Effect of Fiber Type on the Thermomechanical Performance of High-Density Polyethylene (HDPE) Composites with Continuous Reinforcement

Abstract

The thermal, thermomechanical, and viscoelastic properties of continuous unidirectional (UD) glass fiber/high-density polyethylene (GF/HDPE) and ultra-high-molecular-weight polyethylene/high-density polyethylene (UHMWPE/HDPE) tapes are characterized in this paper in order to support their use in extreme environments. Unlike prior studies that focus on short-fiber composites or limited thermal conditions, this work examines continuous fiber architectures under five operational environments derived from Army Regulation 70-38, reflecting realistic defense-relevant extremes. Differential scanning calorimetry (DSC) was used to identify melting transitions for GF/HDPE and UHMWPE/HDPE, which guided the selection of test conditions for thermomechanical analysis (TMA) and dynamic mechanical analysis (DMA). TMA revealed anisotropic thermal expansion consistent with fiber orientation, while DMA, via strain sweep, temperature ramp, frequency sweep, and stress relaxation, quantified their temperature- and time-dependent viscoelastic behavior. The frequency-dependent storage modulus highlighted multiple resonant modes, and stress relaxation data were fitted with high accuracy (R² > 0.99) to viscoelastic models, yielding model parameters that can be used for predictive simulations of time-dependent material behavior. A comparative analysis between the two material systems showed that UHMWPE/HDPE offers enhanced unidirectional stiffness and better low-temperature performance. At the same time, GF/HDPE exhibits lower thermal expansion, better transverse stiffness, and greater stability at elevated temperatures. These differences highlight the impact of fiber type on thermal and mechanical responses, informing material selection for applications that require directional load-bearing or dimensional control under thermal cycling. By integrating thermal and viscoelastic characterization across realistic operational profiles, this study provides a foundational dataset for the application of continuous fiber thermoplastic tapes in structural components exposed to harsh thermal and mechanical conditions.

1. Introduction

Polymer composites are widely used in the aerospace and automotive sectors, primarily due to their high strength-to-weight ratios. Recently, there has been increasing interest in extending their application to extreme environments such as space exploration, superconducting magnet systems, and advanced electronics [1]. Their typical uses include liquid propellant tanks, satellite components, aircraft structural elements, electrical insulation in superconducting magnets, and infrastructure for Arctic operations [2]. The viscoelastic behavior of polymers results in properties that are sensitive to variables like temperature, deformation rate, and frequency, directly influencing the mechanical performance and structural integrity of components over time [3]. Gaining insight into material responses under extreme thermal conditions is vital for accurately predicting mechanical failures and creating designs with reliable service life [4]. The multifunctional requirements of polymers in these demanding environments often need robust thermomechanical performance, especially at low temperatures, where materials usually fall below their glass transition or melting points and exhibit reduced viscoelastic effects [5].

Coefficient of thermal expansion (CTE) plays a critical role in anticipating thermal stresses that arise when different materials are combined and subjected to varying temperatures [1]. In continuous-fiber composites, both the fiber and matrix materials contribute to dimensional changes, with anisotropic behavior commonly observed in fiber CTE [6]. For example, glass fibers contract upon cooling in both directions [7], whereas ultra-high-molecular-weight polyethylene (UHMWPE) can demonstrate a negative CTE, expanding with decreasing temperature [8,9]. Such discrepancies in thermal expansion can lead to complex internal stress states within the composite, potentially affecting interfacial adhesion and overall durability, particularly when considering thermal cycling histories [2].

Material microstructure significantly affects temperature-dependent polymer properties. Generally, polymers become stronger and stiffer while losing ductility at reduced temperatures due to restricted chain mobility and enhanced intermolecular forces [2,10,11]. Stress relaxation behavior also slows significantly as temperature decreases, imposing limitations on the use of polymers, where toughness and ductility are critical [12]. Additionally, polymers exhibit rate- and frequency-dependent mechanical responses; higher frequencies often correspond to increased stiffness and unique vibrational characteristics, such as natural frequencies and damping ratios, which introduce nonlinearities in material behavior [3,13,14]. Thermal lag during cyclic heating further influences the measured thermomechanical properties, underscoring the need to account for these effects in realistic applications [14]. Given their time-dependent nature, understanding the long-term performance and dimensional stability of polymers under extreme environmental exposures remains a key research focus [15]. Factors like prolonged deformation lead to stress-relaxation phenomena, which are also affected by environmental conditions and significantly influence polymer flow mechanisms, especially at low temperatures [4].

While prior studies have examined the thermal and mechanical characteristics of polymer composites, many have focused on short-fiber reinforcements or standard laboratory environments rather than application-relevant extreme conditions [16,17]. Current research on HDPE-based systems using techniques such as DSC, DMA, or TMA often lack comprehensive evaluations performed under these realistic environmental stresses [18]. Continuous-unidirectional-fiber-reinforced thermoplastics, such as GF/HDPE and UHMWPE/HDPE tapes, remain understudied with respect to standardized extreme environmental test protocols [19]. Although dynamic mechanical analysis provides valuable insights into storage modulus and frequency dependence, its integration with formal environmental standards, such as Army Regulation 70-38, is uncommon [20]. Furthermore, the existing literature rarely employs fitted viscoelastic models, such as the Prony series, to capture long-term relaxation behavior and improve predictive capability for continuous-fiber systems [21].

This work forms part of a broader initiative to characterize the behavior of polymer composites under harsh environmental conditions, defined by exposure to extreme stresses that challenge material performance, stability, and longevity [22]. The present study aims to (1) introduce materials tailored for extreme environment applications, (2) characterize thermomechanical properties of continuously reinforced HDPE tapes under these conditions, (3) assess the storage modulus across varying environmental factors, (4) evaluate viscoelastic stress relaxation behavior for long-term performance prediction, (5) examine the effects of frequency on mechanical response, and (6) fit predictive viscoelastic models to experimental data. Five operational conditions derived from Army Regulation 70-38 ensure that our testing and analysis reflect the realistic environmental influences affecting equipment globally [23]. Key measurements include thermal expansion coefficients, elastic moduli, and viscoelastic responses, complemented by viscoelastic modeling to capture complex time-dependent behavior. This study provides a comprehensive evaluation of continuous HDPE fiber-filled composites under relevant extreme conditions, offering valuable material information for their deployment in demanding applications.

2. Materials and Methods

2.1. Materials and Operational Conditions

This study utilizes two types of unidirectional (UD) tapes made from fiber-reinforced high-density polyethylene (HDPE). HDPE, chosen as the matrix material, offers excellent moisture resistance and is known for being a cost-effective thermoplastic [24]. Its ability to perform across a wide temperature range makes it well-suited for use in demanding or extreme environments. In addition to being lightweight and durable, HDPE exhibits strong resistance to chemicals and ultraviolet (UV) radiation [25]. The reinforcement fibers used in this work were glass fiber (GF) and ultra-high-molecular-weight polyethylene (UHMWPE). Both composite tape materials were procured from A+ Composites GmbH (Weselberg, Germany). Technical specifications for these materials are presented in

Table 1. Materials technical information provided by the supplier.

Material	Name	Manufacturer	Fiber Content	Thickness (mm)
GF/HDPE	GF-HDPE-46-51-296	A+ Composites	46 vol. %	0.205
UHMWPE/HDPE	UHMWPE-HDPE-50-51-235	A+ Composites	50 vol. %	0.235

.

Composite materials are increasingly used in the design of structures intended for extreme environments, including cold, hot, humid, tropical, and arid regions.

Table 2. Temperatures selected based on operational conditions using the Army Regulation 70-38 Standard.

Temperature Daily High (°C)	Design Type	Daily Cycle	Ambient Relative Humidity (%RH)
−57	Extreme Cold	Extreme Cold (C4)	Tending toward saturation
−37	Cold	Cold (C2)	Tending toward saturation
−6	Basic	Mild Cold (C0)	Tending toward saturation
24	Basic	Constant High Humidity (B1)	95–100
49	Hot	Hot Dry (A1)	3–8

outlines the selected temperature conditions, along with the associated design types and daily cycle classifications, based on the Army Regulation 70-38 [23]. To characterize the HDPE materials, a temperature range encompassing all five daily cycles was chosen. In this study, achieving and maintaining the specified target temperatures was a primary objective.

2.2. Material Characterization Overview

The material characterization scheme for this work is extensive. This section summarizes the material characterization techniques in this work. Figure 1 shows a diagram indicating the techniques used, the reasoning and objectives behind each method, and the procedures used. For more information on the characterization techniques, refer to Section 2.3 for Differential Scanning Calorimetry (DSC) testing, Section 2.4 for Thermomechanical Analysis (TMA) testing, and Section 2.5 for Dynamic Mechanical Analysis (DMA) testing.

2.3. Differential Scanning Calorimetry (DSC)

Thermal characterization of the materials was conducted using a DSC2500 Differential Scanning Calorimeter (TA Instruments, New Castle, DE, USA). Tzero aluminum pans and lids were used for all measurements. Each material was tested using four replicate samples, and the results were averaged. A heating and cooling rate of 10 °C/min was applied, with an initial and final temperature range of −80 °C to 200 °C for the raw tape and 10 °C to 200 °C for the consolidated panel, as per ASTM standards E793-24 [26], D3418-21 [27], and E1269-11 [28]. Nitrogen was used as purge gas. The DSC procedure followed a standard heat–cool–heat cycle. The first heating run captures the thermal behavior of the as-received material, effectively documenting its original state. The controlled cooling step establishes a consistent thermal history across samples, and the second heating run enables a direct comparison between different materials and processing conditions [29]. The measured melting temperature (T_m) was used to determine appropriate upper limits for subsequent TMA, DMA, and processing parameters. DSC tests were conducted on both the raw (as-received) materials and the consolidated panels to verify that the processing did not alter the material’s microstructure.

2.4. Thermomechanical Analysis (TMA)

Thermomechanical properties were evaluated using a TMA Q400 Thermomechanical Analyzer (TA Instruments, New Castle, DE, USA). This instrument measures dimensional changes in response to temperature, allowing for the calculation of the coefficient of thermal expansion (CTE) based on the sample’s initial length. In accordance with ASTM E831-25 [30], specimen lengths were kept below 10 mm. CTE values for the HDPE systems were determined over a temperature range of −50 °C to 50 °C. For each material, a minimum of three specimens were tested in each principal direction. All tests were conducted using a heating rate of 5 °C/min.

2.5. Dynamic Mechanical Analysis (DMA)

A dynamic mechanical analysis was conducted using a DMA850 (TA Instruments, New Castle, DE, USA) to evaluate both composite materials. Testing included strain sweeps, temperature ramps, frequency sweeps, and stress relaxation experiments on the HDPE-based systems. Input parameters were selected with guidance from ASTM standards D5023-23 [31], D7028-07(2024) [32], E1640-23 [33], and D2990-17 [34]. A three-point bending fixture with a 50 mm span was used to minimize clamping artifacts and ensure an accurate mechanical response. Specimens were machined in both the fiber direction (1-direction) and transverse direction (2-direction), as illustrated schematically in Figure 2.

2.5.1. Strain Sweep Test

Strain sweep experiments were carried out on specimens in both the fiber (1-dir) and transverse (2-dir) directions. Tests were run at 1 Hz and 10 Hz, while the strain amplitude was stepped from 0.001% up to 0.1%. Sweeps were performed isothermally at −50 °C, 30 °C, and 110 °C to establish a strain level that stayed within the linear viscoelastic region (LVR) for all three temperatures. The 10 Hz sweep was conducted only at −50 °C.

2.5.2. Temperature Ramp Test

Temperature ramp tests were conducted on specimens oriented in both the 1-dir and 2-dir. To evaluate the influence of frequency, tests were performed at 1 Hz and 10 Hz. A temperature ramp rate of 5 °C/min was applied, spanning from −70 °C to 140 °C to encompass the five operational conditions listed in Table 2, and to extend beyond the material’s melting temperature (T_m). Strain levels were set at 0.005% for 1-dir specimens and 0.01% for 2-dir specimens, as determined from prior strain sweep results. Prior to testing, a 15 min dwell at the starting temperature ensured thermal equilibrium across all specimens.

2.5.3. Frequency Sweep Test

Frequency sweep tests were carried out on specimens oriented in both the 1-dir and 2-dir to assess the material’s response under dynamic loading conditions relevant to civil and mechanical structures, where operating frequencies can span from a few hertz to several hundred hertz [35]. The tests were conducted over a frequency range of 0.1 Hz to 150 Hz, constrained by the instrument’s capabilities. A 15-min thermal soak was applied prior to testing to ensure uniform specimen temperature. Strain levels of 0.005% and 0.01% were used for the 1-dir and 2-dir specimens, respectively, based on results from the strain sweep tests. Frequency sweeps were performed at −70 °C, −25 °C, 0 °C, 30 °C, and 60 °C.

2.5.4. Stress Relaxation Test

Stress relaxation tests were conducted on specimens oriented in 2-dir. Specimens were thermally equilibrated with a 15 min soak period to ensure uniform temperature throughout before testing. Each relaxation test was run for 30 min. The test conditions, including temperature and applied strain, are summarized in

Table 3. Temperature and strain % values used for stress relaxation tests.

Test	Temperature (°C)								Strain %
Stress Relaxation	−70	−50	−25	0	15	30	45	60	0.01

. The strain value was selected based on equipment limitations, particularly due to the high stiffness of the GF/HDPE material at lower temperatures, while still allowing for a valid comparison between the two material systems.

A master curve was generated using the time–temperature superposition (TTS) principle to estimate the long-term performance of the materials. The shift factors were determined using an Arrhenius model.

2.6. Manufacturing of Specimens

Nominal two-inch-wide tapes (50.8 mm) were manually laid up to fabricate panels measuring 355.6 mm × 355.6 mm (14 in × 14 in). For DMA testing, a [0₁₅] layup was used, while a [0₂₅] layup was selected for TMA testing, in accordance with the relevant ASTM standards. The panels were consolidated using a Monarch thermoforming press (model CMG30H-15CPX, Carver Inc., Wabash, IN, USA). Differential scanning calorimetry (DSC) results were used to establish a suitable processing temperature range. A series of trial runs were conducted to determine the optimal combination of temperature, dwell time, and pressure to produce high-quality consolidation. After each trial, interlayer bonding was visually inspected to confirm uniform consolidation throughout the panel. The final processing parameters for both material systems are summarized in

Table 4. Thermoforming press processing parameters for the HDPE material systems.

Material	Pressure, MPa, (psi)	Processing Temperature, °C (°F)	Dwell Time (min)
GF/HDPE	0.18 (26)	140 (284)	3
UHMWPE/HDPE	0.14 (20)	135 (275)	3

, with English units (psi and °F) reported as required by the Monarch press system.

2.7. Conditioning of Specimens

Prior to testing, all specimens were conditioned for a minimum of 40 h at 23.0 ± 2 °C and 50 ± 10% relative humidity, in accordance with Procedure A of the ASTM D618-21 standard [36].

2.8. Modeling the Viscoelastic Behavior

Polymer materials can exhibit either liquid-like or solid-like behavior at a given temperature, depending on the time scale or rate of deformation [37]. This behavior is known as viscoelasticity. The present work focuses on linear viscoelastic behavior, which applies to materials subjected to small deformations. Within this framework, concepts such as stress relaxation and the time–temperature superposition (TTS) principle are commonly used to describe how polymers respond to mechanical loading over time.

2.8.1. Time–Temperature Superposition (TTS) Principle

The time–temperature superposition (TTS) principle describes the relationship between time scale and temperature, enabling the prediction of long-term material behavior from short-term tests. In stress relaxation experiments, TTS allows for data collected at different temperatures to collapse into a single master curve at a chosen reference temperature, T_ref. This is accomplished by selecting one temperature curve as the reference and horizontally shifting the other curves along the time axis until their ends align, effectively superimposing them. Because density changes during these tests are typically minimal, corrections are generally unnecessary.

The amount each curve is shifted is quantified using a shift factor, which can be plotted as a function of temperature relative to T_ref. Various models exist for determining shift factors; in this work, the Arrhenius model was used, as all test temperatures were below the material’s melting temperature (T_m). The Arrhenius shift factor is defined in Equation (1).

$$ln(a_T)=\left(-\frac{E_a}{R}\right)\left(\frac{1}{T}-\frac{1}{T_{ref}}\right)$$

(1)

where E_a is the apparent activation energy, R is the universal gas constant, T is the absolute temperature, and T_ref is the reference temperature, with both temperatures measured in kelvins. For TTS to be valid, the material must exhibit linear viscoelastic behavior and be homogeneous, isotropic, and amorphous under the conditions tested. In this study, TTS was applied to specimens aligned in the 2-direction, where the mechanical response is assumed to be dominated by the polymer matrix, to estimate long-term performance.

2.8.2. Modeling Stress Relaxation

In a stress relaxation test, the material is subjected to a constant strain, and the resulting decrease in stress over time is measured. This behavior is quantified by the stress relaxation modulus, defined in Equation (2).

$$E_r(t)={\displaystyle \frac{\sigma (t)}{{\epsilon }_o}}$$

(2)

where E_r is the relaxation modulus, ε_o is the applied constant strain, and σ(t) is the stress as a function of time.

The stress relaxation behavior of polymer materials is both time- and temperature-dependent. Relaxation time—the duration required for stress within the material to significantly decay—varies with temperature. At low temperatures, relaxation occurs slowly and over longer periods, while at higher temperatures, it happens more rapidly due to increased molecular mobility.

Various viscoelastic models are available to describe stress relaxation under constant strain. In this study, the Prony series was used to model the relaxation behavior [38]. The Prony series expresses the stress response as a sum of exponential decay functions, as shown in Equation (3).

$$\sigma (t)=\sum _{i=1}^N{\sigma }_i{e}^{-\frac{t}{{\lambda }_i}}$$

(3)

where σ(t) is the stress at time t, σ_i represents the stress coefficients, and λ_i are the relaxation times, typically defined as the ratio of viscosity to storage modulus (μ/E). The experimental data were fit using Prony series models with three and five terms, referred to as Prony3 and Prony5, to evaluate the model’s accuracy and capture the relaxation behavior of the materials.

3. Results and Discussion

3.1. DSC Results

3.1.1. Raw Material

Figure 3 shows the heat flow as a function of temperature for the heat, cool, and heat cycles of the raw materials.

Table 5. DSC results of the first heating cycle.

Material (n **)	Tm-matrix (°C)	ΔHfusion (J/g)	Tm-fiber (°C)	ΔHfusion-fiber (J/g)
GF/HDPE (5)	126.4 ± 0.4	45 ± 5	-	-
UHMWPE/HDPE (4)	126.8 ± 0.2	83 ± 3	144.7 ± 0.3	110 ± 4

** n corresponds to the number of specimens used to average the results.

and

Table 6. DSC results of the cooling and the second heating cycle.

Material (n **)	Tcrystallization (°C)	ΔHcrystallization (J/g)	Tm-matrix (°C)	ΔHfusion (J/g)
GF/HDPE (5)	117.0 ± 0.1	52 ± 5	127.8 ± 0.2	48 ± 4
UHMWPE/HDPE (4)	116.3 ± 0.1	165 ± 5	128.3 ± 0.2	167 ± 4

** n corresponds to the number of specimens used to average the results.

show that the melting temperatures of the HDPE materials are within the same order of magnitude for both heating cycles. The UHMWPE melting temperature obtained in this study is consistent with previously reported values [39,40,41,42], supporting the accuracy of the thermal analysis method used. The step observed in the heat flow curve for UHMWPE/HDPE, with a mid-point value of 62 ± 1 °C, is attributed to the α-transition. The α-relaxation is associated with an inter-lamellar shear process [43]. The α-relaxation temperature for UHMWPE is wide, ranging from 30 to 120 °C, and depends on the crystallite thickness [44]. The α-transition corroborates with the loss modulus (E″) value from the temperature ramp DMA curve. Above 50 °C, a peak in the loss modulus is observed, aligning with the temperature range of the DSC heat flow curve. The heat of fusion (ΔHfusion) and the enthalpy of crystallization (ΔHcrystallization) of both material systems are different. The difference in magnitude can be attributed to the fiber systems’ influence on the heat of fusion [45]. Some studies have shown that T_m can decrease with increased fiber content [46]. This behavior is partly due to nucleation effects, fiber loading, and fiber morphology. Fiber surfaces provide nucleation sites where crystal growth can begin, decreasing the melting temperature [46]. Moreover, increased fiber loading can lead to more nucleation sites [45]. The shape, size, and surface properties have also been shown to influence the nucleation effects impacting the heat of fusion and melting temperature of semi-crystalline materials [47]. When comparing the ΔHfusion of both heating cycles, it can be observed that the ΔHfusion values of the second heating cycle are higher in magnitude for both materials. The increase in ΔHfusion results from the increase in crystallinity produced by the controlled cooling. The change in crystallinity is further discussed in Section 3.1.3.

Table 5 and Table 6 show the melting temperature and enthalpy for the heat–cool–heat cycle. The measured melting temperature of the HDPE composites aligns with reported values for similar material systems, supporting the reliability of the measured values [48,49]. However, the measured values are lower compared to those of pure HDPE, with values ranging from 130 °C and 135 °C [50,51,52]. Table 5 and Table 6 show that the melting temperatures of the HDPE materials are within the same order of magnitude for both heating cycles. The UHMWPE melting temperature obtained in this study is consistent with previously reported values [39,40,41,42], supporting the accuracy of the thermal analysis method used. The heat of fusion (ΔH_fusion) and the enthalpy of crystallization (ΔH_{crystallization}) of both material systems are different. The difference in magnitude can be attributed to the fiber systems’ influence on the heat of fusion [45]. Some studies have shown that T_m can decrease with increased fiber content [46]. This behavior is partly due to nucleation effects, fiber loading, and fiber morphology. Fiber surfaces provide nucleation sites where crystal growth can begin, decreasing the melting temperature [46]. Moreover, increased fiber loading can lead to more nucleation sites [45]. The shape, size, and surface properties have also been shown to influence the nucleation effects, impacting the heat of fusion and melting temperature of semi-crystalline materials [47]. When comparing the ΔH_fusion of both heating cycles, it can be observed that the ΔH_fusion values of the second heating cycle are higher in magnitude for both materials. The increase in ΔH_fusion results from the increase in crystallinity produced by the controlled cooling. The change in crystallinity is further discussed in Section 3.1.3.

Significant changes are observed when comparing the first and second heating cycles of the UHMWPE/HDPE material (Figure 3a and Figure 3c, respectively). The fiber and matrix materials’ melting temperatures (peaks) are present in the first heating cycle. For the second heating cycle, only one peak is present, which may correspond to the mixing of the matrix and fiber material. The first heating cycle provides the properties that result from the manufacturing process of the material (the “as received” raw material). Since the temperature of the first heating cycle surpasses the T_m of both materials, the microstructure may have changed due to the material reaching these higher temperatures [53]. At higher temperatures, the crystalline regions can partially melt and rearrange, impacting the overall material properties [53,54]. The high temperature reached in the first heating cycle could have changed the crystallinity of the UHMWPE fiber, shifting its peak value to the left (lower temperature). For this reason, temperature is an essential parameter if the properties of the fiber material are to be kept constant. As a result, the processing window for the UHMWPE/HDPE material is narrow to avoid changing the microstructure of the fiber material, as shown by the results. For this, a processing range of 125 °C to 145 °C is recommended to process the UHMWPE/HDPE material.

3.1.2. Consolidated Panels

The consolidated panels were subjected to a DSC test to study any change in microstructure resulting from the manufacturing process. Figure 4 shows the UHMWPE/HDPE DSC results for the raw and consolidated panels.

Table 7. DSC results of consolidated panels (first heating cycle).

Material (n **)	Tm-matrix (°C)	ΔHfusion (J/g)	Tm-fiber (°C)	ΔHfusion-fiber (J/g)
UHMWPE/HDPE (3)	128.5 ± 0.2	79 ± 1	144.9 ± 0.3	98 ± 3

** n corresponds to the number of specimens used to average the results.

and

Table 8. DSC results of consolidated panels (cooling and second heating cycle).

Material (n **)	Tcrystallization (°C)	ΔHcrystallization (J/g)	Tm-matrix (°C)	ΔHfusion (J/g)
UHMWPE/HDPE (3)	116.2 ± 0.1	159 ± 2	128.4 ± 0.2	166 ± 1

** n corresponds to the number of specimens used to average the results.

show the melting temperature and enthalpy of fusion and crystallization for the heat–cool–heat cycle of the consolidated panels, respectively.

The DSC results of the first heating cycle show that the heat of fusion is higher for the raw material. The decrease in value may be partly due to another thermal cycle on the material during the composite panel’s processing. Moreover, the α-transition is not observed for the consolidated panel, and may be a result of the control cooling during the consolidation process. For the UHMWPE/HDPE material, the heat of fusion is lower for the consolidated panel, with a percentage difference of 4% for the matrix and 12% for the fiber material.

The first heating cycle of the consolidated panel showed that the structure of the matrix and fiber matrix did not change, as shown in Figure 4a. Two peaks were observed for the consolidated panel, similar to the raw material, showing that the processing temperature selected did not significantly change the microstructure of the composite material. Like the raw material, exceeding the melting temperature changed the micro-structure, as observed by the heat flow data of the second heating cycle in Figure 4b.

3.1.3. Crystallinity Percentage Calculation

Equation (4) was used to calculate the crystallinity percentage of the HDPE matrixes using the first heating cycle, while Equation (5) was used to calculate the crystallinity percentage of the UHMWPE fiber material.

$${\mathsf{\chi }}_c={\displaystyle \frac{\Delta H_m}{\Delta H_{100}\left(1-w_f\right)}}$$

(4)

$${\mathsf{\chi }}_{c-fiber}={\displaystyle \frac{\Delta H_m}{\Delta H_{100}}}$$

(5)

where ${\mathsf{\chi }}_c$ is the degree of crystallinity, $\Delta H_m$ is the melting enthalpy of the material, $\Delta H_{100}$ is the enthalpy of melting for 100% crystallinity, and $w_f$ is the fiber weight percentage. The $\Delta H_{100}$ value for HDPE and UHMWPE is 293 J/g [50,55] and 291 J/g [56], respectively. The fiber volume percentage shown in Table 1 was converted to a fiber weight percentage (wt.%) using the density of the fiber and polymer matrix materials. A density of 2.57 g/cm³ [57,58], 0.97 g/cm³ [59], and 0.95 g/cm³ [60] was used for the glass fiber, UHMWPE fiber, and HDPE matrix material.

Table 9. The degree of crystallinity for all the material systems was determined using the DSC results from the first heating cycle.

Material	GF/HDPE	UHMWPE/HDPE
Fiber wt.%	69.7	50.2
${\mathsf{\chi }}_{c-HDPE}$ % (raw material)	50	57
${\mathsf{\chi }}_{c-UHMWPE/fiber}$ % (raw material)	-	38
${\mathsf{\chi }}_{c-HDPE}$ % (consolidated panel)	59	54
${\mathsf{\chi }}_{c-UHMWPE/fiber}$ % (consolidated panel)	-	34

shows the calculated fiber wt.% and the degree of crystallinity for all the materials and fiber systems. Table 9 shows an increase in crystallinity for the GF/HDPE and a decrease for UHMWPE/HDPE material when comparing the consolidated and raw material results. The crystallinity percentage for the UHMWPE/HDPE matrix and fibers remained within the same order of magnitude for both raw materials and the consolidated panel. The increase in crystallinity of the GF/HDPE may be due to the controlled cooling process during the consolidation of the panels. It has been shown for other semi-crystalline polymers that the crystallization temperature increases with decreasing cooling rates [61,62].

3.2. TMA Results

A temperature ramp was performed in all principal directions to calculate the dimension change as a function of temperature. Figure 5 shows the TMA data for the GF/HDPE and UHMWPE/HDPE materials. The coefficient of thermal expansion (CTE) was computed using a linear regression from −50 °C to 50 °C (below T_m).

Table 10. TMA results of HDPE materials in all principal directions.

Material (n **)	CTE 1-dir (με/C)	CTE 2-dir (με/C)	CTE 3-dir (με/C)
GF/HDPE (3)	10 ± 1	83 ± 5	88 ± 6
UHMWPE/HDPE (3)	−2 ± 9 (6)	149 ± 9 (4)	156 ± 9

** n corresponds to the number of specimens used to average the results.

shows the average CTE and standard deviation for all principal directions and material systems. In principle, all fibers should be oriented in the 1-dir. However, due to flow-induced alignment during manufacturing, the fibers can shift, causing the orientation of fibers within the composite panel [63]. The average CTE in the 2-dir and 3-dir are within the same magnitude for both material systems. The slight difference between the 2-dir and 3-dir can be attributed to the flow-induced alignment resulting from the manufacturing process and the variability of properties within the composite panels.

For the GF/HDPE material, the CTE in the 1-direction corresponds to that of the glass fiber [64]. The CTE in the 2-dir and 3-dir corresponds to that of the HDPE matrix [65,66]. For UHMWPE/HDPE material, the CTE in the 1-direction corresponds to that of the UHMWPE fiber material [67,68]. The CTE in the 2-dir and 3-dir results from a combination of the CTE of the HDPE [66] matrix material and the UHMWPE [69] fiber material, as suggested by the increase in value when compared to the GF/HDPE material. The experimental results in the 2-dir and 3-dir were compared to the values calculated using the rule of mixture, assessing the agreement of results.

3.3. DMA Results

3.3.1. Strain Sweep

A three-point bending strain sweep test was performed on both materials to identify the linear viscoelastic region. Due to the equipment’s limitations and the stiffness of the specimens in the 1-direction, they could not deform above a 0.01 strain percentage. The storage modulus was used as a reference for selecting the linear viscoelastic region. Figure 6 and Figure 7 show the strain sweep results for the GF/HDPE and the UHMWPE/HDPE materials, respectively. The influence of fiber orientation on the material’s storage modulus is evident in both figures. Here, in the 1-dir, the storage modulus is significantly higher than in the 2-dir. For specimens in the 2-dir, the storage modulus is primarily of the matrix material, while for specimens in the 1-dir, it is represented by the fiber material.

Figure 6 shows the storage modulus in the 1-direction as a function of strain for both materials and three temperatures. For all three temperature values, the modulus increases with oscillation strain. When comparing the temperatures, the storage modulus decreases with an increase in temperature. At the same time, an increase in frequency results in an increase in storage modulus, as observed for a temperature of −50 °C. Figure 7 shows the storage modulus in the 2-direction as a function of strain percentage. The storage modulus behaves mostly linearly over the strain range. It can be observed that temperature influences the storage modulus. The storage modulus increases as temperature decreases, becoming stiffer. This behavior reflects the typical thermomechanical response of polymers, where stiffness decreases with increasing temperature. Using the storage modulus as the reference property, the strain percentage selected is 0.01% and 0.005% for the 2-dir and 1-dir, respectively. These values were the input parameters for the temperature ramp, frequency sweep, and stress relaxation test.

3.3.2. Temperature Ramp Test

A three-point bending temperature ramp test was performed to measure the temperature-dependent modulus. The flexural modulus is used as a proxy to compare material properties to tensile properties [70]. Figure 8 shows the temperature ramp test result for both material systems. The temperature ramp was performed from −70 °C to 140 °C. The storage modulus decreases with temperature. The increase in modulus as temperature decreases can be attributed to the increase in stiffness and brittleness due to the extremely low temperatures, as shown for other polymer materials [11]. The modulus steadily decreased up to the T_m. After T_m, the modulus decreases significantly due to the phase change of the matrix material going from a solid elastic to a viscous state. At ambient temperatures (15–25 °C), the storage modulus values from both the strain sweep and temperature ramp tests show close agreement between the 1-direction and 2-direction and fall within the expected range based on manufacturer-provided and previously reported values [52,71]. Moreover, the storage modulus was predicted using composite analysis, which yielded similar results for both specimen orientations. The storage modulus also increases with input frequency value, as seen by others for other polymer systems [13,22]. The results presented in this section show the input frequency and temperature dependency on the storage modulus of polymer materials—specifically, the two material systems studied in this work.

3.3.3. Frequency Sweep Test

A three-point bending frequency sweep test was performed on all material systems to measure the HDPE material systems’ frequency- and temperature-dependent properties. A strain percentage of 0.005% and 0.01% was used for specimens in the 1-dir and 2-dir, respectively. Figure 9, Figure 10 and Figure 11 and Figure 12, Figure 13 and Figure 14 shows the storage modulus, loss modulus, and tanδ as a function of the frequency for GF/HDPE and UHMWPE/HDPE materials, respectively. Similarly to the temperature ramp results, the storage modulus increases with a decrease in temperature. The storage modulus is observed to increase with input frequency for all of the temperatures and directions tested, as seen for pure HDPE material [72] and other composite systems [22,73,74]. These results align with expectations. As the input frequency increases, the material has less time to relax into its initial state. The frequency value increases, reducing the loss modulus’s response, as it cannot keep pace with the applied loading. Consequently, the storage modulus dominates the material’s behavior, increasing stiffness with higher-frequency loading. The measured storage modulus magnitude of the frequency sweep test aligns with the strain sweep and temperature ramp results.

When comparing the storage modulus in the 1-dir of both material systems, it is apparent that the effect of temperature is higher for the UHMWPE/HDPE material. For the GF/HDPE material, the stiffness of the glass fibers may influence the material’s response. In contrast, the material response for specimens in the 2-dir is dominated by the HDPE matrix, showing the effect of temperature on stiffness. The loss modulus and tanδ results show distinct peaks at multiple frequency values for both materials. These peaks can be attributed to the natural frequencies of the HDPE matrix material. These are frequency values at which the material system tends to vibrate without an input-damping force, leading to excessive vibrations and potential damage to components. Similar tests have been performed by others, indicating their value. Edinne et al. [75] conducted a modal analysis of a 20% fiber-filled HDPE, measuring five frequency modes. Bharath et al. [76,77] studied the mechanical behavior of HDPE foams, measuring a natural frequency mode. Xing et al. [78] analyzed the vibration transmission properties of an expanded polyethylene (EPE) polymer, yielding a resonance frequency. Natural-frequency calculations show predictions for the first modes of vibrations ranging from 34 Hz to 72 Hz and 12 Hz to 38 Hz for GF/HDPE 1-dir and 2-dir, respectively, as temperature increases from −70 °C to 110 °C [79]. In the same way, first-mode natural frequencies are predicted to be 36 Hz to 84 Hz and 12 Hz to 26 Hz for UHMWPE/HDPE 1-dir and 2-dir, respectively. These frequency modes are within the range where peaks are observed in the data and trend as predicted, reassuring that these are, in fact, the natural-frequency modes of the GF/HDPE and UHMWPE/HDPE materials.

Figure 11. Tanδ as a function of frequency for GF/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 11. Tanδ as a function of frequency for GF/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

The results from the frequency sweep tests provide valuable insights into material behavior over a range of frequencies. Evaluating the material’s response under these conditions enhances our understanding of the material’s performance in varying environmental scenarios. As input frequency increases, properties such as stiffness and damping typically increase. Excessive vibrations—caused by factors like foot traffic, equipment operation, or high winds—can negatively impact component integrity and functional performance.

Figure 12. Storage modulus as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 12. Storage modulus as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 13. Loss modulus as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 13. Loss modulus as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 14. Tanδ as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

Figure 14. Tanδ as a function of frequency for UHMWPE/HDPE in (a) 1-dir and (b) 2-dir. Each line style represents one specimen tested.

3.3.4. Stress Relaxation Test

A three-point bending stress relaxation test was performed on both material systems at a strain percentage value of 0.01%. The parameters used for testing are shown in Table 3. Figure 15 shows stress as a function of time for both material systems, while Figure 16 shows the shift factor as a function of temperature. Results for the stress relaxation test are for specimens in the 2-direction only. Three specimens were tested per temperature value, and outliers were removed from the master curve dataset.

A master curve at a nominal reference temperature of 30 °C was created using the TTS principle. The activation energy, T_ref, and R² values of the Arrhenius model are shown in

Table 11. Arrhenius fitting parameters.

Material	Strain %	Activation Energy, Ea (kJ/mol)	Reference Temperature, Tref (°C)	R2
GF/HDPE	0.01	95.96	30.55	0.50
UHMWPE/HDPE	0.01	83.24	30.35	0.62

.

The stress relaxation test results appear noisy, in part due to the low strain values selected for the test. Observing the stress development over time, the stress decreases as time increases. This behavior is apparent for both materials as the stress relaxes over time. This behavior is expected, and aligns with similar material systems [80,81]. The stress relaxation is partially caused by the rearrangement of the entangled polymer chains, where the polymer network reorganizes to minimize the elastic energy [80]. High temperatures lead to shorter relaxation times, whereas low temperatures result in longer relaxation times. This behavior results from the free volume between the polymer molecules, which is reduced at low temperatures, restricting their movement [37]. When analyzing the results, it is essential to note that multiple plates were manufactured for testing purposes. Specimens were selected randomly throughout the plates. During testing, variability was observed in the test results during the shifting of the curves and the creation of the master curve. Variability in stress relaxation test results can be attributed in part to (1) the material properties being affected by variability in processing parameters, (2) flow-induced fiber alignment during the manufacturing process, (3) experimental error during the specimen setup, and (4) variability in properties due to the specimen’s location within the manufactured panel. These factors can influence the test results, as well as how the Arrhenius equation fits the experimental data.

The activation energy is higher for the GF/HDPE material. A higher activation energy lowers the reaction rate, as more energy is needed for this reaction to occur. The R² value is higher for the UHMWPE/HDPE material, suggesting that the Arrhenius regression model fits better than the GF/HDPE material. Prony series were used to predict the material’s stress relaxation response.

Table 12. Stress relaxation test: Prony series R² values.

Material	R2-Value Prony3	R2-Value Prony5
GF/HDPE	0.99	0.99
UHMWPE/HDPE	0.97	0.98

shows the R² values for the Prony series using three and five terms. The Prony series parameters are shown in Appendix A (

Table A1. Prony series parameters using three terms.

Parameters	GF/HDPE	UHMWPE/HDPE
${\sigma }_1$	0.0809	0.1143
${\sigma }_2$	0.3046	0.0470
${\sigma }_3$	0.0473	0.0903
${\lambda }_1$	0.0099	2.58 × 10−4
${\lambda }_2$	8.22 × 10−7	3.02 × 105
${\lambda }_3$	8.3309	11.5157

and

Table A2. Prony series parameters using five terms.

Parameters	GF/HDPE	UHMWPE/HDPE
${\sigma }_1$	0.2713	0.0728
${\sigma }_2$	2.57 × 10−7	0.0491
${\sigma }_3$	0.0078	0.0473
${\sigma }_4$	0.0543	0.0553
${\sigma }_5$	0.0401	0.0476
${\lambda }_1$	1.72 × 10−6	9.18 × 10−6
${\lambda }_2$	0.0389	4.36 × 10−4
${\lambda }_3$	0.0011	4.66 × 104
${\lambda }_4$	0.0279	24.1081
${\lambda }_5$	2.7118	0.0785

). The Prony series can predict up to 99% of the material’s response, depending on the number of terms used. The more terms used, the better the prediction, as the model can better capture the complex material response.

The stress relaxation test results can significantly aid in predicting residual stresses on composite material systems. An application would be the prediction of residual stresses due to fixed strains, such as those found in bolted joints and connections. The higher torque of a bolt on a composite part may cause the material to take longer to relax when exposed to extreme conditions. The models used to fit the experimental data can help designers and modelers predict material performance during the manufacturing and deployment of composite structures. For example, designers can predict and mitigate potential issues, such as warping, by understanding how a material relaxes in response to strains caused by thermal events or applied stresses. The predictive capabilities can help ensure that a composite part maintains its structural integrity and functionality over the intended use.

3.4. Comparative Analysis

A comparative analysis of the material properties reveals key differences between the GF/HDPE and UHMWPE/HDPE composites, as shown in

Table 13. Property comparison between the GF/HDPE and UHMWPE/HDPE materials.

Property	GF/HDPE			UHMWPE/HDPE
	1-dir	2-dir	3-dir	1-dir	2-dir	3-dir
Melting Temperature, Tm (°C)	127.8 ± 0.2			128.3 ± 0.2
Processing Temperature, Tprocessing (°C)	Tprocessing > 127.8 ± 0.2			128.3 ± 0.2 < Tprocessing < 144.7 ± 0.3
CTE (με/°C)	10 ± 1	83 ± 5	88 ± 6	−2 ± 9	149 ± 9	156 ± 9
Storage Modulus, E′ (GPa) @ −57 °C	36 ± 3	8.7 ± 0.3	-	40 ± 2	3.9 ± 0.5	-
Storage Modulus, E′ (GPa) @ −37 °C	35 ± 2	8.2 ± 0.3	-	39 ± 2	3.8 ± 0.3	-
Storage Modulus, E′ (GPa) @ −6 °C	34 ± 2	7.1 ± 0.3	-	36 ± 1	3.5 ± 0.2	-
Storage Modulus, E′ (GPa) @ 24 °C	32 ± 2	5.4 ± 0.1	-	31 ± 1	2.9 ± 0.2	-
Storage Modulus, E′ (GPa) @ 49 °C	28 ± 2	3.6 ± 0.1	-	25 ± 1	2.3 ± 0.2	-

. Both materials exhibit similar melting temperatures. However, the UHMWPE/HDPE material requires a narrower processing temperature, indicating more precise processing conditions. In terms of dimensional stability, the GF/HDPE composite exhibits significantly lower coefficients of thermal expansion (CTE) in all directions, particularly in the 1-direction, which is beneficial for minimizing thermal distortion, such as warpage and residual stresses. The UHMWPE/HDPE exhibits higher CTE in the transverse directions, potentially limiting its suitability in applications requiring isotropic thermal behavior.

Dynamic mechanical analysis shows that the UHMWPE/HDPE composite consistently maintains a higher storage modulus in the 1-direction across most of the temperature range, with values as high as 40.3 ± 1.7 GPa at –57 °C, compared to 36.0 ± 2.8 GPa for GF/HDPE. The higher modulus might be attributed to the effect of the fiber’s properties on the overall composite behavior, where the UHMWPE fiber becomes stiffer at lower temperatures. In contrast, GF/HDPE has a higher modulus at 49 °C, as the glass fiber resists thermal softening due to its higher softening temperature [82]. The superior stiffness of UHMWPE/HDPE at extremely low temperatures and ambient conditions suggests its better load-bearing capacity under these conditions. The GF/HDPE system outperforms UHMWPE/HDPE in the transverse direction (2-dir), showing higher stiffness and reduced modulus loss with temperature increase, indicating better dimensional stability across directions. The rate of modulus decay with temperature differs notably between the two systems. The GF/HDPE composite exhibits a more gradual decline in modulus with temperature, particularly in the 2-direction, suggesting better retention of stiffness and dimensional stability at elevated temperatures. In contrast, the UHMWPE/HDPE shows a steeper reduction, indicating greater thermal sensitivity.

Together, these results indicate that UHMWPE/HDPE offers enhanced unidirectional stiffness and low temperature performance, making it well-suited for load-dominant applications with directional reinforcement. In contrast, GF/HDPE exhibits superior transverse stiffness, lower thermal expansion, and lower modulus decay with temperature, which may be advantageous in applications where dimensional control across multiple axes is crucial. This behavior has important implications for performance under thermal cycling, where materials with lower modulus gradients are preferred for maintaining dimensional and mechanical integrity across varying temperatures.

4. Conclusions

The thermal, thermomechanical, and viscoelastic properties of two HDPE composite material systems were successfully characterized under conditions representative of extreme environments. The measured values fall within the expected range for these material systems, supporting the reliability of the testing methods employed. DSC analysis revealed that exceeding the melting temperature of the fiber component in the UHMWPE/HDPE system results in a microstructural change, indicating a narrow processing temperature window. The coefficient of thermal expansion was quantified along three principal directions, demonstrating the significant influence of the fiber’s architecture on the material’s thermal response.

The linear viscoelastic region was identified across the relevant temperature range, allowing for the characterization of the modulus as a function of both temperature and frequency. Frequency sweep results reveal distinct natural frequency modes in both systems. Stress relaxation behavior was evaluated, and master curves were generated for a strain input of 0.01%. A five-term Prony series provided a strong fit (R² = 0.99), effectively capturing the complex time-dependent response. Shift factor analysis indicated potential heterogeneity or anisotropy in the 2-direction, likely attributable to the material system’s characteristics and manufacturing-induced variations.

The TMA and DMA results confirm the expected thermal expansion and viscoelastic behavior of the tested HDPE-based composite materials. Importantly, the relatively low thermal expansion coefficients and retention of storage modulus at sub-ambient temperatures suggest that both materials maintain structural integrity under fluctuating thermal conditions. However, the modulus drop observed for UHMWPE/HDPE at higher temperatures may pose limitations in applications involving elevated thermal cycling, as well as a change in microstructure when exposed to higher temperatures for prolonged periods. Moreover, higher CTE values for the UHMWPE/HDPE can lead to higher thermal and residual stresses during temperature changes. Overall, these findings suggest that, while both materials demonstrate promising stability in extreme environments, the GF/HDPE material system may offer enhanced dimensional and mechanical stability under more severe thermal loads. Overall, these findings provide key insights into the performance of continuous fiber-reinforced HDPE systems under extreme conditions, supporting the development of material characterization protocols essential for engineering applications in harsh environments.