On this page
A good red gold sol looks, in Richard Zsigmondy's words, clear "like good red wine". It does not settle, and seems to be a true solution. Yet it is full of tiny lumps of metal. Zsigmondy proved this by building a microscope that let him see and count particles far too small to see in the ordinary way. This post covers his story, the ultramicroscope, his size method worked through, and what students can take from it.
Who was Richard Zsigmondy?
His Nobel Prize facts page gives his birth as 1 April 1865 in Vienna and his death on 24 September 1929 in Göttingen. The 1925 prize was awarded a year late, in 1926, "for his demonstration of the heterogenous nature of colloid solutions and for the methods he used, which have since become fundamental in modern colloid chemistry".
His Nobel biography says he ran experiments in a small home laboratory as a boy, studied in Vienna and Munich, and worked as an assistant in Berlin. Work on lustre colours for glass and china drew him to colloids and to the Schott glass works in Jena, where he stayed until 1900. In 1907 he became professor and director of the Institute of Inorganic Chemistry at Göttingen.
What question was he trying to answer?
In his Nobel lecture, Zsigmondy says industrial problems with gold-based glass and ceramic colours started him off. Purple of Cassius, a gold dye known since 1665, behaved like a chemical compound; Berzelius had thought it was one. Zsigmondy made very pure red gold sols by reducing gold chloride with formaldehyde, mixed them with colloidal stannic acid, and made the purple from them. So the purple was a mixture of fine gold and stannic acid that behaved like a compound. He wrote the lesson down: "A colloidal mixture may sometimes behave like a chemical compound and has frequently simulated one."
Other chemists called his clear gold liquids mere suspensions. That left open his first question in the lecture: what size are colloidal particles, and how many atoms does one gold particle contain?
How does the ultramicroscope work?
The 1926 presentation speech gives the principle: a strongly lit sample is viewed through a microscope from the side, at right angles to the light beam. A particle much smaller than the wavelength of light cannot form an image, but it scatters light, so it appears as a bright point on a dark background, like dust in a sunbeam across a dark room.
The idea came from Zsigmondy, and he developed it with H. Siedentopf, an optician at the Zeiss firm. Their paper in Annalen der Physik has a title that mentions its first test, gold ruby glass. A narrow slit of light lit a thin sheet of the sample. He says the slit ultramicroscope showed gold particles down to about 10 nanometres (he wrote 10 mμ) in sunlight. He later designed an immersion version.
Worked example: finding particle size by counting
The speech describes his method: mark off a small known volume, count the particles in it, use the known mass of gold per volume to get the mass of one particle, then assume a sphere of normal density to get the size. Here it is with sample numbers for a diluted red sol.
Data. Gold content 1.0 × 10⁻⁶ g per cm³. On average, 4 particles are counted in an illuminated volume of 8.0 × 10⁻¹¹ cm³.
Particles per cm³. 4 / (8.0 × 10⁻¹¹) = 5.0 × 10¹⁰.
Mass of one particle. 1.0 × 10⁻⁶ / 5.0 × 10¹⁰ = 2.0 × 10⁻¹⁷ g.
Volume of one particle. Gold's density is about 19.3 g/cm³ (PubChem lists 19.282). V = 2.0 × 10⁻¹⁷ / 19.3 = 1.04 × 10⁻¹⁸ cm³.
Diameter. For a sphere, V = πd³/6, so d = (6V/π)^(1/3) = (1.98 × 10⁻¹⁸)^(1/3) cm = 1.26 × 10⁻⁶ cm, about 13 nm.
Atoms per particle. One gold atom has mass 197 / (6.022 × 10²³) = 3.27 × 10⁻²² g, so the particle holds about 2.0 × 10⁻¹⁷ / 3.27 × 10⁻²² ≈ 6 × 10⁴ atoms.
So the particle is an aggregate of tens of thousands of atoms, not a single big molecule. Notice the cube root: if you count twice as many particles, each has half the mass, but its diameter shrinks only by a factor of 2^(1/3), about 1.26.
What else did he find?
The nucleus method. Particles too small even for the ultramicroscope could be grown: in a reducing gold solution, new gold deposits on each tiny "nucleus" until it becomes visible. The speech says this reached particles about 1.5 nm across.
Red to blue. In the lecture, the gold particles carry a negative charge that keeps them apart. Add salt and the charge is removed, the particles clump, and the colour shifts from red to blue as the number of particles drops.
Coagulation theory. He asked Marian von Smoluchowski for a theory of the coagulation rate, then tested it; the particles had to be almost touching before they stuck.
If the idea of seeing scattered light rather than an image is hard to picture, the Fermi Simulation Lab has interactive simulations for ideas that are easier to see than read.
Mistakes students make
"Colloid particles can be seen in a microscope." Not in an ordinary one. The ultramicroscope shows scattered points of light, not images.
Mixing him up with Tyndall. The light cone itself is the Tyndall effect, covered in our post on John Tyndall's light beam. Zsigmondy turned the cone into single countable points.
Syllabus. Surface chemistry is not listed in the JEE (Main) 2026 syllabus; colloids appear in school science. Check your own year's syllabus.
What can a student take from how he worked?
Settle an argument with a measurement. Instead of debating "solution or suspension", he found a way to count.
Take industry problems seriously. A puzzle about coloured glass led to a Nobel Prize.
Read before you repeat. He wrote that if he had known of Faraday's earlier gold work, it "would have saved me much unnecessary work."
Next step: redo the worked example with 10 particles in the same volume and check that the diameter falls to about 9 nm.
