Rheological Properties of Functionalized Smart Resins for Transport Applications ^†

Abstract

Hydrogen is a promising alternative to fossil fuels, but its efficient storage presents significant challenges. Polymer composite vessels, especially those made from carbon fiber-reinforced plastic (CFRP), are gaining attention, due to their high strength-to-weight ratio for storing compressed or cryogenic hydrogen. The latest Type V tanks, which lack internal liners, rely solely on fiber composites for both structural integrity and gas containment, enhancing the storage volume-to-weight ratio and supporting recycling. However, this linerless design faces the challenge of preventing gas permeation. Epoxy resins, widely used in aerospace carbon fiber-reinforced composites (CFRCs), offer excellent processability and load-bearing capabilities. The addition of high-aspect-ratio nanofillers can enhance the gas barrier properties, which are essential for preventing hydrogen leakage, while also improving the mechanical, electrical, and thermal properties of the nanocomposites. This study focuses on epoxy-based composites with expanded graphite, aiming to optimize their physical properties and processing for Type V tanks, using a rheological framework to evaluate their processability and multifunctionality in transport applications.

1. Introduction

Hydrogen is recognized as a clean energy carrier, with promising applications in aviation, where it could replace kerosene and fuel cell electric vehicles (FCEVs) [1]. However, effective hydrogen storage remains a key challenge, especially for transportation. While hydrogen has a high gravimetric energy density (120 MJ/kg), its low volumetric density (0.01 MJ/L) requires large storage tanks, impractical for mobile uses [2]. Liquid hydrogen offers a higher density, but requires cryogenic storage at −253 °C. Alternatively, compressed hydrogen storage in high-pressure vessels (350–700 bar) is viable, but poses safety risks, particularly in aerospace applications. Traditional metal-based tanks (Type I) are being replaced by composite overwrapped pressure vessels (COPVs), including Types II, III, and IV, which use carbon fiber with metal or polymer liners. The latest Type V tanks, which are liner-free, reduce the container weight by 10–20% and offer recycling advantages [3]. However, without a liner, the composite must function as a permeation barrier, making it challenging to widely adopt Type V vessels in the commercial market. This technology remains an active area of research due to these technical barriers [4]. An intermediate solution is cryocompressed storage, which combines high pressure (50–700 bar) and low temperatures (25–110 K) to maximize storage density, allowing for more hydrogen per volume. Cryocompressed systems (CcH2) reduce boil-off in liquid hydrogen storage and increase gravimetric density. Lightweight composites, which offer higher strength and a lower density than stainless steel, can cut tank weight by 25–30% and confer smart properties [5,6,7,8]. The ideal materials for these tanks need high strength, fracture toughness, stiffness, and low hydrogen permeability, with carbon fiber-reinforced composites (CFRCs) commonly used due to their tensile strength and stiffness. Epoxy, phenolic, polyester, and vinyl ester resins are often paired with carbon fibers, with an added inner liner, to reduce hydrogen ingress [9]. High-density polyethylene (HDPE) is widely favored for tank liners due to its performance in gas storage applications, while fillers like fibers or nanoparticles further lower permeation by creating a tortuous diffusion path [4]. For manufacturing, understanding curing kinetics and the viscosity of selected formulations is essential, and filament winding is a preferred technique for producing CFRP containers. Carbon fiber is tension-wrapped around a mold, soaked in resin, and cured in a rotary furnace. Resin viscosity is a critical factor in determining the suitability of materials for such a process [10]. In the field of aerospace, there is a growing interest in using graphene or carbon nanotube-reinforced CFRP vessels, as these provide better hydrogen permeation resistance, an enhanced mechanical performance, and reduced weight [11]. This study aims to develop epoxy-based composites filled with expanded graphite for Type V hydrogen storage vessels. By characterizing these materials, the study evaluates their processability and multifunctional potential, using a rheological framework to assess their application in hydrogen storage systems.

2. Materials and Methods

An epoxy resin system made from Diglycidyl ether Bisphenol A (DGEBA) and 4,4-diaminodiphenyl sulfone (DDS) was selected because of its high-performing behavior. Expanded graphite (EG1045) was loaded as a nanofiller to decrease the composite permeability of the hydrogen. First, the precursor (DGEBA) and EG were mixed with a centrifugal planetary mixer (THINKY, Tokyo, Japan). Then, the hardener (DDS) was dissolved in the resin using an oil bath at 120 °C for 1 h 15 min, as shown in Figure 1a. After a degassing step (2 h at 80 °C), the mixture was cured into a mold at 150 °C for 1 h and 220 °C for 3 h (see Figure 1b).

A differential scanning calorimeter (Mettler DSC 822, Zurich, Switzerland) was used for the dynamic and isothermal cure experiments. Dynamic DSC scans were performed from −50 to 350 °C at 10 °C min⁻¹ (nitrogen flow of 20 mL min⁻¹). The isothermal experiments were carried out at temperatures ranging from 160 °C to 220 °C for 100 min. After each isothermal scan, the sample was rapidly cooled to 30 °C and then reheated at 10 °C min⁻¹ to 350 °C to determine the residual heat of the reaction (ΔH_residue). The total heat generated during the curing process (ΔH_tot) is given by the sum of the heat released during the isothermal step (ΔH_i) and the residual heat of the reaction (ΔH_residue). The degree of cure α is described in Equation (1).

$$\alpha \left(T,t\right)={\displaystyle \frac{1}{{\mathsf{\Delta }H}_{tot}}}{\int }_0^t\left({\displaystyle \frac{dH}{dt}}\right)dt={\displaystyle \frac{{\mathsf{\Delta }H}_i\left(t\right)}{{\mathsf{\Delta }H}_{tot}}}$$

(1)

The maximum achievable curing degree (α_max) was calculated via dynamic DSC tests on the uncured and fully cured samples using the following equation:

$${\alpha }_{max}=\left(1-{\displaystyle \frac{{\mathsf{\Delta }H}_{cured}}{{\mathsf{\Delta }H}_{uncured}}}\right)\ast 100$$

(2)

The viscoelastic properties have been analyzed using a strain-controlled rotational rheometer (ARES-TA instruments, New Castle, DE, USA) with a plate–plate geometry on the fluid complete formulation. Figure 2 shows the customized caved-in bottom plate (cavity thickness 1 mm, diameter 10 mm) used to avoid spillage.

Dynamic temperature scans were performed with a 5 °C/min heating rate and 1 rad/s frequency from 50 °C to 200–220 °C. The mechanical properties of the samples were measured with a dynamic mechanical thermo-analyzer (PerkinElmer-DMA 8000, Waltham, MA, USA). A variable flexural deformation in dual cantilever mode was applied from −50 °C to 300 °C on solid bars with dimensions of 2 mm × 10 mm × 35 mm (0.1 mm displacement amplitude, 1 Hz frequency, and 1 °C/min scanning rate). Micrographs of the epoxy nanocomposites were obtained using Scanning Electron Microscopy (SEM-Phanton, Waltham, MA, USA). All the samples were positioned on a carbon tab previously attached to an aluminum stub (Agar Scientific, Stansted, UK) and covered with a 250-thick gold film using a sputter coater (Agar mod. 108 A, Stansted, UK). Nanofilled sample sections were cut from solid samples and were etched before observation using SEM. Thermal conductivity analyses were performed using two thermal flow sensors with an integrated temperature sensor (Hukseflux FHF01, Delft, The Netherland). The analysis was carried out at an average temperature of 60 °C on disk-shaped samples (5 mm diameter, 1 mm thickness) placed between two heated plates with a temperature difference of 20 °C. Fick’s law was applied for the measurement of conductivity:

$$J=-k{\displaystyle \frac{dT}{dx}}$$

(3)

J is the heat flow measured by the sensor and x is the direction of the thickness. A program written in LABVIEW^® allows for the acquisition of temperature and flow signals and the calculation of thermal conductivity k.

3. Results

3.1. Thermal Characterization

The DSC tests provide data regarding the curing and glass transition temperatures. Figure 3 shows the thermograms of the unfilled epoxy formulation before and after the curing treatment. The uncured fluid formulation manifests in a Tg of around 4 °C and begins to react at temperatures higher than 120 °C. The graph clearly shows an onset of the exothermic peaks at ~150 °C, while the end temperature of the reaction is around 315 °C. After the curing treatment, as expected, the thermogram no longer shows the exotherm peak.

As shown in

Table 1. Maximum curing degree for different filler amounts evaluated from DSC.

Filler Amount (%)	αmax (%)
0	100.0
1	98.4
3	97.9
7	95.8
9	94.5

, the samples solidified with the same curing cycle (at 150 °C for 1 h and 220 °C for 3 h) display a reduction in the maximum achievable degree of cure as the expanded graphite amount increases, but a value lower than 94.5% (in the case of 9% EG-filled samples) is not observed. This suggests that the curing protocol adopted is effective for such resin systems.

The isothermal DSC curves show an initial peak and a sudden decay; the same trend is found in the curing rate dα/dt (Figure 4a). The curing kinetics of the cross-linking reactions have been analyzed by applying a modified Kamal’s kinetic model [12]. It was used to describe the curing kinetics of the resin system by considering diffusion’s effect. The model applied is displayed in the following equation:

$${\displaystyle \frac{d\alpha }{dt}}=\left(K_1+K_2{\alpha }^m\right){\left(1-\alpha \right)}^n{f}_d\left(\alpha \right)$$

(4)

where $K_1$ and $K_2$ are kinetic constants depending on temperature, according to an Arrhenius function, $m$ and $n$ are reaction orders, and $f_d$ is the diffusion factor. The latter is expressed in Equation (5), with $C$ and ${\alpha }_c$ as the fitting parameters.

$$f_d\left(\alpha \right)={\displaystyle \frac{1}{1+\exp \left[C\left(\alpha -{\alpha }_c\right)\right]}}$$

(5)

The parameters were adjusted to achieve a good description of the curing kinetics. Figure 4b compares the curing rates at a given temperature (isotherm at 200 °C) between the resin systems unloaded and loaded with 7%_wt EG. The reaction is faster throughout the reaction process in the case of the unloaded system.

3.2. Rheological Characterization

Understanding material viscosity and its behavior with temperature is crucial for manufacturing. In the case of the filament winding process, 0.2–2 Pa·s is the acceptable range. Figure 5 displays the evolution of viscosity with temperature for resin systems loaded with 0 and 7%_wt EG.

The curves can be divided into three main parts. At lower temperatures, viscosity decreases with temperature. With increasing temperature, the curing reaction starts to compensate for the effect of temperature, and viscosity approaches a plateau. After a certain temperature, viscosity increases exponentially, with the curing reaction being the only controlling effect. The filler addition increases system viscosity and anticipates the curing reaction onset.

3.3. Post-Cure Characterization

The mechanical properties were evaluated through DMA tests (see Figure 6). At low temperatures, from −50 °C to 150 °C, the unfilled resin’s modulus ranges from 3.3 to 1.6 GPa. These values are slightly higher for the filled one. The glass transition temperature for such a material can be detected from the tanδ peak, at around 240 °C for both samples.

Among the smart properties, thermal conductivity was also investigated for the different filler percentages. As shown in Figure 7, thermal conductivity significantly increases as the filler amount increases. This could be due to the high filler dispersion even for higher filler amounts, as demonstrated by the SEM micrographs in Figure 7 Inset.

4. Conclusions

This study aimed to characterize an epoxy resin system filled with expanded graphite to develop polymer composites for lightweight hydrogen storage vessels (Type V). The primary focus was on enhancing the physical properties of epoxy-based composites to assess their suitability for hydrogen storage applications. Thermal analyses were carried out to investigate the curing reaction and kinetics, while rheological assessments across varying temperatures demonstrated the material’s adaptability for manufacturing processes. The final composite exhibited promising mechanical and smart properties, underscoring its potential for hydrogen storage. Key challenges, such as reducing hydrogen permeability and addressing thermomechanical fatigue, remain critical in optimizing performance. Future research will prioritize a detailed evaluation of these polymer composites’ mechanical and gas barrier properties under the typical conditions of hydrogen storage in transport applications.