48 Thin Solid Films.

XX ( 1992) 4X 54

Anodizing titanium for space applications

A. K. Sharma
Thrrmul Systems Division, ISRO Srr~dlire Cwire. Vimuntrpurtr Po.vr. Bmgrrbwc~ 56001 7 (lnttiu)

(Received June 27. 1991; accepted August 29, 1991)

Abstract

The work reported in this paper concerns a study on the anodizing of titanium in dilute sulphuric acid electrolyte at constant
voltages. The effects of various parameters. 0.~. bath composition, temperature, voltage and anodizing time, on the anodic film
formation have been investigated to evaluate the optimum anodizing conditions. The anodic current density was found io
increase with increasing applied voltage and with electrolyte temperature. The optical properties. 1.~. solar absorptance 2, and IR
emlttance c,~ of the anodic coatings obtained under direrent operating conditions have been evaluated. The effects of relative
humidity, high temperature heating, thermovacuum performance and thermal cycling tests on the physical and optical properties
of anodic coatings have been examined.

1. Introduction (4) To provide thermul control. Anodized titanium

acts as an excellent thermal control surface as it pro-
The inherent low density, high strength-to-weight vides stable optical properties to its components.
ratio and ability of titanium to withstand higher tem- In the present paper a process of obtaining a uniform
peratures (up to 900 “C) have focused attention for its anodic film on titanium alloys is described. The process
increasing use in aerospace and allied fields. Full uti- described herein is simple and economic, it utilizes
lization of titanium in aerospace industries is, however, ordinary chemicals, operates at room temperature and
prevented by the tendency it has to gall and seize and gives stable oxide coatings within 3-5 min. The effect
also by its reactivity towards atmospheric oxygen. To of various parameters on the formation of coatings
overcome these limitations and to meet some other have been investigated in detail and the optimum condi-
functional requirements, a suitable surface treatment of tions have been evaluated. As the anodic coatings ob-
titanium is of the utmost importance. tained by the present process provide stable optical
Anodization is an electrochemical process used to properties even at higher temperatures, these are ex-
obtain a protective oxide film on the metallic job. tremely suitable for reliable thermal control applica-
Anodization on titanium is carried out for various tions. The effects of humidity, thermal cycling and
applications [ 1,2]. repetitive high temperature heating on these coatings
(1) To provide decorativefinishing. As anodization on have been examined in detail and the process is found
titanium under different conditions results in a range of suitable for space applications.
stable attractive colours, it is used to produce a decora-
tive coating on the articles.
(2) To protect the metal from atmospheric, corrosion. 2. Experimental details
Titanium is a reactive metal; it is rapidly attacked by
atmospheric oxygen. Anodizing provides a uniform sta- Anodic oxide coatings on Ti-6Al-4V alloy (5.5%
ble protective film on the metal. 6.75% Al, 3..5%-4.5% V. 0.30% Fe (maximum),
(3) To reduce the friction on sliding surfuces. As an 0.0125% H, (maximum) (by weight)) was investigated
anodic film is generally harder than bare metal, it helps on samples sized 100 mm x 60 mm x 0.5 mm. The sur-
in reducing friction on moving parts. In addition, an face roughness of the bare substrate was 0.60 + 0.05 pm.
anodic coating, owing to its porous nature, improves The following sequence of operations was adopted: ( 1)
lubricity by providing a base for retention of lubricants. ultrasonic solvent degreasing in LR grade methyl ethyl
As titanium has a severe tendency to gall, lack of ketone for 5-10 min; (2) electrolytic alkaline cleaning
lubricity poses serious problems in applications involv- in a solution formulated and operated as summarized
ing the contact of sliding surfaces in various forming in Table 1; (3) water rinse; (4) acid pickling in a solution
applications. containing 150 ml nitric acid ( 70%) I- ’ and 10 ml

0040-6090/92/$5.00 (’ 1992 Elsevier Sequoia. All rights reserved

 A. K. Sharrna /Anodizing Ti for space applications 49

TABLE 1. Electrolytic alkaline cleaning: bath formulation and oper- As formed the anodically grown oxide films are often
ating conditions not uniform. The outer layer may be contaminated with
anions of the electrolyte and/or contain a considerable
Sodium hydroxide (NaOH) 25 g I-t
Sodium metasilicate (NazSiO 3 . 5H20 ) 18 g 1-~ amount of hydrogen. This can influence the optical
Sodium carbonate (Na2CO3) 16 g 1-~ properties of the oxide film and therefore the calculated
Disodium hydrogen otthophosphate 9g1 film thickness. To minimize the non-uniformity in the
(NazHPO4) oxide film both the electrolyte and the test coupons
Sodium dodecyl sulphate 0.5 g 1-
were agitated continuously throughout the anodizing
(CH3(CH2)I 1OSO3 Na)
Temperature 75 80°C process. The electrolyte agitation was carried out by the
Anode Stainless steel continous flow of clean compressed air (pressure, 40-
Voltage 6 7V 60 kPa; through 2 mm holes) from the bottom of the
Agitation Mild by job movement anodizing tank, while the test specimens were agitated
Time 5-7 min
mechanically by reciprocal movement of an anode bus
bar (6 cm s-l). The agitation prevents the temperature
layering at the specimen-electrolyte interface which is
TABLE 2. Anodizing bath formulation and operating conditions largely responsible for non-uniform film growth. It also
helps in dissipation of absorbed hydrogen at the outer
Sulphuric acid (H2SO4) 120 200g1-1 (165) layer of the film surface. To reduce further the possibil-
(specific gravity, 1.84)
ity of contamination of the outer layer of the oxide film
Temperature 15-38 °C (23)
Cathode Ti by anions of the electrolyte, the test specimens were
Voltage 10-90 V (15) immediately withdrawn from the anodizing bath after
Electrolyte agitation By air (pressure, the power was switched off and rinsed thoroughly in
40- 60 kPa) running water. The delay in withdrawal of an anodized
Job agitation By anode rod movement
article from an anodizing solution after switching the
(6 cm s - ' )
Time 4 - 5 min power off leads to the dissolution of the outer layer of
Appearance Coloured coatings the oxide film in the electrolyte. This results in increased
porosity and contamination of the outer film by the
Optimum conditions are shown in parentheses. electrolyte anions. The cathode may be any inert metal,
but titanium is most convenient. This type of elec-
hydrofluoric acid (40%) 1-1 for 1-2 min; (5) water rinse; trolytic process does not have much throwing power
(6) anodizing in a dilute sulphuric acid electrolyte as and it is necessary to have a cathode of approximately
indicated in Table 2; (7) water rinse; (8) dip in hot water double the size of the surface area of the anode.
(80-85 °C) for 30-40 s and dry in air. All the chemicals
used were analytical or reagent grade. Deionized or
distilled water was used for solution preparation and for 3. Results and discussion
rinsing. A schematic representation of the experimental
set-up used for titanium anodizing is shown in Fig. 1. 3.1. Colour
When titanium is immersed in an electrolytic solution

~
as an anode and the current is drawn, the oxygen
rl~ Foe
MCr.HANICAL A~TATN~N generated at the anodic surface combines with the
reactive titatnium to form titanium oxide. This trans-
parent oxide increases in thickness in relation to the
amount of voltage applied with time. As oxide thickens,
its resistance to the passage of current also increases. At
TITAI~JIq
any given voltage this oxide film grows to a very
TEST SPECIMEN specific thickness and then stops when the resistance
ELECTROLYTE
TITANIUM TANK {CATalOg)
has increased to a point where the current has decayed
to a value at which only a few OH- ions are available
..>'-.~ .-~. to support continued film growth. This phenomenon of
-."" .'S-" voltage-controlled oxide thickness indicates that colour
is also voltage controlled.
This thickness vs. resistance relationship dictates that
the area of oxide produced by any specified voltage will
Fig. 1. Schematic representation of experimental set-up for titanium not pass current at a lower voltage. In other words, an
anodizing. area that is anodized at 50 V need not be masked for
50 ,4. K. Slmrma / AtoMizing Ti./or ,Vutce (q~phcations

anodizing adjacent areas at 30 V. It follows that mul- WHITE LIGHT

tiple colour anodizing is to be carried out with decreas- ~ I f R M A L \ THESE REFLECTED RAYS DIFFER
~ IN PHASE AND WHEN BROUGHT
ing voltage without masking the areas already anodized / ",, TOGETHER INTERFERENCE TAKES
at higher voltages. J f N'x~PLACE
The colours that are produced on titanium after
anodizing are known as interference colours [3]. There
are no pigments or dyes. The colours are generated in a
transparent oxide film grown on the metal surface.
Interference colours are produced by thin transparent
?I
YU SI J

i~
S"~SJ
I~
J ANODIC OXIDE FILM

TITANIUM METAL
films on a reflective surface. In nature, these colours can ' \,;SORPT,ON
be seen in an oily, wet street and in the iridescent
colours of some insects.
l-Anodic oxide film thickness.
The oxide film has ability to reflect, refract and x-Wave Length in the medium.
absorb light. White light falling on the oxide film is B-Angle between any internally reflected ray and
the surface normat,
partly reflected and partly transmitted. The transmitted
Phase difference between two successive rays
fraction reaching the metal surface is again partly ab-
sorbed but largely reflected back to oxide film. A phase ~= @ d cos
shift occurs during this process together with multiple
reflections. The degree of absorption and multiple Path difference between two successive rlys
reflection depends on the thickness of the film. The nX = 2d Cos El
coloured light, i.e. light of restricted wavelength, results
Fig. 2. Reflection of light at anodic oxide lihn of titanium.
from the optical interference of light waves [4]. The
process is shown schematically in Fig. 2.
The various colours produced on anodized titanium, 3.2. Effect of temperature
depending on applied voltage, are given in Table 3. The The current density-time curves for titanium anodiz-
oxide films grown on reactive metals (titanium, nio- ing at constant voltage at temperatures of 15, 23, 30,
bium and tantalum) in general have a higher refractive and 38 C are shown in Fig. 3. As expected, at higher
index than diamond. This accounts for the brilliance of temperatures the anodizing current density tends to be
the colour. higher owing to the higher conductivity of electrolyte

T A B L E 3. Physico-optical properteis of anodic tilms of titanium

Sample Operating Colour 2,..... no Film :C' ~:~R'* :q/qR

number voltage of oxide (A,) thickness
(V) film (A)
1 2 Silver 2788 0,482 0.2 I 2.30
2 6 Light brown 3418 3.5499 241 0.583 0.21 2.78
3 10 Golden brown 4205 2.9055 362 1t.657 0.21 3.13
4 15 Purple blue 5255 2.6757 491 0.716 11.21 3.41
5 20 Dark blue 6095 2.5983 586 11.738 0.21 3,51
6 25 Sky blue 7145 2.5457 702 0.733 0.21 3.49
7 30 Pale blue 8195 2.5151 815 0.714 0.21 3.40
8 35 Steel blue 9245 2.4957 926 0.687 /).21 3.27
9 40 Light olive 10295 2.4824 1036 0.659 0.21 3.14
10 45 Greenish 11345 2.4729 1147 0.670 0.22 3.05
yellow
11 50 Lemon yellow 12290 2.4665 1246 0.647 0.21 0.22 3.08
12 55 Golden 12990 2.4627 1319 0.664 0.22 3.02
13 60 Pink 13865 2.4588 1410 0.663 0.22 3.01
14 65 Light purple 15440 2.4584 1573 0.659 0.22 3.t1/)
15 75 Blue 17330 2.4488 1769 0.658 0.22 2.99
16 15 Silver)' 0.486 0.21 2.31
( heat transparent
treated b)

"Average values in entire solar or IR region.

bAt 850 ' C in vacuum.
 A. K. Sharma Anodizing Ti for space applications 51

600

400

~-~ 600
4C
30O z
)-
j-

400
3eoC i..-

i 200 35V

i '100 , 30oC
i 200 )0V
25V
20V
• 23"C 15V
. 1SiC 10V
i I i I i !
i I i l I I I
I 3 5 7 I 3 5 7

TIME IN MINUTES TIME IN MINUTES

Fig. 3. Variation of anodic current density with time for various Fig. 4. Variation of anodic current density with time for various
operating temperatures: electrolyte, 165 g I ~ H2SO4; voltage, 15 V. applied voltages: electrolyte, 165 g I-~ H2504; temperature, 23 °C.

[5, 6]. At temperatures greater than 50 °C, the bath draws

an unusually higher current and the coating obtained is
often patchy and discontinuous, probably because of the
etching action of electrolyte [6]. The preferable temper-
ature range for titanium anodizing is 15-38 °C.

3.3. Effect of voltage 1,00

The next variable studied was the operating voltage.
Figure 4 shows that the anodic current density increases
with increasing applied voltage. A very high current
density was noticed initially. The rate of reduction in
3oo
current density with time is very fast at the beginning; it
slows down gradually and becomes near to constant at
the later stages. A voltage greater than 80 V frequently
resulted in streaking and pitting of the coating.
200
3.4. Effect of electrolyte concentration
The effect of the sulphuric acid concentration on the
anodic current density-time curve at a constant voltage • 200 g/L
of 15 V and at 23 °C is shown in Fig. 5. Because of the
higher conductivity of the concentrated electrolytes
[5, 6], the current density-time curves lie in the higher
current density region. The addition of a little oxalic acid ! i i I I i i
was found to improve the smoothness and uniformity of 1 3 5 7
the. coatings; this is probably because it helps in the TIME IN MINUTES
reduction of dissolution of coatings in sulphuric acid. Fig. 5. Variation of anodic current density with time for various
Because of the introduction of the so-called " c o m m o n concentrations of H2SO4: temperature, 23 °C; electrolyte, H2SO4
ion effect" the anodic current density is lower for ( t ) , H2SO4 in presence of oxalic acid (©).
52 A. K. Sharma / Anodizing Ti./br ~q)ace application,s

v1
the mixed electrolytic system sulphuric acid oxalic
ISV
acid than observed only for sulphuric acid electrolyte. (HT)
35
The optimum concentration of anodizing electrolyte is
shown in Table 2.
75V
t0V i
3.5. Optical properties
The passive thermal control system utilizes the optical
5V
properties of its components. The equilibrium tempera- ~5V
Z5V
ture of a spacecraft and its subsystems is controlled by
the ratio of solar absorptance c%to IR emittance aiR- The
optical properties of all the specimens were measured
before and after humidity, thermovacuum performance,
thermal cycling and repetitive high temperature heating
tests. A solar reflectometer, version 50, model SSR-ER,
and emissometer, model RD-1, from Devices and Ser-
vices Co., U.S.A., were used for measurement of solar
absorptance and IR emittance respectively [7]. These
instruments provide an average value of solar absorp-
tance and IR emittance digitally in the entire solar or IR ~ Zv

region. These average values of solar absorptance and IR ~

emittance are of importance in thermal design of various
systems. The optical properties of titanium anodized at
different operating voltages are listed in Table 3. 0.2 0.62 1.04 'l.&4 1.80 2.30 0.2 0.42 1.0& 13,6 1.80 2.30
The solar spectra of anodic films were recorded on a
WAVE LEN6TM Jura)
Varian Cary 2300 spectrophotometer, Australia, in the
0.2 2.3 gm region. IR spectra of films were recorded in Fig. 6. S o l a r r e f l e c t a n c e o f t i t a n i u m anodized at d i f f e r e n t o p e r a t i n g
the 2.5-50 gm region with a Perkin Elmer 580 B IR voltages.
spectrophotometer. No difference in the IR spectra of
anodic films obtained at different operating voltages is, 100
however, observed. The solar and IR spectra of anodic
oxide films are shown in Figs. 6 and 7 respectively.
/Z

/
3.6. Measurement of fihn thickness
The thickness of anodic oxide films was calculated 6o
using the following optical expression [8]: ,,--,

t = ;t.... /4no ~,o

where t (,~) is the film thickness, )~.... is the wavelength

of maximum film absorbance and no is the refractive 20
index of the film (ordinary rays).
The refractive index of the titanium oxide film which
101 m J I I l l I
varies with wavelength was calculated with the help of
2.5 3 S 7 10 20 50
the following equation [9]:
2.441 x 1 0 7 WAVE LENGTH Ijum)
n02 = 5.193 + Fig. 7. I R reflectance of anodized titanium: . anodized at
0.803 x l 0 7
)~2na x - -
different voltages; , anodized and heat treated ( 15 V, 850 C).
2. . . . no and thickness of anodic films obtained at
different operating voltages are presented in Table 3. 3. 7. Effect o f humidity
The interference method of thickness measurement This test is designed to check the effect of humidity and
becomes invalid for ultrathin films because of their high Indian on-shore temperature which is likely to be
island nature (agglomeration of thin film island encountered at launch sites; this in turn reveals the
growth). Therefore the thickness of anodic films below resistance of the coating to corrosion. The test was
200 ,~ (e.g. as obtained at 2 V) could not be measured. conducted in a thermostatically controlled Heraeus
The variation in anodic film thickness with applied Votsch type 08,500 (F.R.G.) humidity chamber for 96 h.
voltage is shown in Fig. 8. The relative humidity in the chamber was maintained
 A. K. Sharma / Anodizing Ti for space applications 53

/
high temperature heating on the optical properties of
anodic oxide coatings of titanium. Titanium sheets are
often used as heat shields to protect the satellite-subsys-
tems from the enormous heat generated by jet engines
1600 - during firing. As the passive thermal control utilizes
optical properties of its components, for reliable thermal
o
control evaluation it is necessary to examine the effect of
repetitive high temperature heating on the optical prop-
1200 erties of anodic film. The specimens were heated to 850 °C
in vacuum (10-STorr) at a controlled heating rate of
t/)
10 °C min-1 for 3 h. The anodic oxide coating was found
Oq
to change or lose its colour above 550 °C. The stabilized
800
specimens were then subjected to repetitive heating under
p--
5"
a vacuum of 10 -5 Torr at 850 °C for 1 h and their optical
properties and physical appearance after subsequent
heatings were examined. No change in physical appear-
ance and no degradation in the optical properties after
tO0 repetitive heating were, however, observed.

I 1 t I i I I I I I 4. Conclusions
15 30 ~5 60 75

APPLIED VOLTA[iE IV) Titanium was anodized in dilute sulphuric acid elec-
trolyte under constant voltage conditions. The main
Fig. 8. Variation of anodic film thickness with applied voltages:
conclusions of this study are given below.
temperature, 23 °C; electrolyte, 165 g 1-1 H2SO 4.
(1) The colours that are produced on titanium after
anodizing are interference colours. The thickness-resis-
at 95% +__0.5% at 50 _+ 1 °C. Test specimens were then tance relationship allows multiple colour anodizing to
visually examined under a magnification of 16 × and be carried out with decreasing voltage without masking
their optical properties (~s and ~IR) were measured. No the areas already anodized at higher voltages.
change in optical properties and no sign of any kind of (2) The anodic current density was found to increase
degradation, e.g. corrosion spots, patches or discol- with increasing applied voltage for a constant value of
oration of coatings, was observed. the electrolyte temperature and to decrease with de-
creasing electrolytic temperature for a constant value of
3.8. Effect of thermal cycling and thermovacuum the applied voltage.
performance tests (3) Because of the higher conductivity of the concen-
A spacecraft experiences extreme temperatures in or- trated electrolytes, the anodic current density was found
bit. This test is designed to evaluate the effect of cycling to increase with increasing concentration of sulphuric
temperature on the coatings. Anodized specimens were acid for a constant value of the electrolyte temperature
subjected to a thermal cycling test in a Brabender Real and applied voltage. The addition of a little oxalic acid
Test Umweltsimulatoren hot and cold chamber (F.R.G.). of the sulphuric acid improved the appearance and
A total of 1000 cycles were applied. A cycle consisted of uniformity of the oxide film.
lowering the temperature to liquid nitrogen temperature (4) The oxide coatings obtained in sulphuric acid
(about - 196 °C), a dwell of 5 min, raising the tempera- electrolyte are highly stable. The humidity, repetitive
ture to 150 °C, and a dwell of 5 min. The first ten cycles heating, thermal cycling and thermovacuum perfor-
were carried out in a thermovacuum chamber under a mance tests have no effect on the physical and optical
vacuum of 10 -5 Torr (thermovacuum hot and cold soak properties of the coatings. These coatings are therefore
performance test). These cycles were of 1 h each instead extremely suitable as a thermal control surface for
of 5 min with the lower temperature limit of - 180 °C. space applications.
After thermal cycling all the specimens were inspected
visually under a magnification of 16 x and their optical
properties were measured. No degradation was observed. Acknowledgments

3.9. Effect of repetitive heating The author wishes to record his sincere thanks to A.
This test is designed to evaluate the effect of repetitive V. Patki, Group Director, H. Narayanamurthy, Division
54 A. K. Sharma / Anodizing Till{~r ,~paC{" applications

Head, and H. Bhojaraj, Section Head, Mechanical Sys- 2 L. Young, Anodic Oxide Films, Academic Press, London, 1961.
tems, for their constant encouragement during the 3 B. Seeley, Met. Finish., 84 (8) (1986) 11.
4 J. B. Cotton and P, C. S. Hayfield, Trans. Inst. Alet. Finish., 45(2)
study. Thanks are also due to J. Md. Mohideen for his
(1967) 48.
assistance and cooperation during this work. 5 V. Balasubramanian. S. John and B. A. Shenoi, SurL Teclmol., 19
(1983) 293.
6 S. John, V. Balasubramanian and B. A. Shenoi, Sm:ll Technol., 26
( 1985)207.
References 7 A. K. Sharma, Trans. Inst. Met. Finish., 67 (3) (1989) 87.
8 H. K. Pulker, Thin Fihn Science and Technology, Vol. 6, ()mtin,gs
1 Hel Badekas and Chr. Panagopoulos, Sutj~ Coating Technol., 31 on Glass, Elsevier, Amsterdam, 1984, 293 pp.
(1987) 381. 9 ,1. R. De Vote, .1. Opt. Soc. Ant., 41 (1951) 417.