| Group 3 in the periodic table | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| |||||||||
| ↓ Period | |||||||||
| 4 |  21 Transition metal | ||||||||
| 5 |  39 Transition metal | ||||||||
| 6 |  71 Lanthanide | ||||||||
| 7 | Lawrencium (Lr) 103 Actinide | ||||||||
|
| |||||||||
|
Legend
| |||||||||
|
| |||||||||
Group 3 is the first group of transition metals in the periodic table. This group is closely related to the rare-earth elements. Although some controversy exists regarding the composition and placement of this group, it is generally agreed among those who study the matter that this group contains the four elements scandium (Sc), yttrium (Y), lutetium (Lu), and lawrencium (Lr). The group is also called the scandium group or scandium family after its lightest member.
The chemistry of the group 3 elements is typical for early transition metals: they all essentially have only the group oxidation state of +3 as a major one, and like the preceding main-group metals are quite electropositive and have a less rich coordination chemistry. Due to the effects of the lanthanide contraction, yttrium and lutetium are very similar in properties. Yttrium and lutetium have essentially the chemistry of the heavy lanthanides, but scandium shows several differences due to its small size. This is a similar pattern to those of the early transition metal groups, where the lightest element is distinct from the very similar next two.
All the group 3 elements are rather soft, silvery-white metals, although their hardness increases with atomic number. They quickly tarnish in air and react with water, though their reactivity is masked by the formation of an oxide layer. The first three of them occur naturally, and especially yttrium and lutetium are almost invariably associated with the lanthanides due to their similar chemistry. Lawrencium is strongly radioactive: it does not occur naturally and must be produced by artificial synthesis, but its observed and theoretically predicted properties are consistent with it being a heavier homologue of lutetium. None of them have any biological role.
Historically, sometimes lanthanum (La) and actinium (Ac) were included in the group instead of lutetium and lawrencium, and this option is still commonly found in textbooks. Some compromises between the two major options have been proposed and used, involving either the shrinking of the group to scandium and yttrium only, or the inclusion of all 30 lanthanides and actinides in the group as well.
History
Discoveries of the elements
The discovery of the group 3 elements is inextricably tied to that of the rare earths, with which they are universally associated in nature. In 1787, Swedish part-time chemist Carl Axel Arrhenius found a heavy black rock near the Swedish village of Ytterby, Sweden (part of the Stockholm Archipelago). Thinking that it was an unknown mineral containing the newly discovered element tungsten, he named it ytterbite. Finnish scientist Johan Gadolin identified a new oxide or "earth" in Arrhenius' sample in 1789, and published his completed analysis in 1794; in 1797, the new oxide was named yttria. In the decades after French scientist Antoine Lavoisier developed the first modern definition of chemical elements, it was believed that earths could be reduced to their elements, meaning that the discovery of a new earth was equivalent to the discovery of the element within, which in this case would have been yttrium. Until the early 1920s, the chemical symbol "Yt" was used for the element, after which "Y" came into common use. Yttrium metal, albeit impure, was first prepared in 1828 when Friedrich Wöhler heated anhydrous yttrium(III) chloride with potassium to form metallic yttrium and potassium chloride. In fact, Gadolin's yttria proved to be a mixture of many metal oxides, that started the history of the discovery of the rare earths.
In 1869, Russian chemist Dmitri Mendeleev published his periodic table, which had an empty space for an element above yttrium. Mendeleev made several predictions on this hypothetical element, which he called eka-boron. By then, Gadolin's yttria had already been split several times; first by Swedish chemist Carl Gustaf Mosander, who in 1843 had split out two more earths which he called terbia and erbia (splitting the name of Ytterby just as yttria had been split); and then in 1878 when Swiss chemist Jean Charles Galissard de Marignac split terbia and erbia themselves into more earths. Among these was ytterbia (a component of the old erbia), which Swedish chemist Lars Fredrik Nilson successfully split in 1879 to reveal yet another new element. He named it scandium, from the Latin Scandia meaning "Scandinavia". Nilson was apparently unaware of Mendeleev's prediction, but Per Teodor Cleve recognized the correspondence and notified Mendeleev. Chemical experiments on scandium proved that Mendeleev's suggestions were correct; along with discovery and characterization of gallium and germanium this proved the correctness of the whole periodic table and periodic law. Metallic scandium was produced for the first time in 1937 by electrolysis of a eutectic mixture, at 700–800 °C, of potassium, lithium, and scandium chlorides. Scandium exists in the same ores that yttrium had been discovered from, but is much rarer and probably for that reason had eluded discovery.
The remaining component of Marignac's ytterbia also proved to be a composite. In 1907, French scientist Georges Urbain, Austrian mineralogist Baron Carl Auer von Welsbach, and American chemist Charles James all independently discovered a new element within ytterbia. Welsbach proposed the name cassiopeium for his new element (after Cassiopeia), whereas Urbain chose the name lutecium (from Latin Lutetia, for Paris). The dispute on the priority of the discovery is documented in two articles in which Urbain and von Welsbach accuse each other of publishing results influenced by the published research of the other. In 1909, the Commission on Atomic Mass, which was responsible for the attribution of the names for the new elements, granted priority to Urbain and adopting his names as official ones. An obvious problem with this decision was that Urbain was one of the four members of the commission. In 1949, the spelling of element 71 was changed to lutetium. Later work connected with Urbain's attempts to further split his lutecium however revealed that it had only contained traces of the new element 71, and that it was only von Welsbach's cassiopeium that was pure element 71. For this reason many German scientists continued to use the name cassiopeium for the element until the 1950s. Ironically, Charles James, who had modestly stayed out of the argument as to priority, worked on a much larger scale than the others, and undoubtedly possessed the largest supply of lutetium at the time. Lutetium was the last of the stable rare earths to be discovered. Over a century of research had split the original yttrium of Gadolin into yttrium, scandium, lutetium, and seven other new elements.
Lawrencium is the only element of the group that does not occur naturally. It was first synthesized by Albert Ghiorso and his team on February 14, 1961, at the Lawrence Radiation Laboratory (now called the Lawrence Berkeley National Laboratory) at the University of California in Berkeley, California, United States. The first atoms of lawrencium were produced by bombarding a three-milligram target consisting of three isotopes of the element californium with boron-10 and boron-11 nuclei from the Heavy Ion Linear Accelerator (HILAC). The nuclide 257103 was originally reported, but then this was reassigned to 258103. The team at the University of California suggested the name lawrencium (after Ernest O. Lawrence, the inventor of cyclotron particle accelerator) and the symbol "Lw", for the new element, but "Lw" was not adopted, and "Lr" was officially accepted instead. Nuclear-physics researchers in Dubna, Soviet Union (now Russia), reported in 1967 that they were not able to confirm American scientists' data on 257103. Two years earlier, the Dubna team reported 256103. In 1992, the IUPAC Trans-fermium Working Group officially recognized element 103, confirmed its naming as lawrencium, with symbol "Lr", and named the nuclear physics teams at Dubna and Berkeley as the co-discoverers of lawrencium.
Dispute on composition
The rare-earth elements historically gave very many problems for the periodic table. In 1871, Russian chemist Dmitri Mendeleev (inventor of the periodic table) attempted to put them in the same groups as other elements, but further investigation of the rare earths made it clear that they did not show the necessary valences for those placements to make sense. In 1902, Czech chemist Bohuslav Brauner suggested that the rare earths all belonged in one place on the periodic table: he called this the "asteroid hypothesis", since just as between Mars and Jupiter there is an asteroid belt instead of a planet, so below yttrium there would be all the lanthanides instead of a single element.
With the measurements of the ground-state gas-phase electron configurations of the elements, and their adoption as a basis for periodic table placement, the older form of group 3 containing scandium, yttrium, lanthanum, and actinium gained prominence in the 1940s. The ground-state configurations of caesium, barium and lanthanum are [Xe]6s1, [Xe]6s2 and [Xe]5d16s2. Lanthanum thus emerges with a 5d differentiating electron and on these grounds it was considered to be "in group 3 as the first member of the d-block for period 6". A superficially consistent set of electron configurations was then seen in group 3: scandium [Ar]3d14s2, yttrium [Kr]4d15s2, lanthanum [Xe]5d16s2 and actinium [Rn]6d17s2. Still in period 6, ytterbium was erroneously assigned an electron configuration of [Xe]4f135d16s2 and lutetium [Xe]4f145d16s2, which suggested that lutetium was the last element of the f-block. This format thus results in the f-block coming between and splitting apart groups 3 and 4 of the d-block.
However, later spectroscopic work found that the correct electron configuration of ytterbium was in fact [Xe]4f146s2. This meant that ytterbium and lutetium—the latter with [Xe]4f145d16s2—both had 14 f-electrons, "resulting in a d- rather than an f- differentiating electron" for lutetium and making it an "equally valid candidate" with [Xe]5d16s2 lanthanum, for the group 3 periodic table position below yttrium. This would result in a group 3 with scandium, yttrium, lutetium, and lawrencium. The first to point out these implications were the Russian physicists Lev Landau and Evgeny Lifshitz in 1948: their textbook Course of Theoretical Physics stated "In books on chemistry, lutetium is also placed with the rare-earth elements. This, however is incorrect, since the 4f shell is complete in lutetium." After Landau and Lifshitz made their statement, many physicists likewise supported the change in the 1960s and 1970s, focusing on many properties such as the crystal structure, melting points, conduction band structure, and superconductivity in which lutetium matches the behaviour of scandium and yttrium, but lanthanum is distinct. This form requires no split blocks.
Some authors had also arrived at this conclusion via other means. In 1905, before lutetium had been discovered, Swiss chemist Alfred Werner already placed lanthanum in a different column from scandium and yttrium because of its distinct chemical behaviour. French engineer Charles Janet had also placed lutetium under yttrium in 1928. The Soviet chemist Chistyakov noted in 1968 that secondary periodicity was fulfilled in group 3 only if lutetium was included in it rather than lanthanum. However, the chemical community largely ignored these conclusions. The philosopher of science Eric Scerri suggests that a factor may have been that several authors who proposed this change were physicists.
The American chemist William B. Jensen collected many of the above arguments in a concerted 1982 plea to chemists to change their periodic tables and put lutetium and lawrencium in group 3. Besides those physical and chemical arguments, he also pointed out that the configurations of lanthanum and actinium are better considered as irregular, similar to how thorium was even then universally treated. Thorium has no f-electrons in its ground state (being [Rn]6d27s2), but was and is universally placed as an f-block element with an irregular ground-state gas-phase configuration replacing the ideal [Rn]5f27s2. Lanthanum and actinium could then be considered similar cases where an ideal f1s2 configuration is replaced by a d1s2 configuration in the ground state. Since most of the f-block elements in fact have an fns2 configuration, and not an fn−1d1s2 configuration, the former is strongly suggested as the ideal general configuration for the f-block elements. This reassignment similarly creates a homologous series of configurations in group 3: in particular, the addition of a filled f-shell to the core passing from yttrium to lutetium is exactly analogous to what happens down every other d-block group.
In any case, ground-state gas-phase configurations consider only isolated atoms as opposed to bonding atoms in compounds (the latter being more relevant for chemistry), which often show different configurations. The idea of irregular configurations is supported by low-lying excited states: despite not having an f-electron in its ground state, lanthanum nevertheless has f-orbitals of low enough energy that they may be used for chemistry, and this affects the physical properties that had been adduced as evidence for the proposed reassignment. (Scandium, yttrium, and lutetium have no such low-lying available f-orbitals.) The irregular configuration of lawrencium ([Rn]5f147s27p1 rather than [Rn]5f146d17s2) can similarly be rationalised as another (albeit unique) anomaly due to relativistic effects that become important for the heaviest elements. These irregular configurations in the 4f elements are the result of strong interelectronic repulsion in the compact 4f shell, with the result that when the ionic charge is low, a lower energy state is obtained by moving some electrons to the 5d and 6s orbitals which do not suffer such large interelectronic repulsion, even though the 4f energy level is normally lower than the 5d or the 6s one: a similar effect happens early in the 5f series.
In 1988, a IUPAC report was published that touched on the matter. While it wrote that electron configurations were in favour of the new assignment of group 3 with lutetium and lawrencium, it instead decided on a compromise where the lower spots in group 3 were instead left blank, because the traditional form with lanthanum and actinium remained popular. This could be interpreted either as shrinking group 3 to scandium and yttrium only, or as including all lanthanides and actinides in group 3; the latter interpretation is in effect a return to Brauner's asteroid hypothesis. In either case, the f-block appears with 15 elements, despite quantum mechanics dictating that it should have 14. Such a table appears in many IUPAC publications; despite being commonly labelled "IUPAC periodic table", it is not actually officially supported by IUPAC. The following IUPAC Red Book (a collection of nomenclature rules for inorganic chemistry) of 1990 displayed the periodic table in three formats, 8-column (based on Mendeleev's original), 18-column, and 32-column; the first two showed the compromise with all lanthanides and actinides in group 3, but the third showed lutetium and lawrencium alone under yttrium in group 3. In 2005, the 8-column and 32-column forms were dropped from the IUPAC Red Book, leaving only the compromise version in 18-column format.
This compromise did not stop the debate. Although some chemists were convinced by the arguments to reassign lutetium to group 3, many continued to show lanthanum in group 3, either because they did not know of the arguments or were unconvinced by them. Most research on the matter tended to support the proposed reassignment of lutetium to group 3. However, in chemistry textbooks, the traditional form continued being the most popular up to the 2010s, although it gradually lost some ground to both the new form with lutetium and the compromise form. Some textbooks even inconsistently showed different forms in different places. Laurence Lavelle went further, defending the traditional form with lanthanum in group 3 on the grounds of neither lanthanum nor actinium having valence f-electrons in the ground state, giving rise to heated debate. Jensen later rebutted this by pointing out the inconsistency of Lavelle's arguments (since the same was true of thorium and lutetium, which Lavelle placed in the f-block) and the evidence for irregular configurations. Scerri, who has published widely on this issue, has noted that Jensen's case based on physical and chemical properties is not conclusive because of its selectivity, pointing to other choices of properties that seem to support lanthanum in group 3 instead of lutetium. Nonetheless, he has also consistently supported lutetium in group 3 on the basis of avoiding a split in the d-block, and has also referred to the fact that electron configurations are approximations and the problem of thorium.
In December 2015 an IUPAC project, chaired by Scerri and including (among others) Jensen and Lavelle, was established to make a recommendation on the matter. Its preliminary report was published in January 2021. It concluded that none of the criteria previously invoked in the debate gave a clear-cut resolution of the question and that ultimately the question rested on convention rather than being something that was objectively scientifically decidable. As such, it suggested "a degree of convention" to be used for "selecting a periodic table that can be presented as the best compromise table that combines objective factors as well as interest dependence", for presentation to "the widest possible audience of chemists, chemical educators and chemistry students". Three desiderata were given: (1) all elements should be displayed in order of increasing atomic number, (2) the d-block should not be split into "two highly uneven portions", and (3) the blocks should have the widths 2, 6, 10, and 14 in accordance with the quantum mechanical basis of the periodic table. The block assignment was admitted to be approximate, just like the assignment of electron configurations: the case of thorium was specifically remarked on. These three desiderata are only fulfilled by the table with lutetium and lawrencium in group 3; the traditional form of group 3 with lanthanum violates (2), and the compromise form of group 3 with all lanthanides and actinides violates (3). As such, the form with lutetium in group 3 was suggested as a compromise.
The project ended in December 2021. Currently, IUPAC's website on the periodic table still shows the 1988 compromise, but mentions the group 3 problem and project to resolve it, and writes "Stay tune[d]".
Characteristics
Chemical
| Electron configurations of the group 3 elements | |||
|---|---|---|---|
| Z | Element | Electron configuration | |
| 21 | Sc, scandium | 2, 8, 9, 2 | [Ar] 3d1 4s2 |
| 39 | Y, yttrium | 2, 8, 18, 9, 2 | [Kr] 4d1 5s2 |
| 71 | Lu, lutetium | 2, 8, 18, 32, 9, 2 | [Xe] 4f14 5d1 6s2 |
| 103 | Lr, lawrencium | 2, 8, 18, 32, 32, 8, 3 | [Rn] 5f14 6d0 7s2 7p1 |
Like other groups, the members of this family show patterns in their electron configurations, especially the outermost shells, resulting in trends in chemical behavior. Due to relativistic effects that become important for high atomic numbers, lawrencium's configuration has an irregular 7p occupancy instead of the expected 6d, but the regular [Rn]5f146d17s2 configuration turns out to be low enough in energy that no significant difference from the rest of the group is observed or expected.
Most of the chemistry has been observed only for the first three members of the group; chemical properties of lawrencium are not well-characterized, but what is known and predicted matches its position as a heavier homolog of lutetium. The remaining elements of the group (scandium, yttrium, lutetium) are quite electropositive. They are reactive metals, although this is not obvious due to the formation of a stable oxide layer which prevents further reactions. The metals burn easily to give the oxides, which are white high-melting solids. They are usually oxidized to the +3 oxidation state, in which they form mostly ionic compounds and have a mostly cationic aqueous chemistry. In this way they are similar to the lanthanides, although they lack the involvement of f orbitals that characterises the chemistry of the 4f elements lanthanum through ytterbium. The stable group 3 elements are thus often grouped with the 4f elements as the so-called rare earths.
The stereotypical transition-metal properties are mostly absent from this group, as they are for the heavier elements of groups 4 and 5: there is only one typical oxidation state and the coordination chemistry is not very rich (though high coordination numbers are common due to the large size of the M3+ ions). This said, low-oxidation state compounds may be prepared and some cyclopentadienyl chemistry is known. The chemistries of group 3 elements are thus mostly distinguished by their atomic radii: yttrium and lutetium are very similar, but scandium stands out as the least basic and the best complexing agent, approaching aluminium in some properties. They naturally take their places together with the rare earths in a series of trivalent elements: yttrium acts as a rare earth intermediate between dysprosium and holmium in basicity; lutetium as less basic than the 4f elements and the least basic of the lanthanides; and scandium as a rare earth less basic than even lutetium. Scandium oxide is amphoteric; lutetium oxide is more basic (although it can with difficulty be made to display some acidic properties), and yttrium oxide is more basic still. Salts with strong acids of these metals are soluble, whereas those with weak acids (e.g. fluorides, phosphates, oxalates) are sparingly soluble or insoluble.
Physical
The trends in group 3 follow those of the other early d-block groups and reflect the addition of a filled f-shell into the core in passing from the fifth to the sixth period. For example, scandium and yttrium are both soft metals. But because of the lanthanide contraction, the expected increase in atomic radius from yttrium to lutetium is in fact more than cancelled out; lutetium atoms are slightly smaller than yttrium atoms, but are heavier and have a higher nuclear charge. This makes the metal more dense, and also harder because the extraction of the electrons from the atom to form metallic bonding becomes more difficult. All three metals have similar melting and boiling points. Very little is known about lawrencium, but calculations suggest it continues the trend of its lighter congeners toward increasing density.
Scandium, yttrium, and lutetium all crystallize in the hexagonal close-packed structure at room temperature, and lawrencium is expected to do the same. The stable members of the group are known to change structure at high temperature. In comparison with most metals, they are not very good conductors of heat and electricity because of the low number of electrons available for metallic bonding.
| Name | Sc, scandium | Y, yttrium | Lu, lutetium | Lr, lawrencium |
|---|---|---|---|---|
| Melting point | 1814 K, 1541 °C | 1799 K, 1526 °C | 1925 K, 1652 °C | 1900 K, 1627 °C |
| Boiling point | 3109 K, 2836 °C | 3609 K, 3336 °C | 3675 K, 3402 °C | ? |
| Density | 2.99 g·cm−3 | 4.47 g·cm−3 | 9.84 g·cm−3 | ? 14.4 g·cm−3 |
| Appearance | silver metallic | silver white | silver gray | ? |
| Atomic radius | 162 pm | 180 pm | 174 pm | ? |
Occurrence
Scandium, yttrium, and lutetium tend to occur together with the other lanthanides (except short-lived promethium) in the Earth's crust, and are often harder to extract from their ores. The abundance of elements in Earth's crust for group 3 is quite low—all the elements in the group are uncommon, the most abundant being yttrium with abundance of approximately 30 parts per million (ppm); the abundance of scandium is 16 ppm, while that of lutetium is about 0.5 ppm. For comparison, the abundance of copper is 50 ppm, that of chromium is 160 ppm, and that of molybdenum is 1.5 ppm.
Scandium is distributed sparsely and occurs in trace amounts in many minerals. Rare minerals from Scandinavia and Madagascar such as gadolinite, euxenite, and thortveitite are the only known concentrated sources of this element, the latter containing up to 45% of scandium in the form of scandium(III) oxide. Yttrium has the same trend in occurrence places; it is found in lunar rock samples collected during the American Apollo Project in a relatively high content as well.
The principal commercially viable ore of lutetium is the rare-earth phosphate mineral monazite, (Ce,La,etc.)PO4, which contains 0.003% of the element. The main mining areas are China, United States, Brazil, India, Sri Lanka and Australia. Pure lutetium metal is one of the rarest and most expensive of the rare-earth metals with the price about US$10,000/kg, or about one-fourth that of gold.
Production
The most available element in group 3 is yttrium, with annual production of 8,900 tonnes in 2010. Yttrium is mostly produced as oxide, by a single country, China (99%). Lutetium and scandium are also mostly obtained as oxides, and their annual production by 2001 was about 10 and 2 tonnes, respectively.
Group 3 elements are mined only as a byproduct from the extraction of other elements. They are not often produced as the pure metals; the production of metallic yttrium is about a few tonnes, and that of scandium is in the order of 10 kg per year; production of lutetium is not calculated, but it is certainly small. The elements, after purification from other rare-earth metals, are isolated as oxides; the oxides are converted to fluorides during reactions with hydrofluoric acid. The resulting fluorides are reduced with alkaline earth metals or alloys of the metals; metallic calcium is used most frequently. For example:
- Sc2O3 + 3 HF → 2 ScF3 + 3 H2O
- 2 ScF3 + 3 Ca → 3 CaF2 + 2 Sc
Biological chemistry
Group 3 metals have low availability to the biosphere. Scandium, yttrium, and lutetium have no documented biological role in living organisms. The high radioactivity of lawrencium would make it highly toxic to living cells, causing radiation poisoning.
Scandium concentrates in the liver and is a threat to it; some of its compounds are possibly carcinogenic, even through in general scandium is not toxic. Scandium is known to have reached the food chain, but in trace amounts only; a typical human takes in less than 0.1 micrograms per day. Once released into the environment, scandium gradually accumulates in soils, which leads to increased concentrations in soil particles, animals and humans. Scandium is mostly dangerous in the working environment, due to the fact that damps and gases can be inhaled with air. This can cause lung embolisms, especially during long-term exposure. The element is known to damage cell membranes of water animals, causing several negative influences on reproduction and on the functions of the nervous system.
Yttrium tends to concentrate in the liver, kidney, spleen, lungs, and bones of humans. There is normally as little as 0.5 milligrams found within the entire human body; human breast milk contains 4 ppm. Yttrium can be found in edible plants in concentrations between 20 ppm and 100 ppm (fresh weight), with cabbage having the largest amount. With up to 700 ppm, the seeds of woody plants have the highest known concentrations.
Lutetium concentrates in bones, and to a lesser extent in the liver and kidneys. Lutetium salts are known to cause metabolism and they occur together with other lanthanide salts in nature; the element is the least abundant in the human body of all lanthanides. Human diets have not been monitored for lutetium content, so it is not known how much the average human takes in, but estimations show the amount is only about several micrograms per year, all coming from tiny amounts taken by plants. Soluble lutetium salts are mildly toxic, but insoluble ones are not.
