Glass, Carbon or Aramid: A Selection Matrix for FRP Strengthening

The three fibres used in structural strengthening are not interchangeable, and the difference is not simply “carbon is stronger”. A glass system that would fail a shear check can be correct for confinement; a carbon system that solves the flexural deficit can be the wrong purchase on a job where stiffness is not the problem.

The selection logic follows from what the intervention is asked to deliver: stiffness, strength, ductility, or durability under a specific exposure.

The properties that actually drive the decision

Property	Glass (GFRP)	Carbon (CFRP)	Aramid (AFRP)
Tensile strength (typical laminate)	moderate	high	high
Modulus of elasticity	low–moderate	high	low–moderate
Elongation at rupture	highest of the three	lowest	high
Density	higher than carbon	lowest	low
Alkali resistance	poor without protection	good	moderate
UV resistance	moderate	good	poor
Fatigue performance	moderate	very good	good
Relative material cost	1×	4–8×	2–4×
Typical use	confinement, shear, corrosion-resistant wraps	flexure, stiffness-critical, prestressed	impact, blast, special cases

Two rows carry most of the weight. Modulus decides how much of the composite’s strength can be engaged at the strains the structure can tolerate; elongation decides whether the composite fails before or after the concrete crushes.

Why modulus is the first question

Concrete in a strengthened member typically reaches its usable strain long before a high-modulus composite reaches its tensile capacity. A carbon laminate with a rupture strain of around 1.5% will be doing useful work when the concrete is at 0.2–0.3% strain; a glass laminate with 2.5% elongation needs more deformation to develop the same force, and in a stiffness-critical application it simply cannot deliver.

That is why carbon dominates flexural strengthening of slabs and beams: the design is usually strain-limited, and the high modulus converts a thin laminate into meaningful moment capacity without changing the section depth.

Conversely, for column confinement, the composite works in hoop tension at relatively low strain, and the requirement is a continuous jacket that fails at a controlled strain after the concrete has been allowed to dilate. Glass performs this duty well and at a fraction of the cost. Using carbon for confinement is not wrong; it is usually an expensive way to buy capacity that the mechanism cannot use.

Durability decides the fibre in aggressive exposure

Glass fibres are attacked by the alkaline pore solution of concrete if moisture reaches the fibre–matrix interface. Modern sizing and matrix systems address this, and glass systems are used successfully in contact with concrete — but the qualification depends on the specific system, not on the fibre type in the abstract. Where the exposure includes moisture, chlorides or thermal cycling, the system’s environmental reduction factor is the number to interrogate, not the datasheet’s dry tensile strength.

Aramid has poor UV resistance and needs protection; it also absorbs moisture and has weak compressive performance. Its real strengths are impact and blast resistance, where its elongation and toughness are valuable. In ordinary building strengthening it is a specialist material.

Carbon is chemically stable in concrete and in most building exposures, with two practical exceptions: it is electrically conductive, so carbon dust and offcuts near live equipment are a hazard, and it can form a galvanic couple with steel where the two are in direct contact in the presence of an electrolyte. Isolation at steel interfaces is a detailing requirement, not a theoretical concern.

Form matters as much as fibre

The same fibre sold as a wet-layup fabric, a pultruded laminate and a precured strip behaves differently on site.

• Wet layup — dry fabric saturated on site. Adaptable to complex geometry, easy to wrap corners, and heavily installation-dependent. Quality varies with the applicator more than with the product.
• Pultruded plates — factory-made, high fibre volume, stiff, and therefore efficient in flexure. Poor at wrapping; requires straight surfaces and careful end anchorage.
• Precured strips and shells — consistent factory quality, useful for columns and silos, less adaptable to irregular sections.
• Near-surface mounted (NSM) bars — the composite is set into grooves cut into the cover. Protects the reinforcement from fire and impact, uses less material, and suits negative moment regions over supports where external bonding is impractical.

On a project where the fire strategy is aggressive, NSM has a substantial advantage: the cover provides protection that an externally bonded laminate does not, and the protective system becomes a smaller part of the budget.

A selection workflow that survives review

1. Define the deficit. Flexural capacity, shear capacity, confinement/ductility, stiffness, or durability. Be specific — “strengthen the beam” is not a deficit.
2. Check the limit state. Most FRP flexural designs are governed by debonding strain or by concrete crushing, not by composite rupture. If the limit is debonding, buying a stronger fibre changes nothing.
3. Choose the form based on geometry, access, fire requirement and the anchorage detail that is actually buildable.
4. Choose the fibre based on modulus requirement, ductility need and exposure, then confirm with the system’s environmental reduction factors.
5. Confirm the assembly — primer, saturant, protective coating and fire protection as a tested combination, not as components from three suppliers.
6. Price per unit of capacity delivered, not per kilogram or per square metre. A glass system at a third of the cost that requires four times the material and a thicker protective build-up is not cheaper.

Step 6 is where procurement arguments go wrong. Unit-rate comparison across fibre types is meaningless without the design output each system produces for the same member.

Common selection errors

• Glass specified for a stiffness-critical slab. The material is adequate in strength and inadequate in modulus, and the design compensates by adding plies until the fire protection becomes the dominant cost.
• Carbon specified for confinement. Expensive capacity that the confinement mechanism cannot mobilise. The exception is very large columns or where the same wrap must also carry shear.
• Aramid specified without UV protection. The system degrades in exactly the exposed locations where wraps are most visible.
• Wet layup specified for a geometrically complex detail with no allowance for the labour and inspection that adaptivity requires.
• Mixed-vendor assembly. The design assumes a tested system; the site receives compatible-looking components. Performance is then unknown, not merely uncertain.

The honest summary is that fibre selection is a minor decision compared with getting the substrate, the anchorage and the fire detail right. Carbon, glass and aramid all work in the applications they suit. What does not work is choosing a fibre for its headline tensile strength and then treating the rest of the system as an afterthought.