She shared the Nobel Prize in Physics in 1903, the same year she completed her doctoral dissertation. She, her husband Pierre Curie, and Henri Becquerel, were awarded the prize for their pioneering work on radioactivity. She was awarded the 1911 Nobel prize in Chemistry for her work in the discovery and isolation of the element radium. The name Curie lives on in the periodic table and among scientific units: the discoverers of element 96 named it curium, and a standard unit of radioactivity is called the curie.
The fact that Marie Curie remains the only person to have won two Nobel Prizes in different sciences is sufficient testimony to the significance of her work and her place in the history of science. The fact that she did this work at a time when women were nearly non-existent in academic or industrial laboratories has given Curie nearly mythic status. Her achievements are so compelling that they captured the attention even of Hollywood (Madame Curie, 1943, starring Greer Garson). She is the female scientist nearly everyone can name.
Before discovering elements, though, Curie carried out the work reported in this selection, laying important foundations for further work on radioactivity. Neither Curie nor anyone else in 1898 had any idea of the nature of radioactivity or its cause. The work described here is a model observational survey of a new field. Explanations of radioactivity would only come after the systematic accumulation and organization of information about the phenomenon.
I have studied the conductance of air under the influence of the uranium rays discovered by M. Becquerel, and I examined whether substances other than compounds of uranium were able to make the air a conductor of electricity. In this research I employed a parallel-plate condenser; one of the plates was covered with a uniform layer of uranium or of another finely pulverized sample. (Diameter of the plates 8 cm; separation 3 cm.) One establishes a potential difference of 100 volts between the plates. The absolute value of the current which traversed the condenser was measured by means of an electrometer and a piezoelectric quartz.
I examined a large number of metals, salts, oxides, and minerals.,  The following Table gives, for each substance, the magnitude of the current i in units of 10-11 amperes. The substances which I studied but omitted from the Table are at least 100 times less active than uranium.
All the uranium compounds studied are active, and are, in general, more active to the extent that they contain more uranium.
Amperes Uranium containing some carbon 24x10-12 Black oxide of uranium (U2O5) 27 " Green oxide of uranium (U3O8) 18 " Uranates of ammonium, potassium, or sodium (approximately) 12 " Hydrated uranic acid 6 " Uranyl nitrate, uranous sulfate, uranyl potassium sulfate (approximately) 7 " Artificial chalcolite (uranyl copper phosphate) 9 " Thorium oxide in a layer 0,25 mm thick 22 " Thorium oxide in a layer 6 mm thick 53 " Thorium sulfate 8 " Potassium fluoxytantalate 2 " Potassium fluoxyniobate and cerium oxide 0,3 " Pitchblende of Johanngeorgenstadt 83 " " of Cornwallis 16 " " of Joachimsthal and of Pzibran 67 " Natural chalcolite 52 " Autunite 27 " Various thorites from 2 to 14 " Orangite 20 " Samarskite 11 " Fergusonite, monazite, xenotime, niobite, eschynite from 3 to 7 " Cleveite very active
The compounds of thorium are very active. Thorium oxide surpasses even metallic uranium in activity.
It is remarkable that the two most active elements, uranium and thorium, are the ones which possess the greatest atomic weight.
Cerium, niobium, and tantalum appear to be slightly active.
White phosphorus is very active, but its action is probably of a different nature than that of uranium and thorium. In fact, phorphorus is not active in the form of red phosphorus or in the state of phosphates.
All the minerals which demonstrate activity contain active elements. Two minerals of uranium, pitchblende (a uranium oxide) and chalcolite (uranyl copper phosphate) are much more active than uranium itself. This fact is most remarkable, and suggests that these minerals may contain an element much more active than uranium. I prepared chalcolite from pure reagents according to the procedure of Debray; this artificial chalcolite is no more active than other uranium salts.
Absorption.-- The effects produced by active substances increase with the thickness of the sample layer. This augmentation is very weak for the compounds of uranium; it is considerable for thorium oxide, which thus seems partially transparent to the rays it emits.
To study the transparence of various substances, one places a thin plate of them over the active layer. The absorption is always very strong. However, the rays pass through metals, glass, and paper of slight thickness. Here are the fractions of radiation transmitted through a sheet of aluminum of thickness 0,01 mm:
One can see that compounds of the same metal emit rays absorbed to an equal extent. The rays emitted by thorium are more penetrating than those emitted by uranium; finally, thorium oxide in a thick layer emits rays much more penetrating than those which it emits from a thin layer.
mm 0,2 for uranium, ammonium uranate, uranous oxide, artificial chalcolite 0,33 for pitchblende and natural chalcolite 0,4 for thorium oxide and thorium sulfate in a 0,5-mm layer 0,7 for thorium oxide in a 6-mm layer.
Photographic images.-- I have obtained good photographic images with uranium, uranous oxide, pitchblende, chalcolite, and thorium oxide. These bodies act at a small distance, whether through air, through glass, or through aluminum. Thorium sulfate gives weaker images, and potassium fluoxytantalate very weak images.
Analogy to the secondary rays of Röntgen rays.-- The properties of the rays emitted by uranium and thorium are very similar to those of the secondary rays of Röntgen rays, studied recently by M. Sagnac. I have ascertained, moreover, that the uranium, pitchblende, and thorium oxide emit, under the action of Röntgen rays, secondary rays which, from the point of view of discharging electrified bodies, generally have a greater effect than secondary rays from lead. Among the metals studied by M. Sagnac, uranium and thorium come to be placed beside and beyond lead.
To interpret the spontaneous radiation of uranium and thorium, one could imagine that all space is constantly traversed by rays analogous to Röntgen rays but much more penetrating and unable to be absorbed except by certain elements with high atomic weight such as uranium and thorium.
Paints containing radium were used on watches with luminous dials early in the 20th century.
This work was done at the Municipal School of Industrial Physics and Chemistry. [Curie's original note; Pierre Curie had been head of the laboratory at this Paris institution since 1882.--CJG]
See Becquerel 1896, annotated in the previous chapter.
The ability to make air a conductor of electricity was quickly recognized as a property of radioactive substances and of X-rays too, for that matter. Today we recognize this conductance of air as due to ionization: a ray knocks an electron from a molecule in the air, leaving behind a positively charged atom or molecule and free electrons. As these charged particles move, their electrical charge moves with them; thus we have an electric current (motion of electrical charges) conducted through air.
The size of the current is a measure of the extent of ionization of the air near the radioactive substance, and therefore a measure of the activity of the substance (strength of the radioactivity). View a photo of the Curies with their electrometer (at Musˇe Curie).
The uranium used for this study was given by M. Moissan. The salts and oxides were pure products from the laboratory of M. Étard at the School of Physics and Chemistry. M. Lacroix was willing to procure for me some mineral samples of known provenance from the collection of the Museum. Certain rare and pure oxides were given by M. Demarçay. I thank these gentlemen for their courtesy. [Curie's original note --CJG]
Becquerel had found radioactivity in some uranium-containing crystals. A natural question was whether other substances were also active. As she wrote in her dissertation [Curie 1903], "... I made experiments to discover whether substances other than compounds of uranium and thorium were radio-active. I undertook this research with the idea that it was scarcely probable that radio-activity ... should belong to a certain kind of matter to the exclusion of all other."
The list of scientists who supplied Curie with samples includes two others who discovered or isolated elements. Henri Moissan had isolated fluorine in 1886, for which he would be awarded the Nobel prize in chemistry in 1906. Eugène Demarçay had discovered the rare earth metal europium in 1896. Demarçay would provide spectroscopic confirmation of the Curies' discovery of radium later in 1898 [Demarçay 1898].
Curie gave a bit more information in her dissertation [Curie 1903] about the substances she tested for activity, most of which she omits from this paper because they proved to be inactive. They included all the metals and non-metallic elements easily obtainable; a large number of rocks and minerals (e.g., naturally occuring inorganic substances not necessarily chemically purified); and an inventory of rare elements such as gallium, germanium, niobium, rubidium, and several rare earth metals.
Uranium forms a complicated series of compounds with oxygen including UO2, UO3, and several compounds with oxygen-to-uranium ratios between 2 and 3, including U3O8. As a result, substances with oxygen-to-uranium ratios anywhere between 2 and 3 are possible, but these are usually mixtures (solid solutions) of the various compounds.
A uranate is a compound containing the complex anion UO42-, so the compounds named here are (NH4)2UO4, K2UO4, and Na2UO4.
Uranic acid is H2UO4; hydrated uranic acid has several units of water a part of the crystal, H2UO4.H2O.
The complex cation UO22+ is named uranyl. So uranyl nitrate is UO2(NO3)2 and uranyl potassium sulfate K2UO2(SO4)2. Most likely there substances were used in the form of their common hydrates, UO2(NO3)2.6H2O and K2UO2(SO4)2.2H2O. The latter is the material in which Becquerel first found radioactivity [Becquerel 1896]. Uranous sulfate is U2(SO4)3.
Chalcolite, also known as torbernite, is a copper uranyl phosphate with 8 to 12 units of water in the crystal structure: Cu(UO2)2(PO4)2.8-12H2O. Curie makes the distinction between artificial chalcolite, which she prepared herself, and natural chalcolite, a mineral found in nature, because the two materials gave widely different results, as discussed below the table.
Thorium oxide is ThO2.
Thorium sulfate is Th(SO4)2. This compound can include various numbers of units of water in the crystal structure.
Cerium has two common oxides, CeO2 and Ce2O3.
Pitchblende is an amorphous black ore of uranium, consisting chiefly of uranium oxides. It can contain appreciable quantities of thorium as well, and, as the Curies would learn, traces of polonium and radium. Pitchblende is not a compound, i.e., not a substance of fixed chemical composition. It is not altogether surprising, then, to see pitchblende from different mines exhibiting different activity.
Autunite is a hydrated calcium uranyl phosphate: Ca(UO2)2(PO4)2.10H2O.
Thorite is a mineral consisting primarily of thorium silicate, ThSiO4.
Orangite is a thorium oxide.
Samarskite is a complex mineral. It may be thought of as an yttrium niobium oxide, YNb2O6, in which several other elements can substitute for the yttrium or niobium. For the purposes of this study, it is important to note that uranium and thorium are among the elements that can substitute for yttrium; indeed, samarskite is considered an important ore of uranium.
Fergusonite is another yttrium niobium oxide, Y2O3.Nb2O5, or yttrium tantalum oxide, Y2O3.Ta2O5. Monazite is a rare earth phosphate mineral which can have considerable substitution of thorium for the rare earth metal and of silicon for the phosphorus. Xenotime is yttrium phosphate, YPO4. Niobite, also known as columbite, is FeNb2O6, but manganese can substitute for iron and tantalum for niobium. Eschynite can be thought of as a calcium titanium oxide, CaTi2O6, but several elements including thorium can substitute for calcium and niobium can substitute for titanium. It is worth noting that uranium and thorium are chemically rather similar to the rare earth metals, so it would not be surprising to find some uranium or thorium in natural samples of these minerals, whether or not they are recognized variations of the minerals.
Cleveite is a mineral composed mainly of UO2, with significant substitution of rare earth metals such as yttrium for the uranium.
This statement is more profound and much less obvious than it may appear. The uranium compounds and minerals in the table contain uranium in a variety of chemical conditions including metal, salt, and oxide. The valence of uranium ranges from 0 in uranium metal to 3 in uranous sulfate to 6 in the uranyl compounds. Many properties of these substances differ because of the different chemical states of the uranium. But the radioactivity of these substances does not seem to depend on the chemical state of the uranium, but only on the quantity of uranium.
This observation combined with the next two can be summarized by saying that radioactivity is an atomic property [Curie 1903]. That is, radioactivity is a property of a particular element (or rather, of a particular nucleus or isotope to use later terminology); its strength depends only on the quantity of the active element and is independent of its chemical state.
Before Curie carried out this research, radioactivity had been found only in substances which contained uranium. Gerhard Schmidt independently discovered radioactivity in thorium and reported his work just weeks before Curie's paper was presented (March 24 vs. April 12).
Thorium and uranium are no longer at the end of the periodic table, but uranium remains the heaviest naturally occuring element. Curie's observation at the time suggested, but did not prove, a relationship between radioactivity and heavy elements. That relationship is now well established: all the elements with atomic number greater than that of bismuth (31 elements at latest count: elements 84-112, 114, and 116) are radioactive.
Curie's dissertation [Curie 1903] does not include this conclusion, but rather states: "All the minerals which showed radio-activity contained uranium or thorium." The slight activity observed for these elements was most likely due to impurities of uranium or thorium or perhaps even smaller traces of the more strongly radioactive polonium or radium.
Curie expanded this point in her dissertation [Curie 1903]. White phosphorus is "active" in that it causes the air near it to conduct electricity. But its activity is different than that of thorium and uranium in two important respects. First, the activity of white phosphorus is accompanied by a discernible chemical change (oxidation), but that of uranium and thorium exhibits no such concomitant change. Second, the activity of phosphorus is limited to a specific chemical form (elemental white phosphorus), while that of uranium and thorium occurs in any and all of it chemical forms. Thus Curie could distinguish between the activity of uranium and thorium (radioactivity) and the ionization of air caused by a different phenomenon by careful comparison, even without understanding the cause of radioactivity.
If these minerals have more radioactivity than could be accounted for by the amount of uranium they contain, there must be another radioactive substance present. The comparison between natural and artificial chalcolite is particularly strong evidence for this conclusion. The artificial chalcolite, prepared from pure materials, contains only copper, uranium, oxygen, phosphorus, and hydrogen; the natural may contain traces of other elements as well. Any difference in activity, then, must be due to these traces.
Curie had good reason to suspect that this activity was due to an element as yet undiscovered, for she had tested most if not all of the known elements for activity. And she already had a powerful tool with which to pursue the new element: after any chemical reaction which would separate some elements in a sample from others, she could test the resulting fractions for activity and know that the unknown element was in the more active fraction. The prospect of discovering a new radioactive substance was so promising that Pierre Curie set aside his own line of research to assist Marie in this work. Their work was rewarded by discovery of not one but two new elements, polonium and radium. [Curie & Curie 1898; Curie, Curie, & Bémont 1898]
Curie turned her attention to an aspect of radioactivity which made them interesting in the first place: their ability to penetrate opaque materials at least to some extent. A more detailed examination of penetrating power of radioactivity from a single source led Ernest Rutherford to distinguish between two different kinds of rays, which he labeled α and β [Rutherford 1899]. Like this paper, Rutherford's is a careful phenomenological study, systematically accumulating and organizing information about radioactivity before attempting any explanation of it.
The heading mm is obviously mistaken: the numbers represent fraction of activity remaining after passage through aluminum of a given thickness. At first one might think that the thicker the layer, the more active it should be. After all, the thicker the layer, the more active atoms are present. But the rays, whatever they are, can penetrate materials only to a certain extent. Thus, a sample of greater thickness will appear to be more active only if its rays are strong enough to penetrate the extra thickness of the sample.
Scientific conclusions are stronger, their validity more certain, when they can be derived from more than one line of evidence, more than one kind of measurement. Thus the observations on the relative ease of absorption of the rays reinforce some of the conclusions already based on the amount of activity: rays from the same element are the same in penetrating power, regardless of the chemical state of the element; rays from different elements are different in penetrating power as well as in quantity; the two minerals which Curie had already suspected harbor a new radioactive element have rays whose penetrating power is different from those of uranium and thorium.
Among the hot topics in physics at the close of the 19th century were X-rays (Röntgen rays) discovered in 1895 [Röntgen 1895], radioactivity (sometimes called Becquerel rays or uranium rays) discovered in 1896, and electrons (known at the time as cathode rays and corpuscles) whose study continued after significant progress in 1897 [Thomson 1897]. There were a lot of rays under investigation. Some of them seemed to share several properties, and investigations of one kind of ray sometimes led to discoveries of another. (See previous chapter.) So it was not surprising that Curie would make a comparison to X-ray phenomena.
The secondary rays she mentions are not X-rays themselves, but are most likely electrons ejected by metals exposed to X-rays. Like radioactive rays, electrons could penetrate thin layers of metal foil and make air conduct electricity. (See chapter 16.) In fact, these secondary rays have a different source than radioactivity, but they are the same particle (electron) as the ray Rutherford labeled β.
This speculation is not the correct explanation of radioactivity, but it was not outlandish. Pioneers of radioactivity research were puzzled for a long time by the source of the energy required to emit rays, whatever their nature. This speculation posited a source of energy outside the atom. Furthermore, the notion of space being permeated by something undetectable or nearly so was not foreign to physics at this time, for the concept of an all-pervading material "aether" as a medium for electromagnetic waves still persisted.
At any rate, Curie clearly distinguished her observations from her speculation; her speculation would rise or (in this case) fall to the extent that it agreed with subsequent observations.