Why PET Can Be Both a Bottle Material and a Textile Fibre

A clear drink bottle and a polyester sports shirt look like completely different materials. One is rigid enough to hold a fizzy drink; the other is flexible enough to stretch around your shoulders. Yet both can be made from the same polymer: poly(ethylene terephthalate), or PET.

PET can be both a bottle material and a textile fibre because its chemical structure does not determine just one physical form. The same repeating polymer structure can be processed in different ways to produce different shapes, molecular arrangements, and mechanical properties.

The important distinction is that polyester is a polymer family, while PET is one specific member of that family. Understanding this distinction explains why the material has such different applications.

01Polyester is the family; PET is a specific polymer

Start with the word polyester.

A polyester is a polymer whose main chain contains repeating ester linkages. These linkages connect sections of the polymer together.

The International Union of Pure and Applied Chemistry (IUPAC) defines polyesters by the presence of carboxylic ester linkages in their polymer backbones (IUPAC Gold Book).

An ester linkage has the general structure:

\[ \mathrm{-C(=O)-O-} \]

Notice that the carbonyl group, \(\mathrm{C=O}\), is directly connected to a single-bonded oxygen.

That combination identifies an ester. A carbonyl group by itself is not enough.

PET belongs to this family because its repeating structure contains ester linkages.

What makes PET different?

PET stands for poly(ethylene terephthalate).

It is formed from two monomers:

• Ethylene glycol, also called ethane-1,2-diol.
• Terephthalic acid, also called benzene-1,4-dicarboxylic acid.

Both monomers have two reactive functional groups, allowing them to connect repeatedly and form long polymer chains.

The repeating unit of PET is:

\[ \left[ \mathrm{-O-CH_2-CH_2-O-C(=O)-C_6H_4-C(=O)-} \right]_n \]

Here, \(n\) represents the number of repeating units in the polymer chain.

There are three useful structural features to recognise:

1. The \(\mathrm{-CH_2-CH_2-}\) section comes from ethylene glycol.
2. The \(\mathrm{C_6H_4}\) section comes from terephthalic acid and contains a benzene ring.
3. The \(\mathrm{-C(=O)-O-}\) connections are ester linkages.

The ester linkages are part of the main polymer chain, rather than groups hanging off the side.

This is the chemical feature that places PET in the polyester family.

If you want to practise reconstructing the monomers from this structure, see How to Draw PET from Ethylene Glycol and Terephthalic Acid.

A useful distinction: not all polyesters are PET

It is tempting to treat the words polyester and PET as interchangeable. They are not.

Think of polyester as a broad category. PET is one particular polymer within it.

Other polymers can contain ester linkages but have different repeating units because they are made from different monomers.

For example, poly(butylene terephthalate), or PBT, is another polyester. It contains the terephthalate section but has a four-carbon diol-derived section instead of PET’s two-carbon section.

This gives us a useful comparison:

The physical form does not determine the polymer family.

A bottle can be made from PET or polyethylene. A textile fibre can be made from PET or nylon.

You need to examine the chemical structure to identify the polymer.

02How can the same PET become a fibre or a bottle?

Now we reach the interesting part.

If PET has the same repeating chemical structure in both products, what actually changes?

The answer is mainly the manufacturing process and the resulting arrangement of the polymer chains.

To understand this, it helps to follow the historical development of PET.

First came the textile fibre

In 1941, British chemists John Rex Whinfield and James Tennant Dickson patented PET. Its early development focused on its ability to form strong synthetic fibres.

The material became commercially associated with polyester textile products, including fibres sold under names such as Terylene.

Decades later, engineers developed ways to use PET in packaging. A US patent for a biaxially oriented PET bottle was published in 1973, listing Nathaniel Wyeth and Robert Roseveare as inventors.

These developments are important because they show that the bottle application did not require an entirely new polymer structure. It required a different way of processing PET.

Sources: British Plastics Federation history and 1973 PET bottle patent.

How PET becomes a textile fibre

Imagine starting with small PET pellets.

The pellets are heated until the polymer melts. The molten PET is then pushed through a metal plate containing many tiny holes, called a spinneret.

As the material emerges, it forms thin continuous filaments.

These filaments cool, are pulled through rollers, stretched, and wound onto spools.

The stretching process is particularly important.

Why stretch the filaments?

Before stretching, the polymer chains are relatively disordered. They are not all pointing in the same direction.

When the filament is pulled, the long chains become more aligned with the direction of stretching.

This is called molecular orientation.

Think of a bundle of long threads. If the threads are tangled and randomly arranged, they can move around more easily. If they are pulled into a roughly parallel arrangement, the bundle becomes more effective at resisting a pulling force along its length.

The analogy is not exact: polymer chains are covalently bonded along their backbones and interact with neighbouring chains. However, it gives you a useful picture of why alignment affects strength.

In a PET fibre, the chains are preferentially oriented along the filament’s length.

This gives the fibre good tensile strength, meaning it can resist being pulled apart.

It can still bend and flex because it is extremely thin. A strong individual filament can form a flexible textile when many filaments are woven or knitted together.

How PET becomes a bottle

A bottle needs different properties.

It must maintain its shape, resist mechanical damage, and hold liquids under pressure. For carbonated drinks, it must also limit the escape of carbon dioxide.

Simply spinning PET into a fibre obviously won’t produce a bottle.

Instead, manufacturers use a process called stretch blow moulding.

First, PET is injection-moulded into a small object called a preform. It resembles a thick test tube, with the finished bottle’s threaded neck already formed.

The preform is then reheated until its body can be stretched.

It is placed inside a bottle-shaped mould.

A stretching rod moves down through the preform, extending it along its length. Compressed air is introduced, forcing the softened PET outwards against the mould walls.

The polymer is therefore stretched in two directions:

• Along the bottle’s length, called the axial direction.
• Around the bottle’s circumference, called the hoop direction.

This is known as biaxial orientation. The prefix bi- means two.

Why stretch in two directions instead of one?

Because the bottle must withstand forces in more than one direction.

If the polymer were stretched only along its length, the chains would become preferentially aligned in that direction. The material would not have the same molecular orientation around the circumference.

Biaxial stretching creates a more balanced arrangement of chains in the bottle wall.

The mould also determines the final shape, while cooling helps the polymer retain that shape.

This process produces a material with useful mechanical strength and gas-barrier properties.

The important point is that the polymer’s repeating chemical structure remains PET throughout this process. It is the arrangement of the chains that changes.

Technical descriptions of these processes are available in this review of melt-spun fibres and Introduction to Polymer Science and Technology.

03Molecular orientation is not the same as crystallinity

There is another layer to this explanation.

PET is a semi-crystalline polymer.

This means its structure contains two types of regions:

• Crystalline regions, where polymer chains are arranged in relatively ordered patterns.
• Amorphous regions, where the chains are more randomly arranged.

A polymer does not have to be completely crystalline or completely amorphous.

In fact, PET’s useful properties depend partly on the balance between these regions.

What does crystallinity contribute?

Crystalline regions restrict the movement of polymer chains because the chains are more closely and regularly packed.

This can increase stiffness and strength and can reduce the movement of small molecules through the polymer.

However, there is a trade-off.

Highly crystalline PET tends to become opaque because larger or more extensive crystalline regions scatter visible light. Less crystalline, largely amorphous PET can be transparent.

For a clear beverage bottle, manufacturers need to control the amount and development of crystallinity rather than simply maximise it.

This helps balance transparency, strength, and barrier performance.

The relationship between crystallinity and PET’s optical and mechanical properties is discussed in this American Chemical Society overview of PET.

Orientation versus crystallinity

These two terms are easy to confuse.

A polymer can have substantial molecular orientation without being highly crystalline.

Likewise, a polymer can contain crystalline regions without all its chains being aligned in one overall direction.

This distinction is particularly important when comparing PET fibres and bottles.

Fibres are mainly oriented along their length. Bottle walls are oriented in two directions. Both can contain crystalline and amorphous regions.

How could scientists test whether a bottle is biaxially oriented?

The manufacturing process suggests that the chains should be oriented in two directions. But how could we test that prediction rather than simply assume it?

One method is polarised infrared spectroscopy.

In a simplified version of the experiment:

1. An infrared source directs light towards a polariser.
2. The polarised light interacts with a small PET sample placed against an infrared measurement element.
3. A detector measures the infrared absorption.
4. The polarisation direction is changed, and the measurements are compared.

The prediction is that randomly oriented chains should produce a different pattern of directional absorption from preferentially aligned chains.

If the measured absorption changes with the direction of polarisation, this provides evidence of molecular orientation.

Researchers have used polarised attenuated total reflection Fourier-transform infrared spectroscopy (ATR-FTIR) to study PET films and stretch-blown bottles. Their measurements demonstrated biaxial orientation and also revealed that orientation can vary through the thickness of a material (research study).

There is an important limitation: measuring one location or one direction does not necessarily describe the entire bottle wall. The orientation can vary between the neck, body, and base.

The broader lesson is that processing history affects the internal structure, and internal structure affects the properties we observe.

04Worked examples

05Questions and solutions

Question 1

A polymer has the repeating unit:

\[ \left[ \mathrm{-O-CH_2-CH_2-O-C(=O)-C_6H_4-C(=O)-} \right]_n \]

Identify the polymer family and the specific polymer. Explain how the repeating unit supports your answer.

Solution 1

The polymer belongs to the polyester family, and the specific polymer is PET.

The repeating unit contains ester linkages, identified by the \(\mathrm{-C(=O)-O-}\) connections in its main chain. These are the defining structural feature of polyesters.

The ethylene glycol-derived section and terephthalate-derived section identify the specific polymer as poly(ethylene terephthalate).

It is important to distinguish the general family from the named polymer: PET is a polyester, but polyester is not another name for PET.

Question 2

A manufacturer draws a PET filament to increase its molecular orientation.

The initial filament has:

• Length: 0.40 m
• Diameter: 50 μm

After drawing, its length is 1.00 m.

Illustrative measurement data indicate that the polymer’s density increases by 1.5% during drawing.

Assuming the filament remains cylindrical and no polymer is lost, calculate its final diameter.

Explain why ignoring the density change would produce a slightly different answer.

Solution 2

The final diameter is approximately 31.4 μm.

The density increase means that the polymer occupies slightly less volume after drawing than it did initially. We must account for this rather than assume perfectly constant volume.

Step 1: Calculate the density ratio.

A 1.5% increase means:

\[ \rho_f=1.015\rho_i \]

Step 2: Use conservation of mass.

Mass is constant, so:

\[ \rho_iV_i=\rho_fV_f \]

Rearranging:

\[ V_f=\frac{V_i}{1.015} \]

Step 3: Relate volume to cross-sectional area.

For a cylindrical filament:

\[ V=AL \]

Therefore:

\[ A_fL_f=\frac{A_iL_i}{1.015} \]

Rearranging:

\[ \frac{A_f}{A_i} = \frac{0.40}{1.00\times1.015} \]

\[ \frac{A_f}{A_i}=0.3941 \]

Step 4: Calculate the final diameter.

Because cross-sectional area is proportional to diameter squared:

\[ d_f=d_i\sqrt{\frac{A_f}{A_i}} \]

Substituting:

\[ d_f=50\sqrt{0.3941} \]

\[ \boxed{d_f=31.4\ \mu\mathrm{m}} \]

Step 5: Explain the effect of density.

If density were assumed constant, the calculated diameter would be approximately 31.6 μm.

The actual illustrative density increase makes the filament slightly denser and reduces its volume. Therefore, its final diameter is slightly smaller.

This demonstrates why the assumptions behind a physical model matter.

Question 3

An engineer wants to investigate whether the wall of a stretch-blown PET bottle is molecularly oriented in both the axial and hoop directions.

Two strips are cut from the same central region of the bottle wall:

• Strip A is cut parallel to the bottle’s length.
• Strip B is cut around the bottle’s circumference.

Both strips have the same width, thickness, and gauge length. They are clamped in a tensile testing machine and pulled at the same rate.

Assume the material has been processed to produce biaxial orientation.

a) Predict the general mechanical behaviour expected when each strip is stretched.

b) Explain why testing only Strip A would be insufficient to establish biaxial orientation.

c) Identify one limitation of using this experiment to determine the molecular arrangement.

Solution 3

Both strips should demonstrate substantial resistance to tensile deformation, although their measured strengths do not have to be equal.

a) Expected behaviour

Strip A is pulled along the bottle’s axial direction, while Strip B is pulled around its circumference.

Because the PET chains have been oriented in both directions during stretch blow moulding, both strips should benefit from molecular alignment along their respective loading directions.

Neither direction must necessarily have identical tensile strength. The amount of orientation can differ between the axial and hoop directions.

b) Why both strips are needed

Testing Strip A only tells us how the material responds to a force along the bottle’s length.

It provides no direct evidence about the arrangement or mechanical behaviour of the polymer chains around the circumference.

Strip B tests the second direction.

Comparing both directions provides stronger evidence for biaxial orientation than testing one direction alone.

c) Experimental limitation

A tensile test measures mechanical behaviour, not molecular orientation directly.

Differences in crystallinity, local defects, molecular mass, or processing history can also affect the measured force.

Even when the strips have matching dimensions, their tensile results cannot be treated as a perfect measurement of molecular alignment.

A direct structural technique, such as polarised infrared spectroscopy, would provide additional evidence.

Question 4

A textile manufacturer describes a polymer as a polyester.

Its repeating unit is:

\[ \left[ \mathrm{-O-(CH_2)_4-O-C(=O)-C_6H_4-C(=O)-} \right]_n \]

Compare this polymer with PET.

a) Does the polymer belong to the polyester family?

b) Is it PET?

c) Identify the structural difference between the two polymers and explain why the manufacturer’s description is still correct.

Solution 4

The polymer belongs to the polyester family, but it is not PET.

a) Polymer family

The repeating unit contains ester linkages in its main chain.

Therefore, it is a polyester.

b) Specific polymer identity

It is not PET because its repeating unit contains a four-carbon section:

\[ \mathrm{-O-(CH_2)_4-O-} \]

PET contains the two-carbon section:

\[ \mathrm{-O-CH_2-CH_2-O-} \]

The polymer in the question is poly(butylene terephthalate), or PBT.

c) Structural comparison

Both polymers contain the terephthalate-derived benzene ring and ester linkages.

However, they use different diol-derived sections:

• PET is made using ethylene glycol.
• PBT is made using butane-1,4-diol.

The manufacturer’s description is correct because both polymers belong to the polyester family.

This is an important distinction: a polymer can be correctly classified as a polyester without being PET.

Question 5

A manufacturer produces two PET bottle designs using the same polymer grade.

Both bottle designs have:

• The same final dimensions.
• The same wall thickness.
• The same measured degree of crystallinity.
• The same total areal stretch ratio.

However, the directional stretching conditions are different.

The areal stretch ratio is calculated by multiplying the two directional stretch ratios:

\[ \lambda_A=\lambda_{\mathrm{axial}}\lambda_{\mathrm{hoop}} \]

A designer argues that the bottles must have identical molecular orientation and identical mechanical properties because both have an areal stretch ratio of 8.0.

Evaluate this claim.

In your response:

a) Explain why the two bottles can have different molecular orientations despite having equal areal stretch ratios.

b) Predict which bottle should have greater molecular alignment along its length and which should have greater alignment around its circumference.

c) Explain why the information provided is insufficient to determine which bottle will withstand the greater internal pressure before bursting.

Solution 5

The designer’s claim is incorrect. Equal areal stretch ratios do not guarantee identical molecular orientation or identical mechanical properties.

The important issue is that the two directional stretch ratios are different, even though their products are equal.

a) Why the molecular orientations differ

For Bottle A:

\[ \lambda_A=4.0\times2.0=8.0 \]

For Bottle B:

\[ \lambda_B=2.0\times4.0=8.0 \]

Both bottles have the same areal stretch ratio.

However, this single value does not describe how stretching is distributed between the two directions.

Molecular orientation depends on the direction in which the chains are stretched, not simply on the product of the two stretch ratios.

The two bottles can therefore have different directional molecular arrangements.

b) Compare the two directions

Bottle A has a greater axial stretch ratio of 4.0 compared with Bottle B’s ratio of 2.0.

Assuming comparable processing conditions, Bottle A should have greater molecular alignment along its length.

Bottle B has a greater hoop stretch ratio of 4.0 compared with Bottle A’s ratio of 2.0.

Bottle B should therefore have greater molecular alignment around its circumference.

Neither bottle is automatically stronger in every direction.

The important result is that the orientation is distributed differently.

c) Can we determine which bottle has the higher burst pressure?

No. The information is insufficient to make that prediction reliably.

A bottle under internal pressure experiences stresses in both the axial and hoop directions.

Its ability to resist bursting depends on the actual directional mechanical properties of the PET, not simply the total areal stretch ratio.

Other factors, including local wall-thickness variations, defects, molecular orientation, and the distribution of crystalline regions, can also affect failure.

Even though the question specifies equal overall thickness and crystallinity, we still do not have enough information about the actual directional strength of either bottle.

We would need additional mechanical testing and information about the bottle’s stress distribution to make a reliable comparison.

This is the key conceptual point: a single numerical measure of total stretching cannot describe a material whose properties depend on direction.

Two PET products can have identical repeating chemical structures and even identical overall stretch ratios, yet behave differently because their polymer chains are arranged differently.

The next useful step is to connect PET’s ester linkages to the intermolecular forces between polymer chains. This helps explain why polyester fibres behave differently from polyamide fibres such as nylon, even when both are used in clothing.

For that comparison, see How to Compare Hydrogen Bonding in Polyamides and Polyesters.