U.S.
Department of Commerce, Bureau of Standards
RESEARCH
PAPER
RP650
Part
of
Bureau
of
Standards
Journal
of
Research,
Vol.
12,
February 1934
SOME
PHYSICAL
PROPERTIES OF
PLATINUM-RHODIUM
ALLOYS
By
J.
S.
Acken
1
abstract
In a
continuation of the
study
at the
Bureau of the metals of
the platinum
group, a
series of
platinum-rhodium alloys containing approximately
10, 20,
40,
60,
and 80
percent rhodium
was prepared
from pure
platinum
and pure
rhodium.
The
melting
point, hardness, density, electrical resistivity, temperature
coefficient
of
resistance, and
thermal electromotive force against
platinum
were
determined
for
each alloy.
The microstructure of each alloy appeared to
be
that
of
a
solid
solution.
The particular qualifications of
the
alloys
containing
20
to
40
percent
of
rhodium,
for
use as resistance
furnace windings for service
at
high
temperatures,
are discussed.
CONTENTS
Page
I.
Introduction
249
II.
Preparation
and mechanical working :
250
III.
Physical
properties
252
1.
Melting point
252
2.
Hardness
255
3.
Density
256
4.
Electrical resistivity
256
5.
Temperature coefficient of
electrical resistance
256
6.
Thermal electromotive
force against
pure
platinum
256
7.
Structure of
platinum-rhodium alloys 257
IV.
Discussion
257
I.
INTRODUCTION
As part
of the
study
at
the Bureau of
Standards
of the platinum-
group metals
of
high
purity, and to
supplement the determination of
a number of
physical properties of pure rhodium,
the preparation
and
properties of platinum-rhodium
alloys were given
consideration.
Since rhodium
is much less abundant
than platinum and the alloys
high in rhodium
are
not
easily worked, interest
in platinum-rhodium
alloys has
been
limited largely
to
those containing less
than
20 percent
of
rhodium. Alloys containing about 10
percent of rhodium are
used
as
positive
elements in
thermocouples, as
catalytic gauzes
for the
oxidation of ammonia to nitric acid,
and to some
extent as
the heating
element in resistance furnaces. Alloys
containing
3
to 4 percent
of
rhodium are used for
crucibles
and
similar laboratory ware.
Published
information
on
the physical
properties
of platinum-
rhodium alloys
is
scattered and
incomplete. Some
of the
early work
is
of doubtful value since reliable
methods for
the
separation, purifica-
tion,
and analysis of the metals of the
platinum group were
not avail-
1
With
the cooperation
of members
of
the Bureau
staff,
as
indicated
by footnote to
several
sections of this
paper.
249
250
Bureau
of
Standards Journal
of
Research
[Voi it
able until
recently. Some of the more important early
investigations
were
those of Le
Chatelier,
2
Dewar and Fleming,
3 4
Holborn
and
Wien,
5
and Barus.
6
Later, when somewhat better
methods for
purification had been developed, Burgess,
7
Sosman,
8
and
Adams
9
studied the use of platinum-rhodium alloys for measuring high
temperatures. Subsequently,
Neville
10
described
in detail a method
for the preparation of
the
10 percent rhodium
alloy
of
high
purity
for use
in thermocouples.
Data
on
hardness,
tensile strength,
electrical resistivity,
and temper-
ature coefficients of electrical resistance of
platinum-rhodium
alloys
containing
3.5, 10, 20,
and 50
percent
of
rhodium were
presented by
Carter;
11
Von
Wartenberg
12
determined the melting
point of
the 10
percent alloy, and Muller
13
the
melting
points of a series
of platinum-
rhodium alloys.
In
the present
investigation the
"melting point",
density, hardness,
electrical resistivity, temperature
coefficient of
electrical resistance, and
thermal electromotive
force against platinum
were determined for a
series
of alloys which covered the range from
pure platinum
to
pure rhodium.
II.
PREPARATION
AND MECHANICAL WORKING
The alloys were prepared from platinum and rhodium purified
in
the Chemistry Division
of
the Bureau
by
the
methods
described
by
Wichers, Gilchrist,
and
Swanger.
14
Spectroscopic analysis
of
the
platinum
and
rhodium showed that the
total amount of
metallic
impurities in each
was
definitely
not
over 0.01 percent, probably
not
over
0.001
percent.
Platinum-rhodium
alloys of a nominal rhodium content of
10,
20,
40,
50, 55,
60,
and
80
percent, respectively, were prepared in amounts
of
50 to
75
grams
by
melting in a high-frequency induction furnace.
16
For all of the alloys,
the constituent metals were prepared
in the
form
of
pellets made by compressing the spongy metal and
sintering
the
compressed
pellets
on a block of
lime with
an oxy-hydrogen torch.
The alloys
containing 60 percent or less of rhodium were melted in
air
in zirconium-oxide
crucibles and the 80 percent rhodium alloy in
vacuo in a thorium-oxide crucible. The alloys containing
10
and
20
percent of rhodium were cast into
graphite
molds; the others were
allowed to freeze in the crucible. All appeared to
be
homogeneous.
2
H. Le Chatelier,
De
La
Mesure des Temperatures
Elevees
par les Couples
Thermo-Electriques. Jour,
de
Phys.,
(2),
vol.
6,
pp.
23-31,
1887.
3
J. Dewar
and J.
A. Fleming, On the Electrical Resistance of Pure Metals, Alloys and Non-Metals at
the Boiling Point of Oxygen. Phil. Mag.,
(5),
vol.
34,
pp.
326-337, 1892.
4
J. Dewar and
J.
A. Fleming, The Electrical Resistance of Metals
and
Alloys
at
Temperatures Approach-
ing
the Absolute Zero.
Phil.
Mag.,
(5),
vol.
36, pp.
271-299, 1893.
5
L.
Holborn and W. Wien, Ueber die Messung hoher Temperaturen. Ann.
Physik. vol.
47, pp.
107-134
1892.
6
C. Barus,
Thermoelectrics of Platinum-Iridium
and of
Platinum-Rhodium.
Phil. Mag.,
(5),
vol.
34,,
pp.
376-381, 1892.
'
G. K. Burgess, On
Methods of Obtaining Cooling Curves.
B.S. Bull., vol.
5, pp.
199-225,
1908.
8
R.
B.
Sosman, The
Platinum-Rhodium Thermoelement from
0°
to
1,755°.
Am.
Jour.
Sci.,
(4),
vol.
30,
pp.
1-15,
1910.
9
L. H. Adams,
Calibration Tables
for Copper-Constantan and
Platinum-Rhodium Alloys. Jour.
Am.
Chem. Soc,
vol.
36, pp.
65-72; 1914.
10
R. P. Neville,
The preparation of Platinum and of Platinum-
Rhodium
Alloy
for Thermocouples.
Trans.
Amer. Electrochem.
Soc, vol.
43,
pp.
371-384,
1923.
»
F. E.
Carter, The
Platinum Metals
and
Their Alloys, A.I.M.E., Proc,
Inst. Met. Div.,
pp.
759-785,
1928.
12
H,
V.
Wartenberg, tjber
optische
Temperaturmessung blanker Korper, Verhandl. deut.
Physik.
Ges.,
vol.
12,
pp.
121-127,
1910.
13
L. Muller,
Bestimmung
der
Schemlzpunkte von Platinlegierungen.
Ann. Physik
(5),
vol.
7, pp.
9-47,
1930.
14
E.
Wichers,
R.
Gilchrist,
and W. H. Swanger,
Purification of the Six Platinum Metals. Trans.
A.I.M.E.,
vol.
76,
pp.
602-630,
1928.
i*
The
technique
of melting,
casting, and working these
alloys
is
essentially that described
in references
10 and 18.
Ackeri]
Properties
of
Platinum-Rhodium Alloys
251
The
ingots were
hand forged
at
1,100
C
(approximate)
to
bars
about
10 mm
in diameter. The alloys
containing
40
and
80
percent
of
rhodium were extremely coarse
grained in
the initial
state,
i.e.,
furnace
cooled,
and were not forgeable. However, after the ingots
had been
remelted on lime
with an oxy-hydrogen flame, they
could
be forged.
Each
of the hot-forged
bars was swaged
to wire about
1 mm
in
diameter. The
10
percent alloy was swaged
at room
temperature,
the others
at
900
to
1,000
C.
The
20
percent
alloy
could be swaged
at
room
temperature
if
annealed after
each reduction.
Further reduction of
the 1 mm swaged wires
was accomplished
by
drawing through dies.
The alloys containing
more than 40
percent
of
rhodium
were
drawn hot through
tungsten
carbide dies
until
the
diameter was reduced to 0.5
mm. The temperature
of the wire
was
maintained
at about 900 C
for the
first few draughts, and
then
gradually
lowered
for successive draughts
to
about
550 C
for
the
draught through the 0.5
mm die. Reduction
to 0.2 mm
diameter
was accomplished
by
drawing
through
sapphire dies,
at room
tem-
perature and without
annealing. The
alloys
containing
less than
40
percent
of
rhodium were drawn at room temperature
from
1 mm
swaged bar
to
wire of 0.2 mm
diameter.
Determinations
of
the various physical properties
were
made
during
the course
of
the
mechanical working when
the
form
was
suitable
for specimens
for the particular determination
desired.
Before any
determinations
were made the specimens
were cleaned
by
immersion in
molten
potassium bisulphate. This removed
surface
contamination,
especially
iron,
accumulated
during
the
mechanical
working.
The nominal and
the actual compositions of the principal
alloys,
as determined
by
chemical analysis,
16
are given in
table 1. This
series
of alloys was used for most of the
determinations
of properties
but alloys similar, although
of
slightly different
compositions, were
substituted
for the
18.7, 37.5,
and
77.1 percent alloys in the
determi-
nations
of temperature coefficient
of resistance, resistivity, and
density.
Table
1.
—
Composition
of
platinum-rhodium
alloys
Nominal Found by analysis
Rhodium Platinum Rhodium
Platinum
Percent
10
20
40
50
55
60
80
Percent
90
80
60
50
45
40
20
Percent
10
18.7
37.5
51.6
56.6
61.2
77.1
Percent
90
81.3
62.5
48.4
43.4
38.8
22.9
w
The
analyses
were made
by
Raleigh
Gilchrist,
Associate
Chemist.
Ignition at dull-red
heat
with
sodium
chloride in
an atmosphere of
chlorine
was
used
in getting
the alloys,
particularly those high in
rhodium,
into
solution.
The
details
of
the
method
of
analysis
are described in a pending paper.
252
Bureau
of
Standards Journal
of
Research
[Vol. 12
III. PHYSICAL
PROPERTIES
1. MELTING
POINT
Determinations
were made
of
the approximate " melting
point"
of
each
of the alloys
containing
10,
18.7, 37.5, 56.6,
and
77.1 percent
of
rhodium.
Platinum and rhodium are generally believed
to form
a
continuous
series of
solid
solutions
and the melting or freezing
in
each case occurs
over
a
range of
temperature rather than
at a
single
temperature.
The
apparatus
and
procedure
were similar
to
those used in
previ-
ously
reported
determinations of the
melting
points of high-purity
platinum
and rhodium.
17
18
An
ingot
(50
to 75
g)
of the alloy
with
a
hole
drilled
axially from
top to bottom was
placed
in
a thorium-oxide
crucible
fitted with
a
thorium-oxide sight-tube which
was inserted
2000
rtGHT
ptR
C£N
T
O
1900
Id
0:
W
,""
u-«
/
s
3
£
1800
2
<
id
t
{'?
1700
Figure
1.
20 40
60
80
!O0
COMPOSITION,
PER
CENT RHODIUM
Melting
Points"
of
alloys
of
platinum
and rhodium.
in
the
hole.
The
power input
to the
high-frequency
induction
furnace
which
was
used for
melting
the
ingot
was
controlled
so that several
observations
of
the
temperature
could
be made during
melting
or
freezing.
The
alloys
containing
40 percent
or less
of rhodium
were
melted in
air;
those
of higher
rhodium
content
in an
evacuated
furnace
tube,
under
pressure
of less
than 1
mm of
mercury.
Temperatures
were
measured
with
a
Leeds
and
Northrup
optical
pyrometer
which
was
calibrated
by
the
pyrometry
section of the
Bureau
before
the
observations
were
made and
again
after they had
been
completed.
The
calibration
of this
pyrometer
was certified
to
± 20
C
for
temperatures
between
1,700
and
2,000
C. Consequently
the
accuracy
of the
temperature
observations
cannot exceed the
accuracy,
±20
C, of the
pyrometer
calibration.
» Win.
F.
Roeser, F.
R.
Caldwell
and
H.
T.
Wensel,
The
Freezing Point
of
Platinum,
B.S. Jour.
Re-
search,
vol.
6, pp.
1119-1129,
1931.
i
8
Wm.
H.
Swanger,
Melting,
Mechanical
Working
and
Some Physical Properties
of Rhodium, B.S.
Jour.
Research,
vol.
3,
pp.
1029-1040,
1929.
Acken]
Properties
of
Platinum-Rhodium Alloys
253
In each
of the
time-temperature curves summarizing the observa-
tions
during
melting
and
freezing of the different alloys only one break
was detected
instead of
the two that might
be expected at
the solidus
and liquidus
temperatures
of a solid-solution
alloy. In every case,
the
break was
pronounced,
and temperatures
that were constant,
within the
uncertainty
to
which they
are
given,
were obtained
both
on
melting
and on
freezing over periods of from 2
to 5 minutes. For
each of the
alloys there was close
agreement between
the temperatures
at
which
melting and freezing occurred, which indicates
that the range
between
the solidus
and liquidus temperatures is
probably small.
Evidently, at
the
high
temperatures at
which the
observations
were
made,
the change
in the rates of heating or cooling at the
solidus and
liquidus
temperatures was too slight
to be
detected with the
optical
pyrometer
and
the
procedure
employed.
Table
2.
—
"Melting points"
of
platinum-rhodium
alloys
Composition
Arrest on
heating
Arrest
on
cooling
"Melting
point"
»
Rhodium
(percent)
Platinum
(percent)
10
Average
90
°C
f
1,848
\
1,
845
[
1,845
°C
1,845
1,851
1,845
[
1, 845±20
\
1,
900±20
1 1, 945±20
\ 1,940±20
I
1,
960±20
1,846 1,847
18.7
81.3
I 1,899
I 1, 906
1
1, 889
I
1,906
1,890
1,905
1,880
1,899
1,900 1,894
37.5
Average
62.
5
f 1,940
1,946
{
1,
951
1,949
I
1,943
1,953
1,942
1,933
1,942
1,930
1,946 1,940
56.
6
Average
43.4
( 1,955
J
1,
946
1 1,
934
I
1,948
1,936
1,938
1,933
1,945
1,946
1,938
77.1
Average
22.9
f
1,962
1
1, 965
I 1, 957
I
1,967
1,
964
1,963
1,965
1,
954
1,963 1,961
»
The limit of
accuracy (±20C) ascribed to each
"melting
point
"
is the limit of
accuracy of the pyrometer
calibration.
From
3
to 5 sets of
temperature observations
were made during
melting
and during freezing
of each of the alloys. The temperatures
during
the period in which
they
remained constant
(while
melting
or
freezing
occurred)
are given in table
2.
The data show
that the
precision
of the experimental results is well within the limits
of accu-
racyof
the pyrometer calibration. The average
of the arrests during
heating
and
cooling of each alloy, rounded
off to the nearest
5 C,
is
designated
as
the
"
melting point."
Actually
this temperature is
254
Bureau
of
Standards
Journal
of
Research
[Vol.
it
somewhere
between
the
solidus and
the
liquidus
temperatures.
In
figure
1,
the
"
melting
points"
are
summarized
for
the various
platinum-rhodium
alloys,
together
with the melting
(or freezing)
points
of platinum
1,773.5
C,
19
and of rhodium,
1,966
C.
20
The melting
140
6
9
VICKERS
120
r"
®~
x
X
BABY
BRINELL
100
-i
X
a
X
80
if)
/
/
*
*
B
ROC
s
m
:kwell "e'
1
1
kJ
U
7
\m!
tr
2
60
Z
o
Ml
»
(0
u
z .
a
4Q
i
<
X
20
i
1
20 40
60
"
80 100
COMPOSITION, PER
CENT
RHODIUM
Figure
2.
—
Hardness
of
platinum-rhodium
alloys.
points
of all the
alloys
lie
between
those
of
pure platinum
and
pure
rhodium.
When
plotted
on the basis of weight
percent,
the melting
point curve
rises
sharply from
the melting point
of pure
platinum
as
the rhodium
content is
increased
to about 40 percent.
Further
increase
in
the rhodium
content, however,
is accompanied
by
rela-
1
9
See footnote
»
p.
252.
86
Determination
by W. F.
Roeser and H.
T.
to
be
Wensel.
published subsequently in this
journal.
Acken]
Properties
of
Platinum-Rhodium Alloys
255
tively
small increases in melting
point.
If the
results are
expressed
in atomic percent,
a smoother
curve
is obtained and
the
rapid
increase
in
melting point with additions of rhodium
up
to
40 percent,
is
less
apparent.
o
.004
5
2
x.003
o
a.
UJ
Qu
o°
a
.00 2
UJ
CL
<n
2
I
O
,001
O
10
1
2
x
O
a.
o
2
1
1
1
r
DENSITY
AT
25°
C
I
1 1 l
i
1
i
r^
RESISTIVITY AT
20°
C
25
20
10
2
u
a
OL
2
<
(X
o
20 40
60 60
COMPOSITION,
PER CENT
RHODIUM
100
Figure
3.
—
Density,
electrical
resistivity,
and
temperature
coefficient
of
resistance
of
platinum-rhodium alloys.
2.
HARDNESS
The
hardness
was
determined
by
means
of
the Baby
Brinell,
Rockwell, and
Vickers
hardness
testers.
The ingots
previously
used
for
the
melting
point
determinations
were
hot forged
to square
bars
and
annealed;
L
at
1,200
C before
the hardness
impressions
were
made.
256 Bureau
of
Standards Journal
of
Research
[Vol. is
The
results on a weight
percent
basis are shown
graphically in
figure
2.
The
10
percent
rhodium
alloy is decidedly
harder than
pure platinum.
With increased
rhodium contents
above 10 percent
the hardness
increases
at
a slower rate, and apparently
reaches
a
maximum
at about
70
percent
of rhodium.
With further increase in
rhodium
content the hardness
decreases slowly, and
approaches
the
hardness
of pure rhodium.
The Rockwell hardness
values
have also
been
plotted on the
basis of atomic percent,
to show that this method
of
presentation
tends
to smooth out the curves particularly in
the
region
of low
rhodium
contents. The values obtained agree
well
with
the Brinell
hardness numbers
reported
by
Carter,
21
on pure
platinum and
on alloys containing
10,
20,
and
50 percent of rhodium.
3.
DENSITY
22
Density
determinations
were
made
on wires
1 mm
in diameter
which
had
been
annealed
at
approximately
1,500
C.
The
values
shown in
figure
3
indicate
that the density of the platinum-rhodium
alloys
decreases regularly
with
increasing
rhodium content.
4. ELECTRICAL RESISTIVITY
23
The
electrical resistivity
was
measured on wires 1 meter
long
by
1
mm in
diameter, annealed
at a temperature of
approximately
1,500
C.
The
results are plotted in figure
3.
The increase
in resistivity
is
fairly
rapid with increasing rhodium content up
to
20
percent;
above
20
percent
of rhodium the resistivity decreases
gradually to
that of pure rhodium.
A slight
irregularity is noticeable
in the
curve
between
50
and
60 weight
percent
of
rhodium content.
Similar
relations are
apparent when the results are plotted
on the basis
of
atomic percent,
but the curve is smoothed out and
becomes
more
symmetrical.
5.
TEMPERATURE
COEFFICIENT OF
ELECTRICAL RESISTANCE
24
The results of the measurements
which
were
made on
wires
0.2
mm in
diameter,
show (fig.
3)
that
the
addition
of rhodium to
plati-
num
causes a
rapid
decrease
in the
coefficients of
resistance
which
reach
a
minimum
value
for alloys
containing 20
to
40
percent
of
rhodium; with higher rhodium contents
the values
again increase
regularly
and
approach
the
value for pure
rhodium.
If
these
results
are plotted on the basis of atomic
percent,
the curve
becomes
smooth
and nearly symmetrical, as would be
expected
for
a
series of
solid
solution alloys.
6. THERMAL ELECTROMOTIVE
FORCE
AGAINST
PURE
PLATINUM
Some
of the alloys
prepared
for this
investigation
were
included
in
a
series of
10
platinum-rhodium
alloys whose
thermoelectric
prop-
erties over the
temperature
range 100 to
1,200
C
have already
been
published.
25
21
See
footnote
11, p.
250.
22
E. L. Peffer,
Chief,
section
of capacity and
density.
23
A. R.
Lindberg,
junior
electrical
engineer, section
of electrical
resistance.
24
F.
R. Caldwell, Assistant
Physicist,
pyrometry section.
2
«
Frank R. Caldwell, Thermoelectric
Properties of
Platinum-Rhodium
Alloys. B. S.
Jour.
Research,
vol.
10,
pp.
373-380,
1933.
.
1
%
~\
.M>^
Acken]
Properties
of
Platinum-Rhodium
Alloys
257
7.
STRUCTURE OF
PLATINUM-RHODIUM
ALLOYS
26
Two
previous
investigators
27
have stated that
platinum and rho-
dium
form
a
continuous
series
of solid
solutions. In
the present
investigation,
slight
irregularities
in
melting point, Brinell
hardness,
electrical
resistivity, and
thermoelectric force of the alloys
containing
about 55
percent of
rhodium suggested the
possibility that there
might be some
departure from a
simple
solid
solution
at
or near
this
composition.
The presence
of impurities, to account for these irreg-
ularities, could
not be
established
by
chemical and spectrographic
analyses.
Examination
of
the
microstructure
failed
to reveal evidence
of
any
phase except a
solid solution in any of the
alloys used in this
investi-
gation.
The specimens used for
the microscopic examination
were
swaged wires,
1 mmin diameter,
which had
been
annealed
at
1,500
C.
Figure
4 shows the microstructure
of the alloy containing 37.5
percent
of rhodium. This was typical of the
structure of the other alloys
of
the series.
X-ray diffraction patterns made
from filings
of each of the alloys
likewise failed
to
reveal any
significant
difference
in
structure to
account
for
the
irregularity in properties of some of
the
alloys.
Evi-
dence
of the
existence of phases other
than
a
simple solid
solution
was
not
obtained.
IV. DISCUSSION
When
the
technic
for
the
production
of
rhodium
wire was devel-
oped
28
it became possible
to use
rhodium or platinum-rhodium
alloys
of
high
rhodium content for
windings
on furnaces which were
to be
operated
at
temperatures higher
than
could
be
reached with
plati-
num. Pure rhodium
and an alloy containing
80
percent of rhodium
and 20 percent of platinum
have
been
successfully used at this
Bureau
to
maintain
furnace
temperatures
above
the melting point
of
plati-
num.
However, the melting point
curve
shows that
but little advan-
tage can
be
gained
by using alloys containing more than
40 or
50
percent of rhodium. The
40
percent
alloy is much easier
to
work than
the
80 percent alloy or
pure rhodium and hence should be cheaper,
at
least at
present prices
for rhodium and platinum.
It
also has greater
resistivity
and
a lower temperature
coefficient
of
resistance (both
measured
at
ordinary
temperatures) than the alloy of
higher rhodium
content.
M
R.
E. Pollard and R.
Q. Kennedy, Jr.,
Junior Metallurgist and
Assistant Metallurgist,
respectively.
»
See footnotes 11 and
13,
p.
250.
«
See
footnote
18, p.
250.
-34
1
258
Bureau
of
Standards Journal
oj Research
[vot. 12
It will
be
noted that
the
temperature
coefficient of
resistance is
at
a
minimum for
alloys
containing
20 to 40 percent of rhodium
and that
the resistivity is at its maximum between 10 and
40 percent.
In
these
respects
an
alloy
containing
about 20 percent of rhodium
is
evidently superior
to
the ordinary platinum
winding.
Such an alloy
can be
drawn into wire
nearly as easily as
platinum
and weighs
about
15 percent less per unit length. Observations of the 10
and 20 per-
cent
alloys in
service indicate that they deteriorate less
rapidly
than
platinum, under comparable conditions. For all these reasons they
are
preferable
to
platinum for
use
in furnace windings
as long as the
cost
of rhodium
does
not rise very
much
above that of platinum.
Only
the requirement of a
higher operating temperature
would appear
to
justify
the use
of
the
less easily worked alloys of
higher
rhodium
content.
Washington, November
10,
1933.
