1. Introduction
  2. The Elements
  3. Introduction to Boron
  4. Boron-Oxygen Chains
  5. Icosahedral Boron
  6. Mineralogy and Production of Boron
  7. References


The preparation of an article on mercury some time back awakened an interest and curiosity about other metals, so as much information on lead, tin, zinc, copper, aluminium, silver and gold as I could find was assembled into other articles, many of which illustrated interesting chemical or physical subjects. When I happened across boron in the course of these efforts, I realized that I had little familiarity with this element and its compounds, could make little sense of its chemistry, and that boron offered some very interesting lessons. This article is the result. The reader will become much more familiar with the uses of boron, its curious electron-starved chemistry with half-bonds and icosahedral structures, and its similarity to silicon.

The Elements

For two thousand years, "the elements" meant earth, air, fire and water, the four Empedoclean elements recognized by Aristotle. Empedocles was a Sicilian who lived 495-430 BC. Influenced by Pythagorean ideas, he taught the permanence of the four elements, and that their mixing in various proportions produced observed matter. These elements are related to what we now call the states of matter: solid, liquid and gaseous, with perhaps plasma added in, but were distinct from them.

This was by no means the only view of nature. The basis of Epicurean philosophy was the atomism of Democritus and Leucippus, which taught that matter was the concourse of infinitesimally small atoms that by their chance combinations created the visible universe, and that physical and chemical changes were caused by the motion of atoms, not the will of spirits. The types of atoms were not specifically indentified, but had to be sufficient to create the observed variety in things. Epicurus died about 270 BC, and the two views existed side by side until the followers of Aristotle and Christianity suppressed and maligned Epicurean views.

Our present views are much closer to those of Epicurus than to those of Aristotle. There was no way to decide between them in antiquity, but the modern atomic theory comes firmly down on the side of Epicururus. Quantitative experiments of the 18th century gave rise to the modern view of the elements, first clearly enunciated by Lavoisier at the end of the century. Now we had oxygen, hydrogen, mercury, carbon and other substances unresolvable into simpler ones, soon multiplied by the electrochemistry of Davy into an extensive stockpile of simple substances.

However, the view persisted of substances with certain properties, like the old Empedoclean elements, but greater in variety and existing as atoms. Atomism explained the law of constant and multiple proportions, and the properties of ideal gases. This is still the naive view of chemical elements. Indeed, chemistry texts still define a molecule as the smallest amount of any substance that retains the properties of that substance. Carbon is thought of as black, silver as shiny, sulphur as yellow, and mercury as liquid, and these are considered properties of the elements.

It is confusing, then, when the combination of two elements does not produce a sort of mean between them, like the offspring of parents. The active soft white metal sodium that burns in air combines with the heavy, pungent green gas chlorine with strong bleaching powers to produce salt, an innocuous transparent brittle crystalline material with a characteristic taste, and which neither burns nor bleaches in any degree.

Elements, in fact, are all pretty much alike. Their atoms consist of a heavy, positively-charged nucleus held together by strong nuclear forces, that holds an equal charge of light, mobile electrons tightly around it, so the whole is electrically neutral and does not exert strong forces on other atoms. Neutral atoms are all roughly the same size, from hydrogen with one proton and one electron, to uranium with 92 protons and an equal number of electrons. Neon (Z = 10) is 0.320 nm in diameter, xenon (Z = 54) is 0.436 nm. Magnesium (Z = 12) is 0.320 nm in diameter, mercury (Z = 80) is 0.300 nm. Individually, they have no macroscopic properties, certainly not those of ponderable quantities of the elements. Atoms are only the building bricks out of which substances are constructed.

When these generic atoms come together, it is often possible to rearrange the electrons, and always the outermost, least tightly held, ones so that the resulting electrostatic energy is less than that of the two separate atoms. How the atoms behave chemically is determined by the possibilities for the redistribution of charge. A little energy is required to extract an electron from a sodium atom, a little recovered when the electron sticks to a chlorine atom, and a great deal of energy is released when a lot of these ions associate in a crystal lattice. Salt is this lattice and its properties, not the properties of any molecule or atom by itself. The textbook definition of molecule is devoid of meaning.

The properties of substances are the properties of associations of atoms, and the nature of the association is more important than the particular atom. Metals are a notable example of this. Most metals are almost alike in nature: soft, shiny, heavy, nontransparent substances that can be bent and hammered. They are a pudding of positive ions in a sea of electrons, and their gross properties do not depend on the nature of the ion. Any suitable atom can make a metal. Sulphur does not do this because other sorts of association give lower energies than a metallic lattice would. Tin exists in both worlds: at higher temperatures it is a typical metal, while at low temperatures it is a crumbly nonmetallic crystalline substance.

I have spoken for simplicity of minimum energy as determining the equilibrium state. At a given temperature, it is actually the minimum free energy F = U - TS that counts, and the entropy S is just as important as the internal energy U. Oxygen molecules attract one another, so minimum energy results when they are all in a pile in the corner. Entropy is what keeps them flying around at room temperature, even at the expense of extra kinetic energy.

Boron will give us an excellent example of these concepts. The generic atom can make substances of the greatest variety and interest. It will be quite clear that there is no such thing as "boron-ness" in the atoms.

Introduction to Boron

Boron is one of the simplest of atoms. The only simpler ones are hydrogen, helium, lithium and beryllium. Boron has chemical symbol B, atomic number 5, and occurs naturally as 80% B11 and 20% B10. The latter isotope has a high cross section for thermal neutron absorption, 3800 barns. Thermal neutron counters are often filled with BF3 gas. The gamma ray from the neutron capture reaction B10(n,-)B11* followed by decay of the B11* to an α plus Li7 produces ionization which is then detected. Boron is also used in reactor control rods. This is a nuclear property of boron, and has nothing at all to do with its chemistry. The atomic weight of boron is 10.81.

Boron is found in a variety of similar minerals all related to borax, sodium tetraborate, Na2B4O7·10H2O. The name comes from the Arabic buraq, "white." Borax is the same in French and German as in English, but the element is bor. In Spanish, the words are bóraxo and boro. It is a relatively rare element in the earth's crust, representing only 0.001%. In the United States, borax is found in large amounts in California, in Searles Lake brines and in the Mojave desert. It is also found in Turkey, South America and other places. The natural deposits are dried-up lake beds. Molten borax reacts with metal oxides to form borates that dissolve in the melt, so it is a useful as a welding and soldering flux, and in colored enamels for iron. In fact, this was the earliest use of borax, as a pottery glaze. This same property is used for borax bead tests in chemistry, where the characteristic colors produced in a transparent borax drop melted on a loop of platinum wire in a bunsen burner flame are observed. Blue, for example, is the color of cobalt; green, of chromium. The color can differ in oxidizing (blue) and reducing (yellow) flames.

Borax hydrolyzes in water to form a slightly alkaline solution that is good for cleaning, since it emulsifies grease and oil. It also softens water by precipitating calcium borate. A household hint for preparing borax soap for really difficult laundry jobs was as follows. Cut up one pound laundry soap, add to one ounce of borax and a quart of water, and boil until uniform. Allow to cool, apparently in a mold, and cut into bars when solid.

The electron configuration of boron is 1s22s22p. It has only three electrons to work with, so the ion is unpolarizable, and does not hydrate. For this reason, boron is not eager to donate electrons in an electrovalent bond, and can also not accept them easily. Therefore, most of its bonds are covalent, and even forms half-bonds in which only one electron is shared covalently, not the usual two. This gives boron an apparent valence of +6 that we shall see in some interesting compounds. The first ionization potential is 8.30 V, which is not unusually high.

The other commonly-used boron compounds are orthoboric acid, or simply boric or boracic acid, H3BO3, and boron trioxide, B2O3, its anhydride. Note that the formula for boric acid can be written B(OH)3, as boron hydroxide. If boron were a normal metal, the hydroxyl ions would separate in water, creating the trivalent boron ion B+++. This, however, does not happen to the smallest degree, and boron does not form ionic bonds. Moreover, boric acid is not gelatinous like aluminium hydroxide, but crystallizes nicely. Boron trifluoride, BF3 is not an ionic compound like NaF, which has no ions in its crystals. Instead, the hydrogens are lost long before the oxygens. In aqueous solution, boric acid is a very weak acid, weaker even than carbonic. Its first ionization product is 6.4 x 10-10. Boric acid solution (solubility 6.35% at room temperature, 27.6% at 100°C) is used as an antiseptic, especially as an eyewash. Boric acid can be produced from borax by treatment with H2SO4, and boron trioxide by heating boric acid. These are the initial processes in the synthesis of all boron compounds.

Elemental boron was discovered by Davy, Gay-Lussac and Thénard in 1808. They produced metallic potassium by electrolysis, and then used it to reduce borates to impure boron. Davy called the element boracium, the Frenchmen bore. It can also be obtained in impure form by reduction of the oxide B2O3 by magnesium, or in pure form by the reduction of BCl3 by hydrogen on hot filaments. The first pure boron was produced by Weintraub in 1909. Ordinary boron is a brown-black amorphous powder. Pure boron can be made as extremely hard yellow monoclinic crystals that are a semiconductor resembling silicon. The band gap is 1.50 or 1.56 eV. Crystalline boron is an insulator at low temperatures, but becomes a conductor at elevated temperatures, as would be expected as carriers are thermally excited into the conduction band. Fabrication difficulties have so far prevented the use of boron as a semiconductor. The density of crystalline boron is 2.34 g/cc, of amorphous boron, 2.37. It melts at 2300°C and boils at 2550°C (some sources say 2040°C and 4100°C), so it is a very refractory substance. Boron fibers have been used in composite materials because of their great strength.

Boron gives a blue-green flame, and the brown amorphous form is often used in pyrotechical devices for this purpose. The color can be distinguished from the emerald green of copper. Boron is also used in borosilicate glasses, which are 12%-15% B2O3, 80% SiO2, and 2% Al2O3. Sodium, potassium, magnesium and calcium oxides are kept to minimal amounts, since it is their exclusion that gives the glass its desirable properties. "Pyrex" is a common trade name for a borosilicate glass. This glass is chemically resistant, and has a small coefficient of thermal expansion. Boron carbide, B4C, harder than SiC, is formed by decomposing B2O3 with carbon in the electric furnace: 2B2O3 + 7C → B4C + 6CO. It was first suggested for commercial use in 1934, and is an excellent abrasive. Boron nitride, BN, is another very hard compound, used in cutting tools. There are numerous other borides with complex structures. Boron is also used in porcelain enamels for iron, and for tiles and sanitary ware.

An interesting test for boron and borates uses volatile (CH3)3BO3, which burns with a green flame. Combine the solution to be tested with methanol, then add concentrated sulphuric acid to make the methyl borate. Another test uses turmeric paper, which turns reddish brown when moistened with boric acid or borates. If then a drop of ammonium hydroxide or sodium hydroxide is put on the paper, it will turn dark green, blue or black. To make turmeric paper, wash some ground turmeric in water. Then digest the washed powder with alcohol, and filter. Soak unsized white paper in the filtrate, then dry and cut into test strips. The active ingredient in turmeric paper may be curcumin, which seems to act the same way. It is available in an alcoholic 0.1% solution for this purpose.

Borax is necessary in small amounts for plant growth, one of the 16 essential nutrients. In larger amounts it is poisonous to plants, and the range can be small. For peaches, 1 ppm is required, but more than 5 ppm is toxic. If the signs of boron deficiency are noted in plants, a boron supplement can be applied. Borates can be used as non-toxic and non-specific herbicides. Borates are non-toxic to animals. The LD50 (dose at which there is 50% mortality) for humans is about 6 g per kg of body weight. Anything above 2 g is considered non-toxic, and borates are only 2 to 3 times as toxic as aspirin. Therefore, you are pretty safe unless you eat a pound and a half of borax for a snack. Borates are more toxic to insects than to mammals. The boranes and similar gaseous compounds are quite poisonous. As usual, it is not an element that is intrinsically poisonous, but toxicity depends on structure.

Boron-Oxygen Chains

We have mentioned that boron would rather lose the H in B(OH)3 than the oxygen. Like silicon, which has a similar persuasion, boron can form chains like -B-O-B-O-B-, where each intermediate boron has a free valence to work with. This is seen in tetraboric acid, O=B-O-B(OH)-O-B(OH)-O-B=O, or H2B4O7, which can be derived from B(OH)3 by dehydration: 4H3(BO)3 → H2B4O7 + 5H2O. Dehydration of boric acid also gives boron trioxide: 2B(OH)3 → B2O3 + 3H2O. Boron trioxide can also be written O=B-O-B=O, where the boron-oxygen chain again appears.

Some compounds can be considered derivatives of the theoretical acid HBO2, called borates. This metaboric acid is boric acid less one water molecule. If it loses another water, it becomes B2O3, which we have already seen. Boron trioxide in water becomes boric acid. The use of borax as a flux depends on reactions like O=B-O-B(ONa)-O-B(ONa)-O-B=O (borax) + NiO → 2NaBO2 + Ni(BO2)2.

Boron-oxygen chains also form linked planar structures that have the appearance of fence wire with hexagonal openings. Oxygen atoms project radially to make bonds with other atoms.

Icosahedral Boron

The icosahedron ("twenty-base") is one of the five Platonic solids, a regular convex polyhedron with 20 faces, 30 edges and 12 corners. In the absence of degeneracy, a symmetrical structure will have the lowest energy, and the regularity is possible because of the identity of the atoms. The tetrahedron, square, octahedron and dodecahedron also appear in molecular structures. It is remarkable that boron finds the icosahedron adapted for its purposes. Excellent drawings of the structures we discuss here are found in Pauling and Hayward. The icosahedral structures are exhibited by compounds known as boranes.

Just after world war II, the U. S. military was interested in developing an advanced aviation fuel for jets and ramjets that would replace the hydrocarbon-based JP-4 and similar fuels. Someone suggested that the then poorly-known boranes would give a better power-to-weight ratio, and Project ZIP was launched in 1952. Government money rained down on research projects dealing with boron, and for roughly a decade there was near hysteria in the scramble for research money, and many things were, incidentally, learned about boron. The boranes turned out to be not the powerhouses that were anticipated, were hard to manufacture and store, tended to start fires, and were also extremely toxic. Military enthusiasm cooled, and the research money went elsewhere. However, we found out the interesting structures of boranes, and a good deal of boron chemistry. Fifty years later, airplanes still burn JP-4.

A boron atom at an icosahedral vertex can make five equal bonds to its nearest neighbors, leaving one bond to stick out radially. Boron-boron bonds are half-bonds of length 0.180 nm. If we put one boron atom at each vertex, each with one proton, and assume that all bonds are half-bonds, then we have the neutral structure B12H12. However, the protons would like to have more electrons in their bonds, and overall 6 electrons are lacking for a full single bond. This would give far too large a charge to the ion, so the structure only attracts 2 extra electrons, which are spread over the 12 bonds to the protons. It happens that two extra electrons are ideal for resonance stablization of the ion. The proton bonds wind up at 0.120 nm. The substance potassium dodecaborohydride contains this ion in its crystals.

Boron will not form BH3. Instead, it forms larger borohydrides called boranes. In decaborane, B10H14, only 10 of the 12 vertices of the icosahedron are occupied. The extra 4 hydrogens are each bonded to two adjacent boron atoms, replacing the "missing" ones. The 10 radial hydrogens have bonds 0.118 nm long, while the bridging hydrogens have bonds 0.135 nm long. In tetraborane, B4H10, only four vertices are occupied, and there are 10 bridging hydrogens. Tetraborane has a foul smell. Diborane, B2H6 has no boron-boron bonds, but two bridging hydrogens out of plane in the planar molecule.

Davy might have made a borane when he treated boron with hydrochloric acid. The gas that was evolved burnt with a blue-green flame. Boranes react vigorously with oxygen, so they have been considered as candidates for rocket fuels. They are examples of electron-deficient compounds, since there are more bonding orbitals available than there are electrons to fill them. Carboranes are boranes where some of the borons are replaced by carbons. It is not easy to make boranes, and there was a considerable interval between Davy and the first recognized boranes.

One form of crystalline boron is tetragonal. The unit cell of this lattice consists of 4 icosahedrons of boron and two extra boron atoms, 50 in all. The radial bonds go to other icosahedrons or to the extra atoms, which show tetrahedral bonds. The bond length is 0.180 nm, as in the boranes.

Mineralogy and Production of Boron

Borax was first known from the deserts of western Tibet, where it received the name of tincal, derived from the Sanskrit. Borax glazes were used in China from AD300, and some tincal even reached the West, where the Arabic alchemist Geber seems to mention it in 700. Marco Polo brought some back to Italy in the 13th century. Agricola, around 1600, reports its use as a flux in metallurgy.

In 1777, boric acid was recognized in the hot springs or soffioni near Florence, Italy, and became known as sal sedativum, with mainly medical uses. The rare mineral is called sassolite, which is found at Sasso, Italy. This was the main source of European borax from 1827 to 1872, at which date American sources replaced it.

Borax was found by John A. Veatch in the spring waters of Tehama County, California, in January 1856. By September, borax was being produced from Borax Lake in Lake County. The California Borax Company worked these deposits from 1864 to 1868, when the company failed. John Searles and others discovered borax near Mono Lake in 1870, and this "marsh borax" became the principal source. One of the minerals exploited was ulexite, CaB4O7·NaBO2·8H2O, or "cotton balls," since it is found in soft, fibrous balls. It is common in Chile, Argentina, Nevada and California, where it is found in desert playas. The fibers guide light like fiber optics. Searles worked these deposits until 1896. However, in 1882 colemanite was discovered further south in the Mojave desert, and mining began near Daggett in 1887. This source in the Calico Mountains was much more accessible, since it was near the railway. It was worked by underground mining. In 1903, on exhaustion of the Calico Mountain reserves, mining moved north to Death Valley, and Colemanite began to be mined in open pits. Colemanite is a calcium borate formed by alteration of ulexite. It is named after the San Francisco merchant who first marketed it. Before the discovery of colemanite, South America supplied most of the world's borax, which had been discovered in Chile in 1852, but not worked commercially until much later.

The Death Valley colemanite mines were the origin of a famous trade name. The Harmony Borax Works were established at Furnace Creek in 1883. The product was hauled 166 miles south to Mojave by teams of 20 mules. The size of these teams was probably due to the harsh environmental conditions, in order that the mules should not exhaust themselves in the heat and aridity. They pulled only a wagon and a tank trailer (which was probably their water), which normally could probably have been handled by six mules. "20-Mule-Team Borax" was seen on many store shelves for many years, though now it is difficult to find borax in any supermarket. The teams worked until 1890, when the company failed and the mine was closed.

Kernite is an altered form of borax with only 4 waters of hydration instead of 10. It is also called rasorite, and has been the principal source of borax since 1927. Another source is the waters of Searles Lake, which are worked for other minerals as well, such as trona. Kernite is only found in Kern County, California, in the Kramer district about 3 miles north of Boron, California. The earlier railway siding at Kramer was renamed Boron. These deposits were discovered in 1916. Colemanite had been produced just to the east after 1913. Kramer kernite and Searles Lake brines have now completely superseded colemanite production. The Kramer lake beds are found in the Pliocene Ricardo formation. Most production is now from an open-pit mine opened in 1955. Boron is about halfway between Barstow and Mojave.

Borax is found mainly in arid regions, since borates are soluble to some degree and in a humid region would have been leached away long ago. They were deposited originally from waters associated with vulcanism. In California, this would be the great Tertiary vulcanism typified by sheet basalts. These borax-laden waters evaporated in lakes without an outlet in the arid environment that existed then as well as now. Some of these deposits were covered by later sediment and protected from leaching, like the kernite deposits of Boron. The conversion from borax to kernite is a result of mild metamorphism caused by burial under several hundred feet of overburden.

Where sea waters have evaporated instead of continental waters, as at Stassfurt, Germany, the hard boracite, a magnesium borate and chloride, may be found, associated with halite and gypsum. It is the only hard boron mineral (H = 7), though colemanite (H = 4) is harder than borax minerals.


L. Pauling and R. Hayward, The Architecture of Molecules (San Francisco: W. H. Freeman, 1964). Illustrations 32-36.

M. J. Sienko and R. A. Plane, Chemical Principles and Practices, 2nd ed. (New York: McGraw-Hill, 1974; International Student Edition). pp. 563-567.

W. N. Jones, Inorganic Chemistry (Philadelphia: Blakiston, 1949). pp. 567-580. Table of atomic radii, p. 83.

R. L. Bates, Geology of the Industrial Rocks and Minerals (New York: Dover, 1969). pp. 393-401.

D. E. Garrett, Borates (San Diego, CA: Academic Press, 1998). A recent survey of the industry.

T. Wartik, ed., From Borax to Boranes (Washington, DC: American Chemical Society, 1961). Papers from the 1958 ACS meeting at the height of the borane bubble.

Return to Physics Index

Composed by J. B. Calvert
Created 24 November 2002
Last revised 23 June 2004