46 / 62
FEATURE
A D V A N C E D M A T E R I A L S & P R O C E S S E S | M A R C H 2 0 1 6
4 6
12
MINIMIZING ALPHA CASE DURING VACUUM
FURNACE HEAT TREATING
Understanding detrimental alpha case formation during heat treatment of titanium parts is
increasingly important as titanium use in aerospace and medical applications continues to grow.
Donald Jordan, FASM,* and Virginia Osterman,*
Solar Atmospheres Inc., Souderton, Pa.
A
lpha case is a diffusion reaction that occurs at the
surface of titaniumwhen processing at elevated tem-
perature in atmospheres containing oxygen, nitro-
gen, and/or carbon, with oxygen as the prominent element
associated with alpha case. Oxygen is solution strengthen-
ing at lowconcentrations, but greatly decreases ductility and
forms alpha case at higher concentrations. Thus, alpha case
is brittle and has a detrimental effect on part performance
and longevity. Higher temperatures increase alpha case.
Above 480°C (896°F), air or water vapor begins to produce
alpha case. Temperatures less than 550°C (1022°F) limit oxy-
genmobility and keep the case depth from increasing
[1]
.
To minimize alpha case, high integrity titanium parts
are often heat treated in vacuum furnaces to avoid having
to remove the case by machining or pickling. The majority
of today’s production vacuum furnaces are insulated with
graphite felt. At lower temperatures in vacuum, water vapor
is the principal concern for oxidizing titaniumwhen H
2
O dis-
sociates. It is known in industry that water vapor is difficult to
“pump out” at low temperature in vacuum. As the tempera-
ture increases, water vapor present in the vacuum chamber
will be “driven out,” oxidizing titaniumon the ramp-up to the
cycle hold temperature. At higher temperatures, water vapor
decreases while CO
2
and CO increase, providing additional
sources for alpha case formation
[2]
.
In an attempt to minimize oxidation, industry practi-
tioners slow down the ramp rate and incorporate tempera-
ture holds if outgassing exceeds a certain pressure
[3]
. Sound
practice requires an initial pumpdown to 1 x 10
-4
torr or
lower (AMS2769 specification) and relatively slow ramp at
600°F/hr; if outgassing occurs, hold until the pressure drops
to 2 x 10
-4
torr. Such procedures negatively affect production
time. The current study looks at whether such protocols are
reliably effective in reducing alpha case formation.
Color is often used as a post-welding and heat treat-
ment criterion to indicate presence of alpha case. This study
examines whether there is any correlation between color
and extent of alpha case.
After polishing and etching, alpha case is visible under
a microscope as a white-appearing microstructure zone, or
alpha phase. The literature lists three principal etchants for
revealing titanium alpha case: Kroll’s reagent, 2% HF, and
Kroll’s reagent followed immediately by 2% ammonium bi-
*Member of ASM International
fluoride. Etchant dwell time is an important variable in ob-
taining reliable results.
PROCEDURE
Ti-6Al-4V sheet was cut into 13 coupons approximately
1.5 in. square. Twelve samples were divided into six pairs for
use in six separate heat treat cycles. The 13th sample was re-
tained as the non-heat-treated baseline (virgin) for metallo-
graphic analysis. Additional titanium sheet was used in Test
5 to increase surface area by a factor of nine compared to the
other five tests.
The furnace used for all tests was a cylindrical, vertical
vacuum furnace—10-in. diameter × 18 in. high—with graph-
ite felt insulation and graphite heating elements. An Ametek
residual gas analyzer (RGA) was attached to the furnace. The
RGA captures a sample of residual gases in the furnace hot
zone to provide a trend analysis of relative gas composition
and pressure during the cycle.
Two ramp rates, one relatively slow and the oth-
er fast, and two hold temperatures were used for five
test cycles. All cycles began after an initial pumpdown to
1 x 10
-4
torr. The one-step fast ramp rate was 1200°F/hr to
the hold temperature. The two-step slow ramp rate was
300°F/hr to 900°F, then 600°F/hr to the hold temperature. The
two hold temperatures were 1450°F and 1750°F with a one-
hour hold time. The sixth test cycle used a three-step ramp
rate as follows: 300°F/hr to 600°F, one hour hold; 300°F/hr to
900°F, one hour hold; 600°F/hr to 1450°F, one hour hold.
Two coupons were hung on separate molybdenum
wires attached to the lid of the furnace for each cycle. One
coupon was intended to be color analyzed using a Hunt-
erLab spectrophotometer. The second coupon was used
for metallographic analysis of alpha case. Specimens were
etched using either Kroll’s reagent, 2%HF, or Kroll’s followed
immediately by 2% ammonium bifluoride. Comparisons
were made as to which etchant best delineated alpha case
along with the effect of dwell time.
RESULTS AND DISCUSSION
The appearance of coupons from all tests reveals that
color is not a distinguishing material attribute (Fig. 1). It is
more accurate to say that the samples vary in reflectivity