Periodic table

  Metals,   metalloids,   nonmetals, and   elements with unknown chemical properties.
Classification may vary depending on author's focus.

According to their shared physical and chemical properties, the elements can be classified into the major categories of metals, metalloids and nonmetals. Metals are generally shiny, highly conducting solids that form alloys with one another and salt-like ionic compounds with nonmetals (other than noble gases). A majority of nonmetals are colored or colorless insulating gases; nonmetals that form compounds with other nonmetals feature covalent bonding. In between metals and nonmetals are metalloids, which have intermediate or mixed properties.[28]

Metal and nonmetals can be further classified into subcategories that show a gradation from metallic to non-metallic properties, when going left to right in the rows. The metals may be subdivided into the highly reactive alkali metals, through the less reactive alkaline earth metals, lanthanides and actinides, via the archetypal transition metals, and ending in the physically and chemically weak post-transition metals. Nonmetals may be simply subdivided into the polyatomic nonmetals, being nearer to the metalloids and show some incipient metallic character; the essentially nonmetallic diatomic nonmetals, nonmetallic and the almost completely inert, monatomic noble gases. Specialized groupings such as refractory metals and noble metals, are examples of subsets of transition metals, also known[29] and occasionally denoted.[30]

Placing elements into categories and subcategories based just on shared properties is imperfect. There is a large disparity of properties within each category with notable overlaps at the boundaries, as is the case with most classification schemes.[31] Beryllium, for example, is classified as an alkaline earth metal although its amphoteric chemistry and tendency to mostly form covalent compounds are both attributes of a chemically weak or post-transition metal. Radon is classified as a nonmetallic noble gas yet has some cationic chemistry that is characteristic of metals. Other classification schemes are possible such as the division of the elements into mineralogical occurrence categories, or crystalline structures. Categorizing the elements in this fashion dates back to at least 1869 when Hinrichs[32] wrote that simple boundary lines could be placed on the periodic table to show elements having shared properties, such as metals, nonmetals, or gaseous elements.

Categorisation as described here can vary among authors

The elements of the periodic table shown here are divided into nine categories; six for the metals, and two for nonmetals, and a metalloid category. The nine categories (or sets) correspond to those found in the literature for the applicable part of the periodic table. Different authors may use different categorisation schema depending on the properties of interest.

An individual category is not necessarily exclusive according to its name, boundary, or shared properties. For example, while beryllium in Group 2 is colored as an alkaline earth metal, it is amphoteric rather than alkaline in nature. The heavier members of group 3 are shown as a lanthanide and an actinide, yet both are also transition metals. The transition metals in group 3; the lanthanides; and the later actinides are alkalic in nature just like the alkali metals proper in Group 1. In this instance the alkali metals are so named as they represent the most alkaline of the alkalic metals.

Scientists need not, "lose sleep over the hard cases as long as a classification system is beneficial to economy of description, to structuring knowledge and to our understanding, and hard cases constitute a small minority."[31]

The difference between the two categories is more of degree than kind.[33] The subject metals are light, reactive, and mostly of low mechanical strength, melting and boiling points. The group 1 members are easily cut with a knife. Only Be and Mg have any structural uses. The chemistries of the two categories of metals resemble one another to a large degree.[34] Most of these metals form basic oxides (Be is amphoteric).

   Moderate to high density metals, with high melting and boiling points; many have high hardness, mechanical strength and corrosion resistance. Those from Rf onwards are synthetic. Chemically, they show variable valency and a strong tendency to form coordination and brightly colored compounds. The oxides are basic, amphoteric or acidic, depending on the oxidation state.

Among the transition metals, the noble metals (generally Ru, Rh, Pd, Ag, Os, Ir, Pt, and Au) correspond to the noble gases,[35] although concern has been raised about the inclusion of silver due to its greater chemical reactivity compared to the other seven noble metals.[36]

   They closely resemble Ca, Sr and Ba, but many are heavier, and they form mostly pale colored compounds. The lanthanides are similar to one another, and hard to separate for that reason. They are abundant but widely dispersed; commercially viable concentrations are thus rare. The oxides are strongly to moderately basic.

   Soft, dense and reactive metals, those from Am onwards are synthetic. The early actinides (Th to Am) show some similarities to transition metals and have basic or amphoteric oxides; the late actinides are more like the lanthanides. Many actinides form colored compounds.

   Soft (or brittle) metals of low strength, and with melting points lower than the transition metals (far lower in the case of Hg, which is a liquid). Crystalline structures tend to show directional bonding, with generally greater complexity or fewer nearest neighbours than other metals. Chemically they are characterised, to varying degrees, by covalent bonding tendencies, acid-base amphoterism and anionic species such as aluminates, stannates, and bismuthates. They can also form Zintl phases (typically brittle, colored and semiconducting intermetallic compounds).

   In-between elements with a mix of metallic, non-metallic, and intermediate properties. They look like metals but are brittle and only fair electrical conductors. They mostly behave chemically as nonmetals. Metalloids form weakly acidic or amphoteric oxides.

   They appear colorless, colored, or (under white light) metallic-looking. Most are solid or gaseous. While the solids are brittle most of these are also known in malleable, pliable or ductile forms.

The halogens namely fluorine, chlorine, bromine, and iodine are characterised by their acridity and toxicity, in their natural forms. The remaining reactive nonmetals, hydrogen, carbon, nitrogen, oxygen, phosphorus, sulfur and selenium, being sandwiched between the strongly electronegative halogen nonmetals and the weakly nonmetallic metalloids, are (on an overall basis) moderately nonmetallic in nature. They, or most of them, attract various category names of their own such as biogen,[37] CHONPS,[38] intermediate,[39] light,[40] or other nonmetals.[41]

The reactivity of the reactive nonmetals ranges from relatively inert (N) to highly reactive (Cl, F). They form acidic, and neutral (H₂O, CO, NO, N₂O) oxides.

    Colorless, odorless, non-flammable gases with very low chemical reactivity. The first noble gas compound, of approximate composition XePtF₆, was not prepared until 1962; compounds of He and Ne are as yet unknown.

Many other names for sets of elements are in common use, and yet others have been used throughout history. These include, for example, rare earth metals, noble metals, and refractory metals.

A group or family is a vertical column in the periodic table. Groups usually have more significant periodic trends than periods and blocks, explained below. Modern quantum mechanical theories of atomic structure explain group trends by proposing that elements within the same group generally have the same electron configurations in their valence shell.[42] Consequently, elements in the same group tend to have a shared chemistry and exhibit a clear trend in properties with increasing atomic number.[43] In some parts of the periodic table, such as the d-block and the f-block, horizontal similarities can be as important as, or more pronounced than, vertical similarities.[44][45][46]

Under an international naming convention, the groups are numbered numerically from 1 to 18 from the leftmost column (the alkali metals) to the rightmost column (the noble gases).[47] Previously, they were known by roman numerals. In America, the roman numerals were followed by either an "A" if the group was in the s- or p-block, or a "B" if the group was in the d-block. The roman numerals used correspond to the last digit of today's naming convention (e.g. the group 4 elements were group IVB, and the group 14 elements were group IVA). In Europe, the lettering was similar, except that "A" was used if the group was before group 10, and "B" was used for groups including and after group 10. In addition, groups 8, 9 and 10 used to be treated as one triple-sized group, known collectively in both notations as group VIII. In 1988, the new IUPAC naming system was put into use, and the old group names were deprecated.[48]

Some of these groups have been given trivial (unsystematic) names, as seen in the table below, although some are rarely used. Groups 3–10 have no trivial names and are referred to simply by their group numbers or by the name of the first member of their group (such as "the scandium group" for group 3),[47] since they display fewer similarities and/or vertical trends.

Elements in the same group tend to show patterns in atomic radius, ionization energy, and electronegativity. From top to bottom in a group, the atomic radii of the elements increase. Since there are more filled energy levels, valence electrons are found farther from the nucleus. From the top, each successive element has a lower ionization energy because it is easier to remove an electron since the atoms are less tightly bound. Similarly, a group has a top-to-bottom decrease in electronegativity due to an increasing distance between valence electrons and the nucleus.[49] There are exceptions to these trends: for example, in group 11, electronegativity increases farther down the group.[50]

A period is a horizontal row in the periodic table. Although groups generally have more significant periodic trends, there are regions where horizontal trends are more significant than vertical group trends, such as the f-block, where the lanthanides and actinides form two substantial horizontal series of elements.[51]

Elements in the same period show trends in atomic radius, ionization energy, electron affinity, and electronegativity. Moving left to right across a period, atomic radius usually decreases. This occurs because each successive element has an added proton and electron, which causes the electron to be drawn closer to the nucleus.[52] This decrease in atomic radius also causes the ionization energy to increase when moving from left to right across a period. The more tightly bound an element is, the more energy is required to remove an electron. Electronegativity increases in the same manner as ionization energy because of the pull exerted on the electrons by the nucleus.[49] Electron affinity also shows a slight trend across a period. Metals (left side of a period) generally have a lower electron affinity than nonmetals (right side of a period), with the exception of the noble gases.[53]

Left to right: s-, f-, d-, p-block in the periodic table

Specific regions of the periodic table can be referred to as blocks in recognition of the sequence in which the electron shells of the elements are filled. Elements are assigned to blocks by what orbitals their valence electrons or vacancies lie in.[54] The s-block comprises the first two groups (alkali metals and alkaline earth metals) as well as hydrogen and helium. The p-block comprises the last six groups, which are groups 13 to 18 in IUPAC group numbering (3A to 8A in American group numbering) and contains, among other elements, all of the metalloids. The d-block comprises groups 3 to 12 (or 3B to 2B in American group numbering) and contains all of the transition metals. The f-block, often offset below the rest of the periodic table, has no group numbers and comprises most of the lanthanides and actinides. A hypothetical g-block is expected to begin around element 121, a few elements away from what is currently known.[55]

Approximate order in which shells and subshells are arranged by increasing energy according to the Madelung rule

The electron configuration or organisation of electrons orbiting neutral atoms shows a recurring pattern or periodicity. The electrons occupy a series of electron shells (numbered 1, 2, and so on). Each shell consists of one or more subshells (named s, p, d, f and g). As atomic number increases, electrons progressively fill these shells and subshells more or less according to the Madelung rule or energy ordering rule, as shown in the diagram. The electron configuration for neon, for example, is 1s² 2s² 2p⁶. With an atomic number of ten, neon has two electrons in the first shell, and eight electrons in the second shell; there are two electrons in the s subshell and six in the p subshell. In periodic table terms, the first time an electron occupies a new shell corresponds to the start of each new period, these positions being occupied by hydrogen and the alkali metals.[56][57]

Periodic table trends (arrows show an increase)

Since the properties of an element are mostly determined by its electron configuration, the properties of the elements likewise show recurring patterns or periodic behaviour, some examples of which are shown in the diagrams below for atomic radii, ionization energy and electron affinity. It is this periodicity of properties, manifestations of which were noticed well before the underlying theory was developed, that led to the establishment of the periodic law (the properties of the elements recur at varying intervals) and the formulation of the first periodic tables.[56][57] The periodic law may then be successively clarified as: depending on atomic weight; depending on atomic number; and depending on the total number of s, p, d, and f electrons in each atom. The cycles last 2, 6, 10, and 14 elements respectively.[58]

There is additionally an internal "double periodicity" that splits the shells in half; this arises because the first half of the electrons going into a particular type of subshell fill unoccupied orbitals, but the second half have to fill already occupied orbitals, following Hund's rule of maximum multiplicity. The second half thus suffer additional repulsion that causes the trend to split between first-half and second-half elements; this is for example evident when observing the ionisation energies of the 2p elements, in which the triads B-C-N and O-F-Ne show increases, but oxygen actually has a first ionisation slightly lower than that of nitrogen as it is easier to remove the extra, paired electron.[58]

Atomic number plotted against atomic radius, excluding the noble gases.[n 3]

Atomic radii vary in a predictable and explainable manner across the periodic table. For instance, the radii generally decrease along each period of the table, from the alkali metals to the noble gases; and increase down each group. The radius increases sharply between the noble gas at the end of each period and the alkali metal at the beginning of the next period. These trends of the atomic radii (and of various other chemical and physical properties of the elements) can be explained by the electron shell theory of the atom; they provided important evidence for the development and confirmation of quantum theory.[59]

The electrons in the 4f-subshell, which is progressively filled from lanthanum (element 57) to ytterbium (element 70),[n 4] are not particularly effective at shielding the increasing nuclear charge from the sub-shells further out. The elements immediately following the lanthanides have atomic radii that are smaller than would be expected and that are almost identical to the atomic radii of the elements immediately above them.[61] Hence lutetium has virtually the same atomic radius (and chemistry) as yttrium, hafnium has virtually the same atomic radius (and chemistry) as zirconium, and tantalum has an atomic radius similar to niobium, and so forth. This is an effect of the lanthanide contraction: a similar actinide contraction also exists. The effect of the lanthanide contraction is noticeable up to platinum (element 78), after which it is masked by a relativistic effect known as the inert pair effect.[62] The d-block contraction, which is a similar effect between the d-block and p-block, is less pronounced than the lanthanide contraction but arises from a similar cause.[61]

Such contractions exist throughout the table, but are chemically most relevant for the lanthanides with their almost constant +3 oxidation state.[63]

Ionization energy: each period begins at a minimum for the alkali metals, and ends at a maximum for the noble gases

The first ionization energy is the energy it takes to remove one electron from an atom, the second ionization energy is the energy it takes to remove a second electron from the atom, and so on. For a given atom, successive ionization energies increase with the degree of ionization. For magnesium as an example, the first ionization energy is 738 kJ/mol and the second is 1450 kJ/mol. Electrons in the closer orbitals experience greater forces of electrostatic attraction; thus, their removal requires increasingly more energy. Ionization energy becomes greater up and to the right of the periodic table.[62]

Large jumps in the successive molar ionization energies occur when removing an electron from a noble gas (complete electron shell) configuration. For magnesium again, the first two molar ionization energies of magnesium given above correspond to removing the two 3s electrons, and the third ionization energy is a much larger 7730 kJ/mol, for the removal of a 2p electron from the very stable neon-like configuration of Mg²⁺. Similar jumps occur in the ionization energies of other third-row atoms.[62]

Graph showing increasing electronegativity with growing number of selected groups

Electronegativity is the tendency of an atom to attract a shared pair of electrons.[64] An atom's electronegativity is affected by both its atomic number and the distance between the valence electrons and the nucleus. The higher its electronegativity, the more an element attracts electrons. It was first proposed by Linus Pauling in 1932.[65] In general, electronegativity increases on passing from left to right along a period, and decreases on descending a group. Hence, fluorine is the most electronegative of the elements,[n 5] while caesium is the least, at least of those elements for which substantial data is available.[50]

There are some exceptions to this general rule. Gallium and germanium have higher electronegativities than aluminium and silicon respectively because of the d-block contraction. Elements of the fourth period immediately after the first row of the transition metals have unusually small atomic radii because the 3d-electrons are not effective at shielding the increased nuclear charge, and smaller atomic size correlates with higher electronegativity.[50] The anomalously high electronegativity of lead, particularly when compared to thallium and bismuth, is an artifact of electronegativity varying with oxidation state: its electronegativity conforms better to trends if it is quoted for the +2 state instead of the +4 state.[66]

Dependence of electron affinity on atomic number.[67] Values generally increase across each period, culminating with the halogens before decreasing precipitously with the noble gases. Examples of localized peaks seen in hydrogen, the alkali metals and the group 11 elements are caused by a tendency to complete the s-shell (with the 6s shell of gold being further stabilized by relativistic effects and the presence of a filled 4f sub shell). Examples of localized troughs seen in the alkaline earth metals, and nitrogen, phosphorus, manganese and rhenium are caused by filled s-shells, or half-filled p- or d-shells.[68]

The electron affinity of an atom is the amount of energy released when an electron is added to a neutral atom to form a negative ion. Although electron affinity varies greatly, some patterns emerge. Generally, nonmetals have more positive electron affinity values than metals. Chlorine most strongly attracts an extra electron. The electron affinities of the noble gases have not been measured conclusively, so they may or may not have slightly negative values.[69]

Electron affinity generally increases across a period. This is caused by the filling of the valence shell of the atom; a group 17 atom releases more energy than a group 1 atom on gaining an electron because it obtains a filled valence shell and is therefore more stable.[69]

A trend of decreasing electron affinity going down groups would be expected. The additional electron will be entering an orbital farther away from the nucleus. As such this electron would be less attracted to the nucleus and would release less energy when added. In going down a group, around one-third of elements are anomalous, with heavier elements having higher electron affinities than their next lighter congenors. Largely, this is due to the poor shielding by d and f electrons. A uniform decrease in electron affinity only applies to group 1 atoms.[70]

The lower the values of ionization energy, electronegativity and electron affinity, the more metallic character the element has. Conversely, nonmetallic character increases with higher values of these properties.[71] Given the periodic trends of these three properties, metallic character tends to decrease going across a period (or row) and, with some irregularities (mostly) due to poor screening of the nucleus by d and f electrons, and relativistic effects,[72] tends to increase going down a group (or column or family). Thus, the most metallic elements (such as caesium) are found at the bottom left of traditional periodic tables and the most nonmetallic elements (such as neon) at the top right. The combination of horizontal and vertical trends in metallic character explains the stair-shaped dividing line between metals and nonmetals found on some periodic tables, and the practice of sometimes categorizing several elements adjacent to that line, or elements adjacent to those elements, as metalloids.[73][74]

With some minor exceptions, oxidation numbers among the elements show four main trends according to their periodic table geographic location: left; middle; right; and south. On the left (groups 1 to 4, not including the f-block elements, and also niobium, tantalum, and probably dubnium in group 5), the highest most stable oxidation number is the group number, with lower oxidation states being less stable. In the middle (groups 3 to 11), higher oxidation states become more stable going down each group. Group 12 is an exception to this trend; they behave as if they were located on the left side of the table. On the right, higher oxidation states tend to become less stable going down a group.[75] The shift between these trends is continuous: for example, group 3 also has lower oxidation states most stable in its lightest member (scandium, with CsScCl₃ for example known in the +2 state),[76] and group 12 is predicted to have copernicium more readily showing oxidation states above +2.[77]

The lanthanides positioned along the south of the table are distinguished by having the +3 oxidation state in common; this is their most stable state. The early actinides show a pattern of oxidation states somewhat similar to those of their period 6 and 7 transition metal congeners; the later actinides are more similar to the lanthanides, though the last ones (excluding lawrencium) have an increasingly important +2 oxidation state that becomes the most stable state for nobelium.[78]

From left to right across the four blocks of the long- or 32-column form of the periodic table are a series of linking or bridging groups of elements, located approximately between each block. In general, groups at the peripheries of blocks display similarities to the groups of the neighbouring blocks as well as to the other groups in their own blocks, as expected as most periodic trends are continuous.[79] These groups, like the metalloids, show properties in between, or that are a mixture of, groups to either side. Chemically, the group 3 elements, lanthanides, and heavy group 4 and 5 elements show some behaviour similar to the alkaline earth metals[80] or, more generally, s block metals[81][82][83] but have some of the physical properties of d block transition metals.[84] In fact, the metals all the way up to group 6 are united by being class-A cations ("hard" acids) that form more stable complexes with ligands whose donor atoms are the most electronegative nonmetals nitrogen, oxygen, and fluorine; metals later in the table form a transition to class-B cations ("soft" acids) that form more stable complexes with ligands whose donor atoms are the less electronegative heavier elements of groups 15 through 17.[85]

Meanwhile, lutetium behaves chemically as a lanthanide (with which it is often classified) but shows a mix of lanthanide and transition metal physical properties (as does yttrium).[86][87] Lawrencium, as an analogue of lutetium, would presumably display like characteristics.[n 6] The coinage metals in group 11 (copper, silver, and gold) are chemically capable of acting as either transition metals or main group metals.[90] The volatile group 12 metals, zinc, cadmium and mercury are sometimes regarded as linking the d block to the p block. Notionally they are d block elements but they have few transition metal properties and are more like their p block neighbors in group 13.[91][92] The relatively inert noble gases, in group 18, bridge the most reactive groups of elements in the periodic table—the halogens in group 17 and the alkali metals in group 1.[79]

The four kainosymmetric orbital types filled among the known elements, one per row: 1s, 2p, 3d, 4f.

The 1s, 2p, 3d, 4f, and 5g shells are each the first to have their value of ℓ, the azimuthal quantum number that determines a subshell's orbital angular momentum. This gives them some special properties,[93] that has been referred to as kainosymmetry (from Greek καινός "new").[58][94] Elements filling these orbitals are usually less metallic than their heavier homologues, prefer lower oxidation states, and have smaller atomic and ionic radii.[94]

The above contractions may also be considered to be a general incomplete shielding effect in terms of how they impact the properties of the succeeding elements. The 2p, 3d, or 4f shells have no radial nodes and are smaller than expected. They therefore screen the nuclear charge incompletely, and therefore the valence electrons that fill immediately after the completion of such a core subshell are more tightly bound by the nucleus than would be expected. 1s is an exception, providing nearly complete shielding. This is in particular the reason why sodium has a first ionisation energy of 495.8 kJ/mol that is only slightly smaller than that of lithium, 520.2 kJ/mol, and why lithium acts as less electronegative than sodium in simple σ-bonded alkali metal compounds; sodium suffers an incomplete shielding effect from the preceding 2p elements, but lithium essentially does not.[93]

Kainosymmetry also explains the specific properties of the 2p, 3d, and 4f elements. The 2p subshell is small and of a similar radial extent as the 2s subshell, which facilitates orbital hybridisation. This does not work as well for the heavier p elements: for example, silicon in silane (SiH₄) shows approximate sp² hybridisation, whereas carbon in methane (CH₄) shows an almost ideal sp³ hybridisation. The bonding in these nonorthogonal heavy p element hydrides is weakened; this situation worsens with more electronegative substituents as they magnify the difference in energy between the s and p subshells. The heavier p elements are often more stable in their higher oxidation states in organometallic compounds than in compounds with electronegative ligands. This follows Bent's rule: s character is concentrated in the bonds to the more electropositive substituents, while p character is concentrated in the bonds to the more electronegative substituents. Furthermore, the 2p elements prefer to participate in multiple bonding (observed in O=O and N≡N) to eliminate Pauli repulsion from the otherwise close s and p lone pairs: their π bonds are stronger and their single bonds weaker. The small size of the 2p shell is also responsible for the extremely high electronegativities of the 2p elements.[93]

The 3d elements show the opposite effect; the 3d orbitals are smaller than would be expected, with a radial extent similar to the 3p core shell, which weakens bonding to ligands because they cannot overlap with the ligands' orbitals well enough. These bonds are therefore stretched and therefore weaker compared to the homologous ones of the 4d and 5d elements (the 5d elements show an additional d-expansion due to relativistic effects). This also leads to low-lying excited states, which is probably related to the well-known fact that 3d compounds are often coloured (the light absorbed is visible). This also explains why the 3d contraction has a stronger effect on the following elements than the 4d or 5d ones do. As for the 4f elements, the difficulty that 4f has in being used for chemistry is also related to this, as are the strong incomplete screening effects; the 5g elements may show a similar contraction, but it is likely that relativistic effects will partly counteract this, as they would tend to cause expansion of the 5g shell.[93]

Another consequence is the increased metallicity of the following elements in a block after the first kainosymmetric orbital, along with a preference for higher oxidation states. This is visible comparing H and He (1s) with Li and Be (2s); N–F (2p) with P–Cl (3p); Fe and Co (3d) with Ru and Rh (4d); and Nd–Dy (4f) with U–Cf (5f). As kainosymmetric orbitals appear in the even rows (except for 1s), this creates an even–odd difference between periods from period 2 onwards: elements in even periods are smaller and have more oxidising higher oxidation states (if they exist), whereas elements in odd periods differ in the opposite direction.[94]

The discovery of the elements mapped to significant periodic table development dates (pre-, per- and post-)

In 1789, Antoine Lavoisier published a list of 33 chemical elements, grouping them into gases, metals, nonmetals, and earths.[95] Chemists spent the following century searching for a more precise classification scheme. In 1829, Johann Wolfgang Döbereiner observed that many of the elements could be grouped into triads based on their chemical properties. Lithium, sodium, and potassium, for example, were grouped together in a triad as soft, reactive metals. Döbereiner also observed that, when arranged by atomic weight, the second member of each triad was roughly the average of the first and the third.[96] This became known as the Law of Triads.[97] German chemist Leopold Gmelin worked with this system, and by 1843 he had identified ten triads, three groups of four, and one group of five. Jean-Baptiste Dumas published work in 1857 describing relationships between various groups of metals. Although various chemists were able to identify relationships between small groups of elements, they had yet to build one scheme that encompassed them all.[96] In 1857, German chemist August Kekulé observed that carbon often has four other atoms bonded to it. Methane, for example, has one carbon atom and four hydrogen atoms.[98] This concept eventually became known as valency, where different elements bond with different numbers of atoms.[99]

In 1862, the French geologist Alexandre-Émile Béguyer de Chancourtois published an early form of the periodic table, which he called the telluric helix or screw. He was the first person to notice the periodicity of the elements. With the elements arranged in a spiral on a cylinder by order of increasing atomic weight, de Chancourtois showed that elements with similar properties seemed to occur at regular intervals. His chart included some ions and compounds in addition to elements. His paper also used geological rather than chemical terms and did not include a diagram. As a result, it received little attention until the work of Dmitri Mendeleev.[100]

Julius Lothar Meyer's periodic table, published in "Die modernen Theorien der Chemie" (1864)[101]

In 1864, Julius Lothar Meyer, a German chemist, published a table with 28 elements. Realizing that an arrangement according to atomic weight did not exactly fit the observed periodicity in chemical properties he gave valency priority over minor differences in atomic weight. A missing element between Si and Sn was predicted with atomic weight 73 and valency 4.[101] Concurrently, English chemist William Odling published an arrangement of 57 elements, ordered on the basis of their atomic weights. With some irregularities and gaps, he noticed what appeared to be a periodicity of atomic weights among the elements and that this accorded with "their usually received groupings".[102] Odling alluded to the idea of a periodic law but did not pursue it.[103] He subsequently proposed (in 1870) a valence-based classification of the elements.[104]

Newlands' periodic table, as presented to the Chemical Society in 1866, and based on the law of octaves

English chemist John Newlands produced a series of papers from 1863 to 1866 noting that when the elements were listed in order of increasing atomic weight, similar physical and chemical properties recurred at intervals of eight. He likened such periodicity to the octaves of music.[105][106] This so termed Law of Octaves was ridiculed by Newlands' contemporaries, and the Chemical Society refused to publish his work.[107] Newlands was nonetheless able to draft a table of the elements and used it to predict the existence of missing elements, such as germanium.[108] The Chemical Society only acknowledged the significance of his discoveries five years after they credited Mendeleev.[109]

In 1867, Gustavus Hinrichs, a Danish born academic chemist based in America, published a spiral periodic system based on atomic spectra and weights, and chemical similarities. His work was regarded as idiosyncratic, ostentatious and labyrinthine and this may have militated against its recognition and acceptance.[110][111]

Periodic table of elements. Vienna, 1885. University of St Andrews

Mendeleev's periodic table from his book An Attempt Towards a Chemical Conception of the Ether

The first version of Mendeleev's periodic table, 1 March 1869 (N.S.): An attempt at a system of elements based on their atomic weights and chemical similarities. Here the periods are presented vertically, and the groups horizontally.

Periodic table at the Chemical Auditorium of the Gdańsk University of Technology from 1904

Russian chemistry professor Dmitri Mendeleev and German chemist Julius Lothar Meyer independently published their periodic tables in 1869 and 1870, respectively.[112] Mendeleev's table, dated March 1 [O.S. February 17] 1869,[113] was his first published version. That of Meyer was an expanded version of his (Meyer's) table of 1864.[114] They both constructed their tables by listing the elements in rows or columns in order of atomic weight and starting a new row or column when the characteristics of the elements began to repeat.[115]

The recognition and acceptance afforded to Mendeleev's table came from two decisions he made. The first was to leave gaps in the table when it seemed that the corresponding element had not yet been discovered.[116] Mendeleev was not the first chemist to do so, but he was the first to be recognized as using the trends in his periodic table to predict the properties of those missing elements, such as gallium and germanium.[117] The second decision was to occasionally ignore the order suggested by the atomic weights and switch adjacent elements, such as tellurium and iodine, to better classify them into chemical families.

Mendeleev published in 1869, using atomic weight to organize the elements, information determinable to fair precision in his time. Atomic weight worked well enough to allow Mendeleev to accurately predict the properties of missing elements.

Mendeleev took the unusual step of naming missing elements using the Sanskrit numerals eka (1), dvi (2), and tri (3) to indicate that the element in question was one, two, or three rows removed from a lighter congener. It has been suggested that Mendeleev, in doing so, was paying homage to ancient Sanskrit grammarians, in particular Pāṇini, who devised a periodic alphabet for the language.[118]

Henry Moseley (1887–1915)

Following the discovery of the atomic nucleus by Ernest Rutherford in 1911, it was proposed that the integer count of the nuclear charge is identical to the sequential place of each element in the periodic table. In 1913, English physicist Henry Moseley using X-ray spectroscopy confirmed this proposal experimentally. Moseley determined the value of the nuclear charge of each element and showed that Mendeleev's ordering actually places the elements in sequential order by nuclear charge.[119] Nuclear charge is identical to proton count and determines the value of the atomic number (Z) of each element. Using atomic number gives a definitive, integer-based sequence for the elements. Moseley predicted, in 1913, that the only elements still missing between aluminium (Z = 13) and gold (Z = 79) were Z = 43, 61, 72, and 75, all of which were later discovered. The atomic number is the absolute definition of an element and gives a factual basis for the ordering of the periodic table.[120]

Mendeleev's 1871 periodic table with eight groups of elements. Dashes represented elements unknown in 1871.

Eight-group form of periodic table, updated with all elements discovered to 2016

In 1871, Mendeleev published his periodic table in a new form, with groups of similar elements arranged in columns rather than in rows, and those columns numbered I to VIII corresponding with the element's oxidation state. He also gave detailed predictions for the properties of elements he had earlier noted were missing, but should exist.[121] These gaps were subsequently filled as chemists discovered additional naturally occurring elements.[122] It is often stated that the last naturally occurring element to be discovered was francium (referred to by Mendeleev as eka-caesium) in 1939, but it was technically only the last element to be discovered in nature as opposed to by synthesis.[123] Plutonium, produced synthetically in 1940, was identified in trace quantities as a naturally occurring element in 1971.[124]

The popular[125] periodic table layout, also known as the common or standard form (as shown at various other points in this article), is attributable to Horace Groves Deming. In 1923, Deming, an American chemist, published short (Mendeleev style) and medium (18-column) form periodic tables.[126][n 7] Merck and Company prepared a handout form of Deming's 18-column medium table, in 1928, which was widely circulated in American schools. By the 1930s Deming's table was appearing in handbooks and encyclopedias of chemistry. It was also distributed for many years by the Sargent-Welch Scientific Company.[127][128][129]

With the development of modern quantum mechanical theories of electron configurations within atoms, it became apparent that each period (row) in the table corresponded to the filling of a quantum shell of electrons. Larger atoms have more electron sub-shells, so later tables have required progressively longer periods.[130]

Glenn T. Seaborg, in 1945, suggested a new periodic table showing the actinides as belonging to a second f-block series.

In 1945, Glenn Seaborg, an American scientist, made the suggestion that the actinide elements, like the lanthanides, were filling an f sub-level. Before this time the actinides were thought to be forming a fourth d-block row. Seaborg's colleagues advised him not to publish such a radical suggestion as it would most likely ruin his career. As Seaborg considered he did not then have a career to bring into disrepute, he published anyway. Seaborg's suggestion was found to be correct and he subsequently went on to win the 1951 Nobel Prize in chemistry for his work in synthesizing actinide elements.[131][132][n 8]

Although minute quantities of some transuranic elements occur naturally,[3] they were all first discovered in laboratories. Their production has expanded the periodic table significantly, the first of these being neptunium, synthesized in 1939.[133] Because many of the transuranic elements are highly unstable and decay quickly, they are challenging to detect and characterize when produced. There have been controversies concerning the acceptance of competing discovery claims for some elements, requiring independent review to determine which party has priority, and hence naming rights.[134] In 2010, a joint Russia–US collaboration at Dubna, Moscow Oblast, Russia, claimed to have synthesized six atoms of tennessine (element 117), making it the most recently claimed discovery. It, along with nihonium (element 113), moscovium (element 115), and oganesson (element 118), are the four most recently named elements, whose names all became official on 28 November 2016.[135]

The periodic table in 32-column format

The modern periodic table is sometimes expanded into its long or 32-column form by reinstating the footnoted f-block elements into their natural position between the s- and d-blocks, as proposed by Alfred Werner in 1905.[136] Unlike the 18-column form, this arrangement results in "no interruptions in the sequence of increasing atomic numbers".[137] The relationship of the f-block to the other blocks of the periodic table also becomes easier to see.[138] William B. Jensen advocates a form of table with 32 columns on the grounds that the lanthanides and actinides are otherwise relegated in the minds of students as dull, unimportant elements that can be quarantined and ignored.[139] Despite these advantages, the 32-column form is generally avoided by editors on account of its undue rectangular ratio compared to a book page ratio,[140] and the familiarity of chemists with the modern form, as introduced by Seaborg.[141]

Within 100 years of the appearance of Mendeleev's table in 1869, Edward G. Mazurs had collected an estimated 700 different published versions of the periodic table.[139][143][144] As well as numerous rectangular variations, other periodic table formats have been shaped, for example,[n 9] like a circle, cube, cylinder, building, spiral, lemniscate,[145] octagonal prism, pyramid, sphere, or triangle. Such alternatives are often developed to highlight or emphasize chemical or physical properties of the elements that are not as apparent in traditional periodic tables.[144]

Theodor Benfey's spiral periodic table

A popular[146] alternative structure is that of Otto Theodor Benfey (1960). The elements are arranged in a continuous spiral, with hydrogen at the centre and the transition metals, lanthanides, and actinides occupying peninsulas.[147]

Most periodic tables are two-dimensional;[3] three-dimensional tables are known to as far back as at least 1862 (pre-dating Mendeleev's two-dimensional table of 1869). More recent examples include Courtines' Periodic Classification (1925),[148] Wringley's Lamina System (1949),[149] Giguère's Periodic helix (1965)[150] and Dufour's Periodic Tree (1996).[151] Going one further, Stowe's Physicist's Periodic Table (1989)[152] has been described as being four-dimensional (having three spatial dimensions and one colour dimension).[153]

The various forms of periodic tables can be thought of as lying on a chemistry–physics continuum.[154] Towards the chemistry end of the continuum can be found, as an example, Rayner-Canham's "unruly"[155] Inorganic Chemist's Periodic Table (2002),[156] which emphasizes trends and patterns, and unusual chemical relationships and properties. Near the physics end of the continuum is Janet's Left-Step Periodic Table (1928). This has a structure that shows a closer connection to the order of electron-shell filling and, by association, quantum mechanics.[157] A somewhat similar approach has been taken by Alper,[158] albeit criticized by Eric Scerri as disregarding the need to display chemical and physical periodicity.[159] Somewhere in the middle of the continuum is the ubiquitous common or standard form of periodic table. This is regarded as better expressing empirical trends in physical state, electrical and thermal conductivity, and oxidation numbers, and other properties easily inferred from traditional techniques of the chemical laboratory.[160] Its popularity is thought to be a result of this layout having a good balance of features in terms of ease of construction and size, and its depiction of atomic order and periodic trends.[103][161]

Simply following electron configurations, hydrogen (electronic configuration 1s¹) and helium (1s²) should be placed in groups 1 and 2, above lithium (1s²2s¹) and beryllium (1s²2s²).[162] While such a placement is common for hydrogen, it is rarely used for helium outside of the context of electron configurations: When the noble gases (then called "inert gases") were first discovered around 1900, they were known as "group 0", reflecting no chemical reactivity of these elements known at that point, and helium was placed on the top of that group, as it did share the extreme chemical inertness seen throughout the group. As the group changed its formal number, many authors continued to assign helium directly above neon, in group 18; one of the examples of such placing is the current IUPAC table.[163]

The position of hydrogen in group 1 is reasonably well settled. Its usual oxidation state is +1 as is the case for its heavier alkali metal congeners. Like lithium, it has a significant covalent chemistry.[164][165] It can stand in for alkali metals in typical alkali metal structures.[166] It is capable of forming alloy-like hydrides, featuring metallic bonding, with some transition metals.[167]

Nevertheless, it is sometimes placed elsewhere. A common alternative is at the top of group 17[159] given hydrogen's strictly univalent and largely non-metallic chemistry, and the strictly univalent and non-metallic chemistry of fluorine (the element otherwise at the top of group 17). Sometimes, to show hydrogen has properties corresponding to both those of the alkali metals and the halogens, it is shown at the top of the two columns simultaneously.[168] Another suggestion is above carbon in group 14: placed that way, it fits well into the trends of increasing ionization potential values and electron affinity values, and is not too far from the electronegativity trend, even though hydrogen cannot show the tetravalence characteristic of the heavier group 14 elements.[169] Finally, hydrogen is sometimes placed separately from any group; this is based on its general properties being regarded as sufficiently different from those of the elements in any other group.

The other period 1 element, helium, is most often placed in group 18 with the other noble gases, as its extraordinary inertness is extremely close to that of the other light noble gases neon and argon.[170] Nevertheless, it is occasionally placed separately from any group as well.[171] The property that distinguishes helium from the rest of the noble gases is that in its closed electron shell, helium has only two electrons in the outermost electron orbital, while the rest of the noble gases have eight. Some authors, such as Henry Bent (the eponym of Bent's rule), Wojciech Grochala, and Felice Grandinetti, have argued that helium would be correctly placed in group 2, over beryllium; Charles Janet's left-step table also contains this assignment. The normalized ionization potentials and electron affinities show better trends with helium in group 2 than in group 18; helium is expected to be slightly more reactive than neon (which breaks the general trend of reactivity in the noble gases, where the heavier ones are more reactive); predicted helium compounds often lack neon analogues even theoretically, but sometimes have beryllium analogues; and helium over beryllium better follows the trend of first-row anomalies in the table (s >> p > d > f).[172][173][174]

Although scandium and yttrium are always the first two elements in group 3, the identity of the next two elements is not completely settled. They are commonly lanthanum and actinium, and less often lutetium and lawrencium. The two variants originate from historical difficulties in placing the lanthanides in the periodic table, and arguments as to where the f block elements start and end.[175][n 10][n 11] It has been claimed that such arguments are proof that, "it is a mistake to break the [periodic] system into sharply delimited blocks".[177] A third variant shows the two positions below yttrium as being occupied by the lanthanides and the actinides. A fourth variant shows group 3 bifurcating after Sc-Y, into an La-Ac branch, and an Lu-Lr branch.[28]

Chemical and physical arguments have been made in support of lutetium and lawrencium[178][179] but the majority of authors seem unconvinced.[180] Most working chemists are not aware there is any controversy.[181] In December 2015 an IUPAC project was established to make a recommendation on the matter.[182]

La and Ac below Y

Lanthanum and actinium are commonly depicted as the remaining group 3 members.[183][n 12] It has been suggested that this layout originated in the 1940s, with the appearance of periodic tables relying on the electron configurations of the elements and the notion of the differentiating electron. The configurations of caesium, barium and lanthanum are [Xe]6s¹, [Xe]6s² and [Xe]5d¹6s². Lanthanum thus has a 5d differentiating electron and this establishes it "in group 3 as the first member of the d-block for period 6".[184] A consistent set of electron configurations is then seen in group 3: scandium [Ar]3d¹4s², yttrium [Kr]4d¹5s² and lanthanum [Xe]5d¹6s². Still in period 6, ytterbium was assigned an electron configuration of [Xe]4f¹³5d¹6s² and lutetium [Xe]4f¹⁴5d¹6s², "resulting in a 4f differentiating electron for lutetium and firmly establishing it as the last member of the f-block for period 6".[184] Later spectroscopic work found that the 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.[184] Lanthanum has the advantage of incumbency since the 5d¹ electron appears for the first time in its structure whereas it appears for the third time in lutetium, having also made a brief second appearance in gadolinium.[185]

In terms of chemical behaviour,[186] and trends going down group 3 for properties such as melting point, electronegativity and ionic radius,[187][188] scandium, yttrium, lanthanum and actinium are similar to their group 1–2 counterparts. In this variant, the number of f electrons in the most common (trivalent) ions of the f-block elements consistently matches their position in the f-block.[189] For example, the f-electron counts for the trivalent ions of the first three f-block elements are Ce 1, Pr 2 and Nd 3.[190]

Lu and Lr below Y

In other tables, lutetium and lawrencium are the remaining group 3 members.[n 13] Early techniques for chemically separating scandium, yttrium and lutetium relied on the fact that these elements occurred together in the so-called "yttrium group" whereas La and Ac occurred together in the "cerium group".[184] Accordingly, lutetium rather than lanthanum was assigned to group 3 by some chemists in the 1920s and 30s.[n 14] Several physicists in the 1950s and '60s favoured lutetium, in light of a comparison of several of its physical properties with those of lanthanum.[184] This arrangement, in which lanthanum is the first member of the f-block, is disputed by some authors since lanthanum lacks any f-electrons. It has been argued that this is not a valid concern given other periodic table anomalies—thorium, for example, has no f-electrons yet is part of the f-block.[191] As for lawrencium, its gas phase atomic electron configuration was confirmed in 2015 as [Rn]5f¹⁴7s²7p¹. Such a configuration represents another periodic table anomaly, regardless of whether lawrencium is located in the f-block or the d-block, as the only potentially applicable p-block position has been reserved for nihonium with its predicted configuration of [Rn]5f¹⁴6d¹⁰7s²7p¹.[192][n 15]

Chemically, scandium, yttrium and lutetium (and presumably lawrencium) behave like trivalent versions of the group 1–2 metals.[194] On the other hand, trends going down the group for properties such as melting point, electronegativity and ionic radius, are similar to those found among their group 4–8 counterparts.[184] In this variant, the number of f electrons in the gaseous forms of the f-block atoms usually matches their position in the f-block. For example, the f-electron counts for the first five f-block elements are La 0, Ce 1, Pr 3, Nd 4 and Pm 5.[184]

Markers below Y

A few authors position all thirty lanthanides and actinides in the two positions below yttrium (usually via footnote markers). This variant, which is stated in the 2005 Red Book to be the IUPAC-agreed version as of 2005 (a number of later versions exist, and the last update is from 1 December 2018),[195][n 16] emphasizes similarities in the chemistry of the 15 lanthanide elements (La–Lu), possibly at the expense of ambiguity as to which elements occupy the two group 3 positions below yttrium, and a 15-column wide f block (there can only be 14 elements in any row of the f block).[n 17] However, this similarity does not extend to the 15 actinide elements (Ac–Lr), which show a much wider variety in their chemistries.[197] This form moreover reduces the f-block to a degenerate branch of group 3 of the d-block; it dates back to the 1920s when the lanthanides were thought to have their f electrons as core electrons, which is now known to be false. It is also false for the actinides, many of which show stable oxidation states above +3.[198]

32-column periodic table, with bifurcated group 3

In this variant, group 3 bifurcates after Sc-Y into a La-Ac branch, and a Lu-Lr branch. This arrangement is consistent with the hypothesis that arguments in favour of either Sc-Y-La-Ac or Sc-Y-Lu-Lr based on chemical and physical data are inconclusive.[199] As noted, trends going down Sc-Y-La-Ac match trends in groups 1−2[200] whereas trends going down Sc-Y-Lu-Lr better match trends in groups 4−10.[184]

The bifurcation of group 3 is a throwback to the Mendeleev eight column-form in which seven of the main groups each have two subgroups. Tables featuring a bifurcated group 3 have been periodically proposed since that time.[n 18]

The definition of a transition metal, as given by IUPAC in the Gold Book, is an element whose atom has an incomplete d sub-shell, or which can give rise to cations with an incomplete d sub-shell.[201] By this definition all of the elements in groups 3–11 are transition metals. The IUPAC definition therefore excludes group 12, comprising zinc, cadmium and mercury, from the transition metals category. However, the 2005 IUPAC nomenclature as codified in the Red Book gives both the group 3–11 and group 3–12 definitions of the transition metals as alternatives.

Some chemists treat the categories "d-block elements" and "transition metals" interchangeably, thereby including groups 3–12 among the transition metals. In this instance the group 12 elements are treated as a special case of transition metal in which the d electrons are not ordinarily given up for chemical bonding (they can sometimes contribute to the valence bonding orbitals even so, as in zinc fluoride).[202] The 2007 report of mercury(IV) fluoride (HgF₄), a compound in which mercury would use its d electrons for bonding, has prompted some commentators to suggest that mercury can be regarded as a transition metal.[203] Other commentators, such as Jensen,[204] have argued that the formation of a compound like HgF₄ can occur only under highly abnormal conditions; indeed, its existence is currently disputed. As such, mercury could not be regarded as a transition metal by any reasonable interpretation of the ordinary meaning of the term.[204]

Still other chemists further exclude the group 3 elements from the definition of a transition metal. They do so on the basis that the group 3 elements do not form any ions having a partially occupied d shell and do not therefore exhibit properties characteristic of transition metal chemistry.[205] In this case, only groups 4–11 are regarded as transition metals. This categorisation is however not one of the alternatives considered by IUPAC. Though the group 3 elements show few of the characteristic chemical properties of the transition metals, the same is true of the heavy members of groups 4 and 5, which also are mostly restricted to the group oxidation state in their chemistry. Moreover, the group 3 elements show characteristic physical properties of transition metals (on account of the presence in each atom of a single d electron).[84]

Although all elements up to oganesson have been discovered, of the elements above hassium (element 108), only copernicium (element 112), nihonium (element 113), and flerovium (element 114) have known chemical properties, and conclusive categorisation at present has not been reached.[77] Some of these may behave differently from what would be predicted by extrapolation, due to relativistic effects; for example, copernicium and flerovium have been predicted to possibly exhibit some noble-gas-like properties, even though neither is placed in group 18 with the other noble gases.[77][206] The current experimental evidence still leaves open the question of whether copernicium and flerovium behave more like metals or noble gases.[77][207] At the same time, oganesson (element 118) is expected to be a solid semiconductor at standard conditions, despite being in group 18.[208]

Currently, the periodic table has seven complete rows, with all spaces filled in with discovered elements. Future elements would have to begin an eighth row. Nevertheless, it is unclear whether new eighth-row elements will continue the pattern of the current periodic table, or require further adaptations or adjustments. Seaborg expected the eighth period to follow the previously established pattern exactly, so that it would include a two-element s-block for elements 119 and 120, a new g-block for the next 18 elements, and 30 additional elements continuing the current f-, d-, and p-blocks, culminating in element 168, the next noble gas.[210] More recently, physicists such as Pekka Pyykkö have theorized that these additional elements do not exactly follow the Madelung rule, which predicts how electron shells are filled and thus affects the appearance of the present periodic table. There are currently several competing theoretical models for the placement of the elements of atomic number less than or equal to 172. In all of these it is element 172, rather than element 168, that emerges as the next noble gas after oganesson, although these must be regarded as speculative as no complete calculations have been done beyond element 123.[211][212]

The number of possible elements is not known. A very early suggestion made by Elliot Adams in 1911, and based on the arrangement of elements in each horizontal periodic table row, was that elements of atomic weight greater than circa 256 (which would equate to between elements 99 and 100 in modern-day terms) did not exist.[213] A higher, more recent estimate is that the periodic table may end soon after the island of stability,[214] whose centre is predicted to lie between element 110 and element 126, as the extension of the periodic and nuclide tables is restricted by proton and neutron drip lines as well as decreasing stability towards spontaneous fission.[215][216] Other predictions of an end to the periodic table include at element 128 by John Emsley,[3] at element 137 by Richard Feynman,[217] at element 146 by Yogendra Gambhir,[218] and at element 155 by Albert Khazan.[3][n 19]

The Bohr model exhibits difficulty for atoms with atomic number greater than 137, as any element with an atomic number greater than 137 would require 1s electrons to be travelling faster than c, the speed of light.[219] Hence the non-relativistic Bohr model is inaccurate when applied to such an element.

The relativistic Dirac equation has problems for elements with more than 137 protons. For such elements, the wave function of the Dirac ground state is oscillatory rather than bound, and there is no gap between the positive and negative energy spectra, as in the Klein paradox.[220] More accurate calculations taking into account the effects of the finite size of the nucleus indicate that the binding energy first exceeds the limit for elements with more than 173 protons. For heavier elements, if the innermost orbital (1s) is not filled, the electric field of the nucleus will pull an electron out of the vacuum, resulting in the spontaneous emission of a positron.[221] This does not happen if the innermost orbital is filled, so that element 173 is not necessarily the end of the periodic table.[217]

The many different forms of periodic table have prompted the question of whether there is an optimal or definitive form of periodic table.[222] The answer to this question is thought to depend on whether the chemical periodicity seen to occur among the elements has an underlying truth, effectively hard-wired into the universe, or if any such periodicity is instead the product of subjective human interpretation, contingent upon the circumstances, beliefs and predilections of human observers. An objective basis for chemical periodicity would settle the questions about the location of hydrogen and helium, and the composition of group 3. Such an underlying truth, if it exists, is thought to have not yet been discovered. In its absence, the many different forms of periodic table can be regarded as variations on the theme of chemical periodicity, each of which explores and emphasizes different aspects, properties, perspectives and relationships of and among the elements.[n 20]