SURFACE PROPERTIES OF HIGH PURITY ALUMINIUM POLISHED BY ANODIC AND CHEMICAL PROCESSES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78-04861A000400030010-5
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 13
• Document Creation Date: December 20, 2016
• Document Release Date: June 6, 2006
• Sequence Number: 10
• Publication Date: August 1, 1955
• Content Type: REPORT

7r,Y1
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
SURFACE PROPERTIES OF HIGH PURITY ALUMINIUM
POLISHED BY ANODIC AND CHEMICAL PROCESSES
Metall 6 (1952) 346 - 350
(From German)
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
SURFACE PROPERTIES OF HIGH PURITY ALUP INIMa-
POLISHED BY ANODIC AND CHUJI1ICAL PROCESSES
W, HEILING AND F. BAUvtANN
MTALL.6 (1952) 346-350 (FROM GERMAN
The mechanical polishing of high purity aluminium presents special
difficulties owing to its high plasticity. It is almost impossible to
prevent elevations on the surfaces being pushed into the depressions,
causing overlaps and deformation in lattice zones near the surface, The
increased temperature caused by,mechanical polishing is favourable to
renewed formation of the oxide-film immediately after processing. The
polishing medium remains in the pores of this. layer and in the overlap
joints and cannot be removed entirely by the usual degreasing agents. If
such a surface receives anodic treatment, a clouded oxide film appears.
Even with high purity altinird mi clear oxide films can only be obtained by
a preliminary "polishing" by anodic or chemical means. Table f summarizes
acme of the familiar processes for anodic polishing.
In the last few years a number of processes for the chemical
polishing of aluminium have become known. The principal constituent of
most of the polishing solutions is phosphoric.acid. For example, under
the name "Alupol II - polishing.bath" the following mixture of acids is
given as being suitable for polishing:
NY% phosphoric: acid: 5 acetic acid: % nitric acid.
In addition'to phosphoric acid. and sulphuric acid other acid mixtures
contain nitric. acid. as well as certain metallic salts, e.g. copper nitrate.
A. polishing solution was developed at the Vereinigte Aluminium 7erke,
Grevenbroich (VAw), consisting essentially of ammonium difluoride.
The effect of all polishing processes is the removal of the oxide
film and impurities and also the smoothing of the surface. So far no
details are known about the degree of smoothing resulting from the
individual methods. However, in order to judge the quality of polishing
processes, it, is definitely necessary to know the degree of. preliminary
mechanical polishing required and also . the. state of the surface after
polishing. The present work is,therefore concerned with examining
known methods of measurement and observation with regard.. to their
suitability for indicating the state of the surface and also with
determining approximately the polishing effect of certain processes
which have different modes of operation.
In ora'er to remove as..far as possible the influence of the
metallide on'the'measurement results, examinations were carried out
exclusively on l.?affinaL; and Reflectal of the following composition:
'Raffin.al: Si '0.004 %
Fe 0..001 %
C u .0. 0005 %
Zn 0.002%
Remainder Al.
Reflectal:
Mg 0.5%
Fe 0.001%
Si 0.0045%
Cu 0.001 ;a
Zn 0.002 jo
Remainder A.1
K For Tables, see end.
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
Some of the specimens were simply ground first with 6/0 emery, some
were first ground and polished, followed by the i\Ta.2SO+/NaOH bath and
subsequent anodic polishing-after-removal of the -oxide film; others were
chemically polished by the Erftwerk process.... Reflection measurements
were taken, to determine the surface changes and 'photographs were made
with the Fanphot and the,. electron microscope. (the electron microscope
photographs were taken in' rof. V'. Berrie's' Institute at Dusseldorf).
Photographs were also produced by the microinterference process developed
by Rl hle.,
R +'LECT ION 111EA.SUMiENTS
(.l) ..
Figure 1 illustrates the apparatus :developed by J. Elze and Grtss
for measuring the brightness of surfaces. It consists essentially of a
double optical bench.with a graduated circle; one of the arms.remains
fixed while the other can be rotated. through any desired ,angle which can
be read off. The fixed arm carries the light source and.` a condenser lens.
During measurement these parts are' housed in a!con.tainer',..closed on all
sides, so as to avoid disturbance. Prom. stray, light; in,.,the front of the
container there is an opening; serving. as a stop, through which the beam
of light passes. The rotating arm carries a second convergent lens, 'a
shutter, and .a-,:hotocell. All the sliding ,parts have, a.,fine, lateral
adjustment.. The slide -which carries .tLe stop in },rent of .the photocell
has in. addition a fine adjustment for vertical movement so that the. stop
opening can be brought exactly into the path of the ray of light. The specimen holder is located on the ;o-u.iometer axis. Figure 2 shows the
path of the rays. The apparatus has an accuracy of ]%., ' 'The above-
mentioned authors carried out an interesting experiment to determine the
brightness ? of a surface from the measurement of scattering,' using stops
of varying.;.diametgr. At scme future time we shall give an account of a
ccznparison of the.se.,.values, \rl-dch are called the close-.range spattering
angle reflection, ~-rith the :results of other measuring methods. 1'1e,have
been content; to start with, to measure the regular reflection as a'standard
for the surface brightness, under the conditions set out by-the authors.
These values are shown in Tables 2 and 3 and also, accompany the photographs
(Figures 4-17).
OBSERVAT ION OF THE SURF!' CE
The resolving power of thy, optical microscope is limited, owing. to
the undulatory nature. of light, where` white light is: used it is about
0.4p for. good immersion systems. The necessary magnification.,
appiopriate" to this resolving power,. is then about 1000:1.:
On the other hand, the electron. microscope has a much higher
resolving power - it can resolve dov'n to 1-2 m so that 150,000
magnifications of a surface are possible. Such high magnifications were
not required for these investigations. The highest were a tenth power
lower. The electron microscope is.als'o greatly superior to the optical
microscope as regards definition in death. If,. in spite of all these'
advantages, the electron microscope photographs produced during our
investigations gave to some extent incorrect :information on the effect of
the various polishing processes - as could be. proved' 'by comparison with
the light-optical photographs of the same surfaces - this was entirely
due to the photo caphic technique. A. -lexiglass' replica of each surface
to be studied was obtained and then obliquely shad owed with silicon
"For references see end.
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
monoxide vaporised ' in..a vacuum. The ' vapour-deposited film was removed
from the plexi ,lass and' irradiated in the, electron. microscope., Since
the photographs obtained in this way showed no structure at all for
polished test pieces it can be sus,aected that after the polishing process
a transparent film of aluminium oxide having a smooth surface is
formed and the plexiglass replica was formed.f'rcm this, It is
advisable therefore to employ a .different method of 'making, replicas
in which an artificial oxide film is produced on the aluminium surface
and' is then irradiated in the e:.egtron microscope after the aluminium
has been dissolved. by concentrated .sublimate solution. There are no
photographs in existence, produced by this process, so there is no
possibility of making canparisons.
/.nother very simple method of surface testing is the micro-
interference process developed by.,Rtthle. This is based in principle on
the phenomenon of interference or., equal thicknesses. A semi-transparent,
plans-~aralle1 glass slip, vapour-coated, tWith aluminium, . is laid on the
experimental surface. The air gap between the surface of the specimen
and the glass slip is made into the shape of a tired ,e by, introducing a
thin aluminium foil between the upper end of the' ;lass, slip and the.
experimental surface. Care must be taken here to ensure that the glass
slip, with the coated side turned towards the surface of the specimen,
is firmly attached, at the upper and lower ends. iA ray of light falling
vertically on to the semi-transparent layer is partly reflected, partly
transmitted and then reflected at the surface under examination. This
results in a difference in, ,the length of the light,path,, giving rise to
interference phenomena. X" both faces of the wedge are perfectly level
it is possible to observe' dark bands parallel to the edge of 'the wedge
at those points where the thickness of the wedge measures an odd multiple'
of a quarter wave-length of t.ie light employed, If, however, the lower
wedge face, in our , case: the experimental surface, is xoughi the bands
are no longer straight but more or less irregular 'lines can be. observed
which nevertheless follow the points of equal thickness of the wedge.,'
Thus we obtain a contour pattern, similar to those knownin cartography,
which is shown in Figure 3.
Figure fit.. shows a Reflectal surface . )re-treated by grinding with
6/0 emery. and ?oliship.g,.,.quite distinct grinding marks can be recognized.
Figure 5 shows magnified 100 times, aphotograph obtained by the
microinterference..process of the same surface, k scratch can be
distinguished ,on the hand side. Sincethe convexity of the lines at
any, time extends to, the . next..,line., the depth of the scratch is about
half the length of a light wave eo that wit'n the sodium light employed.
in this. case it is 0.;3 Another.noteworthy point is the serration
of the lines, Which oa~i be seen even more distinctly in. Figure 5.
Figure 6 is. also a 100 x magnification of the s ene surf ace, ..but so as to
obtain a, greater distance between the lines, the..,angle of the wedge
was made about 10 times as small as in Figure 5. This serration is a
consequence of the grinding marks which, accordingly, have a depth of
about 1/10 wave-length.
FigLa c .,7 shows for. comparison an electron :microscope photograph'
of the same surface 15,000 x.- Here, too, the`huch higher resolving`'
power of the electron microscope is demonstrated.
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
In the case of chemical polishing the grinding marks disappear
ccmpletely after a polishing period ' of 10 sec. as is shown in Fig. 8.
Consequently the appropriate -interference photograph,.ig.9, also shows
no serration of the.ines. Polishing for 15 se xy brings ~ no fundamental
change, which of course, to, .judge- from the results of the. reflection
measurements, was only to be expected (Fig,10).
It is interesting to see that the electron microscope. photograph
of the same surface. (Fi g.11) . shows a considerably better polishing effect
than Fig.10, particularly when it is remembered that the magnification in
Fig. 11 is twice as t e .t ' as, to Fig.10, . viz. X = 2200 as, against X = 1000..
The reason for this` surprising'. fact would-seem to lie in the fact mentioned
earlier that in ; ohotagraphs taken with the electron microscope it is' the
surface of the oxide film which forms after polishing which is photographed,
while with the li ght--optical photographs, owing to the transparence of
the oxide film the aliviinium surface is: visible.
All the photographs shown until now were of specimens which had not
been anodised. It appears, however, that conditions are quite similar
in the case of anodised specimens.
The course of events in the removal of the oxide film.
differs from that iri chomionl..polishing. Thus, Fig. 12
shows the Panphoto reproduction V = 1000:1 of a .affinal surface after
one removal operation. As a result of the moderate attack in this
process the grinding marks are still clearly visible although not so
definitely marked as in Fig,,.,
Similarly the'inter'erence photograph, Fig 13, still shows the
serrations which havebeendescribed. After three removal operations the
grinding marks have disappeared (Fig.l.) and the illustration of the
surface now resembles Fig.8.with the exception of the grain boundaries
which can be seen there. tf course the same effect can be established
by means of the reflection measurements. The corresponding interference
photograph (Fig-15) is now to a.':certain extent like Fig.9 and no
serrations can be seen.
Figures 16 and 17 illustrate Raffinal specimens i po shed in the
Na2SOL/NaOH electrolyte after preliminary polishing and fine grinding
as well as preliminary chemical polishing. It can be seen fromn.these
photographs and also from the. : reflection values that the surface has
greater reflectivity 'than after treatment with the other polishing
processes. The difference is most marked in the .interference photographs
and the values for the regular reflection. in the Erftwerk polishing
process preliminary grinding, with 6/c.emery is in general sufficient,
Experience has show that in man r;. cases even. this can. be dispensed with,,.
Fig.18 illustrates how widely ? the,- surfaces .can :differ after mechanical,
chemical and anodic polishing.' Comparison with the values for regular
reflection shows that this is?definitely influenced not only by the
depth of the grooves but also .;by, their frequency. Further study will
be devoted to experiment a.,for`determining and examing:this relationship.
SUM WARY
In order to study the `polishing effect of anodic and chemical' ..
polishing processes observations of high-purity aluminium surfaces were
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
made by optical microscope, electron microscope, and by the multiple
interference process. The regular reflection was also measured by means
of the gloss-meter developed by Elze and Grlxss. Irregularities in the
surface were measured by means of interference photographs and compared
with the values for regular reflection. Comparison of all the photographs
and figures yields a good picture of the change in the surface of high-
purity aluminium brought about by anodic and chemical polishing.
This work deals solely with the measurement and representation of
the effects of the polishing process. It does not deal with the limits
of anodic and chemical polishing. Development of chemical polishing is
still in full course. There is good reason to hope that the polishing
effect of this process can be increased considerably. An account will be
given later of the limits of anodic and chemical polishing processes.
Elze and GrUss "Metalloberflbche" 6 (1952), A 17-23.
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
T -.
'
O
T
~
t
I
SI
~?y
CVMa'
r L `
Q ~
7 NN Q
t"
r-1 z
G
.
!Cy
O
Lf1
t
$ k
~
? ` k
N ~
V
N
#
Q)
zrz 0 0
ro 1
m .. ?? -
rd 9
i f
u1
1 ON r4 0
&Q
-
rd ca
f
., >.
0 !
C
V
0 3
C T
r4
l ? f!1
0.
O
j~
c6 0 C
r-4
C
R
j
-
r
N
Ej
a?-1
O : Ul
r"I L"
LC1 1
r-1
I .~
HC)
Kl ," ,
1 it
-
p Imo]
'r~
a,
c M?M
`i`
as
0
000 r~
00
H 4
N
pp
O tZ1
Q...
.I PR g
Il ?l k
i
1
7
~y 9
co
-%
I'D ..- ~.
I
-
s
'
i1
1
I
,
r~?~
N
I
~x U
)
rI ?
-{
'U
%
1
r-I CP
i
Lc
)
Hr_i
4
1-1
r-1
i
I ..
I
LIl
Pr
'
?.cv
i
t
C\I
'
I ti p
-I ?
?
'
c
1
r-i
-~
,
..~
.~
r
N
0 H
co (D
'~
1
I
r-I
ll
CS
1
I
l
H
i -f
Q a
L
N.~
Li~
~
~
CSI
~ ??VV
?y
tc
?
~\y~`
1
j.
r~ 1
1
r
I0
l
J)
N
I
.
-0 N
I *
k e
I fit, 11H O
rl ,
0
ri ?
r-I
A A
?
O
19
a
; 1
3
4
00
-F) C'i
.L?
.W
CM
4-
co
.
U
co
-
0
?
a
w
w
a
w
co
P74
P4
~
l
CV
CU
N
Pl
a
a
0
4.)
lU
~
U
0
Q
p1
'y
r
P.,
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
?b'
a~
4-5
mom'
d
to S -r-4
a~ia~
?r 1
-
4.3 r-1
b
43 rd O rd
W r--I
Ul
A 1 H a
Q
.Q
Cg
1
1
i
r-
?
MM'1
r-I
?
{
-ri
? y
a
0
A
w
w
pp
N1
K1 p
9 15
li
0
i
C
LCIN
Al
r_q
~,
~
o x
a
-
cu -W
O
ri
W
Lfki
~
ry
U
N
ri -t r
y~++y{~
~
y
[
14
y
y
cc r-4 n
4i
00
Pq
xi
I
rd
4)
P.4 i
?
~4p r-I
O
Q)C
w?
?H
P "0
Y7
43
w 0
b
cr
a
.r
A
rd
0
~C
02
rq
O
0
Pa
~-
p
' 0
H
44
c
w
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
T.AB LE r.
Re ular^ reflection in jo ccmpared with a
silver mirror
(with 97/ regular absolute reflection)
Raffinal, mechanically ground and polished..... 47.6 %
N anodically polished and GS-anodised .... 84.0 %
chemically and anodically polished
GS-anodised .... 84.0 %o
Reflectal (0.$mg), mechanically ground and
polished:
(a) 5,-,a--V Erftwerk-process chemical polishing,
GS-anodised .. .
81.6 %
10 sec. ditto ....
81.1
c 15 sec. ditto
81.6 %
Reflectal (0.3 big), mechanically ground and
polished
a)
1 x WGX, GS-anodised ...................
69.3
(dull)
b)
2 x WGX, GS-anodised ...................
73-9-A
c;+
3 x WGX, GS-anodised. ...................
75.6 %
TABLE L.
Regular reflection of ground and ground
and polished sheet metal specimens of
high-purity aluniniura
Ground Ground and Polished
Raffinal,
Anodically polished and GS-anodised 77. V 83.90
Chemically polished for 15 secs.
GS-anodised
R aff i nal .......... 81.0/.
81.7%
Reflectal (0.3f MIg)
soft ............ 81.E 81.J
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
Fig.l: Optical bench for reflection measurement.
Specimen
/ Stop
Fiq.2: Ray paths in the optical bench
Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
 Approved For Release 2007/10/23: CIA-RDP78-04861A000400030010-5
Flg.4: Optical microscope photograph of a mechanically
pre-polished reflectal specimen (0.5% Mg)
M - 1000:1. 47.6% regular reflection (magnification).
Fig.S: Interference photograph of the pre-polished
reflectal specimen (Figure 4).
N = 100:1, 47.6% regular reflection.
Fig.6: Interference photograph as in Figure 5 with
magnified interference band interval, 47.6%
regular reflection
m