Date of Award


Degree Type


Degree Name

Master of Science


Engineering Science

Major Professor

C. R. Brooks

Committee Members

E. E. Stansbury, J. G. Albert


Calorimeters of various types exist by which certain thermal energy of substances can be measured. A program of dynamic adiabatic calorimetry was initiated at The University of Tennessee in 1950 to further existing work on specific heat data. The calorimeters have continuously been improved in both construction and the operational methods.

This thesis constitutes a phase of the program in the use of dynamic adiabatic calorimeters. Modifications were made on previously existing calorimeters to an extent that specific heat data on various metal systems could be obtained to temperatures as high as 850°C. The calorimeter was then used to determine the specific heat of copper, nickel, and a nickel-20 weight per cent chromium alloy.

The calorimeter used in this investigation consists of an adiabatic surrounding into which a highly polished specimen is suspended. The specimen is heated with a direct current power. The temperature of the surroundings is forced to follow the temperature of the specimen. This is accomplished by use of platinum: platinum-13 weight per cent rhodium thermocouples welded on shields; the emf of these thermocouples activates a control circuit which supplies power to the heaters of the shields. The maintenance of adiabatic conditions causes all of the power fed to the specimen to be consumed in heating it. The power supplied to the specimen is determined by measuring the voltage drop across the heaters and the current through the heater. Temperatures are determined from emf readings taken on a potentiometer from thermocouples. The time necessary to raise the temperature of the specimen a preset amount is measured. From the voltage and current measurements and the time necessary to heat the specimen over a particular temperature interval, the average value of the specific heat can be obtained. The system is held at 10-4-10-5 Torr to prevent oxidation and convectional heat transfers.

Following the design and construction of the new calorimeter, it was used to obtain specific heat data, from which the following results were obtained.

1. The specific heat data of copper and nickel were obtained from 100° to 800°C. Reproducibility of ±0.7 per cent was obtained on copper where the heating rate ranged from 1.5° to 6.0°C./min. and samples of 172.794 grams and 368.778 grams were used. A reproducibility of ±0.2 per cent was obtained for nickel where only the heating rate was varied (2.2° to 6.4°C./min.). The high degree of reproducibility allows the conclusion that accurate data can be obtained to 800°C. There are no apparent reasons why data could not be obtained to higher temperatures on the same calorimeter.

2. The specific heat of nickel experienced a magnetic transformation at its Curie temperature of 355°C. Data were obtained to 800°C. such that predicted data without the magnetic transformation could be compared in a manner that allowed the determination of the energy and entropy of the transformation. These values were 954 joules/gm.-atom and 1.967 joules/gm.-atom °K., respectively. Comparison of the above values to theoretical predictions, based on magnetic models, indicate that 0.3 nickel atoms have two holes in the 3d level with an electron spin of one.

3. The specific heat data of the nickel-20 weight per cent chromium alloy experienced an anomaly at approximately 575°C. Analogies were drawn between the experimental data and work of other investigators to support the theory that short-range order was causing the anomaly.

4. The problems encountered in the operation of the calorimeter indicated that inconel should not be used in a vacuum system that is to operate at high temperatures.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."