Professional Documents
Culture Documents
285 1 Online Viscosity
285 1 Online Viscosity
285 1 Online Viscosity
CrossMark
View Export
Online Citation
On particular measurements the viscosity of liquid cobalt by the method of torsional vibrations
Solid phase epitaxial studies using vacuum deposition on heated silicon substrates
Appl. Phys. Lett. (August 2008)
Reference Data for the Density and Viscosity of Liquid Aluminum
and Liquid Iron
R. Michael Banish
Center for Materials Research, University of Alabama in Huntsville, Huntsville, Alabama 35899
Akira Nagashima
Yuzuru Sato
Department of Metallurgy, Graduate School of Engineering, Tohoku University, Sendai 980-8579, Japan
William A. Wakeham
University of Southampton, Highfield, Southampton SO17 1BJ, United Kingdom
共Received 4 May 2005; revised 13 July 2005; accepted 16 September 2005; published online 10 February 2006兲
The available experimental data for the density and viscosity of liquid aluminum and
iron have been critically examined with the intention of establishing a density and a
viscosity standard. All experimental data have been categorized into primary and second-
ary data according to the quality of measurement specified by a series of criteria. The
proposed standard reference correlations for the density of the aluminum and iron are
characterized by standard deviations of 0.65% and 0.77% at the 95% confidence level,
respectively. The overall uncertainty in the absolute values of the density is estimated to
be one of ⫾0.7% for aluminum and 0.8% for iron, which is worse than that of the most
optimistic claims but recognizes the unexplained discrepancies between different meth-
ods. The standard reference correlations for the viscosity of aluminum and iron are
characterized by standard deviations of 13.7% and 5.7% at the 95% confidence level,
respectively. The uncertainty in the absolute values of the viscosity of the two metals is
thought to be no larger than the scatter between measurements made with different
techniques and so can be said to be ⫾14% in the case of aluminum and ⫾6% in the case
of iron. © 2006 American Institute of Physics. 关DOI: 10.1063/1.2149380兴
Key words: aluminum; density; iron; melt; reference data; viscosity.
0047-2689Õ2006Õ35„1…Õ285Õ16Õ$40.00 285 J. Phys. Chem. Ref. Data, Vol. 35, No. 1, 2006
286 ASSAEL ET AL.
4.1.2.2. The Beckwith, Kestin, and 14. Percentage deviations of the viscosity primary
Newell Analyses. . . . . . . . . . . . 293 data for liquid iron from the values calculated
4.1.3. The Rotating Cylinder Viscometer. . . . 293 by Eq. 共20兲, as a function of the temperature. . . . . 299
4.1.4. The Draining Vessel Method. . . . . . . . . 294
4.1.5. The Oscillating 共Levitated兲 Drop
Method. . . . . . . . . . . . . . . . . . . . . . . . . . 294 1. Introduction
4.1.6. Acoustic Wave Damping. . . . . . . . . . . . 295
4.2. Data Compilation. . . . . . . . . . . . . . . . . . . . . . . 295 There is an increasing use of mathematical models to
4.3. Viscosity Reference Correlation. . . . . . . . . . . . 295 simulate a variety of processes involving liquid metals such
5. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 as ‘‘cast to shape,’’ primary and secondary metal production,
6. Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . 299 powder production by spray forming, and welding. Depend-
7. References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 ing on what aspect of the process is modeled, a need for
viscosity data of relevant alloys exists. Historically there are
wide discrepancies in the viscosity data reported for the me-
List of Tables tallic elements and alloys 关Iida and Guthrie 共1988兲兴. For ex-
1. Data sets considered for the density of liquid ample there is a spread of about 400% in the reported values
aluminum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 for molten aluminum and about 100% for molten iron.
2. Data sets considered for the density of liquid These discrepancies prompted this work to review the val-
iron. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 ues in the literature and attempt to derive recommended val-
共vii兲 Explicit quantitative estimates of the uncertainty of In such a case bubbles of gas will detach from the tip of the
reported values should be given, taking into account capillary. The density of the liquid specimen can be deter-
the precision of experimental measurements and pos- mined by measuring the difference in the overpressure re-
sible systematic errors. quired to form a hemispherical bubble of the inert gas at the
共viii兲 Owing to the desire to produce low-uncertainty refer- tip of the capillary at different depths in the liquid. The tech-
ence values, limits must be imposed on the uncer- nique is not as accurate as the pycnometric method but al-
tainty of the primary data sets. These limits are deter- lows density measurements at high temperatures. Its main
mined after critical evaluation of the existing data disadvantages are the corrections to the spherical shape, the
sets. high precision of the maximum bubble pressure measure-
These criteria have been successfully employed to the propo- ment and the presence of nonwetting effects in the gas/liquid
sition of standard reference values for the viscosity and ther- metal systems that can lead to spreading away from an inside
mal conductivity of fluids over a wide range of conditions, nozzle perimeter and across the surfaces of the capillary
with uncertainties in the region of 1%. tube. When measurements are carried out on aluminum, alu-
In the case, however, of the liquid metals, it was felt and mina can be formed by reaction with oxygen in the atmo-
agreed that these criteria needed to be relaxed slightly, espe- sphere; this can accumulate in the capillary tube and reduces
cially since the uncertainty of the measurements is much the effective radius of the tube.
higher, primarily due to: 共i兲 the difficulties associated with The Sessile-Drop technique employs a liquid drop of
the techniques employed in such high temperatures and 共ii兲 known mass resting on a plate or substrate. Provided the
Primary data
Smith et al. 共1999兲 99.99 X-Ray 共Rel兲 — D
Yatsenko et al. 共1972兲 99.996 Sessile Drop 共Abs兲 1.5 D
Levin et al. 共1968兲 ⬎99.996 Sessile Drop 共Abs兲 1.0 D
Coy and Mateer 共1955兲 99.997 Bubble Pressure 共Abs兲 — P
Gebhardt et al. 共1955兲 99.996 Archimedean 共Abs兲 — P
Secondary data
Sarou-Kanian et al. 共2003兲 99.99 Levitation 共Rel兲 1.5 D
Nasch and Steinemann 共1995兲 — ␥-ray 共Abs兲 0.75 E
a
Abs⫽absolute; Rel⫽relative.
b
D⫽diagram; E⫽equation; P⫽points.
3.2. Data Compilation surface tension effects are identified to be the main reason
for their high values. This data set was not included in the
Tables 1 and 2 present the data sets found for the measure- primary data set.
Primary data
Brillo and Egry 共2004兲 99.995 Elect. Levitation 共Abs兲 1.0 P
Sato 共2003兲 99.99 Pycnometer 共Abs兲 0.5 E
Lucas 共1972兲 ⬎99.96 Archimedean 共Abs兲 0.3 D
Saito et al. 共1969兲 ⬎99.9 Levitation 共Abs兲 1.5 D
Frohberg and Weber 共1964a兲 99.98 Bubble Pressure 共Abs兲 — D
Kirshenbaum and Cahill 共1962兲 ⬎99.9 Archimedean 共Abs兲 0.05 P
Lucas 共1960兲 99.9 Archimedean 共Abs兲 — D
Secondary data
Wille et al. 共2002兲 99.99 Levitation 共Rel兲 — D
Nasch and Steinemann 共1995兲 — ␥-Ray 共Abs兲 0.75 E
Adachi et al. 共1971兲 99.97 Archimedean 共Abs兲 — D
Watanabe 共1971兲 ⬎99.9 Bubble Pressure 共Abs兲 0.7 D
a
Abs⫽absolute; Rel⫽relative.
b
D⫽diagram; E⫽equation; P⫽points.
where c 1 ⫽2377.23 kg m⫺3 , c 2 ⫽0.311 kg m⫺3 K⫺1 , and 4.1.1. The Capillary Rheometer
T ref⫽933.47 K is the melting point of aluminum 关Preston-
Thomas 共1990兲兴. The standard deviation of the above equa- The capillary rheometer is generally thought to be the best
tion at the 95% confidence level is 0.65%. method for the measurement of the viscosity of liquids 关Iida
For iron 共temperature range 1809–2480 K兲 and Guthrie 共1988兲兴 and is based upon the time for a finite
volume of liquid to flow through a narrow bore tube under a
⫽c 3 ⫺c 4 共 T⫺T ref兲 , 共2兲 given pressure 共Fig. 5兲. The relation between viscosity and
I⌬⫽
R 3H
2
共 T 兲 1/2 1⫹冋冉
冊 冉
1 R
4 H
3 4 R 1
a 0⫺ ⫹
2 H p 冊
冉 冊 冉 冊 册
Figure 6 shows the experimental arrangement for an oscil-
lating vessel viscometer and details of this particular design 3 9 R a2 63 45 R a 4
⫹ ⫹ ⫺ ⫺ ... , 共7兲
are given elsewhere 关Brooks et al. 共2001兲兴. The sample is 8 4 H 2p2 128 64 H 4 p 4
contained within an alumina crucible 共105 mm long by 14
mm internal diameter兲, which is screwed into a molybdenum
where
lid and suspension rod and suspended on a torsion wire. A
再 冎 再 冎
rotary solenoid is used to impart oscillatory motion to the
crucible and an optical pointer with a diode array is used to 冑共 1⫹⌬ 2 兲 ⫹1 1/2
冑1⫹⌬ 2 ⫺1 1/2
measure the time constant and decrement of the system. The a 0 ⫽ 共 1⫺⌬ 兲 ⫺ 共 1⫹⌬ 兲
2 2
sample is heated by a two zone furnace. These authors claim
an uncertainty of measurement of ⫾9% within a 95% con- 3 3 1
⫽1⫺ ⌬⫺ ⌬ 2 ⫺ ⌬ 3 ..., 共8兲
fidence limit. A major contribution to the uncertainty is the 2 8 16
再 冎 再 冎
冑1⫹⌬ 2 ⫹1 1/2
冑1⫹⌬ 2 ⫺1 1/2 ⫽T 0 /T; p⫽1/共 2 共 ⌬⫹ 共 1⫹⌬ 2 兲 1/2兲兲 1/2; q⫽1/共 2p 兲 ,
a 2⫽ ⫹ 共14兲
2 2
and
⌬ 1 1
⫽1⫹ ⫹ ⌬ 2 ⫺ ⌬ 3 ..., 共9兲 x⫽R 共 2 / 共 T 兲兲 1/2, 共15兲
2 8 16
where T 0 is the oscillation period and 2 ⌬ 0 is the logarith-
⌬ 3 5 mic decrement for the empty cup. The other quantities are as
a 4 ⫽a 2 / 共 1⫹⌬ 2 兲 1/2⫽1⫹ ⫺ ⌬ 2 ⫺ ⌬ 3 ..., 共10兲 defined in the previous section.
2 8 16
Comparison of the oscillating viscometer with the capil-
and lary viscometer reveals some differences. Iida and Guthrie
p⫽ 共 / T 兲 1/2R. 共11兲 共1988兲 suggest that the finite length of the oscillating cylin-
der is inadequately weighted in the Roscoe treatment and
In Eq. 共7兲, R is the cylinder radius, H the height of the liquid should also include the effect of the liquid meniscus on the
within it, I is the moment of inertia of the suspension, 2⌬ height of liquid, suggesting that a correction factor, , is in-
the logarithmic decrement between consecutive swings, is troduced into the 共uncorrected兲 formula of 1⫾0.04. Brock-
the fluid density, the viscosity, and T is the period of os- ner et al. 共1979兲 actually demonstrated and corrected for the
cillation. meniscus effect. Various studies 关Iida and Guthrie 共1988兲;
In his review of the equations Ferriss et al. 共2002兲, found Kimura and Terashima 共1994兲; Kimura and Terashima
that Eq. 共8兲 was printed incorrectly in the original paper by 共1995兲; Sasaki et al. 共1995兲; and Sato and Yamamura
Roscoe 共1958兲, and should read (⫺3⌬/2), as given above. 共1999兲兴, have suggested that wetting of the crucible may also
4.1.2.2. The Beckwith, Kestin, and Newell Analyses. The be important, and that if the metal does not wet the crucible
analyses due to Kestin and Newell 共1957兲, and Beckwith and it may slip during the oscillation and thus provide greater
Newell 共1957兲 identify the ‘‘small,’’ ‘‘intermediate,’’ and damping. Aspect ratio of the sample may be important in
‘‘large’’ cup regimes according to the magnitude of certain both these assessments, and further work is required to jus-
dimensionless parameters 共Fig. 7兲. Ferriss et al. 共2002兲 has tify modification of the Roscoe equation.
shown that these values for the NPL design correspond to the
‘‘large cup’’ regime.
The corresponding operating equation due to Beckwith 4.1.3. The Rotating Cylinder Viscometer
and Newell, in the form of Brockner et al. 共1979兲, is
冋 册
HR 4 1 B Cq For the rotating cylinder technique the torque on a cylin-
A 共 p⫺⌬q 兲 ⫺ 2 ⫹ 3 ⫽2 共 ⌬⫺⌬ 0 / 兲 , der rotated in a liquid is related to the viscosity of the fluid.
2I x x x Viscometers of this type consist of two concentric cylinders
共12兲
共see Fig. 8兲 named a bob 共internal兲 and a crucible 共external兲.
where The viscosity is determined from measurements of the torque
⫽ 冉 1
⫺
1
r 21 r 20 冊 M
8 4 nh
, 共16兲
Material
Form in contact
Purity Technique Uncertainty of with Density Equation
Reference 共mass %兲 employeda quoted/% datab sample sourcec employed
Primary data
Sato 共2004兲 99.99 OV 1 P Graphite L Roscoe
Mills 共2002兲 99.995 OV 2.5 D Alumina L Roscoe
Wang and Overfelt 99.995 OV 4 P Graphite M Wang and Overfelt
共2002兲
Yamasaki et al. 共1993兲 99.99 OV — D — — Roscoe
Arsent’ev and Various OV 5 D Alumina — Shvidkovskiy
Polyakova 共1977a兲
Pakiewicz 共1970兲 99.999 OV 1 P Graphite L Roscoe
Rothwell 共1961–1962兲 99.9 OS ⬍1 P Graphite L Brewer modified Andrade
Gebhardt and Becker 99.996 OV — P Carbon M Calibration
共1955兲
Secondary data
Efimenko et al. 共1988兲 — — — D — — —
Arsent’ev et al. 共1987兲 Various OV — D — — —
Kisun’ko et al. 共1983兲d — OV 4.5 D L —
than the scatter between measurements made with different the formation of the oxide of aluminum on the surface of the
techniques and so can be said to be ⫾14% in the case of melt and its subsequent entrainment in the body of the liquid.
aluminum and ⫾6% in the case of iron. It is exceedingly unlikely that any experiments have suc-
In the particular case of aluminum it is necessary to add ceeded in achieving a partial pressure of oxygen in the atmo-
one caveat to this statement of uncertainty. Some of the ex- sphere that approaches the 10⫺23 that is necessary to avoid
periments that we have characterized as primary have used oxide formation. Accordingly, it is likely that alumina was
an inert gas atmosphere when studying aluminum to prevent present to some extent in all samples. It can also be expected
Material
Form in contact
Purity Technique Uncertainty of with Density Equation
Reference 共mass %兲 employeda quoted/% datab sample sourcec employed
Primary data
Sato et al. 共2002兲 99.8 OV 1 D Alumina M Roscoe
Brooks et al. 共2001兲 99.8 OV 4.5 D Alumina L Roscoe
Kaplun and Avaliani 99.92 OV 5 D Zircon or — Shvidkovskiy
共1977兲 Alumina
Cavalier 共1959兲d 99.996 OV — D Alumina L Calibration
Secondary data
Arsent’ev et al. 共1987兲 — OV — D — — —
Sroka and Skala 共1979兲 — OV — D Alumina — Knappwost
Arsent’ev 共1978兲d — OV — D — — —
Arkharov et al. 共1977兲 99.997 — — P — — —
Frohberg and Cakici 99.91 OV — D — — Modified
共1977兲 Roscoe
Ershov 共1976兲 — OV — P Alumina — —
Borgmann and — OV — D Alumina L Modified
Frohberg 共1974兲 Roscoe
therefore that there would be some differences between the propriate to circumstances created in a well-controlled gas-
use of alumina vessels for containment and that with other eous atmosphere. These are likely to be the circumstances
materials such as graphite. These observations are supported encountered in usual industrial practice.
by the fact that experiments have shown that the apparent
viscosity of aluminum increases as the partial pressure of 5. Conclusions
oxygen in the atmosphere increases and with increasing time
of exposure 关Mills 共2002兲, Vigneau et al. 共1967兲, Shvidk- The available experimental data for the density and vis-
ovskiy 共1955兲兴. For this reason we suggest that our stated cosity of liquid aluminum and iron have been critically ex-
values for the viscosity of aluminum be taken as those ap- amined with the intention of establishing a density and a
Brockner, W., Torklep, K., and Oye, H.A., Ber. Bursenges Phys. Chem. 83, Nozdrev, V.F., Stremousov, V.I., and Takuchev, V.V., Russ. J. Phys. Chem.
1 共1979兲. 53, 677 共1979兲.
Brooks, R.F, Day, A.P., Andon, R.J.L., Chapman, L.A., Mills, K.C., and Nozdrev, V.F., Stremousov, V.I., and Tokuchev, V.V., Zh. Fiz. Khim. 53,
Quested, P.N., High Temperatures-High Pressures 33, 72 共2001兲. 1199 共1979兲.
Cavalier, G., The NPL Symposium on The Physical Chemistry of Metallic Ogino, Y., Borgmann, F.O., and Frohberg, M.G., J. Jap. Inst. Met. 37, 1230
Solutions and Intermetallic Compounds 共HMSO, London, 1959兲, Paper 共1973兲.
4D. Ogino, Y., Morita, Z., and Adachi, A., Technol. Rep. Osaka Univ. 共Japan兲,
Cavalier, G., Compt. Rend. 256, 1308 共1963兲. 21, 399 共1971兲.
Coy, W.J., and Mateer, R.S., Trans. Amer. Soc. Metals 58, 99–102 共1955兲. Ogino, Y., Morita, Z., Maebana, T., Yokoya, K., and Adachi, A., Tetsu
Cummings, D.L., and Blackburn, D.A., J. Fluid Mech. 224, 395 共1991兲. Hagane 56, 59 共1970兲.
Dinsdale, A.T., CALPHAD 15, 317 共1991兲. Orecki, K., Sikora, B., and Zielinski, M., Prace Inst. Hutnicyzh 23, 353
Efimenko, V.P., Kisunk’ko, V.S., Bychkov, Yu.B., and Beloborodov, A.Z., 共1971兲.
Soviet Cast Technology 9, 37 共1988兲. Pakiewicz, M.J., Ph.D Department of Chemistry, Syracuse University, New
Ershov, G.S., and Kasatkin, A.A., Izv. Vyssh. Uchebn. Zaved., Chern. Met- York, 1970.
all., 141 共1976兲. Petrushevskii, M.S., Levin, E.S., and Gel’d, P.V., Zhur Fiz Khim. 45, 3035
Ferriss, D.H., Quested, P.N., Chapman, L.A., and Day, A.P., ‘‘The Choice of 共1971兲.
Equations for the Measurement of Viscosity by the Oscillating Cylinder Popel, P.S., Zam. Russ. Met. 3, 38 共1983兲.
Method,’’ Presented at ECTP, London, 2002. Preston-Thomas, H., Metrologia 27, 3 共1990兲.
Frohberg, M.G., and Cakici, T., Arch. Eisenhüttenw. 48, 145 共1977兲. Rhim, W.-K., Ohsaka, K., Paradis, P.-F., and Spjut, R.E., Rev. Sci. Instrum.
Frohberg, M.G., and Weber, R., Arch. Eisenhuttenw. 35, 877 共1964a兲. 70, 2796 共1999兲.
Frohberg, M.G., and Weber, R., Arch. Eisenhüttenw. 35, 885 共1964b兲. Roach, S.J., Henein, H., and Owens, D.C., Light Metals 2001, edited by J.L.
Gebhardt, E., Becker, M., and Dorner, S., Z. Metallkde. 44, 510 共1953兲. Aujier 共TMS, Warrendale, PA, 2001兲, pp. 1285–1291.
Gebhardt, E., Becker, M., and Dorner, S., Aluminium 31, 315 共1955兲. Romanov, A.A., and Kochegarowv, V.G., Izv. Akad. Nauk, SSSR, Otd, 1, 30
Gebhardt, E., and Detering, K., Z. Metallkde. 50, 379 共1959兲. 共1960兲.