation at constant potential it is therefcre essential to indicate

the duration of electrolysis. It mioit seem preferable to report values of ka

and A , but these data could be used only for cases in which the above treat-

ment is applicablel this would not be so when films are formed (see below) or

when the anodic process cannot be accounted for by equati.on (1).

OXI-DATION IN ACETATE BUFMER.

Results for acetate buffers of different pHts are suamaried in Fig. 5.

B

Thease results can be conveniently discussed by considering the potential-pH

diagram of iron as established by Yourbaix , It should be eshaoized that

(8) M. PNurbaiz, uThermodynamuique des Solutions Aqueuses Dilues" -

Representation Graphique du R&le diu pH at du Potentiel" (Thesis) Delf, Meinem,

1945- English translation by J. M. Agar, Arnold, London, 1949.

9this diagram suppllesa on-W1 Anulibriuz ,nnditicrz , while -perimental results

obviously depend on the kinetics of the reactions being considered.

(9) The construction of potential-pH diagrams is based on a systematic

appIction of the Nernst squation.

The use of the potential-pH diagram i. nevertheless fruitful in the elucidation

of experimental results. Inspection of the diagram fer iron shows that between

pH 4 and 6 the oxidation of iron proceeds with tha formtion of Fefions. The

equilibrium between Fe4" ions and Fe(OH)3

FM(OH)3 + 3 H* + e Fe** + 3 HaO

must also be considered. The corresponding equilibrium potential at 25' is 10

II

• • 'U ' .,, ..-- .. .. ,...•,.•u, ---., .-

I'

(10) P. Delshay, N. Pourbalx, P. Van Rysselberge, J. Chem. Educations z,

683 (195o0.tI

JI

3 i-.O4 - 0.177 pH - o.0591 log aFJ+ (8)

The standard potential in equation (8) varies somewhat according to the

standard free energy one adopts for Fe(O) 2S i.e. according to the soluoility

II

was usted here;

One calculates from (8) that E -0.327 volt for pH 6- 4.5 and a - 1Fe*

There in therefors a relatively large equilibriu concentration of ferrous ion

up to potentials as anodic, say, 0.3 - 0,4 volt, At mere anodic potentials

ferric hydroxide is formed and the curve for pH 4.05 in Fig. 5 levels off Pro-t

gresuively. At pH 699 Pnd 5o37 the behavior of iron isu somewht different from

that at p11 l0l; there is rapid drop in the rate of oxidation between 0.1 and

0.3 volt. This probably results from the formation of ferric hydroxide at the

electrode surface. At 0.3 volts, one has apprvoliately a nFe d* pobut this

value is quite uncertain because the c Fr for Fe(OH)a used in the calculation

corresponds to a solubility product of 6 x ifoat for oferric hydroxide Values

as low as 10--42 have been reported for this datum8 and the resulting value of

a e:ý* at pH 5.37 and 0.3 would be much lower than i0'" (lO- cr so), In

-- - - - - -, , - - ---- - ---

10

conclusion, thG formation of Fe(OH),1 probably causes the rapid drp in the rate

of oxidation at pH 4.99 and 5.37o This effect is far less :prononced at 19

i.O5.

The more or less linear segments between -047 and 0 volt in Fig. 4

should be shifted to more cathodic potentials by 0,177 volt per increase of one

pH unit (see equation (8)). ActualLy there is a shift in the right dfrectionp

but the relative lack of precision of experinental data does not aflow one to

reach any fmthar confldl-aion,

Because of the complexity -f anodic processes which yield insoluble

substances no attempt was wkde to develop a quantitative treatment,

6XtnATio IN BORATE BMPER.

Data for the oxidation in borste buffer of pH 9.0 are iisted i-n Ta.b!A I.

There was virtually no oxidation at potuntials less anodic than 1.0 volt.

Pitting of the electrode was observecl at more anodic potentials. This ob-

e•r-ation ca. be explained as followa. Oxygen is evolved at these very anodic

potentials. Since the surface of the e6ec.trnde is not t'niformu, oxygen evolution

occurs in theie areas of the electrode surface where the overvoltage for this

reaction is lower than in the other parts of the electrode. The evolution of

oxygen is accompanied by the liberation of hydrogen ion and by a local decrease

in the pH of the solution. The solubility of ferric hydroxide is increased

accordingly in these areas where the oxygen is evolved, and there is a high rate

or localized exidationa

fXIDATIOW IN SODIUM 9IOIIDE.

RatPes of anodic oxidation in 0.1 N. sodium hydoxide were too low to be

|I

,€'- - --. . --- so- ~ .-

'32

measurable with preciston. No experiment was iade above pH 13 because of the

danger of attack of the eirculating putu (rubber part).

ACKNdWLEDGWNT,

The support of this investigation by the Office of Naval Research is

gladly acknowledged. The authors are indebted to Messrs. W. Payne and T.

SLeinhardt for their help in the construction of the potentiostat.

IrI

iI

I

1 12

I TABIE III ANODIC UDATION OF ]RON 3I BOATE BUIPSR OF pH 9.0

Potential Rate of Oridationf Volts v'o N.H.E. ugbcm."ihr.

-0.757 0C

+i.098 .1

*1.25o 0.17

#1.549 340

÷1.o76 7 ,19

(1) Smaller thz, eiperimental errors

_ _ _--- -- !a

13

LIST OF FIGURES

i. Schematic diagram of potentiostat. T1 , isolation transformer, 350watts| T2, isolation transform---, 150 watts; T3 and T4.9 OPowerstat" transforers

3 amp. output; Rq., selenium bridge rectifier, Uj4 volta, 0.9 amp.; R2, selenium

bridge rectifier, 168 volts, 2o4 amp.; C% to C4 , electrolytic condenser, 125

iicrofarads, 350 volts; C6 to Co, electrolytic condenser, 250 microfaradm, 350

volhy at P. to 9 watts; r 4. 40 ohm,, M.0 watts; re to r.. 10 ohms# 200

watts& ra to r.¢. xhunta far meter A to give readings of 005., 0.5s, and 5 amp.,

respectively; r 1 1 , 20 1., 20 watte; r~s, 10 K., 20 watts; rs, 5 K., 20 watts;

r14 9 2 4 20 watts; r 1 s, 1K., 20 watts; rj 6 , 500 ohm., 50 wattsj r,1 7 , 200 ohme,

100 wattl rne, 100 ohms, 1-0 watts r%9, 50 ohm, 100 watts; r 9 o, 25 ohms,

200 watts; pl, potentiometer, 16 ol-a, 100 watts; P2, 'Helipotm potentieter,

1000 ohms, 10 turns, O.I; p•, potentiometer, 2000 ohms, wire-wound; P4,

potentiometer, 200 ohms, wire-woundi So, D.P.D.T. switch; 57 and Se limit

witches; 5 push-button switch; b1 , 6 volt storage battenly; Br, "Brown"

amplifier model#356358-l; H, "Brown" reversible motor, model 416750-35

V%,1 0-150 volt, voltmeterl V2, 0-3 vlt, voltmeter; A, 0-4 milliapo, ameter.

tg. 2. Electrolytic ce2J and accijsories,

P, ,Variations of the rate of anodic oxidation of iron (in 0.1 N perchloric

acid with potential. Duration of electrolysis, 2 hours,

Fig. 4 Variations of the rate of anodic oxidation of iron with duration of

electrolysis. Points are experimntal values; curve calculated for An W-%. - .'

Fig. 5o Variations of the rate of oxidation of iron in acetate buffers Ath

potential. Duration of electrolysis, 2 hours.

- - - - - -- a- a - ---

T

R, R

cii

VI S4

(~~~ I s .~

-~ B4 %v 3jy V

1 rrao ig.b,

Fig. P3 Pse S4a. - ~ ~Al- -----

C\J

I ccc

ii'i

1005 (C

10-

Et 5.0 -

cr 0.5

0.1140.1 0.0 -01 -0.2 -0.3

POTENTIAL (VOLTS VS. N.H.E.)

FIG. 3

50I

I"i

m I

1.0

[F I

r€L

0 20 40 60 80 100

TIME (MINUTES)

FIG. 4

- ., ..

00 I i I

501- pH 4.05

"Ii i l' . .0 p H 4 .99

S0.5o .) I iL,\ \ o*

0

+0.7 +05 +0o +0.1 -0.1 -0.3 -0.5 -07POTENTIAL (VOLTS VS. RN{E)

FIG. 5

n I' Tlrmed Gervices technical nformation fenc!

NOTICE: WHEN GOVERNMENT OR OTHER DRAWINGS, SPECIFICATIONS OR OTHER DATAM-U•RD FOR ANY PURPOSE OTHER THAN IN CONNECTION WITH A DEFINITELY RELATED

GOVERNMENT PROCUREMENT OPERATION, THE U. S. GOVERNMEN . THEREBY INCURSNO RESPONSIBILITY, NOR ANY OBLIGATION WHATSOEVER; AND THE FACT TIRAT THEGOVERNMENT MAY HAVE FORMULATED, FURNISHED, OR IN ANY WAY SUPPLIED THESAID DRAWINGS, SPECIFICATIONS, OR OTHER DATA IS NOT TO BE REGARDED BYIMPLICATION OR OTHERWISE AS IN ANY MANNER LICENSING THE HOLDER OR ANY OTHERPERSON OR CORPORATION, OR CONVEYING ANY RIGHTS OR PERMISSION TO bANUFACTURE,USE OR SELL A1NY PATENTED IaVENTION THAT MAY IN ANY WAY BE RELATED THERETO. I

Reproduced 6yDl CUJMENT SERVICE CENTER

KNOTT BIILDInG, DAYTON, 2, OHIO

N'I%..ASSIFIED