Melting Point vs Crystallinity: The Two Numbers Behind Every Semi-Crystalline Polymer’s Performance

Open any property data sheet for a semi-crystalline plastic and two numbers sit there like an ID card: Tm (melting point) and Xc (degree of crystallinity). They look like plain specs, but they are not. The melting point hides a tug of war between enthalpy and entropy. Crystallinity hides the cooling rate of every second on the factory floor. Learn to read these two values and you can work a data sheet the way it was meant to be worked, and you can hold your own with an injection molder who insists the part cannot be made.

This piece is for engineers, product designers, and materials buyers who have decisions to make. Not for anyone collecting trivia.

Why semi-crystalline plastics resist heat better

Polymers fall into two camps. Amorphous materials such as PS, PMMA, and PC have one hard-soft boundary, the glass transition temperature Tg. Semi-crystalline materials such as PE, PP, PET, PA, and PEEK keep that Tg and then add a second transition above it: the melting point Tm, the temperature where the ordered crystalline regions finally come apart.

Think of the temperature scale in three bands:

• Below Tg the chains are frozen solid. The material is glassy and brittle.
• Between Tg and Tm the soft parts have gone mushy, but the crystalline regions are still holding the chain. This is the “crystal holds the line” band.
• Above Tm everything is disordered and the polymer flows.

That middle band is the whole reason these materials work. Polypropylene has a Tg near minus 10°C, which by rights should leave it a soft lump at room temperature. It is not a soft lump. A PP laundry basket survives a 100°C dishwasher because 40 to 60 percent of its structure is crystalline regions acting like rebar threaded through the chain. Higher Tm and higher Xc mean a stiffer skeleton. That is the real reason people reach for semi-crystalline grades when heat is the problem.

Melting point is not a neat point either. It is a band about 10 to 30°C wide. Lamellae come in different thicknesses, the thin imperfect ones melt first, the thick well-formed ones melt last. A DSC melting peak is just many ice cubes of different sizes melting in turn.

How a polymer actually crystallizes

A polymer chain runs several microns long. Getting something that long into a crystal takes folding. The chain kinks back on itself and stacks upright into thin plates called lamellae, each only 10 to 20 nanometers thick. The lamellae then grow outward from nucleation sites into micron-scale blobs called spherulites, which show the familiar black cross under a polarizing microscope. A semi-crystalline material is a two-phase mix of crystal and amorphous regions, and crystallinity is simply the mass fraction that is crystalline.

Not every chain qualifies. Three gates stand in the way:

• Chemical regularity. The repeat units must match. Mess up the chain with random copolymerization and crystallization falls apart. That is exactly how soft POE elastomers work.
• Stereoregularity. Isotactic PP fits the lattice and melts near 165°C. Atactic PP is a waxy byproduct fit only as an additive.
• Chain mobility. Chains too stiff, such as PC or PEI, freeze into glass before they can line up.

The crystallinity ranges say it plainly. HDPE runs 60 to 80 percent. PP sits at 40 to 60 percent. PET and PA6 land around 20 to 40 percent. PS, PC, and PMMA sit at zero. The polypropylene family is the stark example: isotactic PP is a commodity workhorse, atactic PP is waste wax, syndiotactic PP is an elastomer. One shift in stereochemistry, three entirely different materials.

The thermodynamics of melting point

Melting is a first-order phase transition, and the equation is one line:

Tm = ΔHm / ΔSm

Enthalpy, the cohesive pull inside the crystal lattice, divided by entropy, the disorder freedom of the melt. To push the melting point up you have two moves: grow the enthalpy (strong interactions, rigid chains) or shrink the entropy (a melt that was never very free). The strangest case is polyethylene. Its carbon-carbon bonds are not weak, but the chain is so floppy that after melting it becomes too free, and that entropy pins the melting point near 135°C.

Then there is the equilibrium melting point, Tm°, the melting point of an infinitely thick perfect crystal. PE tops out at 146°C in theory (measured 130 to 137). Isotactic PP at 186°C (measured 160 to 170). PEEK at 385°C (measured 343). The gap is the finite thickness of real lamellae:

Tm = Tm° (1 − 2σe / (Δh l))

That is the Gibbs-Thomson relation. Thin lamellae melt lower. Annealing thickens them, which nudges Tm up and narrows the peak.

The melting peak talks back if you pay attention. One sharp peak means uniform lamellae and good annealing. A broad peak or a shoulder means a jumble of thick and thin crystals, usually from fast cooling or copolymerization. Double peaks in PEEK or PPS are most often melt, recrystallize, melt again rather than two real crystal forms, so do not be quick to call them different polymorphs.

Six molecular levers that move the melting point

Line up commercial semi-crystalline plastics by Tm and the molecular design logic shows itself.

Six levers do the lifting:

1. Chain flexibility. Soft chains carry high entropy and stay low on Tm.
2. Backbone rigidity. Aromatic rings lock the conformation. PE to PEEK climbs about 200°C.
3. Hydrogen bonding. PA66 sits roughly 40°C above PA6 only because its hydrogen bonds line up completely.
4. Regularity. HDPE beats LLDPE beats LDPE.
5. Copolymer dilution. The Flory equation runs this; FEP and PFA give up melting point for easier processing.
6. Molecular weight, pressure, and plasticizer. Small corrections that stop mattering above a few tens of thousands of molecular weight.

How crystallinity is measured

Crystallinity is easy to say in one sentence and annoying to measure. Three common methods rest on different physics, so their numbers never sit neatly together.

DSC gets the most use. The formula is Xc = ΔHm / ΔHm°. A few milligrams, ten minutes, a number. Fast-cooled samples must subtract the cold crystallization enthalpy. Glass-fiber-reinforced grades must divide by the resin fraction only.

XRD and density are the other routes. XRD reads peak area ratios and can also name the crystal form. Density is cheap and quick but fooled by voids and fillers. The same resin can read 5 to 15 points apart across the three methods, and each reading is right.

The reporting rule is strict: state the method, the ΔHm° reference, and the thermal history. Skip those and the number compares to nothing. Crystallinity is not an intrinsic property of the material. It is resin times process. The same PP can read 30 percent after fast cooling and 45 percent after annealing, and both are honest. Common ΔHm° references are PE 293, PP 207, PET 117 to 140, PA66 260, and PEEK 130 J/g.

What crystallinity does to properties

Crystallinity is a master valve. Turn it toward flexible and transparent, or toward stiff and heat resistant.

Transparency has an exception worth keeping in mind. Cloudiness comes not from being crystalline but from spherulite size landing on the wavelength of visible light. Shrink spherulites below 100 nm with a clarifying nucleating agent and a 50 percent crystalline PP still passes 90 percent of light. That is the trick behind transparent PP baby cups.

There is a trap most people miss: warpage. Glass fiber drops the overall shrinkage but widens the difference between directions. The mismatch between flow direction and transverse direction shrinkage is the leading cause of warping in semi-crystalline molded parts, and at root it is the anisotropy of crystallinity again.

How processing designs the crystal

Crystallization only happens between Tg and Tm. Too hot and the whole thing is molten, too cold and the chain is frozen. The rate peaks in the middle of that window. Fast cooling leaves no time to crystallize, slow cooling does. That single fact is the entire quench versus slow cool performance split. The Avrami equation, X(t) = 1 − exp(−kt^n), describes it exactly, but for real work three facts carry the weight.

Fast and slow crystallizers lead different factory lives. PE, PP, PA66, and POM crystallize fully at ordinary mold temperatures and reach final properties straight out of the tool. PET and PEEK are slow and stay nearly amorphous without help. A beverage preform is made that way on purpose, while a PEEK part needs a mold temperature of at least 180°C or a post-anneal to reach a usable 30 to 35 percent crystallinity.

Nucleating agents are a four-for-one modifier. Add 0.2 percent of a sorbitol-based clarifier and PP spherulites shrink from 50 microns to 0.1 microns: transparency up, stiffness up, crystallinity up, cycle time down. A beta nucleating agent pulls in a tougher crystal form, which is how automotive bumpers and hot water pipes nearly double their impact strength.

Annealing is the catch-up class for crystallization. Hold the part 20 to 40°C below Tm and the amorphous regions finish crystallizing while thin lamellae thicken. The return is a small Tm rise, better heat and creep resistance, and released internal stress, paid for with an extra 0.2 to 0.5 percent shrinkage. Medical PEEK implants follow a standard anneal precisely to lock in the best crystallinity.

Five real cases and the mistakes people make

A few products show how the two parameters show up in the wild.

• PET beverage bottle. The preform is fast cooled to lock in the amorphous state, then stretch-oriented to grow nano-crystals for transparency and barrier. Hot-fill bottles get a heat-set step, and the neck is deliberately made highly crystalline with a white ring to hold its shape.
• PA66 versus PA6. A 40°C melting point gap decides who takes the 120°C duty. Intake manifolds run PA66, cable ties run PA6.
• PEEK implants. Low mold temperature means a dark, low-crystallinity reject. The right process plus anneal means a 35 percent crystalline, wear-resistant joint.
• Transparent PP. Nano-scale spherulites deliver a PC-like look at low cost.
• New-energy cooling connectors. Moving from PA66 to PPA or PPS comes down to higher Tm plus higher Xc fighting thermal and humidity aging.

The six mistakes that bite engineers:

1. “It melts at 260°C so 250°C is fine.” No. Strength drops off a cliff near Tm. As a rule the long-term unloaded ceiling is about 0.6 to 0.8 of Tm in kelvin, and for loaded parts you read HDT instead.
2. Treating the DSC peak top as the only melting point. Report the extrapolated onset temperature too.
3. Forgetting to subtract cold crystallization on fast-cooled samples, which inflates crystallinity by 10 points.
4. Skipping filler correction on glass-fiber compounds, which understates crystallinity by 30 percent.
5. Comparing numbers from different methods or different ΔHm° references as if they shared one scale.
6. Calling a double melting peak a different crystal form before checking with variable heating rate or XRD.

A semi-crystalline material is really a composite whose crystal-to-amorphous ratio you can tune. The melting point tells you how strong the crystalline regions are, and molecular structure sets that. The crystallinity tells you how much of them there are, and processing writes that. One is fate, the other is fortune. Material selection is reckoning with the fate, and process design is spending the fortune well. Most of the other numbers on a property data sheet fall out from these two.