by Isaac Asimov
The Lewis-Langmuir concept works so well that it still serves in its original form to account for the more simple and straightforward varieties of behavior among the elements. However, not all the behavior is quite as simple and straightforward as might be thought.
For instance, each of the inert gases—helium, neon, argon, krypton, xenon, and radon—has eight electrons in the outermost shell (except for helium, which has two electrons in its only shell), and this is the most stable possible situation. Atoms of these elements have a minimum tendency to lose or gain electrons and therefore a minimum tendency to engage in chemical reactions. The gases would be inert, as their name proclaims.
However, a “minimum tendency” is not really the same as “no tendency,” but most chemists forgot this truth and acted as though it was ultimately impossible for the inert gases to form compounds. This was not true of all of them. As long ago as 1932, the American chemist Linus Pauling considered the ease with which electrons could be removed from different elements, and noted that all elements without exception, even the inert gases, can be deprived of electrons. This deprivation, however, requires more energy in the case of the inert gases than in that of other elements near them in the periodic table.
The amount of energy required to remove electrons among the elements in any particular family decreases with increasing atomic weight, and the heaviest inert gases, xenon and radon, do not have unusually high requirements. It is no more difficult to remove an electron from a xenon atom, for instance, than from an oxygen atom.
Pauling therefore predicted that the heavier inert gases might form chemical compounds with elements that are particularly prone to accept electrons. The element most eager to accept electrons is fluorine, and that seemed to be the natural target.
Now radon, the heaviest inert gas, is radioactive and is unavailable in any but trace quantities. Xenon, however, the next heaviest, is stable and occurs in small quantities in the atmosphere. The best chance, therefore, would be to attempt to form a compound between xenon and fluorine. However, for thirty years nothing was done in this respect, chiefly because xenon was expensive and fluorine very hard to handle, and chemists felt they had better things to do than chase this particular will-o’-the-wisp.
In 1962, however, the British-Canadian chemist Neil Bartlett—working with a new compound, platinum hexafluoride (PtF6)—found that it was remarkably avid for electrons, almost as much as was fluorine itself. This compound would take electrons away from oxygen, an element that is normally avid to gain electrons rather than lose them. If PtF6 could take electrons from oxygen, it ought to be able to take them from xenon, too. The experiment was tried, and xenon fluoroplatinate (XePtF6), the first compound of an inert gas, was reported.
Other chemists at once sprang into the fray, and a number of xenon compounds with fluorine, with oxygen, or with both were formed, the most stable being xenon difluoride (XeF2). A compound of krypton and fluorine, krypton tetrafluoride (KrF4), has also been formed, as well as a radon fluoride. Compounds with oxygen were also formed. There were, for instance, xenon oxytetrafluoride (XeOF4), xenic acid (H2XeO4), and sodium perxenate (Na4XeO6). Most interesting, perhaps, was xenon trioxide (Xe2O3), which explodes easily and is dangerous. The smaller inert gases—argon, neon, and helium—are more resistant to sharing their electrons than the larger ones, and remain inert for all chemists can do even yet.
Chemists quickly recovered from the initial shock of finding that the inert gases can form compounds: such compounds fit into the general picture after all. Consequently, there is now a reluctance to speak of the gases as inert gases. The alternate name of noble gases is preferred, and one speaks of noble gas compounds and noble gas chemistry. (I think this is a change for the worse. After all, the gases are still inert, even if not completely so. The concept noble, in this context, implies “standoffish” or “disinclined to mix with the common herd,” and is just as inappropriate as inert and, moreover, does not suit a democratic society.)
THE RARE-EARTH ELEMENTS
In addition to the fact that the Lewis-Langmuir scheme was applied too rigidly to the inert gases, it can scarcely be applied at all to many of the elements with atomic numbers higher than 20. In particular, refinements had to be added to deal with a very puzzling aspect of the periodic table having to do with the so-called rare earths—elements 57 to 71, inclusive.
To go back a bit, the early chemists considered any substance that was insoluble in water and unchanged by heat to be an earth (a hangover of the Greek view of “earth” as an element). Such substances included what we would today call calcium oxide, magnesium oxide, silicon dioxide, ferric oxide, aluminum oxide, and so on——compounds that actually constitute about 90 percent of the earth’s crust. Calcium oxide and magnesium oxide are slightly soluble and, in solution, display alkaline properties (that is, opposite to those of acids), and so were called the alkaline earths; when Humphry Davy isolated the metals calcium and magnesium from these earths, they were named alkaline earth metals. The same name was eventually applied to all the elements that fall into the column of the periodic table containing magnesium and calcium: that is, to beryllium, strontium, barium, and radium.
The puzzle to which I have referred began in 1794, when a Finnish chemist, Johan Gadolin, examined an odd rock that had been found near the Swedish hamlet Ytterby and decided that it was a new “earth.” Gadolin gave this “rare earth” the name yttria, after Ytterby. Later the German chemist Martin Heinrich Klaproth found that yttria could be divided into two “earths,” for one of which he kept the name yttria, while he named the other ceria (after the newly discovered planetoid Ceres). But the Swedish chemist Carl Gustav Mosander subsequently broke these down further into a series of different earths. All eventually proved to be oxides of new elements named the rare-earth metals. By 1907, fourteen such elements had been identified. In order of increasing atomic weight they are:
lanthanum (from a Greek word meaning “hidden”)
cerium (from Ceres)
praseodymium (from the Greek for “green twin,” after a green line in its spectrum)
neodymium (“new twin”)
samarium (from “sarnarskite,” the mineral in which it was found)
europium (from Europe)
gadolinium (in honor of Johan Gadolin)
terbium (from Ytterby)
dysprosium (from a Greek word meaning “hard to get at”)
holmium (from Stockholm)
erbium (from Ytterby)
thulium (from Thule, an old name for Scandinavia)
ytterbium (from Ytterby)
lutetium (from Lutetia, an old name for Paris).
On the basis of their X-ray properties, these elements were assigned the atomic numbers from 57 (lanthanum) to 71 (lutetium). As I related earlier, there was a gap at 61 until the missing element, promethium, emerged from the fission of uranium. It made the fifteenth in the list.
Now the trouble with the rare-earth elements is that they apparently cannot be made to fit into the periodic table. It is fortunate that only four of them were definitely known when Mendeleev proposed the table; if they had all been on hand, the table might have been altogether too confusing to be accepted. There are times, even in science, when ignorance is bliss.
The first of the rare-earth metals, lanthanum, matches up all right with yttrium, number 39, the element above it in the table (figure 6.2). (Yttrium, though found in the same ores as the rare earths and similar to them in properties, is not a rare-earth metal. It is, however, named after Ytterby. Four elements honor that hamlet—which is overdoing it.) The confusion begins with the rare earth after lanthanum—namely, cerium—which ought to resemble the element following yttrium—that is, zirconium. But it does nothing of the sort; instead, it resembles yttrium again. And the same is true of all fifteen of the rare-earth elements: they strongly resemble yttrium and one another (in fact, they are so alike chemically that at first they could not be separated except by the most tedious procedu
res), but they are not related to any other elements preceding them in the table. We have to skip the whole rare-earth group and go on to hafnium, element 72, to find the element related to zirconium, the one after yttrium.
Figure 6.2. The electron shells of lanthanum. Note that the fourth subshell of the N-shell has been skipped and is empty.
Baffled by this state of affairs, chemists could do no better than to group all the rare-earth elements into one box beneath yttrium and list them individually in a kind of footnote to the table.
THE TRANSITIONAL ELEMENTS
The answer to the puzzle finally came as a result of details added to the Lewis-Langmuir picture of the electron-shell structure of the elements.
In 1921, C. R. Bury suggested that the shells were not necessarily limited to 8 electrons apiece. Eight always sufficed to satisfy the outer shell. But a shell might have a greater capacity when it was not on the outside. As one shell built on another, the inner shells might absorb more electrons, and each succeeding shell might hold more than the one before. Thus the K-shell’s total capacity would be 2 electrons, the Lshell’s 8, the M-shell’s 18, the N-shell’s 32, and so on—the step-ups going according to a pattern of successive squares multiplied by two (that is, 2 × 1, 2 × 4, 2 × 9, 2 × 16, etc.).
This view was backed up by a detailed study of the spectra of the elements. The Danish physicist Niels Henrik David Bohr showed that each electron shell was made up of subshells at slightly different energy levels. In each succeeding shell, the spread of the subshells was greater, so that soon the shells overlapped. As a result, the outermost subshell of an interior shell (say, the M-shell) might actually be farther from the center, so to speak, than the innermost subshell of the next shell beyond it (say, the N-shell), This being so, the N-shell’s inner subshell might fill with electrons while the M-shell’s outer subshell was still empty.
An example will make this clearer. The M-shell, according to the theory, is divided into three subshells, whose capacities are 2, 6, and 10 electrons, respectively, making a total of 18. Now argon, with 8 electrons in its M-shell, has filled only two inner subshells. And, in fact, the M-shell’s third, or outermost, subshell will not get the next electron in the element-building process, because it lies beyond the innermost subshell of the N-shell: that is, in potassium, the element after argon, the nineteenth electron goes, not into the outermost subshell of M, but into the innermost subshell of N. Potassium, with I electron in its N-shell, resembles sodium, which has I electron in its M-shell. Calcium, the next element (20), has 2 electrons in the N-shell and resembles magnesium, which has 2 in the M-shell. But now the innermost subshell of the N-shell, having room for only 2 electrons, is full. The next electrons to be added can start filling the outermost subshell of the M-shell, which so far has not been touched. Scandium (21) begins the process, and zinc (30) completes it. In zinc, the outermost subshell of the M-shell has at last acquired its complement of 10 electrons. The 30 electrons of zinc are distributed as follows: 2 in the K-shell, 8 in the L-shell, 18 in the M-shell, and 2 in the N-shell. At this point, electrons can resume the filling of the N-shell. The next electron gives the N-shell 3 electrons and forms gallium (31), which resembles aluminum, with 3 in the M-shell.
The point is that elements 21 to 30, formed on the road to filling a subshell that had been skipped temporarily, are transitional elements. Note that calcium resembles magnesium, and gallium resembles aluminum. Now magnesium and aluminum are adjacent members of the periodic table (numbers 12 and 13). But calcium (20) and gallium (31) are not. Between them lie the transitional elements, and these introduce a complication in the periodic table.
The N-shell is larger than the M-shell and is divided into four subshells instead of three: they can hold 2, 6, 10, and 14 electrons, respectively. Krypton, element 36, fills the two innermost subshells of the N-shell, but here the innermost subshell of the overlapping O-shell intervenes, and, before electrons can go on to N’s two outer subshells, they must fill that one. The element after krypton, rubidium (37), has its thirty-seventh electron in the O-shell. Strontium (38) completes the filling of the two-electron O-subshell. Thereupon a new series of transitional elements proceeds to fill the skipped third subshell of the N-shell. With cadmium (48) this is completed; now N’s fourth and outermost subshell is skipped, while electrons fill O’s second innermost subshell, ending with xenon (54).
But even now N’s fourth subshell must bide its turn; for by this stage, the overlapping has become so extreme that even the P-shell interposes a subshell that must be filled before N’s last. After xenon come cesium (55) and barium (56), with 1 and 2 electrons, respectively, in the P-shell. It is still not N’s turn: the fifty-seventh electron, surprisingly, goes into the third subshell of the O-shell, creating the element lanthanum (figure 6.3). Then, and only then, an electron at long last enters the outermost subshell of the N-shell. One by one the rare-earth elements add electrons to the N-shell until element 71, lutetium, finally fills it. Lutetium’s electrons are arranged thus: 2 in the K-shell, 8 in the L-shell, 18 in the M-shell, 32 in the N-shell, 9 in the O-shell (two subshells full plus 1 electron in the next subshell), and 2 in the P-shell (innermost subshell full).
Figure 6.3. Schematic representation of the overlapping of electron shells and subshells in lanthanum. The outermost subshell of the N-shell has yet to be filled.
Now at last we begin to see why the rare-earth elements, and some other groups of transitional elements, are so alike. The decisive thing that differentiates elements, as far as their chemical properties are concerned, is the configuration of electrons in their outermost shell. For instance, carbon, with four electrons in its outermost shell, and nitrogen, with five, are completely different in their properties. On the other hand, in sequences where electrons are busy filling inner subshells while the outermost shell remains unchanged, the properties vary less. Thus iron, cobalt, and nickel (elements 26, 27, and 28), all of which have the same outer-shell electronic configuration—an N-subshell filled with two electrons—are a good deal alike in chemical behavior. Their internal electronic differences (in an M-subshell) are largely masked by their surface electronic similarity. And this goes double for the rare-earth elements. Their differences (in the N-shell) are buried under, not one, but two outer electronic configurations (in the O-shell and the P-shell), which in all these elements are identical. Small wonder that the elements are chemically as alike as peas in a pod.
Because the rare-earth metals have so few uses, and are so difficult to separate, chemists made little effort to do so—until the uranium atom was fissioned. Then it became an urgent matter indeed, because radioactive varieties of some of these elements were among the main products of fission; and in the atomic bomb project, it was necessary to separate and identify them quickly and cleanly.
The problem was solved in short order by use of a chemical technique first devised in 1906 by the Russian botanist Mikhail Semenovich Tswett, who named it chromatography (“writing in color”). Tswett had found that he could separate plant pigments, chemically very much alike, by washing them down a column of powdered limestone with a solvent. He dissolved his mixture of plant pigments in petroleum ether and poured this on the limestone. Then he proceeded to pour in clear solvent. As the pigments were slowly washed down through the limestone powder, each pigment moved down at a different rate, because each differed in strength of adhesion to the powder. The result was that they separated into a series of bands, each of a different color. With continued washing, the separated substances trickled out separately at the bottom of the column, one after the other.
The world of science for many years ignored Tswett’s discovery, possibly because he was only a botanist and only a Russian, while the leaders of research on separating difficult-to-separate substances at the time were German biochemists.
But, in 1931, a German biochemist, Richard Willstatter, rediscovered the process, whereupon it came into general use. (Willstatter had received the 1915 Nobel Prize in chemis
try for his excellent work on plant pigments. Tswett, so far as I know, has gone unhonored.)
Chromatography through columns of powder was found to work on almost all sorts of mixtures——colorless as well as colored. Aluminum oxide and starch proved to be better than limestone for separating ordinary molecules. Where ions are separated, the process is called ion exchange; and compounds known as zeolites were the first efficient agents applied for this purpose. Calcium and magnesium ions could be removed from hard water, for instance, by pouring the water through a zeolite column. The calcium and magnesium ions adhere to the zeolite and are replaced in solution by the sodium ions originally present on the zeolite, so soft water drips out of the bottom of the column. The sodium ions of zeolite have to be replenished from time to time by pouring in a concentrated solution of salt (sodium chloride). In 1935, a refinement came with the development of ion-exchange resins. These synthetic substances can be designed for the job to be done. For instance, certain resins will substitute hydrogen ions for positive ions, while others substitute hydroxyl ions for negative ions; a combination of both types will remove most of the salts from sea water. Kits containing such resins were part of the survival equipment on life rafts during the Second World War.
It was the American chemist Frank Harold Spedding who adapted ion-exchange chromatography to the separation of the rare earths. He found that these elements came out of an ion-exchange column in the reverse order of their atomic number, so that they were not only quickly separated but also identified. In fact, the discovery of promethium, the missing element 61, was confirmed in this way from the tiny quantities found among the fission products.