Science Tribune - Article - January 1997
C.S.M. Pouillet, member of the French Academy of Sciences, discovered the "Curie" point in ... 1832
P. Jössang and A. Jössang
Chemistry Laboratory, National Natural History Museum, Paris, France.
(translated from the French by T. Ojasoo)
C.S.M. Pouillet, an outstanding French physicist, discovered the so-called Curie point some 60 years before Pierre Curie. He called the transition temperature from ferromagnetism to paramagnetism the "magnetic limit". The magnetic limit he determined experimentally for Nickel was 350°C approx. compared to the 340°C approx. later found by Curie. Pouillet's discovery was mentioned in most physics books published in France around 1870. Pierre Curie published his work in the early nineties but does not mention Pouillet.
We were surprised to read the following text on page 413 of the "Traité de Physique Élémentaire" published by Ch. Drion and E. Fernet (1) in 1872 :
"Finally, it is Pouillet who made the curious comment that elements with no coercive force (a), such as soft iron, are insensitive to magnets once heated to a certain temperature : they cease to be attracted and become similar to copper or other nonmagnetic substances. Furthermore, the temperature at which this occurs is not the same for all magnetic substances : thus nickel loses its magnetism at 350°C; iron at a cherry red temperature; cobalt at a much higher temperature."
This statement is confirmed by Jamin in his "Cours de Physique de l'Ecole Polytechnique" (2) (p. 552) :
"Mr. Pouillet has gone further; he has studied iron maintained at a high temperature and observed that, when cherry red, it is no longer attracted by a strong magnet placed close by; it then becomes similar to copper and all other nonmagnetic substances."
But where did Pouillet publish his work (b)?
Pouillet was educated at the Ecole Normale, became Director of the Conservatoire des Arts et Métiers, and held the chair in physics at both the Ecole Polytechnique and the Faculty of Sciences. He was elected a member of the Academy of Sciences rather late, in 1837. It seemed logical therefore to look for his work in the Academy proceedings but we found nothing. Previously, he had written a book "Élémens de Physique Expérimentale et de Météorologie" first printed in 1827. A 7th edition was published in 1856, a German edition in 1887. By a stroke of luck, the 1832 edition (3) is to be found in the superb library of the Ecole des Mines in the Luxembourg Gardens (Paris).
The book - which begins with a citation from Bacon's "Novum Organum" - contains the following statements (vol. 1, part 2, chap. IV, pp. 88-89):
"Here is another effect of heat which has not received enough attention : at a cherry red temperature, not only do magnets - steel and iron - lose any magnetism they may possess but they also become unable to acquire any trace of magnetism; during all the time they are held at this temperature, they behave like wood or stone and are quite insensitive to the decomposing action of the most powerful bars.
Thus, magnets - steel and iron - have a magnetic limit and this limit is somewhere near a cherry red temperature. Some rather striking analogies between the distances of the atoms of elements and their magnetic properties had led me to believe that the magnetic limits of elements occur at very different temperatures and I have, indeed, shown experimentally
1°.- that cobalt never ceases to be magnetic or, rather, that its magnetic limit is at a temperature higher than the most dazzling red white;
2°.- that chromium's magnetic limit is just below a dark red temperature;
3°. - that the magnetic limit of nickel is around 350°, at about the melting point of zinc;
4°. - finally, that the magnetic limit of manganese is 20 to 25° below 0°.
Experiments on the five simple magnetic elements, manganese, nickel, chromium, iron and cobalt seem to prove :
1° - that heats affects magnetism only through the more or less great distance it determines between the atoms of elements;
2° - that all substances would become magnetic if one could somehow or other bring their atoms close enough."
What is the "Curie Point" ?
Any substance placed in a magnetic field - for instance, near a magnet - acquires magnetic properties characterized by a magnetic moment per unity of volume, called magnetisation. According to the value and sign of this magnetisation, all substances can be classified into three main categories :
- diamagnetics, which magnetize very weakly but in a direction opposite to the applied magnetic field (they are thus repulsed by a magnet). Examples of diamagnetic substances are : water, copper, lead, silver, bismuth, etc...
- paramagnetics, which magnetize weakly in the same direction as the applied field. Alkali metals, platinum, aluminium, etc ... are paramagnetic at all temperatures.
- ferromagnetics, which magnetize very strongly in the same direction as the applied field; the best example is iron.
The last two categories of substances are therefore attracted by a magnet.
Above a temperature denoted "Curie" temperature (TC), which is often incorrectly called "Curie point" and which is specific to each substance, all ferromagnetics become paramagnetics and consequently only weakly magnetic. In fact, as Curie points out in the introduction to his thesis (see below), it was Michael Faraday - who discovered induction in 1832 - who stated that ferromagnetic elements do not become exactly like "wood or stone". After a passage above the TC, ferromagnetics only retain their original magnetic properties (c); this means that magnets have to be remagnetized.
The following experiment is a good illustration of the influence of temperature on magnetism (4). An iron rod, freely suspended from a horizontal axis, is placed near a magnetic pole. When the rod is attracted by the magnet, it enters a space just above the flame of a Bunsen burner but only stays there a short while. In fact, it falls back to its original position when its ferromagnetism has disappeared. When the rod has cooled down, it is once again attracted by the magnet. The repetition of the overall process gives rise to an oscillator.
Pouillet had found magnetic limits of quite the right order for iron (TC = 770°C), nickel (TC = 358°C) and cobalt (TC = 1130°C) even though one did not know how to measure high temperatures at the time. Pouillet could not know that chromium and manganese form part of a special subclass of weak ferromagnetics - anti-ferromagnetics - whose transition temperature toward paramagnetism is called "Néel temperature"(TN), 202°C for chromium and -178°C for manganese, respectively.
Pierre Curie, who calls paramagnetic elements "weakly magnetic", published his work on the magnetic properties of elements at different temperatures (the actual title of his thesis (5)) in the proceedings of the Academy of Sciences (6) 60 years after the publication of Pouillet's book. We have found no mention of Pouillet by Curie. It is unlikely that Curie was unaware of Pouillet's work because it was mentioned in the physics treatises of the 1870's, in particular in Drion and Fernet's (1). According to a recent biography of Pierre Curie (7), this was in fact the only physics treatise available when Curie performed his studies (1876-78). The only sentence of note by Curie is to be found in the introduction to his thesis (p. 19) :
"It has been known for a long time that iron loses its magnetic properties at red heat, Faraday has shown that iron remains magnetic at high temperatures but only weakly so." (We would say paramagnetic).
Pouillet seems to have been the first person to have a clear idea of the "Curie point", and to have measured it. According to Daguin (8) p. 8 : "Newton was the first to state that iron is no longer magnetic when it becomes red hot." This fact... was confirmed by Cavallo, and by Barlow for the melting of iron (d). Daguin then went on to develop Pouillet's ideas.
It is undeniable that the "Pouillet point" for nickel is about 350°C; that of Curie about 340°C.
The question is whether a discovery should be named after its inventor or after the person who, over 60 years later, made precise, extensive and careful measurements of the phenomenon under study, a phenomenon which is still highly relevant today since "Chemical Abstracts" include about 500 articles per year under the heading "Curie Point".
The authors thank the library of the Ecole des Mines (Luxembourg, Paris) for all facilities given. They also thank Dr. Vladimir Cagan, former head of the Laboratory of Magnetism and Magnetic Materials of the CNRS at Meudon (1991-1994), for details on the Curie and Néel temperatures.
(a) Jamin states (p. 502 of reference 2) : "Then one imagined that there exists in steel a coercive force which is not present in iron, and by this expression, which is but a word, one expresses both the difficulty of separating fluids and of recomposing them ...". It is a question of "north magnetism" and "south magnetism" analogous to the positive and negative charges of electrostatics (Coulomb's concept). It is Oersted (1819) and especially Ampère (1820) who had the idea that a magnetic field is due to circular currents in magnets.
(b) A brief note on the "Curie .... or Pouillet point ?" by Per Jössang was published in the section "En Bref" of La Recherche (n° 289 (July-Aug.), p. 7, 1996).
(c) Once it has crossed the Curie temperature, a ferromagnetic material recovers only the intrinsic properties it had before ever being submitted to a magnetic field. Under these conditions, the total magnetisation of the sample is nil. For iron, the coercivity is so weak that one can, as a first approximation, consider that for no field, there is no magnetisation.
(d) Barlow, Pierre (1780-1862) : well-known English scholar; Cavallo, Tiberius (1749-1809) : physician who was born in Naples and died in London.
1. Drion C, Fernet E. Traité de Physique Élémentaire; Masson, Paris, 1872.
2. Jamin J. Cours de Physique de l'Ecole Polytechnique. 3rd Edn, vol. 1, Gauthier-Villars, Paris, 1871.
3. Pouillet M. Élémens de Physique Expérimentale et de Météorologie. Ed. Béchet Jeune, Place de l'Ecole de Médecine, n° 4, Paris, 1832.
4. Rocard Y. Electricité. Masson et Cie, Paris, 1956.
5. Curie P. Propriétés magnétiques des corps à diverses températures (thèse). Annales de Chimie et de Physique, series 7, n° 5, 289-405, 1895.
6. Curie P. Notes by M.P. Curie, presented by M. Lippmann. Comptes Rendus de l'Académie des Sciences, (Propriétés magnétiques des corps à diverses températures (115, 805, 1892; 116, 136, 1893; 118, 1134, 1894); Sur les propriétés magnétiques de l'oxygène à diverses températures (115, 1292, 1892); Propriétés magnétiques du fer à diverses températures (118, 796 and 859, 1894).
7. Hurwic A. Pierre Curie. Flammarion, Paris, 1995.
8. Daguin PA. Traité Élémentaire de Physique. Vol. 3, Delagrave, Paris, 1867.