Thermophysical Measurements On Liquid Iron and Nickel: High Temperatures-High Pressures January 1987
Thermophysical Measurements On Liquid Iron and Nickel: High Temperatures-High Pressures January 1987
Thermophysical Measurements On Liquid Iron and Nickel: High Temperatures-High Pressures January 1987
net/publication/277665685
CITATIONS READS
19 43
3 authors, including:
G. Pottlacher
Graz University of Technology
113 PUBLICATIONS 1,241 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by G. Pottlacher on 03 June 2015.
Abstract. Wire-shaped samples of nickel and iron, 40 mm long and 0.25 or 0.125 mm in diameter,
were pulse heated in a fast RCL discharge circuit with typical heating rates of10 9 K s~! (exploding
wire technique). From time resolved voltage and current measurements the electric resistivity and the
enthalpy were derived, with the relative volume change of the sample detected by the shadowgraph
method. The temperature was determined by a fast pyrometer. Investigations made in the range
of 2000 to 4000 K yielded molar heat capacities of 44.7 J mol~ l K~* for liquid nickel and
51.9 J mol"1 K~ l for liquid iron. The results are compared with those found by other authors.
1 Introduction
Pulse heating techniques for measuring thermophysical properties of metals at high
temperatures have been successfully applied by a number of authors (see for example
Cezairliyan 1984; Gathers 1986; Lebedev et al 1973; Seydel et al 1980). The
exploding wire technique shows several advantages over other methods:
(i) fast RCL discharge circuits ensure that no disturbances are caused by MHD
instabilities during the relevant diagnostic time interval (the first 4 /as);
(ii) chemical reactions of the liquid specimen with the surrounding medium are more
easily suppressed when the heating rates are fast;
(iii) many metallic elements can be made into wire-shaped specimens of high purity;
(iv) superheated liquid metal states can be easily achieved and temperatures reached
which are higher than those attained by other methods; and
(v) the wire sample is arranged and pulse heated in a relatively simple way.
However, the required time resolution of the diagnostic system places certain
demands with respect to the measuring equipment as well as the design of the
electric discharge circuit and especially the discharge vessel. The experimental setup
used for the investigations described here is based in principle on that described by
Seydel and Fucke (1977), somewhat modified by Gallob et al (1985) in order to
allow the measurement of the surface temperature by a fast pyrometer in addition to
the measurements of the current through the wire and the voltage drop across it.
The technically important metals iron and nickel were investigated up to
temperatures far above their normal boiling points. Seydel and Fucke (1977) had
already obtained data for these two elements, but they derived their temperature
values from the electric parameters of the discharge with the help of the so-called
'integral of action'. Therefore an independent temperature determination by
pyrometry seemed to be desirable.
2 Experimental
Figure 1 shows the block diagram of the coaxially arranged RCL discharge circuit
and the diagnostic system. Wire-shaped samples, 40 mm in length and either
0.25 mm or 0.125 mm in diameter, were pulse heated. To suppress peripheral
discharges, water at a pressure of 1 bar was used as the surrounding medium in a
discharge vessel described earlier by Gallob et al (1985).
20 G Pottlacher, H Jager, T Neger
The energy was stored in a capacitor bank (C = 5.4 pF) which could be charged
up to 8 kV. From the short-circuit ringing period T = 6.2 jus an inductivity
L — 180 nH can be derived; the circuit resistance was found to be 26.3 mi}. Under
these conditions typical heating rates of about 109 K s*"1 can be achieved.
Current measurements were performed by RC integration of the d//df signal of an
induction coil. A coaxially constructed ohmic voltage divider served for the
determination of the voltage drop across the wire. The volume change of the sample
caused by ohmic pulse heating was measured by the shadowgraph method, which
has been described in detail by Gallob et al (1986), together with the fast pyrometer
(operating at 801.7 /jm; photodiode BPX 65, Siemens) used for the determination
of the surface temperature of the heated wire. The time resolution of the whole
measuring system was about 8 ns.
The determination of the various thermophysical properties starts by considering
the measured time-dependent quantities: current, J(t), and ohmic voltage drop, U(T)
[the inductive part is substracted from the U(t) signal]. The quotient yields the
'corrected' resistivity if the volume expansion of the sample is taken into account.
A flow chart of this data reduction procedure is given by Pottlacher et al (1986).
We shall now consider the problem of temperature measurement. For nickel
Seydel and Fucke (1977) carried out complex calculations and considered a number
of material parameters in order to evaluate the temperature by the so-called 'integral
of action' from the voltage and current measurement data. So it seemed desirable
to use an independent method, that of measuring surface radiation. According to
Figure 1. Block diagram of the discharge circuit: 1 capacitor bank, 2 wire sample, 3 charging unit,
4 start pulse, 5 delay, 6 trigger, 7 flashlight, 8 thyratron stage, 9 three-electrode spark gap,
10 photodetector, 11 trigger, 12 time-mark generator, 13 induction coil, 14 voltage divider,
15 pyrometer, 16 Kerr-cell camera, 17 200 MHz oscilloscopes, 18 shielded measuring room.
Thermophysicat measurements on liquid iron and nickel 21
200
100
Figure 2. Temperature dependence of electric resistivity for nickel, with volume expansion
ignored (p0, curve 1), and taken into account (p, curve 2); circles represent the values of Seydel
and Fucke (1977).
22 G Pottlacher, H Jager, T Neger
Figure 2 shows that pQ is nearly constant from 2000 to 4000 K, whereas the
corrected value p (curve 2) first increases slightly and linearly up to about 3200 K,
and afterwards increases very strongly. It should be noted that the values of Seydel
et al (1977) based on the 'integral of action' are in good agreement with those of
this work.
The polynomial least squares fits are:
200
100
0 1 2 3
///MJ kg' 1
Figure 3. Electric resistivity as a function of enthalpy for nickel, with volume expansion ignored
(Po, curve /) and taken into account (p, curve 2); circles represent the values of Seydel and
Fucke (1977).
heating rate the temperature measurement seems to be less accurate than for the
0.25 mm wire which leads to lower enthalpy values. Within the temperature interval
investigated a linear dependence on temperature was found. The values calculated
by Bauhof (1978) match our curves satisfactorily.
The polynomial fits for the enthalpy H are:
curve
= -0.182 + 0.762 x l(T3(r/K), 2000 K < r < 4 0 0 0 K ;
curve 2:
=-0.113 + 0.667 x 10'3(r/K), 2000 K < T < 6000 K.
Using curve 1 we derived for the molar heat capacity the value Cp = 44.7 J mol""11 K"1
Cezairliyan and Miiller (1983) report for 1700 K the value Cp = 38.43 J mol'1 K'1 ;
they also give a review of Cp data reported in the literature. The melting temperature
of nickel found by them was 1729 K (Cezairliyan and Miiller 1984). Margrave (1970)
reports for Cp at melting point 43.05 J mol"1 K"1. Bauhof (1978) used for his
calculations the value Cp — 43.1 J mol"1 K"1 (curve 3). From curve 2 for the
0.125 mm wire we get Cp = 39.2 J mol"1 K~'.
As reported earlier (Gallob et al 1985) the error in heat capacities obtained by
our method is around 25%.
I 2
3.1.4 The melting transition. Table 1 gives some enthalpy and resistivity values at
the beginning and the end of the melting transition. For nickel and iron we
determined these values by examining the characteristic bends in the voltage
oscillogram. Also listed are the values reported by Barin and Knacke (1973),
Cezairliyan and Miiller (1983), Guntherodt et al (1975), Hultgren et al (1973),
and Seydel et al (1979). The latent heat of fusion, A//, obtained by us is compared
with the results of Kelley (1960), Lebedev et al (1971), Margrave (1970), and
Vollmeret al (1966).
3.2 Iron
Data for iron are presented in the same way as for nickel. The agreement with the
values of Seydel et al (1977, 1979) is again good.
24 G Pottlacher, H Jager, T Neger
Table 1 . The enthalpy and electric resistivity of nickel at the melting transition (s solid, 1 liquid).
cm = 0.113 x 10 3 + 0.546 x 10~ 2 (r/K), 2000 K < T < 4000 K (curve 1);
and
100
3.2.2 Dependence of electric resistivity on energy input. The plots in figure 6 can
be described by the following polynomials:
and
Po//ifl cm = 120 + 4.45(tf/MJ kg"1), 0.85 MJ kg"1 <H < 3 MJ kg'1 (curve 1).
Taking into account the volume change we obtained (curve 2):
p/Mfl cm = 0.4 + 240.2(tf/MJ kg- 1 )+ 435.3(ff/MJ kg"1)2- 1272.2(#/MJ kg'1)3
+ 746.6(#/MJkg- J ) 4 , 0<#<0.85MJkg- 1 ,
and
pln&cm= 152.6-36.8(tf/MJkg- I )+15.4(///MJkg- 1 ) 2 ,
0.85 MJ kg'1 <H< 3 MJ kg'1.
200
100
Figure 6. Electric resistivity as a function of enthalpy for iron, with volume expansion ignored
(po, curve 1} and taken into account (p, curve 2); circles represent the values of Seydel and
Fucke (1977)
3.2.3 Dependence of enthalpy on temperature; specific heat. For the same two
different diameters as for nickel (see section 3.1.3) the increase of enthalpy with
temperature was found to be linear. For the 0.25 mm wire we found (figure 7,
3000 4000
r/K
Figure 7. Temperature dependence of enthalpy for iron: I, 0.25 mm diameter wire; 2, 0.125 mm
diameter wire. Circles represent the values of Treverton and Margrave (1971).
26 G Pottlacher, H Jager, T Neger
curve 7):
Table 2. The enthalpy and electric resistivity of iron at the melting transition (s solid, 1 liquid).
References
Darin I, Knacke O, 1973 Thermochemical Properties of Inorganic Substances (Berlin: Springer)
Bauhof H, 1978 Diploma thesis, Universitat Kiel, Kiel, FRG
Cezairliyan A, 1984 in Compendium of Thermophysical Property Measurement Methods volume 1,
eds K D Maglic, A Cezairliyan, V E Peletsky (New York: Plenum) pp 643-668
CezairSiyan A, McClure J L, 1974 /. Res. Nat. Bur. Stand. Sect. A 78 1
Cezairliyan A, Miiller A P, 1983 Int. J. Thermophys. 4 389
Cezairliyan A. Miiller A P, 1984 Int. J. Thermophys. 5 315
Gallob R, Jager H, Pottlacher G, 1985 High Temp.-High Pressures 17 207
Gallob R, Jager H, Pottlacher G, 1986 Int. J. Thermophys. 1 139
Gathers G R, 1986 Rep. Progr, Phys. 49 341
Giintherodt H J, Hauser E, KUnzi H U, Muller R, 1975 Phys. Lett. A 54 291
Hultgren R. Desai P D, Hawkins D T, Gleiser M, Kelley K K, Wagman D D, 1973 Selected Values
of Thermodynamic Properties of the Elements (Metals Park, OH: American Society for Metals)
KeSley K K, I960 US Bur. Mines Bull. 584
Lebedev S V, Sawatimskii A I, Smirnov Yu B, 1971 High Temp. USSR 9 578
Lebedev S V, Savvatimskii A I, Smirnov Yu B, 1973 Sov. Phys. Tech. Phys. 17 1400
Margrave J L, 1970 High Temp.-High Pressures 2 583
Pottlacher G, Neger T. Jager H, 1986 Int. J. Thermophys. 1 149
Seydel U, Bauhof H, Fucke W, Wadle H, 1979 High Temp. -High Pressures 11 35
Seydel U, Fucke W, 1977 Z. Naturforsch. Teil A 32 994
Seydel U, Fucke W, Molier B, 1977 Z. Naturforsch. Teil A 32 147
Seydel U, Fucke W, Wadle H, 1980 Die Bestimmung Thermophysikalischer Daten Flussiger
Hochschmelzender Metalle mil Schnellen Pulsaufheizexperimenten (Diisseldorf: Peter Mannhold)
Treverton J A, Margrave J L, 1971 /. Chem. Thermodyn. 3 473
Vollmer O, Kohlhaas R, Braun M, 1966 Z. Naturforsch. Teil A 21 181
1987 a Pion publication printed in Great Britain