HomeArticlesMaterials Science

Liquid Crystals: The Phase That Flows Like a Liquid and Aligns Like a Crystal

How the nematic director field, Frank elastic energy and the Freedericksz threshold combine to make an LCD pixel work.

mysimulator teamUpdated June 2026≈ 8 min read▶ Open the simulation

A phase between liquid and crystal

Most substances go directly from a rigid crystal to a disordered liquid at their melting point. Some rod-shaped organic molecules do something stranger: they pass through an intermediate mesophase where the molecules have lost their fixed positions — they flow like a liquid — but keep a preferred average orientation, like a crystal. That in-between phase is a liquid crystal, and the simplest, most common version is the nematic phase, where rods point roughly the same direction with no positional order at all.

The local average orientation is described by a unit vector field called the director, n(r), and by an order parameter S that measures how tightly the molecules cluster around that direction: S = 1 for perfect alignment, S = 0 for a fully disordered liquid. Real nematics sit somewhere around S ≈ 0.3-0.7 at room temperature, and S falls continuously to zero at the nematic-to-isotropic transition temperature.

live demo · a director field responding to an applied field● LIVE

Frank elastic energy: what it costs to bend a director field

A uniform director field costs nothing, but any spatial distortion — splay, twist or bend — costs elastic energy, exactly analogous to how bending a rubber sheet costs energy. Frank's continuum theory writes the free energy density as a sum of three independent distortion modes, each with its own elastic constant:

f = ½K₁(∇·n)²  +  ½K₂(n·∇×n)²  +  ½K₃(n×∇×n)²
       splay              twist                bend

K₁, K₂, K₃  = Frank elastic constants (~10⁻¹¹ N, material-dependent)

These three constants, together with the material's dielectric and magnetic anisotropy, are essentially the entire input needed to predict how a nematic layer responds to boundaries, fields and confinement.

The Freedericksz transition: a threshold, not a gradual response

Sandwich a nematic between two plates that anchor the director along the surface, then apply an electric or magnetic field perpendicular to that anchoring. Nothing happens at first — the elastic cost of tilting the director near the fixed boundaries outweighs the field's energetic reward for aligning with it. Only once the field exceeds a sharp Freedericksz threshold does the director in the bulk suddenly begin to tilt toward the field, with the response growing continuously from zero right above threshold — a textbook continuous (second-order-like) transition.

H_c = (π/d) √(K / Δχ)      threshold field, magnetic case

d       = cell thickness (thinner cell → higher threshold, elastic cost dominates)
K       = relevant Frank constant for that deformation geometry
Δχ   = anisotropy of magnetic (or dielectric, for Δε) susceptibility

This threshold, and the tilt angle above it, is precisely the mechanism every LCD pixel exploits: apply a voltage across a thin nematic cell, the director rotates by a controllable amount above threshold, and because the liquid crystal is birefringent — its refractive index differs for light polarised along versus across the director — that rotation continuously changes how much light gets through a pair of crossed polarisers sandwiching the cell.

Defects: where the director field cannot be smooth

A director field cannot always be continuous everywhere — topology sometimes forces a singular point or line where the orientation is undefined, called a disclination. Around a defect the director rotates by a half-integer or integer multiple of 2π as you circle it once, characterised by a topological charge; a +1/2 defect looks like a comet, a −1/2 defect like a three-pronged trident. These defects cost a finite core energy, repel or attract each other depending on their charge, and annihilate in pairs when they meet — watching a nematic cool from the isotropic phase, a dense tangle of defects gradually thins out exactly this way.

Why colour shows up under crossed polarisers

Because a nematic is birefringent, light travelling through a region of varying director tilt picks up a phase difference between its two polarisation components that depends on thickness, tilt angle and wavelength. Viewed between crossed polarisers this turns into vivid, spatially varying colour patterns — the classic Schlieren texture used to identify nematic samples under a microscope, and the same physics, tamed and controlled electrically, that makes an LCD screen display an image.

Frequently asked questions

What actually distinguishes a nematic liquid crystal from an ordinary liquid?

Positional disorder is the same in both — molecules can flow past each other freely. What a nematic keeps that a liquid loses is orientational order: the rod-shaped molecules maintain a preferred average direction, described by the director field, even though their positions are random.

How does an LCD pixel actually control light with a nematic layer?

A voltage across a thin nematic cell rotates the director above the Freedericksz threshold. Because the material is birefringent, that rotation changes the effective phase delay experienced by light polarised along the director versus across it, which continuously varies how much light passes through a pair of crossed polarisers sandwiching the cell — from fully blocked to fully transmitted.

What is a disclination and why does it look like a comet or a trident?

It is a topological defect — a point or line where the director orientation is undefined because the surrounding field cannot be continuously deformed to a uniform state. The visual pattern (comet-like for +1/2, three-pronged for −1/2) directly reflects how the director rotates as you trace a loop around the defect core.

Try it live

Everything above runs in your browser — open Liquid Crystal and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Liquid Crystal simulation

What did you find?

Add reproduction steps (optional)