Ε. K. Co. imbibition paper (gelatin), 0.1 Μ potassium ferrocyanide; developed, H202
1.5 volts, 15 millicoulombs/sq. cm., aluminum cathode
ELECTROGRAPHY AND ELECTRO-SPOT TESTING 205 Ibarz point out t h a t detail is necessarily limited to gross structures in Glazunov's prints because of t h e coarse background structure of filter paper a n d t h e excessive diffusion attending its use. A m m e r m a n n favored a lightly sized, fine grained drawing paper while Jimeno, Bernal, a n d Ibarz preferred a sized visiting card material. T h e latter workers also m a k e mention of a gelatin coated paper, t h e n supplied by t h e G e r m a n firm of Bayer.
Glazunov does not mention t h e strength of the electrolyte b u t states t h a t an " a g e d " ferrocyanide solution, or one t o which a few drops of peroxide h a v e been added gives more satisfactory results. T h u s he indicates a need for ferricyanide ions in addition to t h e ferrocyanide.
A m m e r m a n n (1), commenting on Glazunov's work, noted t h a t prints m a d e with ferrocyanide are initially weak and develop their color largely after printing, t h r o u g h oxidation in t h e wash water. H e states t h a t equally good prints are obtainable with ferricyanide alone. On t h e other hand, H r u s k a (34), a n d Jimeno, Bernal, a n d Ibarz (35) use potas-sium ferrocyanide containing no ferricyanide. T h e latter workers maintain t h a t this is superior t o ferricyanide-containing electrolytes because t h e more electropositive areas are likely t o yield ferric ions which would not print with ferricyanide. I t should be noted, however, t h a t they used higher voltages t h a n Glazunov (2V-4V), a condition favorable to the production of ferric ions.
These conflicting views indicate t h e need for further s t u d y of the iron printing reactions. T h e writers h a v e observed t h a t prints m a d e with ferrocyanide electrolyte at low current densities are likely t o be quite faint b u t are oxidizable t o full intensity with peroxide, dichro-m a t e , or dichro-more slowly, b y exposure t o air. Idichro-mdichro-mersion in ferricyanide also brings u p the blue color. This would indicate t h a t t h e bulk of the iron dissolves as the ferrous ion a n d is probably fixed as t h e nearly colorless ferrous ferrocyanide, oxidizable to blue ferric ferrocyanide or reactive to ferricyanide to form blue ferrous ferricyanide. On this basis, one might be inclined t o use ferricyanide in t h e electrolyte to provide immediate conversion of t h e ferrous ions to t h e blue ferrous ferricyanide, b u t when this was tried, it was found t h a t t h e p a t t e r n s were not always reproducible and differentiation between unlike areas was less pronounced t h a n with ferrocyanide alone. This was traced to t h e fact t h a t ferricyanide reacts fairly rapidly with iron without t h e aid of anodic solution, oxidizing it t o ferrous ions which precipitate as ferrous ferricyanide. P a r t of t h e ferricyanide is reduced t o ferrocyanide in this reaction. T h u s it is possible t o m a k e simple contact prints with ferricyanide, capable of yielding considerable structural detail in which differentiation follows from t h e varying oxidizibility of t h e surface
206 Η. W. HERMANCE AND Η. V. WADLOW
components. For electrographic purposes, however, t h e oxidizing action of ferricyanide is undesirable in t h a t it superposes a chemical a t t a c k on t h e anodic process, t h u s interfering with t h e purely electrical control of solution. T h e best results were obtained using pure ferro
cyanide at low current densities a n d developing t h e resulting pale print in very weak peroxide.
For macrography, it is generally agreed t h a t it is unnecessary t o give the specimen a high polish. Finishing on #400 aloxite paper followed by crocus cloth or t h e equivalent is usually sufficient. I t should be pointed out t h a t polishing tends t o leave a t h i n amorphous layer on t h e surface of a specimen which usually masks its structure a n d m u s t be removed by etching before macrographic examination can be m a d e . In electrography, such layers would constitute an equipotential surface and, if sufficiently thick, might easily mask t h e potential differences of t h e underlying grains, t h u s yielding a structureless, solid color print.
I t is well, therefore, t o compare prints m a d e on t h e polished surface with those made on the same surface after etching. T h e a m o u n t of etching required t o give t h e best electrograph of its macrostructure must be determined experimentally for each material. Some steels electro-graph best after an extremely light etch, t h e prints becoming contrasty and losing detail rapidly as t h e etching becomes deeper. Others yield structure p a t t e r n s only after intense t r e a t m e n t with etchants. Etching agents which t e n d to leave a patina should be avoided a n d in a n y case, the surface of the specimen should be thoroughly swabbed with cotton in running water before printing. Iodine-potassium iodide, 5 % nitric acid in alcoholic solution, or electrolytic etching are suggested.
Summary of Conditions for Electrographs of Ferrous Macrostructure Preparation of specimen: Finish on #400 Aloxite paper, followed with crocus cloth.
Etch, if desired, with iodine-potassium iodide or 5% alcoholic nitric acid.
Swab surface with cotton in running water and print immediately.
Printing medium: Eastman Kodak imbibition paper, immersed at least 10 minutes in the electrolyte.
Electrolyte: 0.1 Μ Potassium ferrocyanide.
Cathode: Aluminum.
Pressure: 25-50 lb./sq. in.
Current density: 0.5-1.0 ma./cm.2)™ 1 e .... . , , _ ,Λ Λ fTo pass 10 to 15 millicoulombs/cm.2
Time: 10-30 seconds J *
Development: Immerse print for 1 minute in 1 ml. superoxol/100 ml. water wash.
General: Place paper without blotting on the cathode surface so that a pool of the electrolyte remains on its surface. Wet the specimen with the electrolyte and lower onto the paper by pivoting about one edge so as to squeeze out air bubbles.
Apply pressure immediately and take up the excess electrolyte around the specimen with blotting paper. Print, develop, and wash in running water for 10 minutes.
ELECTROGRAPHY AND ELECTRO-SPOT TESTING 207 4.8.2. Structure of Nonferrous Metals. Ferrous metals h a v e received b y far t h e greatest a t t e n t i o n in t h e applications of electrography t o structure printing. However, Glazunov states he h a s prepared macro
graphs of zinc structure with potassium ferrocyanide a n d crystal violet.
Precipitation of t h e white zinc ferrocyanide in t h e unsized printing paper is accompanied b y fixation of t h e dye in t h a t area so t h a t on washing a print is obtained (26). Glazunov also records silver structure b y printing with dichromate, a n d nickel with dimethylglyoxime. H e prepared copper macrographs with ferrocyanide, b u t A m m e r m a n n (1) claims difficulty with t h e m e t h o d when used t o print copper oxide inclusions.
T h e writers have used ferrocyanide prints t o record t h e distribution of copper and iron in aluminum base die castings. T h e specimen is first given an etching for several minutes in a 1 0 % caustic b a t h a n d t h e n scrubbed in running water t o remove t h e dark, spongey metal which collects on its surface. I t is printed on gelatin paper soaked in 1 p a r t of 0.5 Μ sodium n i t r a t e t o three p a r t s of 0.1 Μ potassium ferrocyanide.
T h e print is intensified b y development in 0.1 Μ potassium ferricyanide a n d finally washed. Printing is done at 4 volts (aluminum cathode), 15-30 seconds.
Preliminary work has shown some promise in the printing of macro-structures in brasses and other copper alloys with antimony, cadmium, or zinc sulfide impregnated gelatin papers. 0.5 Μ A m m o n i u m acetate is used with 1.5-3.0 volts, against a carbon cathode.
4.S.8. Surfaces. Quite aside from heterogeneities of the mass struc
ture, t h e surfaces of metal specimens m a y exhibit composition irregulari
ties, acquired as a result of initial processing or exposure t o operational conditions. Often t h e surface p u r i t y of a material will determine its suitability for a given use. T h u s an aluminum foil t o be used in an electrical condenser is affected adversely b y traces of iron or other metals rolled into its surface during manufacture. T h e corrodibility of m a n y metal surfaces is markedly increased b y accidental inclusion of particles of more negative metals. On t h e other hand, contamination need not necessarily occur during manufacture. Mechanical failure, particularly where sliding surfaces are concerned, m a y result from the introduction of traces of foreign metals as dusts or filings which cause seizure a n d "freez
i n g . " Electrographic m e t h o d s are especially useful in the identification a n d mapping of such contaminants, where t h e extreme superficiality of a transferred film or particle precludes application of mechanical sampling.
For corrosion protection or t o provide desirable mechanical properties, surfaces m a y receive t h i n coatings of other metals, applied b y electro
plating, hot-dipping, or b y other methods. Organic protective coatings likewise m a y be applied. T h e initial continuity of such coatings as well
208 Η. W. HERMANCE AND Η. V. WADLOW
as their stability toward corrosive agents, their resistance to wear, etc., can often be evaluated quickly a n d conveniently b y electrographic methods.
A complete discussion of the application of electrography to surface studies here is impossible, b u t it is hoped t h a t t h e inclusion of several selected examples will suffice to illustrate the general approach.
Inclusions: Iron and copper inclusions in aluminum, magnesium, zinc, or nickel surfaces are registered readily b y printing with a 0.5 Μ ammonium acetate electrolyte on gelatin paper and developing in ferro
cyanide. A current density should be employed which is sufficient to polarize t h e specimen surface above the couple potential of the more negative inclusion (see p . 178, also Hunter, Churchill, and Mears, 33).
W i t h aluminum or zinc, a faint blue background m a y be obtained due to alloyed iron. Against this, particulate iron or copper will appear as the deep blue or red ferrocyanides. A nickel surface will give t h e apple green color of its ferrocyanide b u t iron and copper will contrast suffi
ciently against this unless t h e y are present in only t h e faintest traces.
When the q u a n t i t y of copper is extremely minute, a more sensitive developing reagent is diethyldithiocarbamate, used in 1 % solution of the sodium or ammonium salt. For copper on an iron surface, 1 % diethyl
dithiocarbamate in 0.5 Μ sodium potassium t a r t r a t e , m a d e ammoniacal with 2 % concentrated a m m o n i u m hydroxide is used.
Iron on a copper surface is not registered by direct printing because the red copper ferrocyanide would mask any faint blue iron patterns.
I n this case, a print is m a d e on unsized paper with a basic nonreactive electrolyte, (Table V,C, electrolyte " A " ) , t h e copper being washed out with ammonia before developing for iron with ferrocyanide. (See Tables V,A, Test N o . 2, and V,B, Proc. N o . 2.) For a further discussion of inclusions in electrography, see H u n t e r , Churchill, and Mears, (33).
Frictional transfer: Through frictional contact or impact, metal m a y be transferred from one surface t o another. Development of t h e resulting p a t t e r n m a y reveal t h e n a t u r e of t h e action involved. Such information often has diagnostic, a n d possibly criminological value.
T h u s it might be important to learn whether or not a certain axe was used t o sever a cable. An electrographic print m a d e on t h e steel blade reproduced t h e p a t t e r n of the fractionally transferred copper, providing strong circumstantial evidence. (See Fig. 14.)
Mechanical transfer also m a y occur between sliding surfaces. T h e study of such transfer m a y yield valuable information concerning wear and t h e t y p e of lubrication required. T h e authors have used prints t o demonstrate transfer from a wiping silver contact to a fixed brass surface when b o t h are protected from atmospheric tarnish b y a lubricant.
ELECTROGRAPHY AND ELECTRO-SPOT TESTING 209
FIG. 1 4 . Hatchet blade used to cut copper cable, with electrograph of trans
ferred copper.
Printing: 0 . 5 Μ sodium potassium tartrate with 1 % sodium diethyldithiocar-bamate
2 0 seconds, 3 volts, aluminum cathode
paper (electrolyte not stated) a n d developed t h e print for copper with dithioxamide.
The porosity and discontinuities of protective coatings: T h e detection a n d mapping of discontinuities in metallic or organic coatings can be accomplished electrographically whenever the base metal is electrolyti-cally soluble a n d capable of entering into suitable color reactions. T h e
Prints were m a d e with 0.5 Μ N a N 03, t h e n developed for silver with photographic developer. Bowden (4), in his studies of sliding friction has used electrography t o record transfer from a copper slider t o a steel plate under dry and lubricated operation. He employed a gelatin coated
210 Η. W. HERMANCE AND Η. V. WADLOW
precise control of t h e solution r a t e a n d t h e reduction of diffusion gives t h e electrographic method definite superiority over chemical printing in this field.
W h e n the coating is an organic finish or a noble metal, selection of t h e electrolyte and color development is fairly simple, for then no problem is presented by solution of the coating itself during electrolysis. W h e n the protective coating is soluble, however, t h e choice becomes more critical.
If possible, t h e electrolyte should exert a selective action, favoring t h e solution of the base metal while retarding solution of t h e coating. Solu
tion of the coating sometimes m a y be retarded b y employing a n elec
trolyte which forms an insoluble, protective film or induces a passive condition. W h e n this is not possible, t h e total metal t a k e n into solution m a y be reduced to a negligible q u a n t i t y through employment of sensitive color reactions requiring a m i n i m u m of t h e base metal. W i t h indicators such as ferrocyanide for iron or diethyldithiocarbamate for copper, t h e q u a n t i t y needed is so small t h a t t h e danger of creating further pores is practically eliminated for all b u t t h e thinnest coatings. As an example, it was found by the authors t h a t 0.5 μg. iron/sq. m m . is t h e minimum q u a n t i t y needed to give a deep blue color discernible with certainty when viewed on pin-point areas. This would be about 0.3 m g . / s q . in. of iron, or about .35 mg. zinc. On a zinc coated steel specimen, because of the 0.3 volt potential difference between Zn a n d Fe, solution of t h e two metals will not proceed a t t h e same rate. I n practice, two t o three times the zinc equivalent of t h e iron m a y be removed, yet a porosity print could be obtained with solution of about 1 m g . / s q . in. Com
mercial coatings range from 10 t o 200 m g . / s q . in., hence t h e metal dis
solved would a m o u n t t o 0 . 5 - 1 0 % of t h e coating thickness.
Since t h e discontinuities of t h e coating usually consist of pinholes or fine scratches, t h e importance of securing i n t i m a t e contact with t h e printing medium cannot be overemphasized. Lateral diffusion m u s t be prevented as far as possible if t h e print is t o convey information as to t h e size a n d shape of t h e breaks in t h e coating. T h e writers have found t h a t these conditions are best m e t b y using high contact pres
sures obtained with a hydraulic press, (see p. 167) a n d a relatively dry printing medium. After soaking in t h e electrolyte, t h e printing p a d is sandwiched between larger pieces of blotting material a n d pressed for a few seconds at about half t h e printing pressure. F o r gelatin papers, 500-1000 lb./sq. in. is a satisfactory pressure, while for unsized papers it can be m a d e considerably higher, u p t o 2500 lb./sq. in. Some caution is necessary in using these high pressures lest t h e pores be closed when t h e coating is a paint film or other organic material subject t o plastic flow, and in such cases, experimentation is advisable. I n general, however,
ELECTROGRAPHY AND ELECTRO-SPOT TESTING 211 the high pressure technique is desirable, since t h e n t h e electrolyte and even t h e printing substance is forced into t h e smallest pores a n d t h e reaction product is so confined t h a t lateral diffusion is greatly retarded.
T h e use of gelatin paper has been generally recommended t o obtain the least diffusion of reaction product a n d hence t h e most detailed reproduction. I n porosity testing, however, there m a y be instances where t h e use of unsized paper with t h e attending diffusion has merit.
Pinholes m a y be so small t h a t their recognition on gelatin paper requires magnification. For research purposes, this m a y be advantageous b u t in practical testing it m a y be desirable t o obtain enough amplification of t h e original pore b y diffusion on t h e print t o m a k e its recognition easy t o t h e unaided eye. I n such cases, t h e use of a hardened, unsized paper such as C.S. & S. #575 will provide t h e desired " b l e e d i n g . " T h e y are, of course, easier a n d cheaper to prepare t h a n t h e gelatin prints.
Tin on iron: For t h e porosity of tin plate, a mixture of equal p a r t s of 0.1 Μ potassium ferro- a n d ferricyanides a n d 0.5 Μ sodium acetate is a suitable electrolyte reagent (29). W i t h o u t t h e acetate, t h e iron ferro
cyanide reaction products t e n d t o precipitate in t h e pores, clogging t h e m a n d giving faint or blank prints. At about 3 volts, a flow of about 50 millicoulombs/sq. in. gives a satisfactory density of t h e blue iron color with negligible diffusion. This would correspond t o t h e solution of about .03 m g . / s q . in. of tin, or a b o u t .0004 mils thickness.
Tin on copper base alloys: Tin develops considerable passivity toward an electrolyte composed of 1 % sodium diethyldithiocarbamate plus 2 % concentrated a m m o n i u m hydroxide. At t h e same time, solution of copper is facilitated and its detection b y this reagent is among t h e most sensitive tests known. Gelatin coated paper m u s t be used, however, because of t h e tendency for t h e colored product t o " b l e e d . " At 3 volts, printing for 20 seconds produces ample density. T h e current, in a series of experiments, averaged a b o u t 100 millicoulombs/sq. in., cor
responding t o solution of about .06 mg. t i n / s q . in. (29).
Chromium plating: Chromium plating is usually quite t h i n a n d it therefore becomes t h e more desirable t o dissolve as little as possible in the production of porosity prints. W h e n t h e impressed voltage is over 2, chromium dissolves rapidly with oxidation to t h e chromate ion. If the voltage does not exceed 1.5, however, in an electrolyte such as a m m o n i u m acetate, oxidation a p p a r e n t l y does not proceed beyond t h e trivalent stage and is accompanied b y t h e rapid formation of a passive oxide film which effectively blocks a n y transfer of t h e coating t o t h e printing medium. Iron, nickel, a n d copper exposed t h r o u g h pores or scratches in t h e chromium m a y be printed successfully a t 1.5 volts if sufficient time is allowed. Passivity is so complete t h a t current
measure-212 Η. W. HERMANCE AND Η. V. WALDOW
m e n t s under these conditions will have little significance since t h e flow is determined largely by t h e number a n d size of t h e breaks in t h e coating rather t h a n t h e total printed specimen area. T h e printing time is best determined experimentally on specimens of t h e pure base metals at 1.5 volts with t h e same printing p a d and pressure as used on t h e specimen.
Gelatin paper works well with 0.5 Μ a m m o n i u m acetate, followed by development with ferrocyanide, diethyldithiocarbamate or dimethyl
glyoxime for iron, copper, or nickel respectively (29).
Lead coatings: The porosity printing of lead-clad steel affords a good example of retarded solution of t h e coating through t h e selection of an electrolyte which forms a thin, insoluble layer on it. At low voltages lead is oxidized only to the divalent ion. W h e n t h e electrolyte contains sulfate ions, a protective layer of lead sulfate quickly forjris a n d further solution becomes negligible. Solution of iron, copper, and most other base metals through imperfections and pores remains unhampered.
Above 3.5 volts, (aluminum cathode) however, lead is oxidized to t h e tetravalent state with formation of nonprotective peroxide. Figure 15A shows the behavior of pure lead a n d iron specimens toward sulfate and nitrate electrolytes. Curve I illustrates t h e almost uniform high r a t e of solution of lead in nitrate while curve I I shows t h e very rapid sealing off of t h e surface, with very little current flow after 5 seconds. Curve I I I , on t h e other h a n d shows t h e uninhibited solution of lead in t h e sulphate electrolyte when t h e voltage is raised to 4. Curves I V and V depict t h e behavior of iron toward these same electrolytes. I t is quite evident t h a t t h e sulfate, while facilitating t h e solution of t h e iron, pro
tects the lead from being dissolved.
For direct printing, t h e authors have used an electrolyte composed of 0.5 Μ sodium sulfate, 1 p a r t ; 0.1 iVf potassium ferricyanide, \ p a r t ; 0.1 Μ potassium ferrocyanide, \ part. This solution works well for both copper and iron bases. T h e voltage is held at 3 or below a n d t h e time ranges from 10 t o 60 seconds, depending on t h e p a d used. Gelatin paper at 500 lb./sq. in. gives excellent detail. Figure 15B shows t h e behavior of lead a n d iron toward t h e mixed electrolyte, which is essen
tially t h e same as toward sulfate alone.
F u r t h e r applications of electrography t o porosity testing of metallic coatings are given in t h e t a b u l a r s u m m a r y (p. 214). I n t h e opinion of t h e authors, several of t h e published methods fail t o t a k e full a d v a n tage of t h e employment of sensitive reagents as well as controlled and selective solution in minimizing a t t a c k of t h e coating.
Organic finishes: Since organic coatings are nonconducting and chemically inert, t h e recording of pores and breaks can be accomplished with almost a n y electrographic printing technique suited t o t h e base
ELECTROGRAPHY A N D ELECTRO-SPOT TESTING 213
3 5 0 3 0 0 | -2 5 0 2 0 0 1 5 0 1 0 0 5 0 _J 0
? 3 0 0 r-Z
*—
I Am
«•«» ^ m
n
2 5 0 2 0 0 1 5 0 1 0 0
Β
^ I R O N
L E A D
2 5 3 0
0 5 10 15 2 0 T I M E IN SECONDS
FIG. 15. The behavior of lead and iron specimens toward various electrolytes.
Pure metals, 1-inch square, aluminum cathode plate, printing pad C . S . & S. #575 and #601.
A : I—Lead, 3 volts, 0.5 Μ sodium nitrate II—Lead 3 volts, 0.5 Μ sodium sulfate III—Lead 4 volts, 0.5 Μ sodium sulfate IV—Iron, 3 volts, 0.5 Μ sodium sulfate V—Iron 3 volts, 0.5 Μ sodium nitrate B: Iron, 3 volts, ferro- ferricyanide-sulfate
Lead, 3 volts, ferro- ferricyanide-sulfate
metal. Glazunov (24) used printing with ferrocyanide to detect faults in varnish films on iron. H e compared chemical contact printing with t h e electrographic m e t h o d and points out t h e greater sensitivity and sharpness obtained b y t h e latter technique. Shaw a n d Moore (43) have described t h e application of electrography t o the testing of paint and other protective coatings on an iron base. T h e y use E . K . Imbibition paper soaked in 5 % K N 03 a n d pressures of 600 lb./sq. in. T h e print is developed in a mixture of potassium ferro- and ferricyanides.
214 Η. W. HERMANCE AND Η. V. WADLOW
4.3.4. Minerals. Certain minerals such as t h e sulfides, arsenides, a n d antimonides of h e a v y metals are fairly good conductors and m a y be printed electrographically, yielding either anions or cations, depending on t h e polarity of t h e specimen. T h e electrographic method was first applied t o minerals b y Jirkovsky (37) a n d later elaborated on b y Wenger, Gutzeit, a n d Hiller (44). Hiller (32) has published a detailed s u m m a r y of t h e techniques and reactions applicable to polished mineral sections.
Table V I I , t a k e n from Hiller's article, lists t h e minerals and t h e elements detectable electrographically, together with t h e conditions for obtaining