News center
Driven by advanced processing technology

The periodic table was chemistry's most important breakthrough

Dec 10, 2023

By Dennis Rouvray

12 February 1994 , updated 10 January 2019

When the French novelist Balzac wrote ‘without numbers, the whole edifice of our civilisation would fall to pieces’, he might have been anticipating an insight by the Russian chemist Dmitri Mendeleyev. On 17 February 1869, Mendeleyev jotted down the symbols for the chemical elements, putting them in order according to their atomic weights. He wrote down the sequence in such a way that they ended up grouped on the page according to known regularities or ‘periodicities’ of behaviour. It was perhaps the greatest breakthrough in the history of chemistry.

Mendeleyev’s ideas totally changed the way chemists viewed their discipline. Now each chemical element had its number and fixed position in the table, and from this it became possible to predict its behaviour: how it would react with other elements, what kind of compounds it would form, and what sort of physical properties it would have.

Soon, Mendeleyev was predicting the properties of three elements – gallium, scandium and germanium – that had not then been discovered. So convinced was he of the soundness of his periodic law that he left gaps for these elements in his table. Within twenty years, all three had been found, and their properties confirmed his predictions almost exactly.


Mendeleyev himself was surprised by how fast his ideas were confirmed. In a prestigious Faraday Lecture to the Royal Institution in London in 1889, he admitted that he had not expected to live long enough ‘to mention their discovery to the Chemical Society of Great Britain as a confirmation of the exactitude and generality of the periodic law’. As news of his remarkable accomplishment began to spread, Mendeleyev became something of a hero, and interest in the periodic table soared.

In all, Mendeleyev predicted 10 new elements, of which all but two turned out to exist. He later proposed that the positions of some pairs of adjacent elements be reversed to make their properties fit into the periodic pattern. He suggested swapping cobalt with nickel and argon with potassium, which he believed had been wrongly placed because their true atomic weights were different from the values chemists had determined. It took until 1913, some six years after Mendeleyev had died, to clear up this ambiguity. By then chemists had gained a much better understanding of the atom, and in that year the physicist Henry Moseley, working in Manchester, showed that the position of an element in the table is governed not by its atomic weight but by its atomic number.

The atomic number of an element defines the number of protons in its atomic nucleus, which in a neutral atom is equal to the number of electrons surrounding it. Moseley proved that the characteristic frequency of the X-rays generated by a particular element is directly related to its atomic number. One source of confusion for Mendeleyev was that the atomic weight that chemists measure is an average of the slightly different weights of all the different isotopes of an element. (Isotopes have the same number of protons, but different numbers of neutrons.)

Mendeleyev’s intuition had been right, however, and atomic number was used successfully to assign a place in an expanded table for the noble gases – helium, neon, argon, krypton, radon and xenon – which had been discovered in the 1890s. These elements are so unreactive that they could not be made to combine with any other element at the time of their discovery, so finding out about their chemical properties was out of the question.

The heavier elements were positioned in a similar way. These mainly comprise the 15-element lanthanide series, discovered from the 1840s onwards, which starts from lanthanum, element 57, and the 15 radioactive actinides discovered in this century, which start with actinium, element 89. The chemistry of each series of these elements changes only very slightly with increasing atomic number, so would have had major problems placing them in his periodic table.

Today, anyone with even a passing involvement with chemistry refers to the periodic table, and apart from the occasional need to add the odd newly discovered artificial element beyond the actinide series it seems to have reached its final form. But this did not stop Leland Allen, a professor of chemistry at Princeton University, from announcing in 1992 that the table should be extended into an extra dimension. Allen recognises that the periodic table is the most powerful organ-isational instrument chemists have, but argues that it offers no definition of chemical bonding; nor does it provide any information about the energy of atoms, even though this tells chemists a lot about how an element is likely to behave.

Allen’s new dimension has to do with the outermost or ‘valence’ electrons of an atom, which are responsible for chemical bonding. Chemists find it convenient to picture an atom as a nucleus surrounded by electrons arranged in concentric layers or ‘shells’ of different energies. Using quantum mechanics, Allen calculated the average energy of the electrons in the valence shell, which he called the ‘configuration energy’. A large CE means there will be a large energy separation between energy levels in atoms, or energy bands in solids. Materials with a large band gap are insulators. Allen has used his calculations to quantify the ‘metalloid’ region which zigzags through the periodic table (see Figure 1), dividing elements that are metals from those that are nonmetals. These metalloids are boron, silicon, germanium, arsenic, antimony and tellurium.FIG-mg19123901.jpg

Allen also aims to predict what type of bonding will occur between particular atoms, and therefore what the properties of the compound formed might be. He selects each row of the periodic table in turn and arranges every possible combination of atoms from the row in a triangular matrix (see Figure 2). The apexes of the triangle correspond to pure ionic, pure covalent and pure metallic bonding. Areas inside the triangle represent varying combinations of these bonding types. Chemically and physically interesting materials are found near the middle of each edge: semiconductors on the M-C edge, polymeric materials on the I-C edge, and so-called Zintl phases, which have unusual conductivities, on the M-I edge. From the CE of each atom, Allen calculates the CE difference for each of the possible combinations of two atoms and then averages the CE difference for each horizontal row in the triangle. In this way he has generated a scale which represents the gradual variation in properties of the compounds inside the triangle and has used this to make some useful predictions of compounds’ behaviour, such as their electronic properties.FIG-mg19123902.jpg

But not everyone thinks such a radical approach is needed. The Russian chemist A. Godovikov and his Japanese coworker Y. Hariya at the Mineralogical Museum in Moscow have looked at the periodic table from another angle in their attempts to make it more predictive and more precise. They classified all the elements according to a number they calculated by dividing the ionisation energy of each atom by the size of the singly charged, positive ion. They found that they could use this ratio to classify all the elements into 13 groups. Compounds within each group have a distinctive crystal chemistry and form particular types of compounds. Some of these groups are identical to the vertical columns of the classic periodic table. But they also identified some new groupings, such as zirconium, niobium, hafnium and tantalum, which are all particularly good at forming complexes.

Whatever the outcome of the debate over how to extend the periodic table of the elements, several chemists have now begun to construct and use periodic tables, not of elements, but of compounds and molecules. In fact, this idea is not new. As long ago as 1862, the English chemist John Newlands, proposed a periodic table for organic molecules. Even Mendeleyev made heavy use of the behaviour of metal oxides and other compounds in deciding where to place particular elements in his table.

One of the leading architects of molecular periodic tables is Ray Hefferlin, a physicist at the Southern College of Seventh-Day Adventists in Collegedale, Tennessee. Hefferlin has developed two kinds of periodic system: ‘physical’ systems, in which all the molecules contain the same number of atoms, and ‘chemical’ systems of molecules with different numbers of atoms.

As long ago as the late 1970s, Hefferlin put forward a complete periodic system for all diatomic molecules, which divides up into 15 three-dimensional blocks. One dimension of each block is obtained by adding up the row numbers in the periodic table of the constituent atoms and the other two come from the column numbers of the two individual atoms (see Figure 3).FIG-mg19123903.jpg

Among the properties that have been observed to be periodic in Hefferlin’s blocks are the distance apart of the two atoms in the molecule, the frequencies at which the molecules absorb various kinds of light, the energy required to remove one electron from the molecule, and a measure of how the molecules partition themselves between octanol and water. Within the past year, Hefferlin’s research team has completed work on a similar but even more massive system for triatomic molecules. For this they needed a total of 25 blocks to accommodate the different combinations of atoms.

It is now possible to infer many of the properties of diatomic and triatomic compounds from these periodic systems. The Chinese scientist Fanao Kong of the Hefei University of Technology in China has even put forward a system for tetra-atomic molecules. It is simpler than Hefferlin’s and is based on adding up the group and the period numbers of all the constituent atoms. The two sets of numbers are used to make a grid whose columns reveal, for example, how metallic character varies.

Chemical periodic tables are usually smaller than these rather grandiose structures. Early this year, Bruce King of the University of Georgia devised a table for neutral osmium carbonyl clusters in which carbon monoxide molecules attach themselves to triangles of osmium atoms. Osmium carbonyl clusters and similar clusters involving closely related elements such as platinum are important because these metals are used as catalysts in the chemicals industry. One of the axes of the table is the number of metal atoms in the cluster, while the other represents the number of metal-metal bonds. This table lists the nine known osmium carbonyl clusters from Os3(CO)12 to Os7(CO)21 and predicts eight new ones that chemists have yet to make.

Some chemists have attempted to construct periodic tables for organic molecules too. Milan Randic of the University of Iowa has focused on isomers of various hydrocarbons, including the octanes, which are components of petrol. His 18-member table is constructed by using a mathematical technique derived from graph theory, and is based on the number of links in the molecular skeleton. He uses the table to predict properties such as density and octane rating, which shows how efficiently petrol burns in a car’s engine.

Meanwhile, Jerry Dias of the University of Missouri in Kansas City has been working on benzenoid hydrocarbons, whose molecules include one or more six-carbon benzene rings and several five-carbon rings. Benzenoids have many uses, ranging from photochromic pigments, fluorescent agents, and building blocks in the synthesis of organic chemicals, to antistatic additives for plastics. Some, such as benzo(a)pyrene, which comes from burning fossil fuels, are carcinogenic. Chemists have synthesised only around 500 benzenoids, less than 0.03 per cent of the theoretical total, but Dias has classified them all into one giant molecular periodic table. The table arranges the molecules according to how compactly the six-membered carbon rings are joined together. He has used it to predict four levels of chemical reactivity in the benzenoids. During the past year or two Dias has extended this work to other families of molecules, including fullerene clusters.

Mendeleyev would have been intrigued by these attempts to extend and expand his ideas. But none of them can match his pioneering attempts at prediction. His work remains the undisputed cornerstone of chemistry.

Dennis Rouvray is research professor of chemistry at the University of Georgia, US.

This article was updated on 10 January 2019 with a new headline. It was previously published on 12 February 1994 with the title, “Elementary, my dear Mendeleyev: Chemistry without the periodic table is as hard to imagine as sailing without a compass. But that hasn’t stopped some chemists from trying to improve it.”