The freezing points of high-purity metals as precision temperature standards. VIIIa. Sb: Apparatus, freezing techniques, and ingot morphology

1968 ◽  
Vol 46 (5) ◽  
pp. 369-400 ◽  
Author(s):  
E. H. McLaren ◽  
E. G. Murdock

This paper describes the apparatus and the freezing techniques that have been developed to determine the liquidus points (630.55 °C) of samples of high-purity antimony to a reproducible precision of ± < 0.0005 °C. Freezing plateaux steady to ± 0.0001 °C for long (hours) durations are readily obtained on freezing curves of high-purity antimony using an outside-nucleated slow induced freezing (ONSIF) technique in a balanced three-winding inconel block furnace. The pressure effect on the freezing temperature of antimony was determined as + 0.000 85 °C for 1 atm, which corresponds to a contraction on Sb solidification of 0.98%; a contraction on freezing is supported by volumetric measurements on the shrinkage pipes in Sb ingots and by observations on decanted and quenched (during freezing) ingots that Sb dendritic solid is more dense than Sb liquid.Metallurgical studies using decanting, quenching, and tracer techniques determined the ingot morphology during freezing and melting to verify that a satisfactory control of the transforming ingot had been attained for liquidus point realizations and to provide auxiliary information on the nature of solute segregation and segregate remelting for the interpretation of alloy melting-range comparisons on several samples of high-purity antimony described in Part VIIIb of this series of papers. Abundant evidence of the fragmentation of masses of large rod dendrites and the surface recontouring of dendrite spines by recalescent remelting was found in both decanted and quenched Sb ingots: the pileup of dendritic solid in the bottom of the crucible does not preclude precise temperature determinations on ONSIF freezes.

1958 ◽  
Vol 36 (9) ◽  
pp. 1131-1147 ◽  
Author(s):  
E. H. McLaren

An investigation of the freezing and melting temperatures of a sample of high purity indium (Cominco 99.999+ %) has been made. Plateaux of essentially constant (< ± 0.0001 °C) temperature with durations for over 1 hour are readily obtained on the cooling curves of induced freezes on this metal. The standard deviation of the plateau temperature (liquidus point) from a series of 24 induced freezes was ±0.0001 °C. The pressure effect on the freezing temperature of indium was found to be 0.0049 °C for 1 atm. Alloy melting ranges were measured following different types of freezing.An extensive intercomparison of liquidus points and alloy melting ranges has been made on a sample of 99.99% cadmium and two samples of 99.999+ % cadmium from different suppliers. The liquidus points of the high purity samples were indistinguishable using precision resistance thermometry but one sample melted over a slightly smaller range of temperature than the other. Both these samples showed minor arrests on melting curves after induced freezing and detailed analyses of the melting contours after various types of freezing indicated some evidence of characteristic structure inside a range of 0.002 °C.


1960 ◽  
Vol 38 (5) ◽  
pp. 577-587 ◽  
Author(s):  
E. H. McLaren ◽  
E. G. Murdock

An investigation has been made of the freezing and melting temperatures of five samples of high purity lead (supplier's analyzed impurity contents < 0.7 to < 4 p.p.m.) including zone refined metal. Using the induced freezing technique, plateaux of essentially constant (< ±0.0001 °C) temperature with durations of over 1 hour are readily obtained on the cooling curves of these samples. A standard deviation in plateau temperature (liquidus point) of ±0.0001 °C was obtained from a series of 30 induced freezes on a particular sample. The pressure effect on the freezing temperature of lead was found to be 0.0080 °C for 1 atmosphere. A value of 327.426 °C (Int. 1948) was determined for the standard liquidus point of pure lead.The liquidus points of the samples were intercompared with a precision of about 0.0002 °C, and alloy melting ranges were examined following different types of freezing with and without overnight anneals near the solidus temperature. Alloy melting range parameters were found to be useful in the selection of the samples of highest purity and at the same time showed that an uncertainty of 0.002 °C in the above value of the liquidus point of pure lead may exist because of residual impurity contents in the purest samples that were examined.


1963 ◽  
Vol 41 (1) ◽  
pp. 95-112 ◽  
Author(s):  
E. H. McLaren ◽  
E. G. Murdock

An investigation has been made of the freezing and melting temperatures of seven samples of high-purity bismuth (analyzed impurity contents <0.5 to 7 ppm) including zone-refined metal. Using a controlled outside nucleation technique, freezing curves having plateaux of essentially constant (< ±0.0001 °C) temperature with long durations are readily obtained. A standard deviation in plateau temperature (liquidus point) of ±0.00025 °C was obtained from a series of 34 freezes on a particular sample. The pressure effect on the freezing temperature of bismuth was determined as −0.0035 °C for 1 atmosphere. A value of 271.375 °C (Int. 1948) was determined for the standard liquidus point of pure bismuth.The liquidus points of the samples were intercompared with a precision of about 0.0002 °C and their alloy melting ranges were measured for selection of the purest samples. Ingot morphologies and solute redistributions during melting and freezing were investigated using decanting, quenching, and tracer techniques.Appendix I gives the results of the latest intercomparison of temperatures realized in a grid of water triple-point cells that is maintained by this laboratory for use in the precision temperature measurements. Appendix II lists the values and pressure–temperature dependencies of the liquidus points of the purest samples of Zn, Pb, Cd, Bi, Sn, and In that have been determined at this laboratory.


1968 ◽  
Vol 46 (5) ◽  
pp. 401-444 ◽  
Author(s):  
E. H. McLaren ◽  
E. G. Murdock

An investigation was made of the freezing and melting temperatures of seven samples of high-purity antimony, five with analyzed impurity contents ranging from <0.3 to <0.7 p.p.m. (wt.) and two with impurity contents of ~10 and ~100 p.p.m. (wt.) respectively. The Sb melts had to be deoxidized in situ to eliminate the effect of dissolved oxygen (gaseous or oxide form), which was found to cause large depressions and instabilities in the liquidus-point temperatures. A variation of 0.0026 °C in liquidus point was found among the five purest samples, but the standard deviation for a single determination of the liquidus point realized on outside-nucleated slow induced freezes (ONSIF) of any given sample was [Formula: see text]. The best Sb samples had nonequilibrium alloy melting ranges of a few centidegrees which are distinctly inferior to melting ranges of only a few millidegrees found on samples of Sn, Zn, and Pb with similar analyzed impurity contents: the inferiority is attributed to residual impurity effects in the Sb samples and not to either lag in detector response or anomalous molecular effects in the melting Sb. A value of 630.553 (°C Int./60) was determined for the liquidus point at standard pressure of the purest NRC Sb sample.After establishing the Sb point, eight silica-sheathed standard resistance thermometers were intercompared at the ice, Sn, Cd, Zn, and Sb points to provide information at the highest precision on stabilities of fixed-point realizations and resistance thermometers, inter-thermometer variations, advantages of particular quadratic resistance–temperature interpolation relations, preferred calibrating procedures and thermometer handling techniques, and temperature-scale stability and reliability over the full range 0–631 °C.


1957 ◽  
Vol 35 (9) ◽  
pp. 1086-1106 ◽  
Author(s):  
E. H. McLaren

An investigation of freezing and melting temperatures with platinum resistance thermometry on high purity zinc, cadmium, and tin has been carried out. Using appropriate techniques, plateaus of essentially constant (< ±0.0001 °C.) temperature with durations of over 1 hour are readily obtained on the cooling curves of these metals. For series of freezes on particular samples, the standard deviations of the respective plateau temperatures were found to be of the order of ±0.0002 °C. It was not possible to distinguish among the plateau temperatures of three samples selected from different distillation batches of New Jersey S.P. zinc. Evidence is presented on the long term stability ([Formula: see text] years) of the plateau freezing temperature of S.P. zinc determined with six standard thermometers. The pressure effects on the freezing temperatures were found to be 0.0043 °C, 0.0062 °C, and 0.0033 °C. per atmosphere for zinc, cadmium, and tin respectively.Thermal analysis of these high purity metals reveals alloy structures and other features associated with nucleation, coring, and annealing phenomena; typical thermal curves are shown.


1958 ◽  
Vol 36 (5) ◽  
pp. 585-598 ◽  
Author(s):  
E. H. McLaren

Extensive thermal analyses have been carried out on eight grades of high purity zinc with the suppliers' analyzed impurity contents in the range < 0.5 to <70 p.p.m. The liquidus points have been intercompared with a precision of about 0.0002 °C. and the alloy melting ranges have been examined following different rates of freezing. A melting range parameter may provide a more sensitive index of the purity than the liquidus temperature since the shape of the melting curve is strongly influenced by segregation of impurities due to coring on freezing. Two grades of Russian high purity zinc have liquidus points and melting ranges so close to those of New Jersey S.P. zinc that it is difficult to discriminate among them using precision resistance thermometry. This shows that a reported discrepancy of about 0.1 °C. in the freezing points of zinc between the Russian and Canadian laboratories is not due to different impurity contents of the zincs that were used.


1960 ◽  
Vol 38 (1) ◽  
pp. 100-118 ◽  
Author(s):  
E. H. McLaren ◽  
E. G. Murdock

Extensive thermal analyses have been made on 10 samples (suppliers' analyzed impurity contents <0.2 to <100 p.p.m.) of high purity tin, including zone-refined metal; liquidus points have been intercompared with a precision of about 0.0002 °C and alloy melting ranges have been examined following different types of freezing with and without overnight anneals near the solidus temperature. Samples of nominal 99.9999% purity tin were found to have such narrow alloy melting ranges that any ambiguity, arising from unknown impurity concentrations, in specifying the liquidus point of pure tin is well inside 0.001 °C; a value of 231.913 °C (Int. 1948) was found for the standard liquidus point of pure tin. An account is given of the supercooling that was observed on the bulk samples and of anomalous structures that were found on melting curves. An appendix gives the results of long-term intercomparisons of the temperatures realized in four water triple point cells.


1987 ◽  
Vol 15-18 ◽  
pp. 1415-1420 ◽  
Author(s):  
Takeo Muroga ◽  
Naoaki Yoshida ◽  
Yoshinari Miyamoto ◽  
Hideoki Murakami

1959 ◽  
Vol 37 (4) ◽  
pp. 422-432 ◽  
Author(s):  
E. H. McLaren

Eleven standard platinum resistance thermometers, including thermometers having three different types of construction, have been intercompared at the triple point of water, boiling point of water, and the liquidus points of high purity tin, cadmium, and zinc. Temperature coefficients determined from measurements at the triple and boiling points of water and the zinc point were used to calibrate the thermometers for the temperature calculations on measurements at the tin and cadmium points. The results show that, although the measurements were made at a precision of about 0.0002 °C at each fixed point, distinctive deviations from quadratic resistance–temperature relations were not found for the 11 thermometers. This verification of the quadratic form for the resistance–temperature relationship realized with these thermometers gives strong support for the use of the liquidus point of high purity indium, tin, or cadmium as a precision alternative to the steam point on the International Temperature Scale.


Sign in / Sign up

Export Citation Format

Share Document