Load-Dedicated Fiber Reinforcement of Additively Manufactured Lightweight Structures

Abstract

This study focuses on a novel lightweight technology for manufacturing variable-axial fiber-reinforced polymer components. In the presented approach, channels following the load flow are implemented in an additively manufactured basic structure and impregnated continuous fiber bundles are pulled through these component-integrated cavities. Improved channel cross-section geometries to enhance the mechanical performance are proposed and evaluated. The hypothesis posits that increasing the surface area of the internal channels significantly reduces shear stresses between the polymer basic structure and the integrated continuous fiber composite. A series of experiments, including analytical, numerical, and microscopic analyses, were conducted to evaluate the mechanical properties of the composites formed, focusing on Young’s modulus and tensile strength. In addition, an important insight into the failure mechanism of the novel fiber composite is provided. The results demonstrate a clear correlation between the channel geometry and mechanical performance, indicating that optimized designs can effectively reduce shear stress, thus improving load-bearing capacities. The findings reveal that while fiber volume content influences the impregnation quality, an optimal balance must be achieved to enhance mechanical properties. This research contributes to the advancement of production technologies for lightweight components through additive manufacturing and the development of new types of composite materials applicable in various engineering fields.

1. Introduction

The importance of lightweight construction has gained significant attention in recent years, particularly in the context of mass reduction, resource conservation, and the reduction of CO₂ emissions. As highlighted by the European Green Deal, transitioning to more sustainable practices is imperative for achieving climate goals and fostering a greener economy [1]. Lightweight materials not only contribute to lower energy consumption during use but also minimize the overall environmental impact of manufacturing processes due to reduced material usage [2]. A particularly high degree of lightweight construction is made possible through variable-axial fiber reinforcement, which allows for the production of tailored structures, as the fibers follow the load flow [3,4]. Variable-axial fiber composites are also widely known as variable angle tow composites [5,6] or curvilinear composites [7] and can be manufactured using automated fiber placement [8,9] or tailored fiber placement [3], for example.

An option for the mold-free production of tailored structures are additive manufacturing (AM) technologies, which enable the development of innovative lightweight constructions with complex topology-optimized geometries that cannot be achieved using conventional manufacturing methods. This technology not only facilitates the creation of material-efficient and mass-saving structures but also aligns with the principles of lightweight construction [10]. Despite its many benefits, the challenge of low strength between layers can hinder performance, necessitating new approaches to fully exploit its potential [11,12]. To enhance the mechanical properties of AM parts, researchers have explored various strategies, including the use of advanced materials [13,14,15,16], modified process parameters [17,18,19,20], and short-fiber integration [21,22,23,24,25]. One promising approach involves the incorporation of continuous fiber (cF) composites, which can significantly improve both stiffness and strength [26]. However, many existing methods still primarily focus on in-build plane reinforcement [27,28,29,30,31,32,33,34], which does not adequately address the three-dimensional stress distribution seen in complex components. Non-planar continuous fiber additive manufacturing (cF-AM) methods allow the three-dimensionally curved deposition of cF composites and thus a variable-axial fiber alignment [35], which improves the mechanical load distribution and significantly increases the lightweight construction potential [3,4,14,36]. However, non-planar cF-AM methods often require complex machine technology and elaborate slicing processes [37,38,39,40,41].

An alternative to the direct deposition of cF composites is to decouple the cF application from the AM process, where a basic structure is first additively manufactured, followed by cF reinforcement. For example, Crescenti et al. inject impregnated cF bundles into component-integrated cavities in the shape of channels that are integrated into the basic structure [42]. Meißner et al. suggested a related method where impregnated cF bundles are pulled through the channels, as shown in Figure 1 [43,44].

An earlier study demonstrated the high structural performance of the innovative fiber composite construction method [43]. However, it was determined that the strength-limiting criterion results from the shear strength at the interface between the additively manufactured basic structure and the integrated cF composite [43]. From this observation, the hypothesis was derived that an increased channel surface area will improve the shear stress distribution between the polymer basic structure and the integrated continuous fiber-reinforced polymer (cFRP), leading to superior mechanical properties of the overall composite [43]. This study examines the validity of the hypothesis by proposing adapted channel cross-section geometries. The extent to which increasing the channel surface area contributes to an improvement in structural integrity is being examined.

The insights gained from this research could pave the way for a novel ultra-lightweight technology and facilitate more sustainable applications of AM in engineering, particularly in fields where high-performance materials are critical. By addressing current limitations and exploring innovative solutions, the aim is to contribute to the development of stronger, more lightweight composite structures for a wide range of applications and to expand the potential of A3M technologies.

2. Materials and Methods

The methodological approach for examining the research hypothesis is illustrated in Figure 2. Based on the initial definition of channel cross-section geometries and fiber volume content (FVC) variations, a plan of experiments is created. This plan includes the determination of a mechanical test scenario. Subsequently, analytical, numerical, experimental, and microscopic investigations are carried out and evaluated.

2.1. Materials and Process Parameters

The component-integrated method was realized at a laboratory scale, as described in Figure 1.

The AM process of the material extrusion of polymers with thermal reaction bonding (MEX-TRB/P) was utilized to produce the basic structures investigated in this study. An acrylonitrile butadiene styrene (ABS-M30-black from the company Stratasys, Eden Prairie, MN, USA) filament was processed on the AM system Stratasys Fortus 900mc. The CAD models were created with SolidWorks 2020 and prepared with the slicer software Insight 18.6 using single contour and ±45° solid rasters with an infill of 100%. The layer height was 0.18 mm, the building chamber temperature was 90 °C, and the nozzle temperature was 320 °C.

A composite of continuous carbon fibers and epoxy (EP) resin was utilized to realize the integrated cFRP. The FVC of the integrated cF composite depends on the number of cF bundles used and varies depending on the channel cross-section geometry (according to Section 2.2). Tenax-E STS40 F13 24K 1600 Tex cF bundles from Teijin Carbon (Chiyoda, Japan) were utilized. Epikote resin MGS LR385, in combination with the hardener MGS LH385 (Hexion specialty chemicals B.V., Rotterdam, The Netherlands), was applied as the matrix material. The resin hardener ratio was 100:35 by weight. The viscosity of the resin–hardener mixture was 1200 mPa∙s at a processing temperature of 20 °C. The cF impregnation and pultrusion were carried out at 20 °C. After the manual pultrusion process, the specimens were stored at 20 °C for 24 h, then treated at 60 °C for 15 h and stored again at 20 °C for at least 48 h.

The mechanical properties of the material components are listed in

Table 1. Mechanical properties of the material components.

Designation	Young’s Modulus [GPa]		Tensile Strength [MPa]		Source
AM basic structure ABS	$E_{ABS}$	1.8	${\sigma }_{m,ABS}$	26	[43]
Continuous carbon fiber	$E_{x,cF}$	240	${\sigma }_{m,cF}$	4300	Data sheet Teijin
Matrix EP	$E_{EP}$	3.3	${\sigma }_{m,EP}$	75	Data sheet Hexion

.

2.2. Channel Cross-Section Geometries and Maximum FVC

To investigate the research hypothesis, adapted channel cross-section geometries are proposed, which exhibit a drop-shaped (DS), narrow (NA), and very narrow (VN) design, respectively, (Figure 3). These geometries are characterized by an increasing ratio of perimeter p to cross-section area A. The objective is to determine how the increasement in channel surface area (at constant channel cross-section area) influences the shear stress distribution at the interface between the additively manufactured polymer basic structure and the integrated cF composite.

In preparation for this study, preliminary considerations for identifying the maximum feasible FVC of the integrated cF composite have been carried out. The pre-investigations indicated that the technological limit of the cF integration process is determined by the force required to pull the impregnated cF bundles through the component-integrated channels. This force increases with rising FVC and is dependent on the channel geometry. A higher channel surface area results in higher friction between the impregnated cF and the channel wall and therefore a higher force for pulling the impregnated cF bundles through the channels. Excessive pulling force can lead to blocking inside the channel or the failure of the pulling mechanism. The FVC is configured based on [45] according to

$${\phi }_{cFRP}=n·{\displaystyle \frac{m_f/L}{A·{\rho }_f}}$$

(1)

where n is the number of cF bundles, m_f/L is the fiber fineness (1600 tex), A is the channel cross-section area (89.3 mm), and ρ_f is the fiber density (1.78 g/cm³). The results of the pre-investigations in this work regarding the maximum achievable FVC are 44.3% (44 bundles) for DS, 30.2% (30 bundles) for NA, and 24.2% (24 bundles) for VN channel geometry, respectively, (Figure 3).

2.3. Plan of Experiments

The plan of the experiments is shown in Figure 4. The influencing variables investigated include two factors—the cross-section geometry of the channels and the FVC of the cF composite within the component-integrated channels. The respective levels of factors contain three specific geometries for the channel cross-sections, while five values are defined for the FVC in each case. The focus of the investigation is on the Young’s modulus and the tensile strength as target values, which enable precise characterization of the mechanical tensile properties under the various test conditions. The numbering of the test specimens, with the corresponding characteristics of the influencing variables, is listed in

Table 2. Numbering of the manufactured tensile test specimens with corresponding characteristics. The influencing variables are the channel cross-section geometry (DS, NA, and VN) and the FVC φ_cFRP, which is determined as dependent on the number n of cF bundles.

No.	n	φcFRP	No.	n	φcFRP	No.	n	φcFRP
DS-1	36	36.3%	NA-1	22	22.2%	VN-1	16	16.1%
DS-2	38	38.3%	NA-2	24	24.2%	VN-2	18	18.1%
DS-3	40	40.3%	NA-3	26	26.2%	VN-3	20	20.1%
DS-4	42	42.3%	NA-4	28	28.2%	VN-4	22	22.2%
DS-5	44	44.3%	NA-5	30	30.2%	VN-5	24	24.2%

, whereby the FVC is determined according to Equation (1).

2.4. Tensile Test Specimens and Test Procedure

Tensile tests are carried out in accordance with the standard ISO 527-5 [46]. The specimen design is subjected to technologically related adjustments. The thickness of the specimens is increased due to the channel geometries to be tested (12.0 mm for DS, 6.0 mm for NA, and 4.5 mm for VN, respectively). Load transfer elements made of glass fiber-reinforced polymer (GFRP) with a [±45°]₈ laminate structure are bonded to the tensile test specimens with the two-component epoxy adhesive DP490 (3M Corp., Saint Paul, MN, USA). Since, in practice, the thickness of the tabs is often one- to four-times the specimen thickness [47,48], a tab thickness of 8 mm was selected.

The CAD models of the developed tensile test specimens are illustrated in Figure 5. An example manufactured specimen is shown in Figure 6 and the experimental setup is described in Figure 7. The Inspect 100 kN universal testing machine from Hegewald and Peschke GmbH (Nossen, Germany) was used to perform the mechanical tests. The tests were executed under controlled conditions at a temperature of 23 ± 2 °C and a relative humidity of 50 ± 10%. The test speed was 2 mm/min. The Young’s modulus was evaluated in the strain interval from ε_x′ = 0.05% to ε_x″ = 0.25%.

During the tensile tests, the force F_x applied by the testing machine and the change in length ∆l of the specimen in the measuring range are recorded continuously. The change in length is detected using the mechanical extensometer MFX 200 from MF Mess- & Feinwerktechnik GmbH (Velbert, Germany) with an initial measuring length l₀ = 50 mm. The strain is calculated using ${\mathsf{\epsilon }}_x=\mathsf{\Delta }l/l_0$ and the tensile stress is determined based on the overall cross-section area A₀ of the specimen in the measuring range according to ${\mathsf{\sigma }}_x=F_x/A_0$.

2.5. Analytical Estimation of the Tensile Behavior

Preliminary, theoretical considerations are carried out according to the rule of mixture [45] to estimate the tensile behavior of the overall composite. The average FVC of the manufactured specimens is determined by

$${\phi }_{eff}={\phi }_{cFRP}·{\displaystyle \frac{A_{cFRP}}{A_0}}$$

(2)

with the channel cross-section area A_cFRP, the overall cross-section area A₀ of the specimen in the measuring range, and is dependent on the FVC φ_cFRP of the integrated cF composite. The Young’s modulus of the integrated cF composite is calculated based on

$$E_{x,cFRP}={\phi }_{cFRP}\cdot E_{x,cF}+(1-{\phi }_{cFRP})\cdot E_{EP}$$

(3)

with the FVC φ_cFRP of the integrated cF composite, the Young’s modulus E_x,cF of the continuous carbon fibers, and the Young’s modulus E_EP of the matrix material. The expected averaged Young’s modulus in the measuring range of the manufactured specimen is obtained according to

$$E_{x,eff,analytical}={\displaystyle \frac{1}{A_0}}·(E_{x,cFRP}·A_{cFRP}+E_{ABS}·A_{ABS})$$

(4)

by considering the overall cross-section area A₀, the Young’s modulus E_x,cFRP of the integrated cF composite, the channel cross-section area A_cFRP, the Young’s modulus E_ABS, and the cross-section area A_ABS of the additively manufactured basic structure.

2.6. Numerical Models

Static mechanical numerical simulations of the tensile tests are carried out to investigate the failure behavior of the specimens under tensile load dependent on the channel cross-section geometry and the FVC of the integrated cF composite. The elastic strain in the additively manufactured ABS basic structure and the shear stress distribution at the interface between the ABS basic structure and the integrated cF composite are evaluated. ANSYS 2023 R2 is used as the finite elements analysis (FEA) software. The FEA models are shown in Figure 8 and the material parameters are listed in the Supplementary Information (Section S1–S3). The symmetry properties of the specimens allow the computation of an eighth model. For this purpose, three symmetry regions are defined. A tensile force F_x = 5 kN is applied to the GFRP tab, which, taking the symmetry conditions into account, corresponds to a total force of 20 kN. Pentahedral elements with 15 nodes and hexahedral elements with 20 nodes are used. The averaged element size, number of elements, and number of nodes of the simulation models are listed in

Table 3. Averaged element size, number of elements, and number of nodes of the simulation models. For VN channel cross-section design, a convergence study is executed.

Channel Cross-Section	Element Size	Number of Elements	Number of Nodes
DS	0.33	490,334	2,175,609
NA	0.33	680,772	3,010,066
VN	1.00	34,652	138,102
	0.50	251,716	1,006,632
	0.33	845,268	3,775,248
	0.25	1,894,216	8,248,954
	0.20	3,782,900	16,181,767
	0.17	6,438,600	27,253,772

for each specimen. To ensure the validity of the simulation results, a convergence study with an increase in mesh fineness is conducted for the specimen with VN channel cross-section geometry.

A further method to check the plausibility of the simulation results is to determine the effective Young’s modulus of the overall composite based on the numerical results and compare it with the analytical considerations. For this purpose, the strain ε_x is verified in the area that corresponds to the measuring range of the extensometer. Taking the overall cross-section area A₀ into account, the Young’s modulus results from

$$E_{x,eff,num}={\displaystyle \frac{F_x}{A_0·{\epsilon }_x}}$$

(5)

2.7. Microscopy

To prepare the micrographs, the tensile test specimens are cut in the middle after completion of the mechanical tests to enable a detailed analysis of the internal structures. The specimens are carefully ground to ensure a precise representation of the cutting surfaces. During the grinding process, a total of five different abrasive papers with grain sizes from 68 µm to 5 µm are used. The microscopic investigations are carried out using the digital microscope VHX-7000 from Keyence Corp. (Osaka, Japan). This method enables the precise localization of the dry, non-impregnated areas within the composite. Such zones indicate insufficient distribution or penetration of the matrix material. As part of the investigations, the quantity of these dry areas is recorded using the microscope-integrated image processing software. The ratio between the impregnated areas A_imp and the total area A_cFRP of the integrated cF composite is determined. This method enables the examination of the impregnation ratio according to

$$q={\displaystyle \frac{A_{imp}}{A_{cFRP}}}$$

(6)

and allows a well-founded assessment of the quality of the composite examined.

3. Results

3.1. Prediction of the Young’s Modulus

The results of the analytical considerations are listed in

Table 4. FVC φ_cFRP of the integrated cF composite according to Equation (1), Young’s moduli E_x,cFRP of the integrated cFRP according to Equation (3), averaged FVC φ_eff of the tensile test specimens according to Equation (2), and analytical determined effective Young’s moduli E_{x,eff,analytical} according to Equation (4).

No.	φcFRP	Ex,cFRP [GPa]	φeff	Ex,eff,analytical [GPa]
DS-1	36.3%	89.1	17.4%	43.7
DS-2	38.3%	93.9	18.4%	46.0
DS-3	40.3%	98.6	19.3%	48.3
DS-4	42.3%	103.4	20.3%	50.6
DS-5	44.3%	108.2	21.3%	52.9
NA-1	22.2%	55.8	13.6%	34.8
NA-2	24.2%	60.5	14.8%	37.7
NA-3	26.2%	65.3	16.0%	40.6
NA-4	28.2%	70.1	17.3%	43.6
NA-5	30.2%	74.8	18.5%	46.5
VN-1	16.1%	41.5	8.4%	22.6
VN-2	18.1%	46.2	9.5%	25.6
VN-3	20.1%	51.0	10.5%	27.6
VN-4	22.2%	55.8	11.6%	30.1
VN-5	24.2%	60.5	12.7%	32.5

. The average FVC φ_eff calculated by Equation (2) is between 8.4% and 21.3%. The Young’s modulus E_x,cFRP of the integrated cF composite determined by Equation (3) varies from 41.5 GPa to 108.2 GPa. The expected average effective Young’s modulus E_{x,eff,analytical} calculated by the rule of mixture according Equation (4) is between 22.6 GPa and 52.9 GPa.

The results of the numerical simulations are shown in Figure 9. The maximum elastic strain ε_x in the tensile direction is nearly uniformly distributed across the measuring area. The simulation results are checked for plausibility by conducting a convergence study. It shows that the results of the elastic strain converge at an average element size of 0.33 mm. This finding serves as the basis for choosing the element size in the numerical models to ensure precise calculations with low computational effort.

Table 5. Numerical results for an average element size of 0.33 mm. Elastic strain ε_x and Young’s modulus E_x,eff,num with the relative deviation ∆E_x/E_{x,eff,analytical} between numerically and analytically determined Young’s modulus.

No.	εx [–]	Ex,eff,num [GPa]	∆Ex/Ex,eff, analytical
DS-1	2.46 × 10−3	43.7	0.00%
DS-2	2.34 × 10−3	46.0	0.01%
DS-3	2.23 × 10−3	48.3	0.00%
DS-4	2.13 × 10−3	50.6	0.00%
DS-5	2.03 × 10−3	52.9	−0.01%
NA-1	3.95 × 10−3	34.7	−0.22%
NA-2	3.64 × 10−3	37.6	−0.22%
NA-3	3.38 × 10−3	40.5	−0.21%
NA-4	3.15 × 10−3	43.5	−0.22%
NA-5	2.96 × 10−3	46.4	−0.22%
VN-1	5.23 × 10−3	22.6	−0.58%
VN-2	4.71 × 10−3	25.6	−0.57%
VN-3	4.28 × 10−3	27.6	−0.56%
VN-4	3.93 × 10−3	30.1	−0.54%
VN-5	3.62 × 10−3	32.5	−0.53%

summarizes the numerical results of all specimen variants. The numerically determined Young’s modulus E_x,eff,num, calculated by means of the elastic strain ε_x in the measuring range of the extensometer according to Equation (5), is between 22.6 GPa and 52.9 GPa. The relative deviation ∆E_x/E_{x,eff,analytical} between the numerically and analytically determined Young’s moduli is very low (between 0.00% and −0.58%), which emphasizes the plausibility of the numerical results.

3.2. Tensile Test Results

3.2.1. Failure Mechanism

The typical failure pattern is shown in Figure 10. It is characterized by the fact that the ABS shows partial damages at the transition to the GFRP load transfer element. The fracture is caused by the separation of the cF composite from the ABS, as shown by the broken basic material, exposing the cF composite. Accordingly, the initial failure occurs at the interface between ABS and cF composite due to excessive shear stress, which is numerically verified and shown in Figure 11. After initial interface shear failure, the load is transferred to ABS, which leads to the permissible stress being suddenly exceeded and finally to total failure of the specimen.

3.2.2. Measurement Data and Characteristic Values

The stress–strain curves of the individual specimens are approximately linear, as shown in Figure 12. The extracted characteristic values are listed in

Table 6. Experimental results of the tensile tests. F_x,max—maximum tensile force; E_x,eff,exp—Young’s modulus; ∆E_x—absolute deviation between experimentally and analytically determined Young’s modulus; ∆E_x/E_{x,eff,analytical}—relative deviation between experimentally and analytically determined Young’s modulus; ε_x,max—strain at failure; σ_x,max—tensile stress at failure.

No.	Fx,max [kN]	Ex,eff,exp [GPa]	∆Ex [GPa]	∆Ex/Ex,eff,analytical	εx,max [%]	σx,max [MPa]
DS-1	17.12	38.57	5.14	11.8%	0.24	92.1
DS-2	16.80	37.80	8.20	17.8%	0.24	90.3
DS-3	21.49	43.49	4.79	9.9%	0.27	115.6
DS-4	19.66	49.90	0.68	1.3%	0.21	105.7
DS-5	18.52	51.00	1.86	3.5%	0.20	99.6
NA-1	20.22	27.91	6.89	19.8%	0.50	138.8
NA-2	25.72	31.09	6.62	17.6%	0.57	176.6
NA-3	16.55	39.61	1.01	2.5%	0.29	113.7
NA-4	19.58	36.48	7.07	16.2%	0.37	134.4
NA-5	19.29	43.27	3.19	6.9%	0.31	132.5
VN-1	24.47	21.65	0.91	4.0%	0.67	144.2
VN-2	23.29	20.54	4.51	18.0%	0.67	137.2
VN-3	25.54	19.62	7.93	28.8%	0.77	150.4
VN-4	26.06	27.39	2.66	8.9%	0.56	153.5
VN-5	22.98	29.37	3.17	9.7%	0.46	135.4

and summarized in Figure 13. The failure of the tensile force F_x,max occurs between 16.55 kN and 26.06 kN. The Young’s modulus E_x,eff,exp can be determined despite the failure occurring outside the measuring range, as it is detected in a low, linear measuring range. It ranges from 19.62 GPa to 51.00 GPa. The absolute deviation ∆E_x to the analytical determined Young’s modulus E_{x,eff,analytical} is between 0.68 GPa and 8.20 GPa, which means a relative deviation ∆E_x/E_{x,eff,analytical} from 1.3% to 28.8 %. The maximum detected strain ε_x,max at failure is between 0.20% and 0.77%. The elongation at break ε_m of the cF is 1.8%. In relation to this value, the cF utilization is between 11% and 43%. The maximum determined tensile stress σ_x,max at failure varies from between 90.3 MPa and 153.5 MPa. As postulated, the tensile test specimens with large channel perimeters achieve higher maximum tensile stress.

The higher performance is due to a more homogeneous shear stress distribution between the ABS and the cF composite, which is accompanied by lower shear stress peaks (Figure 14,

Table 7. Numerically determined maximum shear stress τ_xz,max at the interface between additively manufactured ABS and integrated cF composite.

No.	τxz,max [MPa]	No.	τxz,max [MPa]	No.	τxz,max [MPa]
DS-1	25.5	NA-1	17.6	VN-1	16.1
DS-2	25.8	NA-2	16.9	VN-2	15.3
DS-3	26.1	NA-3	16.3	VN-3	14.6
DS-4	26.4	NA-4	15.8	VN-4	13.9
DS-5	26.6	NA-5	15.3	VN-5	13.4

and Figure 15). For DS specimens, the shear stress τ_xz rises with increasing FVC φ_eff, as a strong notch effect occurs due to the curved channel surface. The notch effect increases as the Young’s modulus E_x,cFRP rises, resulting in a higher stress concentration. For NA and VN specimens, the shear stress τ_xz decreases with rising FVC φ_eff. The channel surfaces are even, allowing the load to be applied over the entire surface of the cF composite. As the Young’s modulus E_x,cFRP increases, the shear stress is distributed more homogeneously in the tensile direction, which reduces the stress peaks.

3.3. Microscopic Results

Figure 16 shows a micrograph of a tensile test specimen with DS channel cross-section geometry, which is characteristic for all further specimens. The micrograph shows the specificity of the additively manufactured basic structure, which is distinct by cavities typical for the MEX-TRB/P process. These voids illustrate the layered structure of the AM. The cF composite structure is clearly recognizable in the component-integrated channel of the tensile specimen. The light-colored sections of the cF stand out, while the darker zones represent the matrix material of the cF composite. The obvious boundary between the ABS and cF composite is remarkable. This boundary is made particularly clear by the stair-step effect of the MEX-TRB/P process. Furthermore, the black areas within the cF composite show the existence of non-impregnated, dry fibers.

Table 8. Results of the microscopy. Impregnation quality q of the integrated cF composite, quantified by the ratio of non-impregnated area to overall area.

No.	q [–]	No.	q [–]	No.	q [–]
DS-1	0.975	NA-1	0.990	VN-1	0.987
DS-2	0.969	NA-2	0.990	VN-2	0.987
DS-3	0.981	NA-3	0.989	VN-3	0.989
DS-4	0.972	NA-4	0.992	VN-4	0.988
DS-5	0.972	NA-5	0.994	VN-5	0.996

shows the impregnation rate q, which is determined according to Equation (6), as an indicator for the impregnation quality by measurement of the non-impregnated areas. Figure 17 illustrates the dependence of the impregnation quality q of the fibers within the component-integrated channels as a function of the FVC φ_eff, which is between 0.969 and 0.990. Since the micrographs were taken in the middle of each specimen and thus always at the same location, the microscopic results are comparable with each other. However, stochastically distributed defects over the specimen length cannot be completely ruled out, so that further investigations (for example, computer tomography) in follow-up studies are advisable.

4. Discussion

The results of this study provide valuable insights into the mechanical behavior of additively manufactured polymer structures reinforced with component-integrated cF composites. The analytical, numerical, experimental, and microscopic findings (summarized in Figure 18) collectively emphasize the critical role of channel geometry as a significant factor limiting the mechanical performance of such composite materials.

The experimental results confirm the hypothesis that the channel geometry significantly impacts tensile properties. Specifically, an increased ratio of channel perimeter to cross-section area reduces shear stress at the interface between the additively manufactured polymer and the integrated cF composite, which correlates with enhanced load-bearing capacity. These findings support earlier studies that have similarly identified the importance of optimizing geometrical features to improve mechanical performance in composite materials [49,50].

However, it is also important to note that an enlarged channel surface area is associated with lower FVC and thus also lower Young’s moduli. Despite the lower FVC, the enlarged channel surface area exhibits higher structural integrity because of the more homogeneous interface shear stress distribution. The channel designs tested offer a path for future work as they not only improve mechanical properties but also provide insight into structural integrity and failure mechanisms under loading. This aligns with the findings of prior research, which consider design strategies to minimize stress concentrations and enhance load distribution in composite structures [51,52].

Furthermore, the observed relationship between FVC and impregnation quality of the cF with matrix material confirms previous conclusions that higher FVC can restrain effective resin transfer due to increased friction within cavities [53,54]. While a higher FVC generally leads to improved stiffness, it can also reduce the matrix material’s availability for good impregnation, which is reflected in the reduced mechanical properties at higher FVCs. This finding suggests a balance must be achieved between maximizing FVC and ensuring optimal impregnation quality to achieve desired mechanical properties.

Figure 19 evaluates the potential of the investigated component-integrated method by classifying the study results in the landscape of FRP technologies and fiber-reinforced AM parts. It shows that the method presented achieves comparable results in terms of tensile strength to established FRP processes such as resin transfer molding and compression molding. Compared to short-fiber-reinforced AM methods, the component-integrated cf-reinforced AM method is more performant. Although there is an expected performance improvement compared to the earlier study [43], its potential has not yet been fully exploited compared to established cF-reinforced AM methods. However, the investigated method has a considerable advantage over most established methods due to the possibility of including a variable-axial cF arrangement inside additively manufactured parts independently from the AM building direction.

This study demonstrated that the structural integrity of the novel cF composite was significantly enhanced. However, it also became apparent that the tensile testing method used did not allow its full potential to be exploited. The tensile strengths of other unidirectional carbon fiber composites with comparable FVC show that the potential is higher [55]. A tensile test specimen geometry should be developed that is tailored to the characteristics of the novel fiber composite. Recent studies show that the choice of test specimen geometry can significantly influence the measured tensile strength of fiber composites [56,57].

The study confirmed the results of the previous study [43], which found that the strength-limiting criterion is the shear strength between the additively manufactured basic structure and the integrated cF composite. Accordingly, this characteristic value is fundamental for the structural mechanical design of components. Common standardized test methods with corresponding specimen geometries are not completely applicable due to manufacturing restrictions, so that an adapted or alternative test method for the experimental determination of shear strength must be developed in the context of future work. The Iosipescu test [58,59] or the pull-out test [60,61], for example, can serve as a basis for developing a suitable shear test methodology.

In summary, this research contributes to the existing state of knowledge by demonstrating how geometrical modifications can lead to significant improvements in the performance of additively manufactured polymer composites reinforced with cF. Future studies should focus on further optimizing channel designs, exploring alternative geometries, and investigating their implications in practical applications. Alternative material combinations could also be promising. This could lead to more robust and efficient composite structures suitable for demanding engineering applications.

5. Conclusions

This study successfully demonstrates a novel ultra-lightweight approach based on variable-axial reinforcement of additively manufactured polymer structures using component-integrated cF composites. By developing and analyzing three channel geometries, the results showed the critical role of channel cross-section geometry for improving the mechanical performance of these composites. The findings reveal a clear relationship between channel geometry and mechanical properties, particularly in behavior under tensile load. Specifically, increasing the surface area of the channels effectively reduces shear stress at the interface between the additively manufactured polymer basic structure and the integrated cF composite. This improvement leads to enhanced load-bearing capacity and demonstrates the importance of elaborated channel design in composite applications.

Moreover, the research highlights the influence of FVC on impregnation quality. While higher FVC can lead to reduced impregnation of the fibers, the correlation between impregnation quality and mechanical properties was not as significant as expected. This suggests that future studies should focus on achieving an optimal balance between FVC and impregnation to maximize mechanical performance.

The practical implications of this research are significant. The enhanced mechanical properties could expand the applicability of AM in industries requiring lightweight materials, such as those of aerospace, automotive, and construction. Future work should aim to further refine the channel geometries and investigate the effects of component design features on mechanical performance to open the way for more robust applications of this innovative method in application-specific scenarios. In addition, a test method needs to be developed to determine the shear strength between the basic structure and the cF composite as it is the performance-limiting parameter of the novel composite.

In summary, the integration of variable-axial cF composites into improved, internal channels of additively manufactured polymer structures represents a promising advancement in the field of polymer composites, with the potential to revolutionize the design and production of high-performance materials.