Robert Bunsen

How to use this course

Robert Bunsen is remembered for a piece of laboratory equipment, but his real legacy is a style of chemistry. He measured gases to a fraction of a percent, built his own instruments when none were good enough, and kept theoretical speculation at arm’s length. With the physicist Gustav Kirchhoff he founded spectral analysis and found two new elements. With Henry Enfield Roscoe he made photochemistry quantitative. Hundreds of students passed through his Heidelberg laboratory, and many of the leading chemists of the next generation learned their craft there.

Each technical lesson opens with a paragraph for the general reader, then gives precise statements, derivations and exercises with folded solutions. The lessons follow the order of his work: arsenic chemistry (1834–1843), gases and furnaces (1838–1857), electrolysis (1841–1860), the burner (1855), photochemistry (1855–1862), spectral analysis (1859–1861), geysers (1846) and calorimetry (1870). The companion course on Kirchhoff treats the physics of spectra and radiation in more depth.

Timeline

Göttingen and the grand tour, 1811–1833

Robert Wilhelm Eberhard Bunsen was born in Göttingen on 30 March 1811. His baptismal record and two of his own handwritten curricula vitae give this date, although he later celebrated on the 31st and many books follow him. His father, Christian Bunsen, was chief librarian of the university and professor of modern philology; his mother was Auguste Friederike Quensel. Robert was the youngest of four sons and, by family account, a stubborn boy whom only his mother could manage.

He finished school at Holzminden in 1828 and entered the University of Göttingen. He studied chemistry with Friedrich Stromeyer, the discoverer of cadmium, mineralogy with Johann Friedrich Ludwig Hausmann, and mathematics with Carl Friedrich Gauss. His Latin dissertation, a description and classification of hygrometers, was a physics thesis; he completed it in 1830 and received the doctorate in 1831, aged about 19.

The tour

A grant from the Hanoverian government paid for a long study tour in 1832 and 1833. He looked at factories and machines as closely as at laboratories. He saw a new steam engine in Kassel, met Friedlieb Runge and Gustav Rose in Berlin, visited Justus von Liebig at Giessen and Eilhard Mitscherlich at Bonn, and in the autumn of 1832 reached Paris, where he worked in the laboratory of Joseph Louis Gay-Lussac. He finished with a tour of industrial plants around Vienna in 1833. The habits formed on this trip lasted: Bunsen always cared about furnaces, gas works and mineral springs as much as about pure chemistry.

Kassel and Marburg, 1834–1851

Bunsen qualified as Privatdozent at Göttingen in 1834. His first notable result was medical. With the physician Arnold Adolph Berthold he showed that freshly precipitated hydrated iron oxide binds dissolved arsenic as an insoluble compound, so that a patient who swallows it soon after arsenic is largely protected. It remained the standard antidote for many decades.

In January 1836 he succeeded Friedrich Wöhler at the Polytechnic School in Kassel, and from 1837 he studied the foul-smelling arsenic compounds known as Cadet’s fuming liquid. The work is the subject of Lesson 04. It nearly killed him twice: once through arsenic poisoning, and once when a sample of cacodyl cyanide exploded and destroyed the sight of his right eye. Sources disagree on the date of the explosion, placing it in 1836 or in 1843.

In 1839 he moved to the University of Marburg, becoming full professor in the early 1840s. There he invented the carbon–zinc battery that bears his name (1841) and a grease-spot photometer for comparing light sources. From 1838 he analysed the gases of iron blast furnaces, first in Germany and then in England with Lyon Playfair. Their 1845 report showed that most of the fuel’s heat left the furnace unused in the waste gas, and proposed recovering it. In one such campaign Bunsen was overcome by furnace gas and had to be dragged clear by a colleague.

In 1846, the year after the eruption of Hekla, he joined a Danish-sponsored expedition to Iceland for about three and a half months. He collected and analysed volcanic rocks and gases, and he lowered thermometers into the tube of the Great Geysir. Lesson 10 explains the theory of geysers that came out of these measurements. His Marburg students included Hermann Kolbe, Edward Frankland and John Tyndall.

Breslau and Heidelberg, 1851–1859

In 1851 Bunsen moved to Breslau, where he taught for three semesters and met the young physicist Gustav Kirchhoff, thirteen years his junior. The friendship shaped both careers. In 1852 Bunsen succeeded Leopold Gmelin in the chair of chemistry at Heidelberg, on the promise of a new laboratory. He stayed there for the rest of his life. In 1854 he helped bring Kirchhoff to the Heidelberg chair of physics.

Heidelberg was installing coal-gas street lighting, and the university piped gas into the new building. Bunsen wanted a hot, clean, nearly invisible flame for analysis. In late 1854 he gave the design principles to the university mechanic Peter Desaga, who built a prototype; when the laboratory opened in early 1855, Desaga had made fifty burners for the students (Lesson 07). Bunsen published the design in 1857 and, as with all his inventions, took no patent.

The 1850s were his most productive decade. He prepared magnesium, aluminium, sodium and other metals by electrolysis of molten salts (Lesson 06). With Roscoe, who came from London to work with him, he measured the chemical action of light (Lesson 08). In 1857 he published Gasometrische Methoden, which set new standards of accuracy in gas analysis (Lesson 05). In the summer of 1859 Kirchhoff suggested that they look at the flame colours of salts through a prism. By October they had a working spectroscope. Within two years the collaboration had produced caesium and rubidium (Lesson 09).

Cacodyl and the radical theory

For the general reader. In the 1830s chemists suspected that organic molecules contain stable groups of atoms, called radicals, that pass unchanged from one compound to another the way an element does. Bunsen gave one of the strongest early proofs. He took an evil-smelling, self-igniting arsenic liquid, worked out its family of compounds, and showed that one group, which he called cacodyl, runs through all of them.

Cadet’s fuming liquid

In 1760 the French pharmacist Louis Claude Cadet de Gassicourt heated potassium acetate with arsenic trioxide and obtained a fuming, spontaneously flammable, garlic-smelling liquid. Its main components are now written

with cacodyl oxide as the product, together with some cacodyl itself, \((\mathrm{CH_3})_2\mathrm{As{-}As}(\mathrm{CH_3})_2\). Bunsen worked with the liquid from 1837 to 1843 in sealed apparatus under carbon dioxide, since it ignites in air.

Jöns Jacob Berzelius greeted cacodyl as a confirmation of the radical theory that he, Liebig and Wöhler had championed since the benzoyl work of 1832. The “free radical” was in fact a dimer, as Auguste Cahours and Alfred Riche later showed, but the chemistry was correct. The long-term consequence came through Edward Frankland and Hermann Kolbe. Following Bunsen’s lead, Frankland made organometallic compounds of zinc, tin and other metals, and in 1852 he observed that each element combines with a fixed number of groups. This was the beginning of the concept of valence.

The arsenic antidote

The 1834 antidote rests on the low solubility of iron(III) arsenite. Freshly precipitated hydrated iron(III) oxide has a large surface and binds dissolved arsenious acid; the product passes through the gut without being absorbed. Today the same chemistry is used to remove arsenic from drinking water with iron oxide adsorbents.

Gas analysis and gas solubility

For the general reader. Before Bunsen, analysing a gas mixture was rough work. He made it a precision science. He measured volumes over mercury, corrected every reading for temperature, pressure and water vapour, burned or absorbed one component at a time, and checked the arithmetic. His methods let him trace heat through blast furnaces and measure how much of each gas dissolves in water.

Eudiometry

A eudiometer is a graduated tube closed at the top, with platinum wires sealed in for a spark. A measured gas sample is mixed with excess oxygen and fired. Water condenses, so the volume contracts; carbon dioxide is then absorbed by a potassium hydroxide bead and the volume drops again. From the contraction and the carbon dioxide one solves for the composition. Every reading is first reduced to 0 °C and 760 mm of mercury, dry:

Absorption coefficients

For his book Bunsen built an absorptiometer and measured the solubility of many gases in water and alcohol over a range of temperatures. He expressed the results through a coefficient still in use.

Bunsen tested both laws carefully and showed where they hold: they are good for sparingly soluble gases such as oxygen, nitrogen and hydrogen, and they fail for very soluble gases such as ammonia and hydrogen chloride, which react with water.

The blast furnace

Between 1838 and 1846 Bunsen sampled gas from different heights inside working iron furnaces. Combining the analyses with the heats of combustion of carbon monoxide and hydrogen, he found that German charcoal furnaces lost over half the fuel’s heat in the top gas, and English coal furnaces over four fifths. The 1845 report with Playfair recommended burning the gas to heat the blast, which became standard practice.

The Bunsen cell and electrolytic metals

For the general reader. The most powerful batteries of the 1830s used platinum electrodes, which made them very expensive. In 1841 Bunsen replaced the platinum with pressed carbon. The cheap, strong battery that resulted powered arc lamps and, in his own hands, broke molten salts apart to give the first pure samples of several reactive metals.

The cell

William Robert Grove’s cell of 1839 had a zinc anode in dilute sulfuric acid and a platinum cathode in concentrated nitric acid, separated by a porous pot. Bunsen kept the chemistry and made the cathode of carbon, moulded from coke and coal. The overall reaction reduces nitric acid at the carbon:

The electromotive force is about 1.9 V, nearly twice that of a Daniell cell, and the internal resistance is low. With a battery of 44 such cells Bunsen ran a carbon arc light. Later he also used zinc and carbon in chromic acid, avoiding the toxic nitrogen oxides.

Electrolytic metals

From 1852 at Heidelberg he electrolysed fused chlorides. He obtained magnesium in quantity, pressed it into wire and showed that burning magnesium is a brilliant light source for photography. He prepared sodium and aluminium from molten chlorides and, with Augustus Matthiessen, lithium, barium, calcium and strontium. With his students he also prepared metals of the cerium group. He used the samples to measure specific heats and so atomic weights (Lesson 10).

In water, any metal much more reactive than hydrogen cannot be deposited, because water is reduced first and the metal, if formed, reacts with water. A molten salt contains no water, so the decomposition potential of the salt itself governs, and the metal collects as a liquid or sponge at the cathode.

The burner and premixed flames

For the general reader. A candle or a plain gas jet burns with a yellow, sooty flame because the fuel meets air only at the flame’s surface. Bunsen’s burner lets air into the gas stream before it burns. The premixed flame is blue, hot and clean, and its sharp inner cone is the simplest model system in combustion science. Its shape gives a way to measure how fast a flame travels.

The design

Gas leaves a small nozzle at the bottom of a vertical tube. The fast jet lowers the pressure around it and draws air in through adjustable side holes. Gas and air mix in the tube and burn at its mouth. With the air holes closed, the flame is luminous and yellow, from glowing soot. With them open, the flame becomes a pale blue cone of premixed combustion surrounded by a faint outer mantle where the remaining fuel burns in the surrounding air. Michael Faraday had used similar principles earlier, and a related device was patented by R. W. Elsner in 1856; Bunsen and Desaga’s design of 1855 became the laboratory standard.

Stoichiometry

For methane, \(\mathrm{CH_4}+2(\mathrm{O_2}+3.76\,\mathrm{N_2})\to\mathrm{CO_2}+2\mathrm{H_2O}+7.52\,\mathrm{N_2}\), so complete combustion needs 9.52 volumes of air per volume of fuel. The equivalence ratio is

A Bunsen burner normally runs rich (\(\phi>1\)): it entrains less than the stoichiometric air, and the outer diffusion flame finishes the job.

The cone

A premixed flame front moves into unburnt gas at a definite speed, the laminar burning velocity \(S_L\), set by the mixture’s chemistry and heat conduction. For stoichiometric methane–air at room conditions \(S_L\approx0.37\) m/s. In the burner the gas flows upward faster than this, so the front cannot stand flat across the tube. It tilts until the component of the flow normal to it equals \(S_L\).

Derivation. On the cone surface the unit normal makes angle \(90^\circ-\alpha\) with the vertical, so the normal component of the velocity \(U\hat z\) is \(U\cos(90^\circ-\alpha)=U\sin\alpha\). The tangential component \(U\cos\alpha\) only carries the front along itself. For the second form, mass conservation says all the gas \(Q=\pi r^2U\) crosses the cone at normal speed \(S_L\): \(Q=S_L\,\pi r\sqrt{r^2+h^2}\), and \(r/\sqrt{r^2+h^2}=\sin\alpha\). The relation fails near the tip and near the rim, where the front curves and loses heat to the tube; at the rim the local velocity drops to zero, which is what anchors the flame. If \(U\) falls below \(S_L\) the flame runs down the tube (flashback); if \(U\) is much too large it lifts off and blows out.

Photochemistry with Roscoe

For the general reader. Light can drive chemical reactions: it fades dyes, darkens silver salts and makes chlorine and hydrogen combine. Bunsen and Henry Enfield Roscoe set out to measure this action as carefully as heat or electricity. Their chief law says that the chemical effect depends only on the total light received: double the brightness and halve the time, and the effect is the same.

The actinometer

Roscoe came to Heidelberg from London to work with Bunsen, who became a lifelong friend. Their joint papers appeared from 1855 to 1862. Their instrument exposed an equal-volume mixture of hydrogen and chlorine, generated electrolytically from hydrochloric acid, to light. The hydrogen chloride formed dissolves at once in water in contact with the gas, so the rate of reaction appears as a steady movement of a water column in a calibrated tube.

They found three things. After exposure began there was a delay, the photochemical induction, before the reaction reached a steady rate. Once steady, the rate was proportional to the intensity of the light absorbed. And the light that had acted was removed from the beam, so that a layer of the gas shielded the gas behind it. They used the instrument to measure the chemical intensity of daylight and expressed the Sun’s chemical power as the volume of the gas mixture it could combine. Later work traced the induction mainly to traces of impurities, and Walther Nernst in 1918 explained the high efficiency of the reaction as a chain: one absorbed photon breaks a chlorine molecule, and each chlorine atom then starts a long sequence \(\mathrm{Cl+H_2\to HCl+H}\), \(\mathrm{H+Cl_2\to HCl+Cl}\).

The law holds over a wide range. In photographic emulsions it fails at very long and very short exposures: “reciprocity failure.” Karl Schwarzschild in 1899–1900 fitted the deviations empirically with an effect depending on \(I\,t^{\,p}\), with \(p\) somewhat less than 1 (he found about 0.86 for the plates he used). The same law, under the name reciprocity, governs exposure settings in photography and dose calculations in phototherapy.

Comparing lights

Bunsen’s grease-spot photometer of 1841 is a paper screen with a translucent spot of grease. Lit from both sides, the spot vanishes when the illuminances on the two sides are equal. With point sources this gives

Flame spectra, caesium and rubidium

For the general reader. Chemists had long used flame colours as rough tests: yellow for sodium, crimson for lithium, violet for potassium. The colours mix and mask each other. Kirchhoff and Bunsen spread the light of a flame through a prism and found that each element gives a fixed set of sharp lines, independent of the compound and the flame. The method was sensitive enough to reveal elements nobody had seen, and in two years it found two.

The method

The non-luminous Bunsen flame was the key: it is hot enough to vaporize and excite salts but emits almost no continuous light of its own. A salt on a platinum wire, held in the flame, gives the spectrum of its metal; chloride, nitrate or sulfate makes no difference. In their 1860 paper “Chemische Analyse durch Spectralbeobachtungen” Kirchhoff and Bunsen established this for lithium, sodium, potassium, calcium, strontium and barium, and showed that the lines appear for quantities far below what any other test could detect.

Two new elements

In 1860 they evaporated mineral water from Dürkheim in the Palatinate, about 44,000 litres in all, removed the alkaline earths and lithium, and looked at the residue. Two blue lines matched no known element. Bunsen named the metal caesium, from the Latin caesius, blue-grey, the colour of a clear sky. Separating it from potassium and the second new element by chloroplatinate precipitation and fractional crystallization gave 7.3 g of caesium chloride. In 1861 dark red lines in the mineral lepidolite revealed rubidium, from rubidus, deep red. Bunsen isolated metallic rubidium in 1862; caesium metal was first obtained in 1882 by the Swedish chemist Carl Setterberg, by electrolysis of the cyanide.

Why lines?

In modern terms, each atom has discrete energy levels, and a transition between levels \(E_u\to E_l\) emits a photon of wavelength \(\lambda=hc/(E_u-E_l)\). In a flame in thermal equilibrium the fraction of atoms in an excited level is given by the Boltzmann factor,

with \(g\) the level degeneracies. Alkali metals have their first excited level only about 1.4 to 2.1 eV above the ground state, low enough for a flame at 2000 K to excite a detectable fraction. That is why flame spectroscopy found the alkalis and why the method still serves in flame photometry for sodium, potassium and lithium in clinical laboratories.

Geysers and the ice calorimeter

For the general reader. Two of Bunsen’s ideas rest on the same simple physics of water. A geyser erupts because the boiling point of water rises with pressure, so deep water can be far hotter than 100 °C without boiling, until a small disturbance sets it off. And ice shrinks when it melts, so the heat given to a block of ice can be read off as a change in volume. Bunsen used the first to explain Iceland’s Great Geysir and the second to measure the specific heats of rare metals from tiny samples.

The geyser

In 1846 Bunsen measured the temperature of the water at several depths in the tube of the Great Geysir, including shortly before an eruption. Near the bottom the water was well above 100 °C, but below the boiling point for the pressure at that depth. Higher up, the margin was smaller.

The boiling point at depth \(z\) follows from \(p(z)=p_0+\rho g z\) and the Clausius–Clapeyron relation, with \(L\) the molar latent heat of vaporization:

The ice calorimeter (1870)

A small test tube is sealed into a glass vessel filled with pure water and mercury, and a mantle of ice is frozen around the tube. The mercury connects to a horizontal capillary. A warm sample dropped into the test tube melts some ice. Water is denser than ice, so the volume shrinks and the mercury thread retreats. The retreat measures the heat. Bunsen calibrated it with known masses of warm water and found the latent heat of fusion by measuring the volume change of freezing directly. A few hundred milligrams of a rare metal were enough for a specific heat.

This was how specific heats arbitrated between candidate atomic weights. Dmitri Mendeleev, who had spent 1859 to 1861 in Heidelberg, revised several atomic weights in building his periodic table, among them that of indium, which he raised from about 76 to about 113. A specific heat measurement decides between such values at once, as Exercise 10.2 shows.

The Heidelberg school, 1860–1889

After caesium and rubidium, Bunsen was among the most famous chemists in Europe. He received the Copley Medal of the Royal Society in 1860, had been elected a foreign member of the Society in 1858, and joined the Prussian order Pour le Mérite for Sciences and Arts in 1864. In 1863 he declined the Berlin chair left by Eilhard Mitscherlich. With Kirchhoff in physics and, from 1858 to 1871, Hermann von Helmholtz in physiology, Heidelberg had a scientific circle few universities could match.

He kept working with his hands. In 1868 he published methods for separating the platinum metals left over after platinum refining, and built a water-jet filter pump to speed the washing of precipitates. An explosion in that work singed his eyebrows and lashes; his hand, which was injured, shielded his eyes. In 1870 came the ice calorimeter, in 1871 a government-sponsored analysis of the mineral waters of Baden, and in 1875 a study of the spark spectra of rare earths. Late in life he measured the specific heats of platinum, glass and water with a steam calorimeter.

Teacher

Bunsen lectured on experimental chemistry himself and spent most of the day in the laboratory with his students. Those who trained or took their doctorates with him at Marburg and Heidelberg include Hermann Kolbe, Edward Frankland, John Tyndall, Henry Roscoe, Adolf von Baeyer, Friedrich Konrad Beilstein, Lothar Meyer, Viktor Meyer, Adolf Lieben, Carl Graebe, Ludwig Carius, Thomas Edward Thorpe and Carl Auer von Welsbach. Dmitri Mendeleev worked in Heidelberg from 1859 to 1861. Two of these students, Lothar Meyer and Mendeleev, built the periodic system, partly on atomic weights measured in Bunsen’s laboratory. One former student summed him up as great as an investigator, greater as a teacher, and greatest as a man and friend.

He never married, saying he never found the time. He took out no patents, on principle. He avoided theoretical controversy; the structural theory of organic chemistry, developed in part by his own students, interested him little. In 1877 he and Kirchhoff received the first Davy Medal of the Royal Society, for spectrum analysis.

Retirement, death and legacy, 1889–1899

Bunsen retired in 1889, aged 78, and gave up chemistry for geology and mineralogy, the interests of his youth. In 1898 the Society of Arts in London awarded him its Albert Medal. He died in Heidelberg on 16 August 1899, aged 88, and was buried in the Bergfriedhof there.

Legacy

• Analytical chemistry: precise gas analysis, gas solubility coefficients, the filter pump, and the burner on every teaching bench.
• Spectroscopy: with Kirchhoff, the method that identified the chemistry of the Sun and stars, found caesium, rubidium, thallium, indium, gallium and helium, and later gave atomic physics its data.
• Physical chemistry: the first quantitative photochemistry and the reciprocity law; calorimetry of small samples.
• Electrochemistry and industry: the carbon electrode, electrolytic metals, magnesium light, and heat recovery from furnace gases.
• Earth science: the steam mechanism of geysers and the chemical classification of volcanic rocks.
• Honours: the lunar crater Bunsen; the German Bunsen Society for Physical Chemistry bears his name.

Sources and further reading

Biography

1. Robert Bunsen, Wikipedia; and Robert Wilhelm Bunsen, German Wikipedia (birth date and honours).
2. Bunsen, Robert Wilhelm Eberhard, Dictionary of Scientific Biography, via Encyclopedia.com.
3. Royal Society of Chemistry, Blazing a trail, historical profile.
4. H. E. Roscoe, Bunsen Memorial Lecture, Journal of the Chemical Society, Transactions 77 (1900).
5. G. Lockemann, Robert Wilhelm Bunsen: Lebensbild eines deutschen Naturforschers, Stuttgart, 1949.
6. APS News, This month in physics history: Bunsen and Kirchhoff, May 2018.

Science

1. R. Bunsen, Gasometrische Methoden, Braunschweig, 1857.
2. R. Bunsen and H. E. Roscoe, Photochemische Untersuchungen, Ann. Phys. (1855–1862).
3. G. Kirchhoff and R. Bunsen, Chemische Analyse durch Spectralbeobachtungen, Ann. Phys. 110 (1860) and 113 (1861).
4. R. Bunsen, Calorimetrische Untersuchungen, Ann. Phys. 141 (1870); summarized in Nature 3 (1871).
5. Bunsen burner and Caesium, Wikipedia (history sections).
6. Reciprocity (photography), Wikipedia: Bunsen–Roscoe law and Schwarzschild’s exponent.
7. The strange case of Dr. Petit and Mr. Dulong, arXiv:1807.02270 (Dulong–Petit law and atomic weights).
8. I. Glassman and R. A. Yetter, Combustion, Academic Press, chapter 4 (laminar flame speed and the Bunsen cone).