METALLOGRAPHY 9, 51-61 (1976) 51

An Improved Etchant for Copper and Copper Alloys

R. M. SLEPIAN AND J. P. PROHASKA

Westinghouse Research Laboratories,
Pittsburgh, Pennsylvania

A new etchant for copper and copper alloys using ferric nitrate and hydrochloric acid
was developed that delineates grain boundaries better than numerous recommended
micro-etchants. The same solution can also be used as a macro-etchant for the same
alloys. For best results proper metallographic procedures prior to etching must be
followed.

Introduction

Metallographers have often had difficulty in obtaining good photo-

micrographs of the grains of copper and some copper alloys. The difficulty
arises because all grains in the sample do not etch at the same rate. Con-
sequently the metallographer has to compromise between using a photo-
micrograph t h a t does not reveal all the grain boundaries or a photomicro-
graph of the sample in the over-etched state that reveals all the grain
boundaries with the faster etching boundaries being much too wide. In
addition the over-etching causes the sample to become pitted by removing
the fine copper oxide particles a n d / o r inclusions. Also any porosity present
in the sample is increased by the long etch. To overcome this problem a
study to develop a more satisfactory micro-etchant for copper and copper
alloys was undertaken.

Metallographic Procedures

The preparation of metallic samples prior to macro-etching and micro-

etching is most important. In general, the more carefully prepared a sample
is with a fine unworkhardened finish, the better the finish etched specimen
will be. This is especially true for soft copper samples, which are prone to
(~) American Elsevier Publishing Company, Inc., 1976
52 R. M. Slepian and J. P. Prohaska

persistent scratches, damaged metal layers, and pull-out of oxide inclu-

sions. The following procedures are used in our laboratory and lead to
excellent, reproducible results.

Macro-Etching Procedure

Samples are carefully ground on 100, 240, and 600 grit water-cooled
papers. Next the samples are polished with 6-micron diamond lapping
compound on white duck cloth using kerosene as a lubricant. After the
samples are thoroughly cleaned, they are etched with the new e t c h a n t - - t h e
composition of which is listed in Table 1 - - b y cotton swabbing with firm
hand pressure for 3-4 minutes. Finally the samples are finished macro-
etched b y immersion in the new etchant for 15-60 seconds, followed by
rinsing in warm water, then alcohol, and dried under warm air.

TABLE 1
Compositions of the Copper Etchants Used

Etchant Composition

New etchant 5 g Fe (NOs)s

25 ml HC1
70 ml Water
ASTM #26 5 g FeCls
10 ml HC1
50 ml Glycerol
30 ml Water
ASTM #28 1 g Fe (NOs)3
100 ml Water
ASTM #30 25 ml NH40H
25 ml Water
50 ml H~O~(3%)
ASTM #31(d) 10 g Ammonium persulfate
100 ml Water
ASTM #40 5 g FeCls
15 ml HC1
60 ml Ethanol (95%) or Methanol (95~)
ASTM #41 2 g Potassium dichromate
8 ml H2S04
4 drops HC1 (just before using)
100 ml Water
New Copper Etchant 53

Micro-Etching Procedure

Samples are carefully cut and ground on a 100 grit belt. Then if size per-
mits the samples are mounted in standard 1, 13, or 1½ inch-diameter
molds, and the resulting microspecimens are processed in the Buehler
"Automet ''® attachment. Grinding is done on 240, 320, 400, and 600 water-
cooled silicon carbide papers using a light pressure of 25-30 pounds for 3-4
minutes. Next the specimens are rough polished with 6-micron and then
1-micron diamond lapping compound with kerosene as a lubricant on duck
cloth using the same pressure and times as in the grinding steps. The
latter diamond step, not usually used in metallographic preparation, adds
to the quality of the sample preparation and also aids in edge retention.
Final polishing is done on a "Microcloth ''® wheel using a .05-micron
alumina-water slurry and a diluted etching solution of 3 parts water and
1 part new etchant. This etch-polish step removes the 1-micron diamond
scratches and disturbed layer. After the microspecimens have been etch-
polished for 3 minutes at 25 pounds pressure, the wheel is flushed with
running water and then attack-polished with the same solution (no abra-
sive or slurry) for about 2-5 seconds, h very light etch should result, The
"Automvt Attachment" is then immediately removed from the final
polishing wheel and flushed thoroughly with hot running water and blown
dry. Final etching is done by immersion in the new etehant for 10--30 seconds
followed by rinsing in warm water, then alcohol, and drying under a warm
air blast,

Results

MACR0-ETCHING

As a macro-etehant the new etchant does a good lob of bringing out the
grains with nice contrast, The etehant is neither very rapid or very slow
and thus is very easy to use and control. On a few occasions thi~ etehant
was able to bring out the grains of samples far better than the usual macro-
etching solutions [11 used for copper and copper alloys, Figure I shows
excellent grain delineations that were obtained on three experimental alloys
using the new etchant and the above recommended sample preparatioaal
techniques. Compositions of the alloys is given in atomic percentages and
magnification of the macrographa is 13×. Figure 5 (below) shows the
100× microetructure of the 05 Cu-5 Sn sample shown macro-etched in
Fig. l(b),
New Copper Etchant 55

(c)

FIG. 1. New etchant used as a macro-etchant for three experimental

copper alloys (13 X): (a) As cast, 97.5% Cu-2.5% Sn; (b) Annealed 825°C
for 48 hours, 95% Cu-5% Sn; (c) Annealed 950°C for 48 hours, 99%
Cu-1% Si.

Figure 2 shows the macro-structure developed by the new etchant on a

section of a centrifugal cast induction motor end ring. The casting had
been machined prior to sectioning and macro-etching. The left side of Fig.
2 is near the inside diameter of the original casting and the right side is
near the 9.utside diameter. The columnar grains and equiaxed grains are
well resolved. By normal macro-etching means using other solutions and
less sample preparation, the grains of this sample are not brought out
nearly as well.
The last macrograph (Fig. 3) shows, at 2X, a brazed sample consisting
of 14 pieces of copper. The top three pieces going horizonally are electrolytic
tough pitch copper 110. Proceeding downwards the two pieces on the right
are joined to a "coarse-grained" solid bar of 110 copper. The solid bar in
turn is joined to a fine-grained hollow tube of oxygen-free high conductivity
copper 102. The remainder of the macrograph consists of 2 bars, 1 tube,
2 bars, 1 tube, 2 bars and ½ of a tube at the bottom of the picture. At the
56 R. M. Slepian and J. P. Prohaska

Fro. 2. New etchant used as macro-etchanton a section of centrifugalcast

induction motor end ring, Phosphorus-deoxidizedcopper 122 (1.5×).

magnification of Fig. 3 the grain structures of the various copper com-

ponents can be seen, and also the thicker brazing alloy areas can be seen
and judged for penetration and porosity. At higher magnification the
thinner brazed joints can be macroscopically evaluated.

MICRO-ETCHING

The need to find a better micro-etchant was brought about when a very
largo number of hollow copper tube samples had to be examined for grain
size~ grain flow, and cracking at the corners and when existing etchants
did not uniformly bring out grain boundaries satisfactorily. After the
literature was surveyed and published etchants were tried and found to
range from unsatisfactory to fair~ experimental etchants were tried until a
satisfactory one -the new etchant--listed in Table 1 was discovered.
This new etchant was compared to six ASTM micro-etchants ['2J for
copper by using them as a final micro-etchant, after identical specimen
preparation~ for the same annealed tough pitcb 110 copper sample. Figure
4 at 200 × shows very clearly the superiority of this new etchant over the
representative six of the 18 mioro-~tchants recommended by the ASTM
for copper. Compositions of the six A•TM otchants are listed in Table 1.
The top photomicrograph of Fig, 4~ taken when the new micro.etchant
was used~ shows light to heavy grain boundaries with just about all the
grain boundaries delineated~ very slight pitthag of the copper oxide particles
 c~

Single Solid
Bar
Hollow Tube

Two Solid Bars

F I c . 3. N e w e t c h a n t u s e d a s m a c r o - e t c h a n t on a b r a z e d s a m p l e of 110 a n d 102 c o p p e r ( 2 X ) . ¢~
58 R. M. Slepian and J. P. Prohaska

New Et¢hant

~,L~ ~ ,. ,..~.J . -- ~ , ~.~ .-.~ . . . . ' ,. ~,*.:.~ "~.,,,.,, "!." ,~:.~.~ ~:~

ASTM Etchant #26 ASTM E~chan~ #28 ASTM Etchsnt ~30

ASTM Et:clmn~: #31(d) ASTM Etchan~: #40 ASTM E~:chant #41

200X

FIG. 4. New etchant compared to six ASTM micro-etchants on the same electrolytic
tough pitch copper 110 sample (200 X, shown here at 6 0 ~ of original size).

(black dots), and uniform white coloring of all grains (desirable for such
single phase metals and alloys). A heavier etch will bring out the grain
boundaries more, increase the pitting, and eventually develop contrast
between grains of slightly different orientations. With ASTM etchants
#26, #28," #31(d), and #41 all the grain boundaries are not brought
New Copper Etchant 59

out. With # 30 the oxides are quite pitted and some grain boundaries and
twins are very heavily attacked. Etchants # 28, # 31(d), and # 40 show
grain contrast. Of all the etchants tried only # 40 approximates the new
etchant. Etchant # 40 is slightly faster acting, attacks twins more rapidly,
and develops grain contrast more quickly.
Figure 5 is a 100 X photomicrograph of the same sample shown macro-
etched in Fig. 1 (b). The 95% Cu-5% Sn as-cast ~-inch diameter sample
was annealed for 48 hours at 825°C prior to sectioning and macroscopic
and microscopic examination. The new etchant as shown in Fig. 5 has
done an excellent job of bringing out the grain boundaries and twin bound-
aries with nice contrast. In addition the etchant has revealed faintly some
remnants of prior as-cast grain boundaries (right center with arrows point-
ing toward the prior boundaries).
The last photomicrograph (Fig. 6) is of a fractured brazed sample at
100X. The copper on the left is oxygen-free high-conductivity 102, the
copper on the right is electrolytic tough pitch 110, and the brazing alloy at
the center of the picture is 80 Cu-15 Ag-5 Zr. The new etchant again has
done an excellent ]ob on this sample of bringing out the metallographic

\
\
1
FIG. 5. New etchant used as a micro-etchant on the same 95 Cu-5 Sn sample shown
in Fig. 1 (b) (100×).
60 R. M. Slepian and J. P. Prohaska

FIG. 6. New etchant used as micro-etchant on brazed joint of oxygen-freehigh

conductivity 102 copper (left) to electrolytictough pitch Copper 110 (right) (100X).

structure and in this case has revealed much information. The fatigue
fracture at the top of this photomicrograph is shown to be transgranular.
The brazed joint is shown to be excellent on the 102 copper side (left), the
thickness of the brazing alloy is seen to be uniform .010 inch wide, and the
joint on the 110 Cu side (right) is shown to be normal except for hydrogen
embrittlement being present to a maximum depth of .005 inch in the
electrolytic tough pitch copper next to the brazing alloy. The thick dotted
and solid wide black boundaries are the metallographic evidence of the
embrittlement.

Conclusion

The new ferric nitrate-hydrochloric acid etchant does a superior job in

revealing all grain boundaries in copper and copper alloy samples with a
minimum of pitting. The same solution can also be used as a macro-etchant
New Copper Etchant 61

for the same alloys. For best results proper metallographic procedures
prior to etching must be followed. It is recommended that this etchant be
added to both ASTM E-340-68 Standard Method for Macro-etching
Metals and Alloys and ASTM E 407-70 Standard Methods for Micro-
etching Metals and Alloys in their next revisions.

References

1. 1971 Annual ASTM Standards, Part 31 E 340-68, pp. 863-879.

2. 1971 Annual ASTM Standards, Part 31 E 407-70, pp. 936-950.

Received March, 1975