MR. R. ADIE ON THERMO-ELECTRICAL CURRENTS &C. 33 V.-On the Thermo- Electrical Currents generated in Elements where Bismuth is used to form the Joint. By R I c H A R D ADI E Liverpool. THEuse of bismuth as a solder for thermo-electrical couples appeared to me to be worthy of trial for a variety of metals in order to show the extent to which the action of couples might be governed by the nature of their joint. The arrangement of the bars and wires of me-* Chem. SOC.Qu.J. v. 136. VOL. V1II.-XO. XXIX. D 34 MR. R. ADIE ON 'CHB THERMO-ELECTRIC CURRENTS ffENERATED tals to be tested was to solder them together by means of srriall pieces of bismuth and to form the other extremities of the bars into a circuit with a galvanometer. To each pair under examination heat was applied first on the right hand side and then on the left hand side of the joint when the effeet on the direction of the thermo-elec- tric current generated was observed by the galvanometer.After this manner 72 thermo-electric couples were examined. The following is a summary of the effects observed. In nine metals when two bars of the same metal were joined by bismuth solderings the positive electrical current flowed in an oppo-site direction to that of the heat current. These metals were gold silver platinum copper zinc cadmium antiniony iron and soft steel. In three metals used singly with bibwuth solderings the direction in which the heat crossed the joint and that of the electricity were the same. These metals were palladium lead and tin.Twenty-eight pairs of different elements soldered by bismuth being combinations of the metals tested singly showed the direction in which the heat erossed the joint to be opposite to that of the electricity as in the case of the nine single metals tested,-total thirty-seven pairs. One pair of different metals soldered by bismuth showed heat and electricity to cross the joint in the same direction as in the instance of the three single metals tested. Thirty-one pairs of different metals soldered by bismuth showed the direction of the passage of the heat across the joint not to govern the electrical current ; they acted according to their ascertained thermo-electrioal relation independently of the side of the jaint on which the heat was applied.On looking over these results I was led to examine the instances of the three single and one double pair of elements to see why they differed from the thirty-seven other cases where the direction of the electrical current was governed by the passage of the heat across the joint. Their peculiarity appeared to me to arise from the tendency the bismuth had to alloy with these metals and thus form joints of a mixture of metals which gave them an indefinite character. I conse-quently sought to arrange these fbur cases with bismuth joints of a definite kind. EXPERIMENT €.-Two slips of palladium were soldered together with a thin film of bismuth for the joint; their other extremities were connected with a galvanometer. When heat was applied the passage of the heat an8 electricity across the joint was in the same direction as I had previously noted.The soldered joint was broken asunder ; IN ELEMENTS WHERE BISMUTH IS USED TO FORM THE JOINT. 85 the palladium surfaces cleaned free from the bismuth solder ; and to form a joint a thin piece of bismuth was placed between the two slips of metal and secured in its position by firm tying by this means a couple was obtained where the bismuth had a definite surface. When heat was applied first on the right and then on the left hand side of the joint the heat and electricity crossed it on opposite direc- tions in the same manner as they had done in the thirty-seven cases where the direction of the electrical current was governed by a bismuth- soldered joint.EXPERIMENT 11.-Two pieces of lead-wire were formed into a couple with a small piece of bismuth tied firmly between them. In this case the passage of the heat and electricity across the joint was in opposite directions. EXPERIMENT 111.-Two pieces of tin-wire gave a result similar to No. 11. EXPERIMENT 1V.-A wire of lead and a wire of tin with a piece of bismuth tied between them for a joint gave a similar result to Nos. I. TI. and 111. These four experiments comprise the cases noted in the summary where the passage of the heat and electricity across the joint was in the same direction; they now show the heat and electricity to cross the joint in opposite directions,-a change which has been effected by avoiding the mixture which soldering of these metals would prodnce.To obtain pieces of bismuth for inserting in the joints for the above experiments the point of a bar of that metal was held in the flame of a candle until large tear-like drops fell from it these were received on a smooth surface below in the form of thin circular discs. The discs admitted of being cut into small pieces. The weight of bismuth put in the joints for the above experiments varied from&th to &th of a grain; yet this small quantity of bismuth sufficed to render active pieces of lead and tin which without the bismuth joint would only afford a trace of a thermo-electrical current. The general result proved by thew experiments is that the source of the thermo-electrical current is the surface of a joint,-a fact which gives additional value to an observation I made and which was published in the Edinburgh Philosophical Magazine for 1848 that a pair of bismuth and antimony bars soldered together by pure bis- muth and long employed to generate a powerful therrno-electric current produced a disintegration of the bismuth at its surface of contact with the antimony.