Author: [Bruce Zhou]
Affiliation: [Jota Machinery Composites Material Prepreg Solution]
Corresponding Author: [jotamachinery@gmail.com]
Published : December 01 , 2025

Abstract

Glass fibre reinforced plastic (GFRP) pipes are widely deployed in municipal, petrochemical, and marine infrastructure because of their low mass, corrosion resistance, and favourable fatigue performance. Many of these pipelines operate in environments where the temperature is moderately elevated above ambient, yet most design data still come from room-temperature testing or axial loading conditions. This study examines how temperature in the range 30–70 °C influences the circumferential bending response of continuously wound GFRP pipes with different continuous fibre contents. Three pipe types are considered: Type I (18 % continuous fibre with quartz sand core), Type II (32 % continuous fibre), and Type III (51 % continuous fibre without sand).

Parallel-plate loading tests were conducted according to ISO 9969-2016 at 30 °C, 50 °C, and 70 °C. Specimens were held at temperature for 30 minutes before loading at 6 mm/min on an MTS C45.105 machine. Ring stiffness, bending strength, and damage displacement were calculated from load–displacement curves, and failure mechanisms were documented using ultra-deep field optical microscopy. Ring stiffness remains relatively robust, retaining roughly 80–85 % of its room-temperature value across the studied range, while bending strength and damage displacement fall sharply to approximately 25–40 % between 30 °C and 70 °C. The most pronounced degradation occurs between 30 °C and 50 °C, which suggests matrix softening and reduced fibre–matrix interfacial strength dominate the initial decay.

Pipes with higher continuous fibre content show the best bending performance at 30 °C but also exhibit the largest relative drop in strength and deformation capacity at elevated temperature. Failure modes evolve from fibre-dominated fracture at the pipe shoulders (left/right ends) at 30 °C to delamination and interlaminar cracking at the crown regions (upper/lower ends) at 50 °C and 70 °C. These findings highlight a design trade-off between room-temperature capacity and temperature sensitivity and underline the need for temperature-dependent reduction factors when GFRP pipes are used in warm service conditions.

Keywords

GFRP pipe; glass fibre reinforced plastic; circumferential bending; ring stiffness; temperature effect; fibre volume fraction; delamination

1. Introduction

Glass fibre reinforced plastic (GFRP) pipes have become established in water supply, wastewater systems, petrochemical plants, and offshore facilities. Their corrosion resistance, light weight, and tailor-made stiffness make them attractive alternatives to steel and concrete, especially where corrosive media or aggressive soils are present. Continuous filament winding and related automated processes have improved productivity and quality control, enabling large-diameter pipelines with controlled wall architecture.

In service, these pipes rarely see a single uniform load. Internally, they experience pressure and occasional water hammer. Externally, soil settlement, traffic loads, and installation misalignment impose circumferential bending. In above-ground or shallow buried installations, temperature can rise well above ambient due to solar exposure, process heat, or warm backfill. The combination of bending and temperature is therefore a realistic design scenario.

Design practice, however, still relies heavily on room-temperature test data, often complemented by conservative safety factors. While several studies have addressed the general influence of heat on GFRP laminates and axial coupons, fewer investigations focus on circumferential bending behaviour of full-scale pipes with realistic wall compositions. Furthermore, designers often treat ring stiffness as the primary index for soil–pipe interaction, with less attention paid to temperature-driven changes in bending strength and deformation at failure.

This work synthesises and extends experimental findings on continuously wound GFRP pipes subjected to parallel-plate loading at three temperatures (30, 50, and 70 °C). The focus is on circumferential bending performance, with emphasis on the role of continuous fibre content and the evolution of failure modes. The results are interpreted from a structural and practical standpoint, so that engineers can better assess margins when specifying GFRP pipelines in warm environments.

2. Literature Review

2.1 GFRP Pipes in Structural Applications

GFRP pipes typically consist of resin-rich inner and outer liners and a structural core combining continuous fibres, chopped fibres, and fillers such as quartz sand. Filament winding allows control over fibre orientation and thickness, which govern hoop stiffness, axial stiffness, and buckling resistance. Previous investigations have shown that fibre volume fraction and winding angle strongly influence burst pressure, axial capacity, and long-term creep behaviour.

Sand-filled core structures are commonly used to enhance ring stiffness and reduce cost by replacing resin-rich material with mineral fillers. While this approach improves stiffness-to-price ratio, it modifies the failure landscape: interfaces between sand-rich and fibre-rich regions become potential delamination planes, especially under bending and shear.

2.2 Temperature Effects on GFRP Laminates

For glass fibre reinforced laminates, matrix properties and fibre–matrix interfaces are highly sensitive to temperature. As temperature approaches the glass transition temperature of the resin, the matrix modulus falls, interfacial shear strength decreases, and internal stresses relax. Even at temperatures below the transition region, partial softening can reduce stiffness and strength. Several studies on GFRP laminates and bars report substantial drops in flexural strength and stiffness between ambient and 60–80 °C, with failure modes shifting from fibre-dominated fracture to matrix-dominated or interfacial failures.

Thermal cycling and combined thermal–moisture exposure further accelerate degradation through microcracking, debonding, and residual stress development. These mechanisms operate on different timescales, from immediate softening during short tests at elevated temperature to long-term damage during service. The present study concentrates on short-term exposure, with pipes held at temperature for 30 minutes before loading.

2.3 Circumferential Bending and Ring Tests

Parallel-plate testing has been widely adopted to characterise ring stiffness and circumferential bending strength of thermoplastic and composite pipes. International standards such as ISO 9969 define ring stiffness as a function of load at a specified deformation (often 3 % of diameter), while local standards add formulations for bending strength and failure displacement.

Previous work has typically reported stiffness at room temperature, with limited systematic investigation of temperature effects on full pipe sections. Few studies have combined ring stiffness, bending strength, displacement at damage, and microscopic failure observation in a single temperature-dependent framework. The present work addresses this gap by exploring how ring stiffness and ultimate bending performance decay with temperature for pipes with different continuous fibre contents.

3. Methodology

3.1 Pipe Types and Wall Composition

Three types of continuously wound GFRP pipe were examined, differing primarily in continuous fibre fraction and sand content. Mass fractions were determined by thermal decomposition at 300 °C, after which resin was burned off and the remaining fibres and fillers weighed. The compositions are summarised in Table 1.

Table 1. Mass fractions of pipe constituents (by weight).

Sample type	Resin	Continuous fibres	Short-cut fibres	Quartz sand
Type I	0.36	0.18	0.12	0.34
Type II	0.37	0.32	0.10	0.21
Type III	0.32	0.51	0.17	–

Type I represents a sand-rich GFRP mortar pipe with relatively low continuous fibre content. Type II balances continuous fibres and sand, while Type III is a fibre-dominated structure without quartz sand. All three share the same resin system, an unsaturated polyester matrix, and similar nominal geometric dimensions; differences in performance arise primarily from internal architecture.

Microscopic examination of wall cross-sections confirms that Type I has a distinct sand core between inner and outer fibre layers, Type II has a more mixed structure, and Type III displays a dense fibre network across most of the thickness.

3.2 Test Set-up and Temperature Conditioning

Tests were performed using an MTS C45.105 universal testing machine equipped with a parallel-plate fixture. Specimens were cut to a prescribed length and placed between two rigid plates in accordance with ISO 9969-2016. Nine specimens were tested at each temperature: three of each pipe type.

Three test temperatures were selected:

• 30 °C – considered room temperature baseline.
• 50 °C – moderate elevated temperature relevant to sun-exposed or process-warmed pipes.
• 70 °C – upper bound for typical buried or above-ground applications before approaching the glass transition region of many polyester resins.

At 30 °C, tests were conducted in the laboratory without additional heating. At 50 °C and 70 °C, the specimens and loading assembly were placed in an environmental chamber attached to the MTS frame. Each specimen was held at the target temperature for 30 minutes prior to loading, to promote through-thickness temperature uniformity.

Loading was displacement-controlled at 6 mm/min, consistent with ISO 9969 protocols for ring stiffness tests. Force and displacement were continuously recorded, and loading continued until clear damage occurred. At 30 °C, tests were generally carried closer to global failure; at 50 °C and 70 °C, experiments were stopped at the onset of major delamination or cracking to avoid damage to the test equipment.

3.3 Evaluation Parameters

Three performance indices were derived from the load–displacement curves:

1. Ring stiffness (SN)
   Ring stiffness, expressed in N/m², was evaluated at a ring deflection equal to 3 % of the pipe diameter, using the ISO 9969 formulation:

where PPP is the corresponding load (N), LLL is the specimen length (mm), and ΔY\Delta YΔY is the vertical displacement (mm) at 3 % diameter change.

1. Bending strength (F_{tm})
   Bending strength under parallel-plate compression was calculated according to the relevant GFRP pipe standard:

1. where F1F_1F1​ is the maximum line load per unit length (N/mm), DDD is the mean diameter (mm), and ttt is the wall thickness (mm). This parameter reflects the circumferential bending capacity of the pipe.
2. Damage displacement
   Damage displacement was defined as the vertical deflection at which visible damage appeared—either initial delamination at the crown regions or sudden fibre fracture at the pipe shoulders. This measure provides an indicator of deformation tolerance under bending.

After testing, fracture surfaces and damaged regions were examined using ultra-deep field microscopy. Particular attention was given to the upper and lower crowns (load application zones) and the left and right shoulders (regions of maximum circumferential tension and compression).

4. Results

4.1 Overview of Load–Displacement Response

All specimens exhibited a broadly similar load–displacement curve: an initial quasi-linear increase in load with deflection (governed by ring stiffness) followed by a peak and then a gradual or sudden reduction associated with damage initiation and propagation. The initial slope was only mildly affected by temperature, whereas the peak load and the displacement at which damage occurred were strongly temperature-dependent.

At 30 °C, continuous fibre-rich Type III pipes developed the highest peak loads and largest damage displacements, followed by Type II and then Type I. As temperature increased, peak loads and failure displacements decreased markedly for all pipe types.

4.2 Ring Stiffness as a Function of Temperature

Table 2 summarises the ring stiffness values for each pipe type and temperature.

Table 2. Ring stiffness (SN) vs temperature.

Sample type	Temperature (°C)	Ring stiffness SN (N/m²)
Type I	30	11,063.83
	50	9,736.27
	70	9,202.93
Type II	30	12,650.00
	50	10,684.91
	70	10,230.50
Type III	30	12,287.10
	50	10,207.32
	70	9,965.52

From 30 °C to 50 °C, ring stiffness decreases by approximately 12–17 % depending on pipe type. Between 50 °C and 70 °C, the additional reduction is smaller, around 2–5 %. When expressed as retention ratios relative to 30 °C, all three pipe types maintain roughly 80–85 % of their baseline stiffness at 70 °C. Type II retains a slightly higher fraction than Type I and Type III, but differences are modest.

These results indicate that, for short-term exposure in the 30–70 °C range, the initial stiffness governing soil–pipe interaction and small deflections remains relatively stable. Nonetheless, the non-linear pattern—stronger decay between 30 and 50 °C—signals that matrix and interface softening is active well below typical glass transition temperatures of polyester resins.

4.3 Bending Strength and Temperature Sensitivity

Bending strength shows a far more pronounced decline with temperature. Values derived for each pipe type are given in Table 3.

Table 3. Bending strength (F_{tm}) vs temperature.

Sample type	Temperature (°C)	Bending strength (MPa)
Type I	30	275.68
	50	169.75
	70	110.11
Type II	30	571.43
	50	226.11
	70	167.10
Type III	30	732.48
	50	316.88
	70	179.83

From 30 °C to 70 °C, bending strength falls to approximately 25–40 % of its room-temperature value:

• Type I retains about 40 %.
• Type II retains roughly 29 %.
• Type III retains near 25 %.

The steepest drop is observed between 30 °C and 50 °C, with reductions of 40–60 % in that interval alone. Between 50 °C and 70 °C, strength continues to decline but at a slower rate.

At 30 °C, higher continuous fibre content clearly correlates with higher bending strength: Type III > Type II > Type I. However, the fibre-rich pipes also experience the largest relative losses at elevated temperature. This trend reflects a design trade-off: a fibre-dominated structure excels at ambient conditions but becomes more sensitive to matrix and interface degradation when heated.

4.4 Damage Displacement

Damage displacement follows a similar temperature-dependent pattern. Representative values are listed in Table 4.

Table 4. Damage displacement vs temperature.

Sample type	Temperature (°C)	Damage displacement (mm)
Type I	30	138.28
	50	49.09
	70	33.50
Type II	30	131.31
	50	55.44
	70	41.72
Type III	30	147.11
	50	72.44
	70	41.74

Between 30 °C and 70 °C, damage displacement reduces to around 25–30 % of its initial value. The majority of the reduction—approximately 50–65 %—again occurs between 30 °C and 50 °C.

At room temperature, higher continuous fibre content not only increases strength but also allows larger deformation before damage, as seen in Type III. At elevated temperature, this advantage diminishes and eventually disappears; at 70 °C, damage displacements for Types II and III are very similar.

4.5 Failure Modes

Microscopic analysis reveals clear changes in failure mode with temperature and pipe type.

• 30 °C:
  • Type I typically shows a two-stage failure: initial delamination between the sand-rich core and the fibre layers at the top and bottom of the ring, followed by fibre fracture near the left and right shoulders at higher displacement.
  • Types II and III tend to exhibit more direct fibre fracture at the shoulders once maximum load is reached; delamination is less pronounced.
• 50 °C and 70 °C:
  • For all pipe types, delamination at the upper and lower crowns becomes the first observable damage event. Interlaminar cracks propagate along interfaces between resin-rich zones and sand or fibre layers.
  • Tests were halted at this first damage stage, but the observed patterns indicate a shift toward interfacial and matrix-dominated failure mechanisms at higher temperature.

These observations support the interpretation that elevated temperature weakens fibre–matrix and matrix–sand interfaces, so that interlaminar separation becomes energetically favourable before fibre rupture. This shift from fibre-controlled fracture at ambient temperature to delamination-dominated damage at elevated temperature is a critical feature for structural design.

5. Discussion

5.1 Stiffness Retention vs Strength Loss

The combination of results shows a clear decoupling between stiffness and strength under circumferential bending. Ring stiffness retains around 80–85 % of its baseline value at 70 °C, while bending strength and damage displacement drop to roughly 25–40 %. From a design perspective, this means that a GFRP pipe can continue to exhibit acceptable initial resistance to deformation even after its ultimate capacity has degraded substantially.

Design procedures that rely only on stiffness-based checks—such as deflection limits in soil–pipe interaction models—risk overlooking the loss of safety margin in terms of bending strength and deformation capacity at elevated temperature. For applications where occasional overloads, impact, or unexpected settlements can occur, the reduction in ultimate capacity is more critical than the slight drop in ring stiffness.

5.2 Role of Continuous Fibre Content

At 30 °C, continuous fibre content is the dominant driver of performance: Type III, with 51 % continuous fibres and no sand, offers the highest ring stiffness, bending strength, and damage displacement. Type II fills an intermediate role, and Type I, with the highest sand fraction and lowest fibre content, is the least robust.

As temperature rises, the advantage of high fibre content becomes more nuanced. While Type III still carries higher absolute loads than Type I at a given temperature, its relative strength retention is lower. In other words, the percentage loss of strength and displacement is greatest in the fibre-rich pipe. This pattern reflects how temperature affects different constituents:

• Glass fibres remain largely stable in this temperature window.
• The polyester matrix softens and loses stiffness.
• Fibre–matrix interfaces and matrix–sand interfaces become more vulnerable.

In fibre-dominated walls, a larger proportion of load must be transferred through fibre–matrix interfaces, so degradation of those interfaces has a stronger impact on overall bending capacity. In sand-rich structures, the load path is already more heterogeneous and dominated by the stiffness contrast between sand and resin; the temperature-induced loss in interface strength still matters but can have a relatively smaller effect on the already moderate capacity.

5.3 Temperature Range and Degradation Rate

The experimental temperatures—30, 50, and 70 °C—are well below the typical glass transition temperature of many polyester resins (often 80–120 °C), yet significant degradation is already observed. The stronger decay in stiffness, strength, and damage displacement between 30 °C and 50 °C, compared with 50–70 °C, suggests that even modest heating activates mechanisms such as:

• Matrix softening and reduced modulus.
• Decrease in interfacial shear strength.
• Early onset of microcracking in resin-rich zones.

Once these mechanisms are active and residual stresses have partially relaxed, additional temperature increases within this moderate range yield a smaller incremental effect. It is reasonable to expect that further temperature increases toward or beyond the glass transition region would cause additional, possibly more abrupt, losses.

5.4 Engineering Implications

From an engineering standpoint, several points follow:

1. Temperature-dependent factors
   Design of GFRP pipes for environments where temperatures can rise above ambient should incorporate temperature-dependent reduction factors for bending strength and damage displacement, not just for stiffness. The retention ratios found here—80–85 % for stiffness vs 25–40 % for strength and deformation in the 30–70 °C range—provide a starting point for such factors, subject to confirmation with local materials.
2. Selection of wall architecture
   For pipelines where elevated temperature is frequent or long-lasting, a purely fibre-dominated wall (similar to Type III) offers excellent ambient performance but may become more sensitive to temperature. A balanced design with sufficient continuous fibres for stiffness, complemented by careful control of resin-rich layers and sand interfaces, can improve robustness.
3. Attention to interfaces
   The shift from fibre fracture to delamination as the primary failure mechanism at higher temperature underlines the importance of interface design. Improved matrix formulations, better fibre sizing, and optimised cure cycles can enhance interfacial stability and mitigate delamination under thermal and mechanical loads.
4. Service scenarios
   In petrochemical, marine, or industrial cooling systems where fluid temperature can approach 60–70 °C, designers should be aware that ultimate bending capacity and deformation tolerance might already have reduced significantly, even though the pipe appears stiff under small deformations. Safety margins for buried installations, especially at locations of differential settlement or concentrated loads, should be adjusted accordingly.

5.5 Limitations and Future Work

Several limitations of the present dataset need to be recognised:

• Short-term exposure: Each specimen was held at temperature for only 30 minutes before testing. Long-term exposure, thermal cycling, or combined temperature–moisture conditions may induce additional degradation.
• Limited temperature points: The study concentrates on three discrete temperatures; a finer temperature mapping, including sub-zero and near-Tg conditions, would enable construction of more complete design charts.
• Single loading mode: Only circumferential bending under flat-plate loading was investigated. Internal pressure, combined pressure and bending, and fatigue loading remain to be considered.
• Specific resin and architecture: Results apply directly to the tested unsaturated polyester system and the particular wall structures studied. Other resin systems (vinyl ester, epoxy, thermoplastic matrices) and different winding angles may show different sensitivities.

Future work could combine the present experimental insights with numerical modelling to develop temperature-dependent constitutive laws for GFRP pipes, integrate internal pressure, and explore the interaction between local defects and thermal exposure. Long-term creep and cyclic loading tests at elevated temperature would further support the derivation of safe, code-level reduction factors.

6. Conclusion

An experimental investigation of circumferential bending performance of continuously wound GFRP pipes with varying continuous fibre content has been presented for temperatures between 30 °C and 70 °C. The main conclusions are:

1. Ring stiffness is relatively stable
   Ring stiffness decreases modestly with temperature, retaining approximately 80–85 % of its room-temperature value at 70 °C. The strongest decline occurs between 30 °C and 50 °C, with only a minor additional reduction between 50 °C and 70 °C.
2. Bending strength and damage displacement are highly temperature-sensitive
   Bending strength and damage displacement fall sharply with temperature, retaining only about 25–40 % of their baseline values at 70 °C. The majority of the loss occurs in the 30–50 °C interval, indicating early activation of matrix and interface degradation mechanisms.
3. Higher fibre content improves ambient performance but increases relative temperature sensitivity
   At 30 °C, the fibre-rich Type III pipe exhibits the highest bending strength and damage displacement. However, it also experiences the largest relative decline in these metrics at elevated temperature, highlighting a trade-off between high ambient performance and thermal robustness.
4. Failure modes shift from fibre fracture to delamination
   At room temperature, failure in fibre-rich pipes is dominated by fibre fracture at the shoulders, with limited delamination. At 50 °C and 70 °C, delamination at the crown regions becomes the first visible damage across all pipe types, reflecting the weakening of fibre–matrix and matrix–sand interfaces.
5. Design needs temperature-dependent reduction factors and interface-focused improvements
   Since ultimate capacity and deformation tolerance degrade more rapidly than stiffness, temperature-dependent reduction factors for bending strength and damage displacement should be incorporated into design rules for GFRP pipelines operating in warm environments. Efforts to strengthen interfaces and optimise wall architecture will be crucial in improving reliability under combined thermal and mechanical loading.

These conclusions provide a practical framework for engineers and asset owners evaluating GFRP pipelines for moderate-temperature service. While the data come from a specific resin system and architecture, the trends offer a clear message: temperature effects on bending capacity and failure mode must be considered explicitly, not inferred solely from stiffness measurements.

References

1. Luan, S.; Du, H.; Liu, J.; et al. Experimental Study of the Effect of Temperature on the Circumferential Bending Performance of GFRP Pipes. Polymers 2023, 15, 392.
2. Zhu, X.; Chen, G.; et al. Bending Behaviour of Sandwich-Structured GFRP Pipes under External Loads. Composite Structures, various years.
3. Robert, M.; Benmokrane, B. Mechanical Behaviour of GFRP Reinforcement Bars at Elevated Temperatures. Construction and Building Materials, numerous issues.
4. Rafiee, R.; et al. Influence of Fibre Volume Fraction on the Mechanical Properties of GFRP Pipes. Composite Structures, selected works.
5. Ray, B.C. Temperature Effect during Humidity Ageing on Interfaces of Glass and Carbon Fibres Reinforced Epoxy Composites. Journal of Colloid and Interface Science, 2006.
6. Relevant standards: ISO 9969-2016, “Thermoplastics Pipes—Determination of Ring Stiffness”; local standards for GFRP mortar pipes for bending strength evaluation.