Balancing Strength and Flexibility: Mechanical Characterization of Carbon Fiber-Reinforced PLA Composites in FDM 3D Printing

Abstract

Fused Deposition Modeling (FDM) is a commonly used 3D printing process characterized by its versatility in material selection; however, FDM’s layer-by-layer process often leads to lower strength properties. This study explores the mechanical properties of FDM 3D-printed composite materials printed with varying nozzle diameters, specifically on the influence of Carbon Fiber-reinforced Polylactic Acid (PLA-CF) on tensile and flexural strength when reinforcing Polylactic Acid (PLA) parts. Composite samples were printed with varying ratios of PLA and PLA-CF, ranging from 0% to 100% PLA-CF in 20% increments, with layer groups stacked vertically, while also using three different nozzle diameters (0.4 mm, 0.6 mm, and 0.8 mm). Tensile testing revealed a proportional increase in strength as PLA-CF content increased, indicating that carbon fiber reinforcement significantly enhances tensile performance. However, flexural testing demonstrated a decrease in bending strength with higher PLA-CF content, suggesting a trade-off between stiffness and flexibility. Mid-range ratios (40–60% PLA-CF) provided a balance between tensile and flexural properties. Finally, atomic force microscopy was utilized to provide a better understanding of the microscale morphology and surface properties of PLA and PLA-CF thin films. The results highlight the potential of PLA-CF/PLA composites to allow for more direct control over the tensile–flexural trade-off during the printing process, as opposed to manufacturing filaments with fixed fiber percentages. These results provide a path for tailoring the mechanical behavior of printed parts without requiring specialized filaments.

1. Introduction

Fused Deposition Modeling (FDM) is one of the most widely utilized 3D printing technologies due to its accessibility, cost-effectiveness, and versatility [1,2]. It operates by depositing thermoplastic materials layer by layer to form three-dimensional objects, offering flexibility in design and material selection [3,4]. Polylactic Acid (PLA), a biodegradable thermoplastic derived from renewable resources, is among the most commonly used FDM materials because of its ease of printing, availability, and affordability [5,6]. However, PLA’s mechanical properties, particularly its tensile and flexural strengths, are often insufficient for applications requiring higher performance or structural integrity [7,8]. Recent advancements in material development have introduced fiber-reinforced thermoplastics, such as Carbon Fiber-reinforced PLA (PLA-CF), which aim to address these limitations. PLA-CF incorporates short carbon fibers into the PLA matrix, enhancing the material’s stiffness and tensile strength without significantly compromising its printability [9,10]. These materials have become increasingly popular for use in functional prototyping and manufacturing lightweight, high-strength components [11]. Numerous prior works have shown that carbon fiber reinforcement can increase stiffness and tensile strength, and yet few studies focus on varying the fiber content within a single build. The current study aims to determine how layering PLA-CF in different proportions and using distinct nozzle diameters affect both tensile and flexural properties [12,13].

Modern consumer-grade FDM printers, particularly those equipped with multi-material capabilities, enable innovative methods for customizing material properties. One such approach involves creating composite structures with alternating layers of PLA and PLA-CF, effectively combining the advantageous properties of both materials [14,15]. By varying the ratio of PLA to PLA-CF and modifying the printing parameters, such as nozzle diameter, it is possible to tailor the mechanical performance of 3D-printed parts to meet specific application requirements [16,17].

While the potential for PLA/PLA-CF composites has been explored, there is a lack of comprehensive studies evaluating the combined effects of material composition and printing parameters on mechanical performance [18,19]. Previous research has primarily focused on optimizing individual parameters, such as fiber content or layer orientation, without addressing the interactions between multiple factors [20,21].

This study seeks to bridge this gap by systematically investigating the mechanical behavior of FDM-printed PLA and PLA-CF composites. Specifically, it evaluates the influence of varying PLA/PLA-CF ratios, core materials, and nozzle diameters on tensile and flexural properties. By employing standardized ASTM D638 and D790 testing protocols, this study aims to identify optimal combinations of these parameters for applications requiring a balance between strength and flexibility [22,23]. This research contributes to the growing field of 3D-printed composite materials by providing insights into how layering strategies and material proportions can be used to enhance mechanical performance during the printing process, as opposed to the need to manufacture specialized filaments for specific applications. These findings have implications for industries such as aerospace, automotive, and biomedical engineering, where lightweight, high-strength components are in high demand [24,25].

This work introduces a novel experimental strategy that distinguishes it from the existing literature in several key aspects: (1) Layered composition strategy using off-the-shelf filaments: Instead of relying on pre-compounded filaments with fixed fiber content, our study demonstrates a layer-by-layer tuning approach by alternating PLA and PLA-CF in specific proportions within a single print. This enables property customization at the part level, which has not been extensively explored or quantified in the literature. (2) Combined effect of material ratio and nozzle diameter: Prior studies typically isolate variables such as fiber content, print orientation, or infill density. Our study simultaneously investigates the combined effect of varying nozzle diameters (0.4 mm, 0.6 mm, and 0.8 mm) and PLA/PLA-CF layer composition, providing a multidimensional understanding of how these parameters interact to influence both tensile and flexural performance. (3) Tensile–flexural trade-off mapping: We provide a detailed mapping of the mechanical trade-offs between stiffness (tensile strength) and flexibility (bending strength) across multiple composite configurations. This analysis offers a practical framework for engineers and designers to select the optimal printing parameters based on specific application requirements. (4) Bimodal AFM characterization: Our study complements macroscale mechanical tests with microscale surface characterization using bimodal atomic force microscopy. This dual-eigenmode technique provides insight into surface morphology and material phase contrast, supporting the mechanical trends observed and highlighting the microscale mechanisms behind performance differences. (5) Cost-efficient customization without specialized feedstock: This approach shows that performance optimization can be achieved without expensive or specialized filament production processes. This has real-world implications for accessible and cost-effective prototyping or production, especially for small labs, educational institutions, and startups.

2. Materials and Methods

The two materials used for the composites in this study consisted of red Bambu Lab PLA basic and black Bambu Lab PLA-CF. The specifications of the materials are provided below in

Table 1. Material properties of 3D printing polymers used.

Property	PLA (Bambu Lab Basic)	PLA-CF (Bambu Lab Carbon Fiber)
Base Material	Polylactic Acid (PLA)	PLA with chopped carbon fiber
Color	Red	Black
Melt Flow Index (MFI)	~6.0 g/10 min @ 210 °C/2.16 kg	~3.0–4.5 g/10 min (estimated)
Carbon Fiber Content	N/A	~15–20 wt%
Carbon Fiber Length	N/A	~100–150 µm (chopped, literature-based)
Density (typical)	~1.24 g/cm3	~1.30–1.35 g/cm3
Print Temperature (used)	220 °C	220 °C
Bed Temperature (used)	55 °C	55 °C

.

The samples were printed using the Bambu Lab X1C 3D printer, with an Automatic Material System (AMS) used to create the layered composites. Samples for tensile tests and three-point bending were prepared according to the ASTM D638 [22] and D790 [23] standards, respectively. Figure 1a shows the dimensions of the tensile samples use, while Figure 1b shows the dimensions of three-point bending samples used. The printing conditions are shown in

Table A1. Printing parameters for samples with PLA-CF cores.

Sample Set	% PLA	% CF-PLA	% PLA	Nozzle Diameter (mm)	Layer Height (mm)	Nozzle Temp (°C)	Bed Temp (°C)	Bed Type	Wall Loops	% Infill	Infill Pattern
4A	0	100	0	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4B	10	80	10	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4C	20	60	20	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4D	30	40	30	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4E	40	20	40	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4F	50	0	50	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
6A	0	100	0	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6B	10	80	10	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6C	20	60	20	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6D	30	40	30	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6E	40	20	40	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6F	50	0	50	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
8A	0	100	0	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8B	10	80	10	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8C	20	60	20	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8D	30	40	30	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8E	40	20	40	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8F	50	0	50	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear

and

Table A2. Printing parameters for samples with PLA basic cores.

Sample Set	% PLA-CF	% PLA	% PLA-CF	Nozzle Diameter (mm)	Layer Height (mm)	Nozzle Temp (°C)	Bed Temp (°C)	Bed Type	Wall Loops	% Infill	Infill Pattern
4B′	10	80	10	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4C′	20	60	20	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4D′	30	40	30	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
4E′	40	20	40	0.4	0.3	220	55	Textured PEI	2	100	Rectilinear
6B′	10	80	10	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6C′	20	60	20	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6D′	30	40	30	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
6E′	40	20	40	0.6	0.3	220	55	Textured PEI	2	100	Rectilinear
8B′	10	80	10	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8C′	20	60	20	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8D′	30	40	30	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear
8E′	40	20	40	0.8	0.3	220	55	Textured PEI	2	100	Rectilinear

.

The printing factors being considered were nozzle diameter, ratio of PLA basic to PLA-CF, and core material. All samples were printed to be 3 mm thick with a 0.3 mm layer height to allow for consistent and easily defined proportions between the two materials (i.e., all samples have 10 total layers). The nozzle diameters used for this study were 0.4 mm, 0.6 mm, and 0.8 mm, which were selected based on common commercially available FDM nozzle diameters capable of printing carbon fiber-embedded filaments. All of the samples were printed as sandwiches, in which the top and bottom layers were printed in one material while the middle layers were printed in the other material. The percentage of layers made of PLA basic ranged from 0% to 100% in intervals of 20%. For instance, a 40% PLA basic sample with a PLA-CF core would have two bottom layers of PLA basic, six core layers of PLA-CF, and two top layers of PLA basic. The 20% interval was selected due to it being the smallest possible interval that would allow for symmetry across the sandwich layers of the samples. For each combination of nozzle diameter and material ratio, two sets of samples were printed: the “standard” set with a PLA-CF core and the “prime” set with a PLA basic core. It should be noted that, for the samples made entirely of one material, the “standard” and “prime” variations are identical, and thus the “prime” variations were omitted. In total, five tensile samples and five three-point bending samples were printed for each combination of nozzle diameter, material ratio, and core material. A full breakdown of the printing parameters for each sample variation can be found in Appendix A.

Tensile tests were performed using the Tinius Olsen Model 100 ST, which collects the force applied (F) and the change in position of the head, which is also the change in length of the sample (ΔL). By quantifying the initial cross-sectional area (A) and length (L_o), the stress (σ) and strain (ε) at each data point can be determined using Equations (1) and (2), respectively. Based on ASTM 638 standards, the rate for tensile testing was 1 mm/min.

$$\sigma ={\displaystyle \frac{F}{A}}$$

(1)

$$\epsilon ={\displaystyle \frac{\Delta L}{L_o}}$$

(2)

By graphing the stress versus strain, the slope of the linear elastic region is the Young’s modulus, or modulus of electricity, and the maximum stress is the Ultimate Tensile Strength.

Three-point bending tests were conducted using the ADMET eXpert 5600 Series, which collects the force applied (F) as well as the distance deflected. The flexural stress (σ_f) can be determined for any load applied using Equation (3) based on the support span for the sample (L), width of the test beam (b), and thickness of the test beam (d).

$${\sigma }_f={\displaystyle \frac{3FL}{2b{d}^2}}$$

(3)

Ultimate Flexural Strength (UFS) is determined as the flexural stress at the maximum load applied. By determining the slope (m) of the linear elastic region of the force versus the deflection plot, Equation (4) can be used to find the flexural modulus (E_f) of the sample. Based on the ASTM D790 standard, the rate for the three-point bending test was 5.5 mm/min.

$$E_f={\displaystyle \frac{L^{3}m}{4b{d}^3}}$$

(4)

For each set of samples, the average UTS and UFS were compared to determine the ideal combinations of PLA basic and PLA-CF for various applications.

Additionally, microscale characterization was performed using the bimodal atomic force microscopy (bimodal-AFM) technique. AFM uses a micro-cantilever, where a laser is reflected from the back of a reflective surface. The reflection is absorbed by a photodiode detector. Using a feedback loop, the distance and dynamics of the cantilever tip and sample surface is modulated. During bimodal AFM (i.e., one of the multifrequency AFM techniques), the cantilever is excited with two frequencies simultaneously: (i) first eigenmode, and (ii) second eigenmode. The first eigenmode excitation is modulated similar to traditional tapping-mode AFM, where the oscillation amplitude is kept constant and any fluctuation dictates the tip-sample distance be adjusted using the feedback loop. The second eigenmode is excited in an open loop. Using this technique, the sample can be scanned once to obtain not only the topography (from the first eigenmode), but also the material composition from the second eigenmode. All of the measurements were performed using an Asylum Research MFP3D AFM system.

PLA and PLA-CF filaments were cut and dissolved in isopropanol alcohol. A dissolved solution of each sample was deposited on a cleaned silicon wafer. The solution was spin coated for 60 s at 1500 RPM to create a thin film. After the spin-coating process, each sample was placed in a environmentally controlled chamber with temperature of 20 °C and relative humidity of 30%. After 24 h, the samples were tested.

3. Results and Discussion

Figure 2 illustrates representative tensile stress–strain curves for PLA–carbon fiber (PLA-CF) composites printed using three different nozzle diameters: 0.4 mm, 0.6 mm, and 0.8 mm. Each subplot corresponds to one nozzle diameter and shows how mechanical behavior evolves with increasing carbon fiber contents (0%, 20%, 40%, 60%, 80%, 100%). In order to quantitively analyze this data, we have focused on the Ultimate Tensile Strength of each part, shown in Figure 3.

Figure 3 compares the UTS of each PLA-CF core sample set. Each bar represents the average UTS across five identical samples per set. The highest average UTS of 42.2 MPa in this set occurred with the samples printed at 20% PLA composition and 0.8 mm nozzle diameter. UTS trended down as percentage of PLA-CF decreased in the 0.8 mm samples. While the smaller (0.6 mm and 0.4 mm)-nozzle-diameter specimens followed this trend for samples with under 40% PLA, they showed upticks in strength as percentage PLA-CF neared 0. This behavior is seen more in the 0.6 mm set, with the highest average strength characteristic of 40.9 MPa in the 100% PLA composition samples. This indicates a tensile structural advantage to higher ratios of PLA-CF, as well as larger nozzle diameter.

Figure 4 shows the Ultimate Tensile Strength of each PLA core sample set. Each bar represents the average UTS across five identical samples per set. The highest average UTS of this group at 42.41 MPa was printed with a nozzle diameter of 0.8 mm and a composition of 20% PLA. However, compared to the PLA-CF core set, UTS does not follow the same trend downward as % PLA decreases. This group shows a more consistent strength across sample sets with both the 0.6 and 0.8 mm nozzle. This is an interesting finding, because both the PLA core and PLA-CF core sets have the same proportion of material in the cross-sectional area, which would imply that the tensile strength trends would be identical.

One possible explanation for this behavior is the influence of material continuity and fiber dispersion in the PLA-CF samples. While pure PLA tends to flow more uniformly during extrusion, resulting in better fusion between adjacent filaments and layers, the addition of chopped carbon fibers can disrupt this continuity. In PLA-CF filaments, carbon fibers can act as barriers to polymer chain entanglement and reduce the degree of interfacial bonding between printed roads. Furthermore, the fibers themselves may align randomly or become discontinuous across layers, introducing weak points in the structure. Even though the overall cross-sectional material volume is consistent across samples, the microscopic continuity—the uninterrupted material flow and cohesion—is compromised in PLA-CF parts, which may explain the deviation in tensile strength trends. This highlights the importance not only of the material composition, but also its distribution and bonding quality when evaluating mechanical performance in 3D-printed composites.

Figure 5 shows the UTS of all PLA-CF core tensile samples sorted by nozzle diameter as well as composite percentage, colored by UTS. The results demonstrate a correlation between increased nozzle diameter and the percentage PLA-CF with larger UTS.

Specimens printed with larger nozzle diameters (0.6 mm and 0.8 mm) and % PLA-CF exhibited superior tensile strength compared to those printed with smaller nozzle diameters (0.4 mm) and pure PLA. The trend again demonstrates that the integration of PLA-CF with non-carbon fiber PLA gives proportional tensile strength increase as the composition ratio changes.

Figure 6 presents the UFS results from three-point bend tests conducted on the PLA-CF core set of specimens. The data indicate that the incorporation of PLA-CF, while beneficial for tensile strength, results in a reduction in flexural strength across all tested samples. However, similarly to the tensile samples, an increase in nozzle diameter contributed to an enhancement in flexural strength. The highest UFS of 77.94 MPa in the 0.8 mm nozzle diameter and 80% PLA samples. The lowest UFS of 62.00 MPa in the 0.4 mm nozzle diameter and 0% PLA samples.

This points to larger nozzle diameters improving the layer adhesion and structural integrity during bending, whereas the stiffening effect of carbon fibers could reduce the material’s ability to flex, lowering the overall flexural strength.

Figure 7 shows the average Ultimate Flexural Strength of PLA core sample sets. The PLA-CF core bending data, similarly to the PLA core bending samples, presents a trend of increased UFS as % PLA increases. However, these data sets show larger fluctuation in UFS across composition ratios. For example, in the 0.8 mm set, the max UFS occurred in the 100% PLA samples with a max strength of 77.86 MPa, while the minimum was 67.00 MPa (40% PLA), with a difference of 10.86 MPa. In the PLA-CF core sample of Figure 6, the 0.8 mm set ranged from 77.94 MPa (80% PLA) to 71.29 MPa (0% PLA), with a difference of 6.65 MPa.

Figure 8 presents the UFS of all PLA-CF core bending samples sorted by nozzle diameter as well as composite percentage, colored by UTS. The results reinforce that, inversely to the UTS in Figure 5, decreasing the % PLA-CF increased the UFS, regardless of nozzle diameter. The plot also represents the smaller change in bending strength across composite percentages seen in this set, particularly with the larger nozzle diameters (0.8 mm).

Figure 9 provides the same data shown previously in 3D bars; however, with the focus on the standard deviation for each sample se. Figure 9a–c shows UTS of PLA-CF core samples as a function of PLA % for three nozzle diameters. Panels (a), (b), and (c) show results for 0.4 mm, 0.6 mm, and 0.8 mm nozzles. Mean UTS values at 0, 20, 40, 60, 80, and 100% PLA are plotted with error bars representing the standard deviations across five specimens. In Figure 9a, the 0.4 mm nozzle yields a maximum UTS of 40.6 MPa at 0% PLA, declines to a low of 38.1 MPa at 60% PLA, and then rises modestly to 39.3 MPa at 100% PLA. Figure 9b follows a similar trend for the 0.6 mm nozzle, beginning at 39.6 MPa, dipping to 37.5 MPa at 60% PLA, and rebounding to 40.8 MPa at full PLA. With the 0.8 mm nozzle shown in Figure 9c, tensile strength peaks at 42.5 MPa at 20% PLA before decreasing steadily to 38.9 MPa at 100% PLA. Across all three nozzle sizes, low-to-moderate PLA additions enhance tensile performance relative to high-PLA formulations, with the most pronounced effect observed for the largest nozzle diameter.

Figure 9d–f shows UTS of PLA core samples as a function of PLA % for nozzles of 0.4 mm, 0.6 mm, and 0.8 mm shown in Figure 9a, Figure 9b, and Figure 9c respectively. Bars show mean UTS at 0, 20, 40, 60, 80, and 100% PLA. In Figure 9a, the 0.4 mm nozzle UTS falls from 40.7 MPa at 0% PLA to 34.4 MPa at 20% PLA, then rises steadily to 39.3 MPa at 100% PLA. The 0.6 mm nozzle in Figure 9b reaches its lowest strength (37.7 MPa) at 40% PLA before recovering to 40.9 MPa at full PLA. In Figure 9c, the 0.8 mm nozzle peaks at 41.6 MPa at 40% PLA and decreases to 39.0 MPa by 100% PLA. Overall, adding moderate PLA content (40% PLA) effectively restores the tensile performance lost at low PLA fractions. The standard deviations shown in Figure 8 represent the repeatability of the data presented earlier in this paper, and we can conclude that we can rely on the mean values for UTS.

Similarly, Figure 9 represents UFS of PLA-CF core samples as a function of PLA content, plotted for nozzles of 0.4 mm, 0.6 mm, and 0.8 mm shown in Figure 9a, Figure 9b, and Figure 9c respectively. In Figure 10a, the 0.4 mm nozzle yields a low of 62.1 MPa at 0% PLA, rises sharply to 66.2 MPa at 20% PLA, then plateaus around 70–71 MPa from 40% PLA onward, reaching 71.8 MPa at 100% PLA. For the 0.6 mm nozzle in Figure 10b, UFS increases steadily from 71.9 MPa at 0% PLA to 75.5 MPa at 100% PLA, with notable gains at each 20% increment. In Figure 10c, the 0.8 mm nozzle shows a similar upward trend, starting at 71.4 MPa, rising through 74.0 MPa at 20% PLA, and peaking at 77.9 MPa for both 80% and 100% PLA. Overall, bending strength improves monotonically with PLA addition, and larger nozzle diameters deliver higher UFS across all PLA fractions. All curves exhibit a low dip at 40–60% PLA followed by a strong recovery at high PLA, with peak strength at 100% PLA. In Figure 10a, the 0.4 mm nozzle rises from 62.1 MPa at 0% PLA to 66.2 MPa at 20% before dipping to 63.4 MPa at 60% and climbing to 71.8 MPa at 100% PLA. The 0.6 mm nozzle in Figure 10b follows a similar pattern, dropping to 67.3 MPa at 40% and rebounding to 75.5 MPa at full PLA, while Figure 10c shows the 0.8 mm nozzle peaking at 78.0 MPa at 100% PLA. Larger nozzle sizes consistently produce higher UFS, and error bars denote standard deviation.

Figure 11 shows cross-sectional views of fractured tensile specimens printed at three nozzle diameters (0.4, 0.6, and 0.8 mm) with PLA-CF content varying from 100% to 0% in 20% increments. Black regions indicate PLA-CF, while red regions correspond to PLA. Each sample was arranged in a “sandwich” configuration, where one material formed the core and the other served as the outer layers. Larger nozzle diameters (particularly 0.8 mm) appear to produce smaller gaps in extrusion that can improve bonding. The regions of PLA-CF show more fractures with no elongation of extrusions, suggesting their more brittle nature. Conversely, PLA sections display deformities and elongations, indicative of higher ductility, but reduced tensile strength. These images support the primary trends discussed, revealing how fiber reinforcement improves tensile properties at the expense of flexural capacity.

To investigate how microstructural characteristics influence mechanical performance in 3D-printed PLA and PLA-CF components, we performed a correlation analysis between three key microstructure metrics—void fraction, mean solidity, and perimeter-to-area ratio—and two primary mechanical properties—Ultimate Tensile Strength (UTS) and Ultimate Flexural Strength (UFS). These metrics were obtained through image analysis of cross-sectional prints, and the resulting relationships are visualized in Figure 12 (six-panel scatter plot) and summarized in

Table 2. Summary of results. (sample numbers with + refer to reversed order of PLA and PLA-CF).

Sample	Nozzle Diameter (mm)	% PLA	Outside Material	Void Fraction (%)	Mean Solidity	Perimeter-to-Area Ratio	UTS ± SD (Mpa)	UFS ± SD
4A	0.4	0	PLA	76.15	0.74	0.40	40.63 ± 0.50	62.00 ± 0.46
4B	0.4	20	PLA	84.30	0.74	0.42	39.02 ± 0.42	66.24 ± 0.78
4C	0.4	40	PLA	79.67	0.75	0.45	38.68 ± 0.29	70.32 ± 1.17
4D	0.4	60	PLA	72.37	0.79	0.48	38.08 ± 0.35	70.92 ± 0.43
4E	0.4	80	PLA	63.84	0.84	0.37	38.54 ± 0.33	70.86 ± 0.56
4F	0.4	10	PLA	56.86	0.90	0.26	39.23 ± 0.16	71.76 ± 0.16
6A	0.6	0	PLA	72.50	0.78	0.42	39.63 ± 0.87	71.94 ± 0.38
6B	0.6	20	PLA	85.37	0.77	0.43	39.51 ± 0.13	73.13 ± 0.91
6C	0.6	40	PLA	82.81	0.82	0.37	39.26 ± 0.19	74.31 ± 0.53
6D	0.6	60	PLA	74.28	0.81	0.42	37.59 ± 0.18	74.10 ± 0.54
6E	0.6	80	PLA	66.82	0.83	0.28	37.58 ± 0.38	74.90 ± 0.47
6F	0.6	100	PLA	54.69	0.83	0.37	40.87 ± 0.59	75.42 ± 0.39
8A	0.8	0	PLA	69.08	0.80	0.42	41.41 ± 0.68	71.29 ± 0.75
8B	0.8	20	PLA	82.41	0.79	0.38	42.41 ± 0.25	73.81 ± 0.67
8C	0.8	40	PLA	81.27	0.83	0.38	40.63 ± 0.50	76.07 ± 0.29
8D	0.8	60	PLA	73.77	0.81	0.37	40.06 ± 0.57	76.65 ± 0.81
8E	0.8	80	PLA	66.28	0.85	0.24	39.01 ± 0.57	77.94 ± 0.29
8F	0.8	100	PLA	56.52	0.91	0.26	38.99 ± 0.28	77.86 ± 0.44
4B+	0.4	20	PLA-CF	81.96	0.75	0.44	34.40 ± 0.77	65.79 ± 0.49
4C+	0.4	40	PLA-CF	82.78	0.76	0.49	36.43 ± 0.40	64.04 ± 0.17
4D+	0.4	60	PLA-CF	73.85	0.83	0.35	38.26 ± 0.35	63.34 ± 0.35
4E+	0.4	80	PLA-CF	64.27	0.82	0.41	39.10 ± 0.39	66.07 ± 1.52
6B+	0.6	20	PLA-CF	87.80	0.75	0.46	38.66 ± 0.36	70.56 ± 0.37
6C+	0.6	40	PLA-CF	81.38	0.82	0.40	37.61 ± 0.61	67.13 ± 0.57
6D+	0.6	60	PLA-CF	73.36	0.83	0.42	38.39 ± 0.28	66.66 ± 0.51
6E+	0.6	80	PLA-CF	64.46	0.82	0.44	37.40 ± 0.40	71.13 ± 0.33
8B+	0.8	20	PLA-CF	90.30	0.82	0.40	39.40 ± 0.43	71.16 ± 0.39
8C+	0.8	40	PLA-CF	82.37	0.80	0.44	41.58 ± 0.97	69.00 ± 0.64
8D+	0.8	60	PLA-CF	73.56	0.77	0.47	39.70 ± 0.63	71.09 ± 0.51
8E+	0.8	80	PLA-CF	65.00	0.82	0.43	40.16 ± 0.38	75.32 ± 0.24

.

The analysis revealed that Ultimate Flexural Strength (UFS) is generally more sensitive to microstructural quality than Ultimate Tensile Strength (UTS). The strongest correlation was found between mean solidity and UFS (r = +0.50), indicating that well-defined and highly fused bead structures significantly enhance resistance to bending. Similarly, perimeter-to-area ratio, a proxy for interfacial irregularity or weak bonding, showed a moderate negative correlation with UFS (r = −0.47), highlighting the detrimental effect of rough or jagged filament deposition on flexural performance.

Void fraction had a moderate negative correlation with UFS (r = −0.33), as expected; increased porosity reduces structural continuity and compromises the ability to resist bending stresses. In contrast, the correlation between void fraction and UTS was weak (r = −0.06), suggesting that tensile strength may be governed more by filament alignment, interlayer adhesion, or material composition than by void content alone.

Interestingly, the correlation between mean solidity and UTS was also weak (r = +0.11), and the correlation between the perimeter-to-area ratio and UTS was modestly negative (r = −0.13). These findings indicate that, while microstructural metrics play a role in tensile performance, their influence is notably more pronounced under flexural loading conditions.

Taken together, this analysis underscores the importance of optimizing microstructural quality—particularly layer deposition orientation and geometric regularity—to improve the flexural performance of FDM-printed components. It also highlights the need to consider different loading modes when selecting printing parameters or material blends for engineering applications.

Figure 13 represents the bimodal AFM study on PLA-CF and PLA thin-film samples. Figure 12a,b represent the topography images. It is clear from images that the fibers are present in Figure 12a. The porous surface is due to the surface reacting to humidity, although it was kept at relative low humidity (~30%). The phase images show the material composition. The brighter colors represent softer materials, and darker colors preset stiffer materials. It is clear that, for PLA-CF samples, humidity is absorbed around the junction of the PLA polymer and CF. Additionally, the PLA sample seems to have a stiffer surface compared to PLA-CF; however, this does not necessarily hold for bulk properties, since AFM is only limited to surface measurements.

The bimodal AFM images of spin-coated thin films (Figure 13) provide microscale insight into material surface morphology and phase stiffness contrast. In PLA-CF films, fibers can be seen disrupting the surface topography, and phase images indicate heterogeneous stiffness—particularly at fiber–polymer interfaces, where moisture accumulation is more pronounced. PLA films, in contrast, display a smoother and more uniform phase contrast, indicative of more homogeneous material properties.

These findings support our hypothesis that fiber-induced microstructural discontinuities contribute to the mechanical behavior observed: specifically, the lower interfacial bonding and increased brittleness in PLA-CF-rich composites. This added analysis strengthens the connection between the mechanical testing results and underlying material characteristics, aligning macro- and microscale observations.

In FDM 3D printing, interlayer bonding and intra-layer cohesion are primarily governed by thermal diffusion, polymer chain entanglement, and surface energy compatibility between layers: (1) Thermal diffusion and interfacial fusion: When a new layer is deposited, its temperature affects how well it fuses with the underlying layer. PLA, being amorphous and more thermally conductive than fiber-filled PLA-CF, allows for better polymer chain diffusion across layer boundaries, which supports stronger interlayer bonding. In contrast, PLA-CF contains short carbon fibers that act as thermal sinks, disrupting heat transfer and lowering the extent of interlayer diffusion. (2) Polymer chain mobility and fiber interference: In PLA-CF composites, carbon fibers introduce mechanical discontinuities and restrict polymer chain mobility, reducing entanglement at the interface. This limited chain interdiffusion leads to weaker adhesion between adjacent roads and layers—particularly when carbon fiber content is high—contributing to brittleness and lower flexural strength. (3) Anisotropy and material compatibility: When PLA and PLA-CF are printed together, the difference in surface energy, melt viscosity, and thermal shrinkage can result in interfacial stress buildup during cooling, potentially leading to microvoids or weak bonding at the interface. These mismatches also explain why the layered strategy with mid-range ratios (40–60%) performed better: it balances stiffness without inducing excessive interfacial incompatibility. (4) Effect of nozzle diameter on thermal mass: Larger nozzles extrude material with higher thermal mass, promoting deeper interdiffusion and better bonding—especially important in composites where fiber interference is a concern. By explicitly considering these mechanisms, our findings demonstrate that mechanical performance is a function not only of material composition, but also of how bonding at the layer and material interfaces evolves during the deposition process.

The findings of this study provide a comprehensive understanding of the mechanical behavior of PLA-CF/PLA composites under different conditions, offering insights into their potential applications and limitations. The key observations from tensile and flexural testing indicate distinct trade-offs in mechanical performance based on the material composition and printing parameters.

3.1. Influence of Material Composition

One of the primary findings of this study is the proportional increase in tensile strength with higher PLA-CF content. This behavior aligns with the existing literature, which highlights the role of carbon fibers in improving stiffness and tensile modulus in polymer composites. The stiffening effect of PLA-CF, attributed to the high aspect ratio of the fibers and their interaction with the PLA matrix, contributes to load transfer efficiency and reduced deformation under tensile loads.

However, the flexural strength results demonstrate a contrasting trend, with higher PLA-CF content leading to decreased bending strength. This phenomenon can be attributed to the inherent brittleness of carbon fibers, which impairs the composite’s ability to absorb energy under bending stress. The findings suggest that applications requiring flexibility, such as biomedical related devices or flexible automotive components, may benefit from a lower PLA-CF content, whereas high-stiffness applications, such as aerospace and tooling, can leverage higher reinforcement ratios.

3.2. Effect of Nozzle Diameter

This study indicates that increasing nozzle diameter improves both tensile and flexural properties, likely due to enhanced layer adhesion and reduced void formation. Larger nozzle diameters promote better fusion between layers seen in the microscopic analysis, mitigating the delamination issues commonly observed in FDM-printed parts. These results align with prior research that emphasizes the importance of optimal extrusion settings in achieving superior mechanical performance.

Interestingly, tensile strength showed a more pronounced improvement with increased nozzle diameter compared to flexural strength. This observation suggests that tensile loading benefits more from improved interlayer bonding, while flexural performance remains sensitive to factors such as fiber orientation and load distribution.

3.3. Layering Strategy and Core Material

The alternating layering strategy employed in this study demonstrates the potential to customize mechanical performance by strategically distributing PLA and PLA-CF layers. The results show that mid-range ratios (40–60% PLA-CF) provide a balance between tensile and flexural properties, making them ideal for applications that require a combination of strength and flexibility.

Further investigation into layer orientation and core–shell configurations could provide additional avenues for enhancing mechanical properties. For instance, employing gradient material distribution or incorporating multi-material printing techniques could optimize performance for complex loading conditions.

3.4. Practical Implications

The implications of these findings are significant for industries exploring the adoption of PLA-CF composites in FDM 3D printing. Aerospace and automotive sectors, which prioritize lightweight and high-strength materials, can benefit from optimized compositions with higher PLA-CF ratios. In contrast, consumer goods and medical applications, which require compliance and impact resistance, may favor compositions with higher PLA content.

Moreover, the ability to tailor mechanical properties through printing parameters allows for the creation of functionally graded structures that meet specific design requirements. This customization capability aligns with the growing trend of personalized manufacturing in the 3D printing industry.

3.5. Limitations and Future Work

The design of our samples aimed to reflect a general-purpose print configuration suitable for consistent comparison across our sample sets. As a result, the mechanical testing incorporated both X–Y and Z build directions. Given this hybrid orientation, it was anticipated that the Ultimate Tensile Strength (UTS) values would fall between the expected extremes: approximately 89 MPa for optimal X–Y orientation and 49 MPa for Z-direction prints. The measured UTS values, ranging from 62 to 72 MPa, align well with this expectation and support the validity of our experimental approach.

An interesting observation from our results is that the PLA samples outperformed the PLA-CF samples in bending tests, which contrasts with trends reported in manufacturer technical data sheets. One likely explanation for this lies in the post-processing conditions. According to the technical specifications, the reported values for both materials were obtained after annealing and drying at 55 °C for eight hours. It is plausible that PLA-CF, which contains carbon fiber reinforcements, benefits more significantly from this thermal treatment due to improved fiber–matrix bonding. Our samples were tested as-printed, without post-processing, which may explain the reduced performance of PLA-CF relative to PLA in our data. This discrepancy highlights the potential importance of annealing, particularly for fiber-reinforced materials, and suggests a direction for future studies.

Despite the promising results, certain limitations must be acknowledged. This study primarily focuses on static mechanical properties, and future research should explore dynamic loading conditions, such as fatigue and impact resistance. Additionally, investigating the effects of environmental factors, such as temperature and humidity, could provide a more comprehensive understanding of the composite’s performance in real-world conditions.

Further research should also explore alternative fiber reinforcements, such as glass or aramid fibers, to compare their mechanical properties with PLA-CF. The development of hybrid composites with synergistic reinforcement strategies could offer enhanced performance across multiple properties. Additionally, a finite element analysis that can study the stress distribution on the samples and individual layers could be valuable for the field.

This study demonstrates that increasing the proportion of carbon fiber-reinforced PLA (PLA-CF) in FDM 3D-printed composites enhances tensile strength while reducing flexural strength, thereby highlighting a trade-off between stiffness and flexibility. Mixing PLA and PLA-CF within a single FDM print is shown to be a viable strategy for customizing both tensile and flexural properties. Higher fiber content increased the Ultimate Tensile Strength, particularly when larger nozzles were used to improve interlayer bonding. In contrast, flexural strength declined at elevated PLA-CF ratios, although mid-range blends of approximately 40–60% PLA-CF provided a more balanced mechanical profile. These findings underscore the value of tailoring fiber composition during printing, rather than relying on pre-blended filaments, and emphasize the significance of nozzle diameter in optimizing interlayer adhesion. Future investigations are encouraged to include microscopic analyses of fracture surfaces and to explore additional print parameters, such as layer orientation and post-processing methods.

From a cost perspective, standard PLA remains the most economical option for 3D printing, typically priced around USD 20–25 per kilogram. Commercially available PLA-CF filaments, which are pre-compounded with a fixed percentage of chopped carbon fibers (commonly 15–20%), are significantly more expensive, ranging from USD 40 to USD 60 per kilogram due to the added material cost and specialized manufacturing. In contrast, customizing the carbon fiber content by blending raw PLA pellets with carbon fiber filler offers potential cost flexibility, particularly for larger-scale or research applications. While the raw materials for blending (PLA pellets and CF powder or short fibers) may be obtained at lower bulk prices, the blending process itself introduces additional costs in terms of equipment (e.g., filament extruders), labor, and quality control. Moreover, achieving uniform dispersion and consistent mechanical properties through in-house compounding can be technically challenging. Therefore, while custom blending allows for tunable properties and potentially optimized cost-performance ratios, it is generally more suitable for experimental setups or high-volume production environments where the overhead can be justified. For typical users or small-scale production, commercial PLA-CF filaments remain the most practical option, despite the higher per-kilogram cost.

4. Conclusions

This study presents a systematic investigation of the mechanical and microstructural behavior of FDM-printed PLA and PLA-CF composites under variations in nozzle diameter and material composition. Tensile and flexural tests were conducted in parallel with cross-sectional image analyses to uncover the underlying microstructural mechanisms influencing mechanical performance.

Our results confirm that increasing nozzle diameter generally improves both Ultimate Tensile Strength (UTS) and Ultimate Flexural Strength (UFS), with the effect being more prominent on flexural performance. The addition of carbon fibers enhances stiffness and strength, but introduces challenges such as increased brittleness and irregular bead formation. Cross-sectional imaging revealed notable differences in porosity, inter-bead bonding quality, and filament shape across samples.

To deepen our understanding, we conducted a quantitative correlation analysis between microstructural metrics—void fraction, mean solidity, and perimeter-to-area ratio—and mechanical properties. The analysis showed that UFS is more sensitive to microstructural quality than UTS. Specifically, mean solidity exhibited a strong positive correlation with UFS (r = +0.50), while the perimeter-to-area ratio and void fraction had moderate negative correlations (r = –0.47 and –0.33, respectively). These findings demonstrate that smoother, well-fused bead geometry and reduced void content enhance flexural performance. In contrast, UTS was only weakly influenced by these metrics, suggesting that tensile strength may depend more on factors such as material composition, interlayer adhesion, and filament orientation.

This study provides an integrated methodology for linking microstructure to mechanical performance in FDM-printed parts. The insights gained here can inform future process optimization and material selection strategies, particularly in applications where flexural reliability is critical. Furthermore, the image analysis pipeline and correlation framework developed in this work offer a valuable tool set for researchers and practitioners working in additive manufacturing.