Circular Slab Track—Structural Analysis of Adapting Composite Materials to Ballastless Track Systems

Abstract

Rail transport is widely regarded as an efficient and environmentally sustainable mode of mobility, although lifecycle emissions from infrastructure can diminish its ecological benefits. This study assesses a novel slab track system design that replaces conventional concrete components with recycled polymeric composite sleepers, supporting circular economy objectives. Analytical calculations (per EN 16432-2 and EN 13230-6) and finite element analysis (FEA) were conducted on a 2.6 m polymeric composite sleeper model under static vertical loading. The results demonstrate that bonded base layers comprising asphalt and hydraulically bound materials reduce bending stresses within the sleeper to 1.307 N/mm², substantially below the 5.50 N/mm² observed without bound layers and well below both characteristic fatigue limits. Laboratory validation via strain-gauge measurements corroborates the numerical model. Despite minor torsional effects from first-batch production, the polymeric composite sleeper design is structurally viable for slab track applications. The methodology is directly transferable to alternative composite designs, allowing material-based adaptation of mechanical performance. These findings support the use of recycled polymeric composite sleepers in slab track systems, combining structural adequacy with enhanced circularity. Further research can base itself on the findings and should incorporate long-term durability testing.

1. Introduction

Rail transport is widely recognized as one of the most efficient and environmentally sustainable modes of transport, primarily due to the low rolling resistance between steel wheels and steel rails, which results in significantly reduced energy requirements compared to road transport. However, while much focus has been placed on the operational efficiency of rail vehicles, the environmental impacts arising from the construction, maintenance, and upgrading of rail infrastructure are often overlooked. These impacts are becoming increasingly relevant, particularly in light of ambitious climate neutrality goals set by the European Union and national governments.

Numerous advantages offered by rail transport include high energy efficiency, reduced environmental impact, increased safety, the ability to move large volumes of passengers and freight, etc. As a consequence, this mode of transportation has become a focus of numerous researchers in recent years. A noticeable interest surrounding rail transport is primarily driven by global efforts to promote sustainable mobility and smart infrastructure. It is not surprising that a diverse range of research topics has emerged, attracting attention from transportation engineers, urban planners, environmental scientists, and policy makers. Hence, a wide range of compelling topics within the realm of rail transport have been explored, including efficient energy use in railways [1], behaviour of the railway substructure under dynamical loading [2], structural solutions for reduction of vibrations and emitted noise with FE analyses [3,4], investigation of influences caused by geometric irregularities in rails [5,6], reliability, and predictive maintenance [7].

Recent studies have shown that the infrastructure component of rail systems can contribute disproportionately to total lifecycle emissions when compared to other transport modes [8]. One key factor is the need for low gradient alignments (≤12.5‰), which necessitate the construction of resource-intensive civil engineering structures such as bridges and tunnels. Additionally, the materials used in track systems, particularly concrete in conventional slab track designs, are associated with high embodied emissions and with causing higher emissions in the production and construction phase of slab track systems [9,10]. This is particularly concerning as the environmental advantage of rail may erode if infrastructure emissions are not addressed. In response, this paper presents the development and technical evaluation of an innovative slab track system designed according to circular economy principles.

The initial concept was to replace the entire slab track plate with recycled polymeric composite materials. However, at present there are no extensive production facilities for such large, pressed components, and therefore a superimposed system was further pursued.

The proposed system replaces conventional concrete components with polymer-based composite elements made at least partly from recycled plastics. The system utilizes recycled polymer composite sleepers, which are already approved by Deutsche Bahn AG for ballasted tracks, now repurposed and integrated into a fully bonded innovative slab-track configuration on asphalt (see Figure 1). Requirements for testing (e.g., dynamically) of polymeric composite sleepers are specified in ISO 12856 [11] and in guidelines issued by DB AG. The aim is to assess whether such a configuration can meet structural requirements while offering environmental advantages through material reuse. The development aligns with broader regulatory trends such as the EU’s Circular Economy Action Plan [12], the revised TEN-T Regulation (Article 46 on climate proofing) [13], and national strategies like Germany’s “Circular Economy strategy” [14]. It also anticipates future material policy developments, including requirements for recycled content in construction products.

The research presented includes analytical calculations of the superimposed sleeper and finite element modelling (FEM) of the sleeper adapted to the innovative system verified through the analytical calculation and validated via laboratory testing of composite components. Particular attention is given to stress distribution within the polymeric composite sleeper and compliance with German and European standards for slab track and recycled materials. Applying FEM to analyse different components of the railway track was also shown for the superstructure-subgrade analysis of ballasted tracks by Fischer et al. [15]. Although this study does not address material durability under environmental stressors and microplastic emissions, future research directions are outlined to consider this aspect of composite use in rail infrastructure.

Ultimately, the objective is to demonstrate that a circular polymeric composite sleeper is adaptable to slab track systems in order to develop a circular slab track.

2. Materials and Methods

The focus of this work lies in the analytical design and numerical analysis (via FEM) of the superimposed sleeper of an innovative slab track system under vertical, static loads induced by railway operation. To account for dynamic loads and wheel force shifts in curves, partial safety factors are applied per DIN EN 16432 [16], in conjunction with DB AG’s guideline (Ril) 820.2020 requirements [17]. Since approved superimposed systems are already in operation on the DB AG network (see Figure 2), the load-bearing asphalt base layer combined with a hydraulically bound layer is assumed inherently capable, the effects of thermal stresses or similar phenomena are therefore not considered.

Asphalt is already a highly circular construction material and is almost entirely reused [18,19]. Given the expected minor deformations under serviceability limits, a linear numerical analysis is appropriate. Superimposed composite components are also treated as linear-elastic according to their material properties investigated via laboratory testing, presuming adherence to the “TA Kunststoffschwellen (polymeric composite sleepers)” requirements [20]. Investigations of durability (e.g., dynamically), weathering, etc., according to ISO 12856 [11] and guidelines are prerequisites and are assumed to be fulfilled.

The polymeric composite sleepers analysed in this study are manufactured according to the cradle-to-cradle principle and consist of around 50% recycled wood fibres, around 20% recycled glass fibre-reinforced plastic (GFRP) fibres, around 20% recycled HDPE, and additives. This study assesses if sleepers made of recycled composite components substituting concrete in the proposed superimposed system are load-bearing according to regulations. Slab track design calculations follow methods by Zimmermann/Eisenmann or Westergaard [16]. According to EN 16432-2 [15] the superimposed elements should be analysed individually according to EN 13230-6 [21]. In all the superimposed slab track system consists of following components (see Figure 3):

• Polymeric composite sleeper (260 × 26 × 16 cm)
• 30 cm asphalt layer (ABC)
• 30 cm hydraulically bound layer (HBM)

Bituminous bonding creates a full surface connection between HBM and ABC, reinforced by bolting of sleepers. Hence, analytical calculations assume System II (fully-bonded) behaviour, consistent with numerical modelling.

The foundation of the analysis following the elastic beam theory according to Winkler is a proportional bedding reaction to an impact [22]. This reaction is influenced by the bedding modulus “C”, which could be determined by field tests, e.g., according to Kerr [23], in the case of slab track systems, C is defined by Deutsche Bahn AG by C = 0.3–0.4 N/mm³ [17].

$$\mathrm{p}\left(\mathrm{x}\right)=\mathrm{C}·\mathrm{w}\left(\mathrm{x}\right)$$

(1)

Regarding the analytical calculation, sleepers are considered as elastic foundation beams, following Winkler and the structural track calculation by Zimmermann [22,24,25]. Since polymeric composite sleepers consist of differing materials and behaviours, the material-independent verification method by Hoffmann [26], which is based on the sleeper analysis according to EN 13230-6 [21], is applied to the ballastless track load case.

A superimposed polymeric composite sleeper measuring 0.26 m × 0.16 m in cross section with a constant rectangular profile and a length of 2.60 m is analysed. Successful verification for this geometry implies larger cross-sectioned superimposed sleepers may also be assumed to be load-bearing.

In ballastless track systems, the bound substructure limits sleeper deflection due to full-surface support compared to a ballast substructure.

2.1. Material Parameters

Due to laboratory tensile testing, the composite material containing >75% filler behaves linearly elastically (see Figure 4), consistent with 3-point bending tests. The slight kink (Figure 5) is ascribed to the measurement procedure with an extensometer, which was retracted after the determination of Young’s modulus.

Young’s modulus and Poisson’s ratio were determined to be of similar magnitude in both the longitudinal and transversal directions of the sleeper. Owing to fibre orientation from the extrusion process, the longitudinal tensile resistance is higher, fitting to the primarily load-bearing direction (see Figure 4). For the FE analysis, Young’s modulus and Poisson’s ratio are the relevant parameters, and isotropic material behaviour was assumed due to the findings.

Due to the composite material’s fibre content and brittle behaviour, tensile resistance and bending resistance differ. The composite’s bending resistance according to laboratory material performance tests is 40.03 MPa [27]. This resistance must be reduced by partial safety factors accounting for permanent loads, material ageing, and unprotected structural components, resulting in the following resistance regarding the ultimate limit state (ULS)

$$f_{char, ULS}=9.50 \mathrm{M}\mathrm{P}\mathrm{a}>{\sigma }_{max}$$

(2)

Regarding the serviceability limit state (SLS) the fatigue resistance is:

$$f_{char, SLS}=6.34 \mathrm{M}\mathrm{P}\mathrm{a}> {\sigma }_{max}$$

(3)

2.2. Analytical Calculation According to Standards

A fundamental parameter is the bedding modulus C. For a ballastless track, a design target on the safe side of C = 0.3 N/mm³ is used [17].

Additionally, the following boundary and load conditions apply:

• Rail: UIC60 E2, dynamic stiffness 40 kN/mm
• Rail seat per System 300, adjusted to base plate of laboratory test (0.39 × 0.16 m²)
• Sleeper length 2.60 m
• Loads per load model LM71, DIN Fachbericht 101, Betonkalender 2015 [28]
• Material parameters: E = 3800 N/mm², ν = 0.33 (previous laboratory tests)

To perform the design for ballastless track systems, corresponding vehicle speeds are considered. According to EN 13230-6 [21], the speed factor for velocities >200 km/h is constant as shown in

Table 1. Load case determination, calculation based on [25].

Load Case	Axle Load (kN)	Speed (km/h)	Sleeper Spacing (mm)	Bedding Modulus [N/mm3]
Slab track	225	>200	650	0.3

.

According to the conditions by Deutsche Bahn AG, the maximum dynamic load on one rail seat is 100.08 kN [28]. The maximum load on the rail seats is derived from the UIC load model LM71 and the corresponding load distribution, as illustrated in Figure 5.

Figure 5. Loads according to guideline 820.2020 of DB [16], according to UIC LM71 [29].

Figure 5. Loads according to guideline 820.2020 of DB [16], according to UIC LM71 [29].

2.3. Finite Element Method (FEM)

Based on the analytical calculation an FEM tool calibrated against analytic results may be used for slab track design according to EN 16432-2 [16]. SIMULIA Abaqus 2023 was used and conducts a three-stage modelling process (pre-processor, solver, and post-processor) to compute stresses and deformations.

Model Setup:

• Axes:

  ○

  x—sleeper direction

  ○

  y—layer depth

  ○

  z—track direction

• Units:

  ○

  Metres

  ○

  Newtons

• Material parameters of the polymeric composite sleeper

  ○

  Composite: E = 3800 N/mm², ν = 0.33

The geometry was meshed in Abaqus using 3D elements of type C3D20R. These 20-node hexahedral volume elements are arranged quadratically, and the application of reduced integration (R) ensures computational efficiency while maintaining accurate stress and deformation distributions. Quadratic elements with additional mid-side nodes were selected to achieve higher accuracy compared to linear elements, which only contain corner nodes.

The mesh design was based on a convergence analysis. For the supporting layers, an element size of 0.10 was adopted. In the finite element model comprising seven sleepers, the same element size of 0.10 was retained for the supporting layers, while a refined element size of 0.05 was applied to the sleepers. Furthermore, to enable a more detailed evaluation of strains at the sleeper surface, a finer discretization with an element size of 0.015 was introduced using the ‘Seed Edges’ function. The elastic bedding was modelled via the ‘Elastic Foundation’ approach.

Based on the findings described in Section 2.1, the sleepers are modelled as linear-elastic and isotropic. Structural layers are also modelled as linear-elastic according to RDO Asphalt [30].Furthermore, a validating laboratory test was conducted. The target was to analyse strain appearing within the polymeric composite sleeper and compare the appearing strains to those of the FEM. To achieve this, an individual sleeper was equipped with strain gauges around the rail seat plate and then subjected to static laboratory loading, with the resulting strains evaluated and compared to those of FEM.

3. Results

3.1. Analytical Calculation

The polymeric composite sleeper exhibits positive and negative bending stresses below the characteristic fatigue resistances, confirming both ultimate and serviceability performance. The analytical calculation is taking the elastic length according to Zimmermann [24] into account. By calculating the elastic length (L) and resulting lever arms according to EN 13230-6 [21] adapted to composite materials by Hoffmann [26], the following maximum bending moment appears at the rail seat. Due to full-surface support on the asphalt base course, the partial safety factor for support misalignment errors, as it is applied in a ballasted superstructure, is not applicable for a slab track with supportive layers:

$$M_{a,pos,i}= {\displaystyle \frac{F_{dyn.}}{2}} · a_{res,i} = {\displaystyle \frac{100.08 \mathrm{k}\mathrm{N}}{2}} · 121.9 \mathrm{m}\mathrm{m}=\mathrm{6,099,876} \mathrm{N}\mathrm{m}\mathrm{m}$$

(4)

At the mid-span of the sleeper the following bending moment appears according to the calculation parameters of EN 13230-6 [21], the partial safety factor is not taken into account, as described above:

$$M_{m,i}=F_{k,i}·m_{F,i} =\mathrm{100,080} \mathrm{k}\mathrm{N} ·\left(-30 \mathrm{m}\mathrm{m}\right) = -\mathrm{3,002,400} \mathrm{N}\mathrm{m}\mathrm{m}$$

(5)

The section modulus needs to be calculated to calculate the bending stresses:

$$W= {\displaystyle \frac{I}{h/2}}= {\displaystyle \frac{\mathrm{88,746,667} {\mathrm{m}\mathrm{m}}^4}{80 \mathrm{m}\mathrm{m}}}=\mathrm{1,109,333} {\mathrm{m}\mathrm{m}}^3$$

(6)

Based on that, the bending stresses could be calculated according to EN 13230-6 [21]. At the rail seats, the following bending stresses appear:

$${\sigma }_{max,rs}= {\displaystyle \frac{M_k}{W}}= {\displaystyle \frac{\mathrm{6,099,876} \mathrm{N}\mathrm{m}\mathrm{m}}{\mathrm{1,109,333} {\mathrm{m}\mathrm{m}}^3}}=5.50 \mathrm{M}\mathrm{P}\mathrm{a}$$

(7)

At the mid-span of the sleeper, the following bending stresses appear:

$${\sigma }_{max,ms}= {\displaystyle \frac{M_k}{W}}= {\displaystyle \frac{-\mathrm{3,002,400} \mathrm{N}\mathrm{m}\mathrm{m}}{-\mathrm{1,109,333} {\mathrm{m}\mathrm{m}}^3}}=2.71 \mathrm{M}\mathrm{P}\mathrm{a}$$

(8)

The maximum bending stress, calculated at 5.50 N/mm², is safely lower than the design fatigue resistance of the composite, thereby validating its structural integrity.

Despite the comparatively small cross-section of the sleeper, which results in elevated stresses, the verification is fulfilled. The maximum bending stress of 5.50 MPa remains below the fatigue limit of the polymeric composite sleeper material. According to the analytical calculation, the newly developed composite material is fundamentally suitable for use in a ballastless track system. Given the proximity of the stresses to the material’s fatigue limit under the Serviceability Limit State, these stresses are further evaluated through a detailed numerical analysis using the finite element method.

3.2. Finite Element Analysis (FEA)

The analysis was designed according to the analytical calculation with one sleeper. This sleeper is based on an elastic bedding. The bedding modulus of C = 0.3 N/mm³ was generated using the “Elastic Foundation” function of Abaqus. In line with the design principles for superimposed components, the bedding modulus is applied directly to the bottom of the sleeper. Since the superimposed element design is based on sleeper designs for a ballast track, the base layers are disregarded.

Figure 6 shows the bending stresses along the longitudinal axis of the sleeper (x-axis).

The appearing stresses are highly comparable to the stresses calculated analytically (s.

Table 2. Tensions sleeper determined via analytical calculation and FEM.

Position	Analytical Calculation [MPa]	FEM [MPa]	Δ [%]
Rail seat	5.50	5.47	0.55
Mid-span	2.71	2.69	0.74

).

To account for the presence of the asphalt concrete (ABC) and hydraulically bound material (HBM) base layers, these were introduced in the subsequent phase of the analysis to evaluate their effect on stress distribution within the superimposed sleeper.

According to the “Guidelines for the Structural Design of Pavement Superstructure with Asphalt Surfacing” (RDO) [30], the following material parameters were set:

• ABC: E = 5000 N/mm², ν = 0.35
• HBM: E = 5000 N/mm², ν = 0.25

The polymeric composite sleeper was superimposed on top of the asphalt layer (see Figure 1).The system is an interlayer bonding system according to EN 16432-2 [16], also the superimposed sleeper is bonded to the asphalt base using both screws as shown in Figure 3 and bitumen. To achieve this within the simulation, the entire system was modelled and merged in Abaqus, enabling a comprehensive analysis of the full assembly.

Analyses have shown, that the supporting layers will face the bending stresses when superimposing a polymeric composite sleeper instead of concrete sleepers [31]. For this reason, the focus here is placed on examining how the bending stresses within the sleeper are altered by the innovative support on asphalt, rather than on ballast with a simple elastic bedding. The loads were applied according to [28] based on LM71 (see Figure 4).

By placing the sleeper on top of the asphalt layer and applying each load according to LM71 [28] on the rail seat area (0.16 m × 0.39 m), the stresses shown in Figure 7 appear.

Due to the additional base layers, which limit sleeper deflection compared to a simple elastic bedding, lower stresses are generated, up to a maximum of 1.230 MPa at the bottom of the sleeper with the highest load (sleeper 4), compared to 5.50 MPa without the base layers.

Regarding the fatigue resistance according to serviceability limit state (SLS), the polymeric composite material is durable:

$${\sigma }_{max}=1.230< f_{char, SLS}=6.34 \mathrm{M}\mathrm{P}\mathrm{a}$$

(9)

In addition, a sensitivity analysis was carried out on the stresses within the sleeper without bond to the asphalt supporting layer. A friction coefficient of 0.2 was assumed, which resulted in a slightly higher stress of 1.307 MPa. Nevertheless, this value remains below the material’s fatigue resistance. This analytical confirmation underscores the suitability of the new composite material for application in ballastless track systems. Further research was conducted regarding bigger superimposed blocks, which showed that increasing the size of the superimposed block leads to reduced stresses within these parts. Consequently, it is advantageous to aim for larger superimposed parts.

3.3. Validating Laboratory Test

The experimental setup for the validation test of the polymeric composite sleeper with strain gauges is shown in Figure 6. As this was the first production batch of the polymeric composite sleeper, slight torsional deformation was observed at the end of the sleeper (see Figure 8). Consequently, this led to variations in the measured strains apart from the strain gauges placed on the neutral axis because loading caused uneven deformation until the torsion was constrained. Multiple sleepers were examined, all exhibiting minor torsion, which explains the observed deviations. However, the production or cooling processes can be adjusted to prevent this effect in future batches. In the present case, the strains occurring along the neutral axis of the sleeper can be used for validation purposes.

In the FEM simulation, the mechanical strains at the sleeper close to the rail seat plate along the neutral axis are 53.33 µm/m. This matches the lab measurements, which showed 52.22 µm/m at the same position. The difference is 2.08%, which shows the high accuracy of the simulation.

4. Discussion

The findings confirm that polymeric composite sleepers are durable when integrated into fully bonded slab track systems. The investigated polymeric composite sleepers under bonded conditions exhibited a maximum bending stress of 1.307 N/mm², significantly lower than the 5.47 N/mm² seen without bound layers, both well below the fatigue resistance of the polymeric composite material.

Studies comparing polymeric composite and concrete sleepers both in ballasted tracks have shown that composite variants offer similar structural performance, with added benefits such as enhanced damping and reduced rail-seat pressure [32].

The study is demonstrating that polymer-based composite sleepers can fulfil slab track structural roles, with the added advantage of the circular reuse principle. The compositional innovation by replacing cementitious elements with composite materials made from recycled plastics strongly aligns with sustainability trends in the industry and political aims.

5. Conclusions

The structural analysis demonstrates that the superimposed polymeric composite sleeper, developed under circular economy principles satisfies fatigue and serviceability requirements under ballastless track loading. Peak bending stresses remain within safe limits compared to material fatigue resistance. Further refinement through numerical methods (FEM) enhances confidence in the design and supports implementation in slab track systems. This study demonstrates that a fully circular ballastless track system comprising recycled polymeric composite sleepers and circular asphalt is structurally viable. Analytical methods and finite element validation confirm that the polymeric composite sleepers, designed with sustainability at their core, meet both ultimate and serviceability performance requirements. Laboratory validation further supports the close alignment between numerical and experimental results. By integrating recycled plastics within the sleepers and utilizing reclaimed asphalt in the supporting base, the system maintains load-bearing capacity while advancing circular economy principles.

In summary, the results substantiate the concept that a slab track assembled entirely from circular materials can fulfil structural demands. Future studies should therefore focus on assessing long-term durability under dynamic loads, environmental resilience, and lifecycle emissions to further validate and optimize such sustainable track designs. Overall, the analyses demonstrated that the composite materials, which need to be tested under dynamic conditions and proven adequate for use in ballasted tracks according to ISO 12856 [11] and the guidelines of DB, are also suitable for application in slab track systems. Compared to ballasted tracks, the stresses within the sleeper are reduced by the bonded layers.

Additionally, experimental (e.g., dynamic behaviour of sleeper on asphalt) and long-term field investigations can be based on the findings presented, to confirm long-term durability.

The findings of this study are applicable to other polymeric composite sleeper designs and can be adapted and assessed according to the specific material properties of those alternative sleepers. Similar to documented cases such as the adaptable bending stiffness of fibre-reinforced and polymer-based composites, this transferability enables systematic scaling of design and verification based on characteristic values like elastic modulus and fatigue resistance.

6. Patents

There are both a national and an international patent application for the system described. The national patent application has been published under DE 10 2023 112 802 A1, and the international patent application under PCT/EP 2024/063030.