On this page
Every pane of plain glass reflects some of the light that hits it, which is why shop windows show your reflection and camera lenses can produce ghost images. In 1938 Katharine Burr Blodgett patented a way to cancel most of that reflection with a coating far thinner than a wavelength of light. This post covers who she was, how her layer-by-layer films worked, a worked example using the numbers in her patent, and the habits behind her inventions.
Who was Katharine Blodgett?
Physics Today records that she was born on 10 January 1898 in Schenectady, New York. As a high-school student she visited General Electric's research laboratory there and met Irving Langmuir, who advised her on a research career. She studied at the University of Chicago, then earned a doctorate at the University of Cambridge under Ernest Rutherford, the first woman to receive a physics PhD at Cambridge. She spent the rest of her career at GE, received the American Chemical Society's Garvan Medal and was inducted into the National Inventors Hall of Fame.
Her name survives in the Langmuir–Blodgett film. Our post on Irving Langmuir and surface chemistry covers his work on single layers of molecules on water; this one is about what she built from them.
How did she build a film one molecule at a time?
Her patent US 2,220,860, Film structure and method of preparation, filed on 16 March 1938 and granted on 5 November 1940, describes the method. Stearic acid dissolved in benzene is dropped onto water containing a barium salt. It spreads into a layer one molecule thick and becomes barium stearate. A drop of oil pushes on the layer to keep it tightly packed, and a waxed silk thread floating on the water makes its edge visible. A glass plate is dipped through the surface again and again, picking up one layer each time.
The patent puts each layer "on the order of one ten-millionth of an inch" thick. Because thickness is set by counting layers, a film can be built to a precise thickness. As it thickens, it shows bright interference colours, so exactly reproducible that the patent proposes using such films as colour standards and as standards for measuring very small thicknesses.
The second trick was the "skeleton" film. If a film mixes barium stearate with stearic acid, soaking it briefly in benzene dissolves the acid and leaves the stearate framework with air in the gaps. The thickness barely changes, but the refractive index drops.
Why does a thin coating stop reflection?
Her second patent, US 2,220,861, Reduction of surface reflection, filed on 16 June 1938, explains. Light reflects from the top of the coating and again from the coating-glass boundary. If the coating's optical thickness is a quarter of a wavelength, the second wave travels an extra half wavelength and the two cancel. The cancellation is complete when the coating's refractive index is the geometric mean of the indices on either side. She described the physics in Physical Review in 1939 in Use of Interference to Extinguish Reflection of Light from Glass.
A worked example with her patent's numbers
The patent uses glass of refractive index 1.52 in air (index 1.00).
Reflection from bare glass. At normal incidence, R = ((n − 1)/(n + 1))² = (0.52 ÷ 2.52)² ≈ 0.0426, about 4.3% per surface. The patent gives "approximately 4.25%" per face and "about 8.5%" for both faces of a window.
Ideal coating index. n(coating) = √(1.00 × 1.52) ≈ 1.233. The patent gives 1.233.
The real film. A skeleton film of cadmium arachidate about forty layers thick, the patent says, has a refractive index of about 1.25.
Thickness needed. For green light of wavelength 550 nm (a sample choice in the middle of the visible range), t = λ/4n = 550 ÷ (4 × 1.25) = 110 nm.
Layer thickness implied. 110 nm ÷ 40 layers ≈ 2.75 nm per layer, close to the patent's one ten-millionth of an inch (2.54 nm).
How much is left? For a quarter-wave coating, R = ((1.52 − n²)/(1.52 + n²))². With n = 1.25, n² = 1.5625, so R = (−0.0425 ÷ 3.0825)² ≈ 1.9 × 10⁻⁴, about 0.02% per surface instead of 4.3%.
Physics Today notes that her films boosted the transmittance of glass to 99%.
Common mistakes:
Using λ/4 instead of λ/4n. The quarter wave is measured inside the film, where the wavelength is λ/n.
Thinking you need a half-wave film. Both reflections here happen at a boundary with a denser medium, so both get the same phase shift; the extra path must be λ/2, which needs a film of λ/4n.
Expecting zero reflection for all colours. A coating tuned to 550 nm reflects a little more at the red and violet ends.
Interference in thin layers builds on the ideas in the NCERT Class 12 chapter Wave Optics; our guide to interference and resultant intensity has practice. If you find it hard to picture two reflected waves cancelling, Fermi's Simulation Lab has interactive simulations for ideas that are easier to see than read.
What can a student take from Blodgett's way of working?
Control the variable that matters. Counting layers turned thickness into an integer she could set exactly.
Turn a side effect into a tool. The interference colours of her films became colour and thickness standards.
Solve the second problem too. Getting the thickness right was not enough; the skeleton film fixed the refractive index.
Write it down precisely. Her patents still give enough numbers to check the physics.
Try it now: find the quarter-wave thickness of a coating with n = 1.38 for 550 nm light, and the reflection it leaves on glass of index 1.52.
