On this page
Join a strip of copper and a strip of another metal into a loop, warm one joint, and a compass needle nearby swings. No battery is involved. The heat itself drives a current. This is the Seebeck effect, the basis of the thermocouple thermometers used in labs, kilns and engines. This post tells the story of the man who found it, what he thought he had found, and how to calculate with a thermocouple.
Who was Thomas Seebeck?
The Neue Deutsche Biographie records that Thomas Johann Seebeck was born in Reval (now Tallinn, Estonia) on 9 April 1770 and died in Berlin on 10 December 1831. He studied medicine in Berlin and Göttingen and took his medical doctorate in 1802. A biographical article by Erik Velmre in the Proceedings of the Estonian Academy of Sciences explains that his father's inheritance let him live as an independent scholar instead of practising as a doctor.
In Jena he moved among the philosophers Schelling and Hegel and became a close collaborator of the writer Goethe on experiments with light and colour. Velmre notes that Seebeck found photoelasticity, the coloured patterns that appear in stressed glass under polarised light, and that the French Institute split a physics prize between him and David Brewster for it. He moved to Berlin in 1818 and became a full member of the Berlin Academy of Sciences.
What did he find in 1821?
In 1820 Oersted reported that an electric current moves a compass needle. Velmre writes that Seebeck heard about it that August and switched from optics to electricity and magnetism. He reported his results to the Berlin Academy in a series of talks, starting on 16 August 1821 and continuing in October 1821 and February 1822.
His finding, in the words of the NDB, was that in a circuit of two different metals, heat makes an electric current flow, and with it magnetism appears. What Seebeck actually watched was the magnetism: a compass needle deflected when the junctions of the closed loop were at different temperatures. So he called the effect "thermomagnetism". Velmre records that it was Oersted who gave the clear explanation, a current driven by the temperature difference, and who coined the word "thermoelectricity".
The full results were reprinted in Annalen der Physik in 1826 under a title that translates as "On the magnetic polarisation of metals and ores by temperature difference". The NDB adds that in 1823 he drew up a thermoelectric series, an ordering of materials by how strongly they behave in such loops.
Why were some minerals so interesting?
Seebeck tested ores as well as metals. Velmre notes that some natural minerals, such as galena (lead sulphide) and pyrite, gave a much stronger effect in contact with copper than many metals did, and sorted them into bismuth-like and antimony-like groups. We now call those materials semiconductors, which puts Seebeck near the start of semiconductor research, about ten years before Faraday's work on silver sulphide.
Worked example: a thermocouple thermometer
Suppose a pair of metals gives an emf proportional to the temperature difference between its two junctions: emf = S × ΔT. The values below are sample numbers of a realistic size; real thermocouples are not perfectly linear and are read from tables.
Take S = 41 µV per °C.
Emf for a 100 °C difference. emf = 41 × 10⁻⁶ × 100 = 4.1 × 10⁻³ V, or 4.1 mV.
Current in Seebeck's kind of loop. A short, thick metal loop might have a resistance of 0.05 Ω. Then I = V/R = 0.0041/0.05 = 0.082 A, or 82 mA. That is enough to swing a nearby compass, which is why Seebeck saw an effect even though the emf is tiny.
Reading a temperature. A meter reads 2.05 mV with the reference junction at 25 °C. ΔT = 0.00205 / 0.000041 = 50 °C, so the hot junction is at 25 + 50 = 75 °C.
A thermopile. Fifty pairs in series, each with a 100 °C difference, give 50 × 4.1 mV = 205 mV, about 0.2 V.
The current calculation uses only Ohm's law, which is in the current electricity unit of the JEE (Main) 2026 syllabus. The thermoelectric effect itself is not named there or in the current electricity chapter of NCERT Class 12 Physics, so treat it as background, and check the syllabus for your exam year.
Common mistakes with thermocouples
Forgetting the second junction. A thermocouple measures a difference. Without knowing the reference temperature, the reading means nothing.
Thinking one metal is enough. A loop of a single uniform metal gives no net emf. You need two different materials.
Assuming the emf is large. It is millivolts. Large currents come only from very low resistance loops.
Treating S as exact. Real thermocouples curve slightly, which is why standard tables are used.
Where is the effect used today?
Velmre lists thermocouple thermometers working from about −270 °C to +2500 °C, thermopiles made of many couples in series, and, since the 1960s, thermoelectric generators that turn heat from radioactive decay into electricity on distant spacecraft where solar cells no longer work. The reverse effect, heating or cooling at a junction when you drive a current through it, was found in 1834 and is the story of Jean Peltier and cooling at a junction.
If you want to rehearse the current-electricity steps behind this, the Fermi Problem Library has a step-by-step solver that explains the reasoning rather than printing an answer.
What can a student take from how Seebeck worked?
Jump on new results. Within a year of hearing Oersted's news, he had found a new effect.
Test widely. He tried metals and many ores, and that breadth found the strongest materials.
Separate what you see from what causes it. He saw magnetism; the cause was current.
Practice routine: find the emf for a 250 °C difference with S = 41 µV per °C, find the current if the loop resistance is 0.02 Ω, and explain in two sentences why a thermocouple needs a reference junction.
Next step: draw a loop of two metals, mark the hot and cold junctions, and label where the emf comes from and where the current flows.
