On this page
Run your finger over Braille and you read the bumps without seeing them. The scanning tunnelling microscope does something similar with a needle so sharp its tip can end in a single atom, and it never quite touches the surface. Gerd Binnig and Heinrich Rohrer built the first one at IBM in Zurich. This post covers who they are, how the instrument works, a worked example of why it is so sensitive, and what students can take from their approach.
Who are Gerd Binnig and Heinrich Rohrer?
Binnig was born on 20 July 1947 in Frankfurt, according to his Nobel facts page. Rohrer was born on 6 June 1933 in Buchs, Switzerland, and died on 16 May 2013, according to his facts page. They shared half of the 1986 Nobel Prize in Physics "for their design of the scanning tunneling microscope"; the other half went to Ernst Ruska for the electron microscope.
In his Nobel autobiography, Binnig writes that he spent more of his student years playing in beat bands than on physics, and found that "actually doing physics is much more enjoyable than just learning it". He joined IBM's Zurich laboratory in 1978. Rohrer's autobiography recalls that during his PhD his instruments were so sensitive to vibration that he "learned to work after midnight, when the town was asleep".
How does a scanning tunnelling microscope work?
The 1986 press release explains it. A voltage is applied between a sharp tip and a conducting surface. Even without contact, a current flows by the quantum tunnel effect, and its strength depends strongly on the gap. A feedback loop keeps the current constant, holding the tip about 10⁻⁷ cm (1 nm) from the surface. Piezoelectric elements scan it across the surface, and recording its up-and-down motion maps the surface atom by atom. The release gives a horizontal resolution of about 2 Å and a vertical resolution of about 0.1 Å.
In their joint Nobel lecture, they write that they had no experience in microscopy or surface science, which "probably gave us the courage and light-heartedness to start something which should 'not have worked in principle' as we were so often told." The original aim was not a microscope at all but measuring properties of a surface over a patch less than 100 Å across.
What happened on the night of 16 March 1981?
Vibration was the enemy. Their first set-up floated the microscope on magnets above a bowl of superconducting lead. The lecture describes measuring at night, "hardly daring to breathe", until they saw the current depend exponentially on the tip-sample distance, the signature of tunnelling: "It was the portentous night of March 16, 1981." They reported this in Applied Physics Letters in 1982, and the first images of surface steps in Physical Review Letters. When they later imaged the silicon surface known as the 7 × 7 reconstruction, Binnig recalled: "I could not stop looking at the images. It was like entering a new world."
Worked example: why is the current so sensitive?
For a simple barrier, the tunnel current falls off as I ∝ e^(−2κd), where d is the gap and κ = √(2mφ)/ħ depends on the barrier height φ. Take a sample barrier φ = 4.0 eV.
Convert φ. 4.0 × 1.602 × 10⁻¹⁹ = 6.41 × 10⁻¹⁹ J.
Find κ. 2mφ = 2 × 9.11 × 10⁻³¹ × 6.41 × 10⁻¹⁹ = 1.17 × 10⁻⁴⁸. Square root: 1.08 × 10⁻²⁴. Divide by ħ = 1.055 × 10⁻³⁴: κ = 1.02 × 10¹⁰ m⁻¹.
Change the gap by 0.1 nm. The ratio of currents is e^(2κ × 10⁻¹⁰) = e^(2.05) ≈ 7.8. The current changes nearly eightfold.
Gap change to double the current. Δd = ln 2 / (2κ) = 0.693 / 2.05 × 10¹⁰ = 3.4 × 10⁻¹¹ m, about 0.03 nm.
Gap change for a 2 per cent current change. Δd = ln 1.02 / (2κ) = 0.0198 / 2.05 × 10¹⁰ ≈ 1 × 10⁻¹² m, about 0.001 nm. Holding the current steady therefore holds the gap steady to a tiny fraction of an atom.
What mistakes do students make with this?
Dropping the 2. The current depends on the square of the wave amplitude, which is why the exponent has 2κ.
Leaving φ in eV inside the square root with m in kg.
Mixing nm and m when you multiply κ by the gap.
Thinking the tip "sees" with light. It senses current; the sample must conduct.
Quantum tunnelling is not in the JEE (Main) 2026 syllabus, but exponentials and logarithms are, so treat this as a maths workout. Matter waves, which make tunnelling possible, are covered in our post on de Broglie's matter waves. If exponential decay or logarithms feel shaky, the Fermi Concept Library has every concept in physics, chemistry, biology, mathematics and computing, each with its own lesson and its place in the map.
What can a student take from how they worked?
Being new to a field can help. They tried what experts said would not work.
Solve the noise problem first. Most of their early effort went into stopping vibration.
Change the goal when the result is better. A plan for local measurements became a microscope.
Next step: redo steps 2 and 3 with a barrier of 5.0 eV and see how the eightfold factor changes.
