Thermo-Electricity, electricity developed by heat, and also the science which treats of the phenomena and mode of production. Prof. Seebeck of Berlin, in 1822, was the first to make any well directed observations upon the subject. He found that when two rods or bars of different metals were soldered together or otherwise held in intimate contact at their ends, and the junction heated, an electrical disturbance took place, and that if the ununited ends were connected by a conductor an electric current was established. Several crystals, while their temperature is rising or falling, also become oppositely electrically excited at their opposite ends. The term pyro-electricity is usually applied to the electrical phenomena which arise from changes of heat in crystals. These phenomena were first observed in tourmaline, a double-refracting silicate crystallizing in hexagonal prisms. (See Tourmaline.) Its electrical manifestations are confined within certain limits of temperature, chiefly between 50° and 300° F., but these limits vary with the length of the crystal. If a crystal of tourmaline is suspended by a thread at its middle, and heated, its ends will be attracted and repelled by electrically excited bodies.
Many other crystals exhibit like phenomena, but less in degree, which in many cases can only be detected by a delicate electroscope. That pole of a crystal at which the algebraic sign of the change of temperature is the same as that of the electricity developed, that is to say, which manifests positive electricity when the temperature is rising, is called the analogous pole, and the other, the antilogous pole. Brazilian topaz becomes electrical when heated, the Siberian variety slightly, the Saxon not at all. When the first two are treated negative electricity appears at both ends of the crystal, while the positive is developed on the lateral faces. Pyro-electricity is chiefly developed in hemihedral crystals. The phenomena of thermo-electricity in metals is most strongly marked when two metals are heated at their junction ; but if a wire of a single metal be tied in a knot, and be heated on one side of the knot, electrical disturbance will take place. When two metals are employed, the strength of the current appears to be in proportion to the difference of temperature of the two metals on each side of the junction, and its direction and also its strength upon the natures of the metals used.
In fig. 1, m n represents a plate of copper, soldered on to a plate of bismuth, op, the middle of which also supports a magnetic needle, beneath the copper plate. If heat be applied at o while the axis of the instrument is in the magnetic meridian, the north pole of the needle will be deflected to the left hand of an observer looking from n to m (see Galvanism, vol. vii., p. 592), which indicates that a galvanic current is passing through the copper from n to m. If however the junction n o is cooled, the current will flow from m to n. In the following list, according to Becquerel, the direction of the current will be from any element to any one following, the intensity being greatest between the first and the last: bismuth, platinum, lead, tin, gold, silver, copper, zinc, iron, antimony. The direction of the current often changes when the couple is heated beyond a certain degree. Thus, in a copper and iron circuit, the current passes from the copper to the iron through the heated part when the temperature is not higher than 570°; above this the curent passes in the opposite direction. The cause of thermo-electric currents is diversity in the molecular structure of the elements, and Becquerel ascribes them to unequal propagation of heat in the different parts of the circuit.
A thermoelectric pile, or battery, in which a series of several couples are joined somewhat like the arrangement in a voltaic pile, or at least with the opposite poles of the elements in contact with each other, was devised by Nobili. A modification of this is shown in fig. 2, in which the lowest plate is bismuth, the next above antimony, the next again bismuth, and so on, the last plate being antimony. These sets of elements are arranged in a copper frame, P, in four vertical series, making in all 20 couples. The terminal plates are connected with binding screws, m and n, by which they may be connected with a resistance measurer or rheostat, or with a sine or a tangent galvanometer. (See Galvanism, vol. vii., pp. 593-5, and Diathermancy, vol. vi., p. 81.) "When the pile is composed of a great number of pairs and connected with a very delicate galvanometer, it may be used to detect the slightest changes of temperature; it is much employed in physical investigations, and will undoubtedly in time have extended practical use in physiology and medicine.