Evaluation of the Influence of Technological Parameters of Selected 3D Printing Technologies on Tribological Properties

Abstract

The article presents the results of tribological research of sample models manufactured using three separate 3D printing technologies: selective laser sintering—SLS, photo-curing of liquid polymer resins—PolyJet Matrix (PJM) and fused deposition modeling—FDM. The impact of process parameters (printing direction, layer thickness, and energy density for SLS) on tribological properties was assessed through linear wear and coefficient of friction. The research was carried out to assess the possibility of using 3D printing for the quick manufacturing of casting models, which has a significant impact on shortening the time of implementation for mass production of the casting process. The research results proved the possibility of controlling the technological process in a manner allowing to produce models with controlled properties, including tribological parameters. In addition, the results for three additive technologies and different materials were compared by using the same friction parameters.

Introduction

The dynamic development of unconventional manufacturing technologies,¹ including 3D printing, determined by the ongoing industrial revolution 4.0,² means that the current research in this area is focused on new materials and their functional properties as well as innovative additive technologies and their practical applications, for instance, in foundry,^3–6 medicine,^7–9 automotive industry,^10,11 the aerospace industry¹² and in the production of complex Freeform surface models.¹³

For this purpose, works are carried out that are aimed at determining not only the mechanical properties,¹⁴ dimensional and shape accuracy,¹⁵ and tribological properties,¹⁶ but also the impact of technological parameters on the quality of the manufactured models and control of the technological process in a manner allowing for an effective quality management¹⁷ of manufactured models.

The study of new materials used in 3D printing is frequently based on the analysis of the initial two with the omission of tribological analysis, which means that this specific issue is not sufficiently discussed in scientific publications. In addition, in the case of 3D printing technology, the analysis of the impact of technological parameters¹⁸ on the properties of the manufactured models is a particularly important issue.

Currently, 3D printing is used in increasingly newer applications—one of the promising ones involves its use for the construction of composite models with the additional use of electrospinning technology^19,20 as well as textiles.^21,22 Studies illustrating such use were presented in Kozior et al.,²³ where the authors examined the possibility of using two 3D printing technologies— PolyJet Matrix (PJM) and fused deposition modeling (FDM) to build composite models, and later to combine them with the electrospinning technology.

Moreover, the work analyzes the influence of technological parameters on the quality of the combined composite model. The presented research results clearly indicate new paths of the 3D printing application, for instance, in medicine or foundry, which certainly requires further extended research in the area of tribological wear, which is covered in part in this publication.

The analysis of the current state of knowledge on the tribological properties of models produced by 3D printing is presented in several scientific papers.²⁴ Research has shown that the tribological performance of 3D-printed parts can be enhanced by optimizing the process parameters. A study by Roy and Mukhopadhyay²⁵ used ABS and PLA thermoplastics to print samples of varying parameters (layer thickness, infill pattern, infill angle, and build orientation), which were later subject to testing for friction and wear in a multi-tribo tester in dry conditions.

The multi-tribo tester's block-on-roller configuration has been selected for tribological investigations. They discovered that PLA and ABS samples showed similar tribological properties under similar printing conditions. Another study by Srinivasan et al.²⁶ evaluated the influence of FDM process parameters (infill density, layer thickness, and infill pattern) on the wear strength of ABS. They found that layer thickness and infill density showed a major impact.

The wear rate and specific wear rate were enhanced with a lower layer thickness, higher infill density, and using the grid pattern. Norani et al.²⁷ determined that the coefficient of friction and wear rate of acrylonitrile butadiene styrene polymer were affected by the nozzle temperature, layer height, and printing pattern. The coefficient of friction and wear rate were assessed using a pin-on-disk tribometer. The pin samples were created with 3D printing using the fused filament fabrication (FFF) method.

A layer height of 0.10 mm and a nozzle temperature of 234°C using the triangle printing pattern was found to minimize the friction coefficient and wear rate. Amrishraj and Senthilvelan²⁸ used twin screw extrusion to prepare ABS composites reinforced with PTFE and nano zirconia, which showed improved wear resistance. The addition of nano zirconia and PTFE was found to be the most significant factor affecting wear, followed by load and sliding distance. ABS reinforced with 2.5% PTFE and 3% nano zirconia exhibited the least wear and coefficient of friction at a velocity of 2 m/s and a load of 10 N, as determined by desirability-based optimization.

Dangnan et al.²⁹ conducted a study on the friction behavior of additive manufacturing (AM) polymers using the UMT Tribo Lab in a ball-on-plate configuration under dry sliding conditions. They manufactured the parts using the Polyjet D photopolymer technique. The results showed that the frictional performance of the AM polymers was strongly dependent on the surface orientation under a low load of 1 N, where surface asperities played a significant role during reciprocating sliding.

However, under higher loads of 5 and 10 N, bulk mechanical properties had a greater impact on the coefficient of friction values compared with surface roughness. They also found that a low shear tribofilm formation caused a reduction in the coefficient of friction at low loads. In addition, they observed that Verogray PJM 3D-printed parts achieved minimum wear in the parallel orientation across the load range. In a separate study, Rudnik et al.²⁴ evaluated the tribological performance of MED610, a bio-medical material suitable for medical and dental applications.

They described the wear process using variable friction parameters in the pressure-rotational speed (P-V) system. While the results showed that MED610 had good abrasion resistance, it was not suitable for heavily loaded friction nodes, as the sample models were destroyed even at relatively low values in the pressure–velocity diagram.

In his research, Lorusso³⁰ found that parts produced using laser powder bed fusion technology possess higher resistance to wear and lower friction coefficient compared with parts manufactured using traditional techniques in similar dry conditions and boundary lubrication. The primary reason for this outcome is the small size of the grains and their high hardness. Porosity was also identified as a critical factor affecting wear behavior. The existence of pores weakens the bond between molten pools, leading to the formation of cracks under stress. Surface waviness and roughness also play a crucial role in determining friction coefficient.

Calignano, in his study,³¹ concluded that the quality of the surface in such parts depends heavily on the STL file parameters, orientation, position, and thermal stress during the printing process. Moreover, the thickness of the powder layer significantly influences roughness, which increases due to partially fused particles adhering to the molten part. Kalani et al.³² conducted research on the suitability of 3D-printed gears made of PA-12 Nylon material for use in polymer gearing.

They utilized an AM technique called Selective Laser Sintering (SLS) to create the gears and subjected them to varying torque values of 1.6, 2.0, and 2.4 Nm subject to various rotations per minute values (rpm) of 800, 1000, 1200, and 1400. The gears were subject to 1 × 105 load cycles, and the researchers examined their thermal and wear characteristics. The gears failed when subject to high torque of 2.4 Nm and high rpm of 1400 due to an increase in contact surface. The study revealed that the durability of the gear decreases along with the increased torque.

Further, the specific wear rate is higher at low rotational speed compared with high speed. At low torque and low rotational speed, the gears operate smoothly without any significant signs of failure. In a separate study, Ma et al.³³ examined the tribological behavior of Polyamide 12 (PA12) composites reinforced with micron-silicon carbide (SiC) and nanometer SiC and SiO₂ particles, prepared using SLS, under seawater-lubricated conditions.

The researchers investigated the seawater absorption, contact angle, hardness, and tribology performance of the composites. The results showed that the addition of micro SiC particles reduced the wear loss and friction coefficient under seawater conditions, whereas the addition of nano-SiC and SiO₂ particles increased these values. The sample printed with recycled powder had a higher friction coefficient but better wear resistance, although with an increase in the depth and width of the wear track in some areas.

In a study by Yu et al.,³⁴ the researchers investigated the tribological properties of composites made of PA12 matrix with added SiC particles, which were manufactured using an SLS system. They also examined pure PA12 for comparison. The results of the study indicated that the composite had improved friction and wear resistance compared with the PA12 matrix.

Specifically, the friction coefficient decreased by ∼10%, and the specific wear rate decreased by 20% after the addition of 10% weight of SiC to PA12. The worn surface of the composite showed deformed asperities, which were able to withstand increased tangential load and resulted in a lower specific wear rate. The study also revealed that the content of SiC particles on the surface decreased after friction tests. Through analysis of energy dispersive spectroscopy (EDS), SEM, and fourier transform infrared spectroscopy (FTIR) results, the researchers concluded that the wear mechanisms were due to abrasive and fatigue modes.

The use of 3D printing technology in the foundry industry is increasing each year. At the same time, research work is being carried out to illustrate the technological possibilities of producing accurate casting models using 3D printing. Papers^3–5 present the preliminary results of the authors' research on the dimensional and shape accuracy of casting models produced with selected 3D printing technologies.

The results of these tests showed that the influence of the printing direction and other technological parameters on the geometric accuracy of the manufactured models and the quality of the technological surface layer is clearly visible. Therefore, this article chose to present the results of tribological tests of samples produced by three 3D printing technologies (SLS, PJM, and FDM) from the following materials: polyamide PA 2200, liquid resin FullCure 720, and ABS, respectively.

It seems that the analysis of dimensional accuracy combined with the presented tribological results conducted preliminarily is going to allow for a more complete assessment of the possibility of using the research results in industrial practice for the quick construction of foundry models, in particular for disposable sand moulds. The analysis of the test results allows to determine the specific cases where the use of specific 3D printing technology is worthwhile to construct casting models of specific technological parameters.

This analysis can be used in industrial applications to optimize production, which fits into the realities of the ongoing industrial revolution 4.0 and activities carried out as part of lean optimization—LEAN manufacturing. The production of casting models using 3D printing technology and appropriately set process parameters can also contribute to reducing electricity consumption and production time as well as the implementation of projects for mass production.

Materials and Methods

This work uses three well-known 3D printing technologies for building sample models: SLS, Photocuring of Liquid Polymer Resins—PJM and FDM. These specific technologies were selected due to the fact that they are the most frequently used 3D printing methods in industrial practice and allow for the production of models from both well-known construction materials such as polyamides and ABS, as well as liquid polymer resins with variable physicochemical properties giving the ability to mix and create new materials.

These technologies are also characterized by high dimensional stability, high strength properties, and dimensions of the working platform enabling the production of small and medium casting models.

Samples preparation

All samples for testing were manufactured in the shape of a ring (Fig. 1) with predetermined technological parameters: printing direction 0° and 90° and two-layer thicknesses. The layer thicknesses were different for all technologies, but each time the smallest values recommended by the manufacturer of 3D printers were used and increased to higher values, as permitted by the manufacturers of 3D printers. Table 1 presents the technological parameters of the sample models. In the case of the SLS technology, an additional analyzed parameter included the variable energy density supplied to the sintered powder layer, also set at two degrees of variation.

FIG. 1.

Samples dimensions.³⁵

Table 1.

Technological Parameters for Manufacturing All Samples

No.	Technology	Technological parameters for manufacturing
		Lt, mm	Pd, °	Ed, J/mm2
1 a, b, c	FDM	0.254	0	N/A
2 a, b, c		0.254	90
3 a, b, c		0.330	0
4 a, b, c		0.330	90
5 a, b, c	PJM	0.016	0
6 a, b, c		0.016	90
7 a, b, c		0.032	0
8 a, b, c		0.032	90
9 a, b, c	SLS	0.100	0	0.056
10 a, b, c		0.100	90
11 a, b, c		0.200	0
12 a, b, c		0.200	90
13 a, b, c		0.100	0	0.076
14 a, b, c		0.100	90
15 a, b, c		0.200	0
16 a, b, c		0.200	90

Ed, energy density; FDM, fused deposition modeling; Lt, layer thickness; Pd, printing orientation; PJM, PolyJet Matrix; SLS, selective laser sintering.

The shape of the test samples was designed in accordance with the manufacturer's recommendations of the T-15 ring-disk type tribological tester used during the tests. The dimensions of the test samples are shown in Figure 1. Approximations of the CAD model were carried out in SolidWorks software using a triangle grid by saving STL files in an appropriate manner.

The approximation is shown in Figure 2, where the total number of triangles with which the model was approximate was 5764. At the same time, the presented figure contains information on the accuracy parameters of the approximation. For each technology, the samples were placed on a platform for building 3D printers with two printing angles: 0° and 90° (Fig. 3), to thoroughly analyze the impact of technological parameters on tribological properties.

FIG. 2.

STL model of the samples with parameters.

FIG. 3.

Model orientation on the 3D printer platform.

Selective laser sintering

The SLS technology is one of the most frequently used in industrial applications. This is due to the exceptionally large number of available materials with high properties. Currently, the SLS technology for plastics has materials based on polyamides PA 6, PA 11, and PA 12, often enriched with additives that reduce flammability (mainly for the aviation industry), increase mechanical properties (additives of carbon and glass fibers), as well as materials for special purposes, for example for foundry applications.

In the case of the SLS technology, 3D printer with the trade name of Formiga P100 (EOS GmbH, Krailling, Germany) was used to build the sample models, and the material in the form of polyamide PA 2200 (based on PA 12) was also used. The selected mechanical properties of the material used are shown in Table 2.

Table 2.

Mechanical Properties of PA 2200 Material

Mechanical properties	Test methods	Value (xz axis)
Tensile modulus	ISO 527	1650 MPa
Tensile strength	ISO527	48 MPa
Strain at break	ISO 527	18%
Shore D hardness (15s)	ISO 868	75
Biocompatible according to ISO 10993-1 i USP/level VI/121°C

The SLS technology is based on the principle of sintering thin layers of material using an energy source in the form of a CO₂ laser. This method of construction involves the need to take into account the impact of numerous technological parameters in the manufacturing process that affect the quality of the manufactured models. Therefore, in the case of SLS technology, this work analyzes the influence three technological parameters determined at two degrees of variability: layer thickness: 0.1 and 0.2 mm, printing direction 0° and 90°, and energy density of 0.056 J/mm² (P—21 W, v—2500 mm/s) and 0.076 J/mm² (P—24 W, v—2100 mm/s). The energy density was determined from the known Equation (1).³⁶

$$Ed=\frac{P}{vh}x$$ (1)

where: Ed—energy density (J/mm²), P—laser power (W), v—laser speed (mm/s), h—hatch distance (0.25 mm), d—diameter of focused beam (0.42 mm), and x—beam overlay ratio (1.68).

PolyJet matrix

Due to the very small thickness of the built layer, as little as 0.016 mm, the technology of photocuring liquid polymer resins is one of the most accurate 3D printing methods. This method allows for the production of models of very complex shapes made of a number of materials on different places of the object simultaneously.

The technological process of 3D printing consists of jetting liquid droplets of resin in the place of the currently constructed cross-section of CAD-3D models followed by initiating the polymerization process using ultraviolet lamps. A 3D printer from Object (currently Stratasys Ltd.) under the trade name Connex 350 was used to build samples in this technology.

The material used for the construction of the samples was liquid FullCure 720 resin with the properties and composition presented in Tables 3 and 4. The samples were made in four variants of the selection of technological parameters, where two-layer thicknesses of 0.016 and 0.032 mm were used, as well as two variants of location on the building platform 0° and 90°.

Table 3.

Mechanical Properties of FullCure 720

Mechanical properties	Test methods	Value
Tensile strength	ASTM D638	50–65 MPa
Flexural strength	ASTM D790	80–110 MPa
Elongation at break	ASTM D638	15–25%
Modulus of elasticity	ASTM D638	2000–3000 MPa

Table 4.

Chemical Composition of FullCure 720

CAS	Components	Percentage
	FullCure 720
—	Acrylic monomer	<30
588-33-5	Isobornyl acrylate	<25
—	Phenol, 4,4′-(1-methylethylidene)bis-, polymer with (chloromethyl)oxirane, 2-propenoate	<15
—	Phosphine oxide, phenylbis(2,4,6-trimethylbenzoyl)-	<2
52408-84-1	Acrylic acid ester	<0.3

Source: Kozior and Kundera.³⁷

Fused deposition modeling

The Fused Deposition Modeling technology known as FDM or FFF is one of the most popular 3D printing methods on the global market. This is due to the very low cost of purchasing and servicing 3D printers, as well as the materials used. In addition, the materials have been used in engineering practice for a number of years and have real possibilities of practical use.

The FDM technology consists of extruding the material in the form of a rod on a spool through the 3D printer head, where it is heated to a temperature slightly lower than the melting point of the specific material, which turns them into a semi-liquid state. The material is later distributed in the place of the currently built cross-section of the model corresponding to the currently produced layer.

In the presented work, a 3D printer—Dimension 1200es (Stratasys Ltd.) and ABS P430 material with properties presented in Table 5 were used to build sample models. The samples in this technology were manufactured with a set layer thickness of 0.254 mm, which is the lowest value permitted by the 3D printer used, which was later increased to 0.330 mm in line with the machine's capabilities. In addition, the samples were made in two variants of location on the building platform 0° and 90°.

Table 5.

Mechanical Properties of ABS P430

Mechanical properties	Test methods	Value
Young's modulus	ASTM D638	2320 MPa
Tensile strength	ASTM D638	37 MPa
IZOD Impact (method A, 23°C)	ASTM D256	106 J/m

Surface texture analysis

Metrological surface measurements were made with a Leica DCM 8 microscope after all tribological tests had been carried out. The device is designed for confocal microscopy combined with interferometry. The device is equipped with lenses with the following magnification strengths: × 5, × 20, × 50, and × 150.

The system allows for the geometric assessment of the surface with a vertical resolution of 2 nm and a horizontal resolution of 140 nm. The test results were processed using MountainLab software. Measurements were carried out for each single sample at the same point, where the measurement area was 6.47 × 4.87 mm; a lens with a magnification of × 5 was used.

Tribological methodology

Tribological tests were carried out using the T-15 tribological tester (Lukasiewicz Research Network—Institute for Sustainable Technologies, Radom, Poland). The device allows to carry out tribological analysis of ring-shaped samples in a frictional pair with a disk-shaped counter-sample made of C45 steel. The view of the tribological stand together with the scheme of the test implementation is shown in Figure 4.

FIG. 4.

Research stand for tribological tests with a test scheme.

The tribometer is designed to evaluate the tribological properties of materials used for: face seals, sliding elements of machines, and metal-polymer connections. The contact geometry of the sample with the cooperating element for the T-15 tester is plane to plane. The tribological tests presented in the article were performed with constant friction parameters for all samples, that is, rotational speed V—0.25 m/s, pressure force Fp—45 N. The following measurements were carried out during the test: friction force, linear wear, and temperature in the test chamber in the friction zone.

Results

A tribometer apparatus consisting of a ring on disk configuration was employed to conduct the tests. As explained in the methodology section, three different 3D printing technologies (FDM, PJM, and SLS) were used to manufacture the samples. The experiment included testing the impact of the 3D printing technique and process parameters by changing the printing orientation, layer thickness, and energy density.

Each orientation (horizontal 0° and vertical 90°) was tested with two different layer thickness (thick and thin), whereas SLS-printed samples were evaluated using two different energy densities (0.056 and 0.076 J/mm²) in addition to various orientations and layer thicknesses. The samples were evaluated based on friction coefficient, linear wear depth, and temperature as a function of sliding time.

A constant normal load of 45 N and velocity of 0.25 m/s were applied across all tests, and dynamic coefficients of friction were obtained through analysis of recorded data. While conducting the tests, Spider 8 (measurement device) was used to record three parameters over time: friction force, displacement (or wear depth), and temperature. The sensors first measured these data, which were then transferred to a computer through the Spider 8. As seen in Figure 5a–c, the virgin data of each sample may be displayed independently against time.

FIG. 5.

Examples of measured and evaluated data following testing (a) friction force, (b) wear depth, (c) temperature, and (d) friction coefficient, all plotted against time.

To determine the friction coefficient (μ), the frictional force (F) values were divided by the normal applied loads (N) in accordance with the μ equation (μ = F/N).³⁸ The initial graphs of wear depth and coefficient of friction in Figure 5b, d, respectively, were very dense, with 3600 measurements. These curves were unsuitable for analysis or comparison with other results in a single graph.

Therefore, we filtered the data to draw a proper curve, resulting in only 45 data points remaining out of a total of 3600. This was carried out by selecting one point out of every 80 measurements, specifically the first point of each 80 measurement set while omitting the remaining 79. This method eliminated unnecessary oscillations around each point and clearly showed the tendency of the curve. Figures 6 and 7 show the plotted data of friction coefficient and wear depth, respectively, after filtration.

FIG. 6.

Friction coefficient behavior at (a) 0° print orientation samples—thicker layers, (b) 90° print orientation samples—thicker layers, (c) 0° print orientation samples—smaller layer thickness, and (d) 90° print orientation samples—smaller layer thickness.

FIG. 7.

Wear depth behavior at (a) 0° print orientation samples—thicker layers, (b) 90° print orientation samples—thicker layers, (c) 0° print orientation samples—smaller layer thickness, and (d) 90° print orientation samples—smaller layer thickness.

Figure 6a–d present the friction coefficient curves of samples prepared using different 3D printing techniques, such as FDM, PJM, and SLS, and three different materials, under different printing orientations, layer thickness, and energy density. In general, the results show that the friction coefficient was mostly consistent for most of the tested conditions, regardless of the build orientation and layer thickness.

However, significant variations were observed in the friction coefficient of PJM samples (FullCure 720) (Fig. 6b, c), which was more than double compared with the other samples. This could be due to the sticky nature of the FullCure 720 material, which enhances adhesion between the sliding surfaces, thus increasing the coefficient of friction. In addition, within each printing orientation sample, it was noticed that the friction of the FullCure 720 material friction coefficient was more fluctuating compared with the tested samples manufactured in other technologies test pieces.

This implies that the material has a more significant impact on the friction coefficient than the printing parameters evaluated (printing direction and layer thickness). In case of energy density, it was observed that SLS samples from polyamide PA 2200 irradiated with an energy density of 0.076 J/mm² had the lowest friction coefficient in all diagrams of Figure 6, regardless of printing orientation and layer thickness, indicating that higher energy density improves the tribological properties of SLS-manufactured components.

Therefore, using SLS technology with higher energy density can be recommended to produce components with better frictional characteristics for PA 2200. On the other hand, it appears that PJM-made components from FullCure 720 material are not suitable for heavy-duty tribological applications due to their unstable and high friction behavior.

The displacement results for the wear depth investigation of the samples are presented in Figure 7a–d. Most of the samples showed a wear depth ranging between 10–40 μm, which increased steadily with the sliding time. However, the FC720 samples were an exception as the wear depth either increased dramatically until the end of the test (e.g., 90° orientation at thicker layer) or showed unstable behavior (i.e., increasing, smoothing out, and later increasing again).

On the other hand, the wear remained stable over the tribology test time in the PA 2200 sample at 0° and thicker layer and ABS sample at 90° and thicker layer. Further, the PA 2200 printed samples reported the smallest wear depth values, indicating its suitability for wear-resistant uses more than the other examined techniques.

To ensure accurate results, multiple tests were conducted on three identical samples under different printing technologies and process parameters. The average value was then calculated for each set of tests. Figure 8a displays the highest friction coefficients obtained during testing under all conditions, whereas Figure 8b compares the average friction coefficient obtained from all tests.

FIG. 8.

Comparison of the average values of (a) maximum friction coefficient, (b) dynamic friction coefficient, (c) wear depth, and (d) temperature, at various print orientations and layer thicknesses.

The average friction coefficient was estimated by analyzing the stable stage of the friction behavior curve. However, some curves did not reach a stable stage, therefore the MODE.SNGL function was used in Microsoft Excel to determine the average value. This function identifies the most frequently occurring number in a dataset, which was extracted as the average friction coefficient.

Figure 8c illustrates the comparison of wear depth achieved by tribological samples produced using various AM methods and different materials. Although wear depth is typically determined by the final point of the curve, it was difficult to obtain a stable measurement due to continuous fluctuations in some cases.

For the purpose of obtaining a more accurate measurement, the final 600 measurements (taken between 3000 and 3600 s) were averaged, and the average was used to determine the wear depth value. Figure 8d displays the temperature development observed in tribological test samples tested under all experimental conditions.

Surface texture

After carrying out tribological tests, measurements of the geometric structure of the surface of the sample models were carried out. The preliminary test results and the analysis of the presented friction results allow to conclude that in general, the most favorable technological parameters of 3D printing in terms of frictional strength are the variant of the location of the sample models on the working platform of the machine with the set printing angle of 0° and the lowest possible thickness of the built layer.

Therefore, the samples were subject to an in-depth analysis of the quality of the surface geometric structure, and the results are presented in Table 6 and in Figures 9–11.

Table 6.

Surface Texture Measurements Results for Fused Deposition Modeling, PJM, and Selective Laser Sintering Technologies

Roughness parameters	Before friction, A	After friction, B	$\Delta C$ ($\Delta C=B-A$ )
FDM-ABS-0254-0-1b
Rp, μm	54.01	36.89	−17.12	−46%
Rv, μm	38.48	29.33	−9.15	−31%
Rz, μm	92.49	66.22	−26.27	−40%
Rc, μm	73.46	29.49	−43.97	−149%
Rt, μm	169.8	119.3	−50.5	−42%
Ra, μm	18.94	7.543	−11.397	−151%
Rq, μm	22.49	10.21	−12.28	−120%
Rsk	0.4469	0.2188	−0.2281	−104%
PJM-FC720-0016-0-5a
Rp, μm	9.053	14.33	5.277	37%
Rv, μm	8.733	18.79	10.057	54%
Rz, μm	17.79	33.12	15.33	46%
Rc, μm	8.636	15.92	7.284	46%
Rt, μm	23.98	51.68	27.7	54%
Ra, μm	2.662	5.002	2.34	47%
Rq, μm	3.416	6.442	3.026	47%
Rsk	0.1334	−0.3861	−0.5195	135%
SLS-PA2200-01-0-056-9a
Rp, μm	32.78	30.84	−1.94	−6%
Rv, μm	36.35	27.66	−8.69	−31%
Rz, μm	69.13	58.51	−10.62	−18%
Rc, μm	40.82	32.51	−8.31	−26%
Rt, μm	91.43	88.89	−2.54	−3%
Ra, μm	12.86	9.574	−3.286	−34%
Rq, μm	15.82	12.03	−3.79	−32%
Rsk	−0.1076	0.1743	0.2819	162%

FIG. 9.

View of the geometric surface structure for a sample 1b—FDM technology, ABS material (Lt—0.254 mm, Pd—0°). FDM, fused deposition modeling; Lt, layer thickness; Pd, printing orientation.

FIG. 10.

View of the geometric surface structure for a sample 5a—PJM technology, FullCure material (Lt—0.016 mm, Pd—0°). PJM, PolyJet Matrix.

FIG. 11.

View of the geometric surface structure for a sample 9a—SLS technology, PA 2200 material (Lt—0.1 mm, Pd—0°). SLS, selective laser sintering.

Analyzing the isometric views shown in Figures 9–11, we can clearly see that for the FDM technology and sample 1b there are traces of abrasion similar in shape to the surface after machining (milling). The nature of the abrasion gradually changes for the PJM technology and sample 5b, and in the case of the SLS technology, the abrasion traces completely disappear, which is mainly due to the tribological properties of the polyamide.

A similar tendency applies to the analysis of the two-dimensional (2D) surface profile, where the graphs clearly show that after the friction process, the surface of models made of ABS material is the most regular, and the largest irregularities in terms of amplitude and their frequency were observed for models of samples made with the use of liquid polymer resins—PJM.

Discussion

The knowledge of the friction coefficient and wear resistance of a material is essential for numerous industrial applications, as they help to determine the suitability of material for specific uses. Typically, friction curves begin with a running-in stage where friction increases rapidly. This is due to the unstable slip between the surface of the 3D-printed part and its corresponding surface (counterpart) caused by the surface roughness of the sample that has not yet been polished.³⁹ Eventually, the curve reaches a steady state where the friction oscillates near a constant level.

The results of the experiment show that the maximum friction coefficient was higher for the samples manufactured using FDM (ABS) and PJM (FC720) technologies compared with those made using the SLS (PA 2200) 3D printing method, irrespective of the energy density used (0.056 or 0.076 J/mm²). However, this trend was not observed for the SLS samples angled at 90° and thicker layer thickness (0.2 mm), where the maximum friction coefficient was significantly higher than that of other SLS samples.

Further, Figure 8a clearly shows that within each 3D printing technology and materials, the maximum friction coefficient values were almost similar, except for the FC720 samples. This difference can be attributed to the adhesion that occurs while sliding due to the properties of FC720 material.

The average friction coefficient followed a trend similar to the maximum friction coefficient, with ABS and FC720 materials showing higher values. In addition, some PJM samples, specifically the 0° orientation at smaller layer thickness (0.016 mm) and the 90° orientation at thicker layer thickness (0.032 mm), showed a significant increase in friction coefficient. Moreover, the samples with smaller layer thickness exhibited less friction than those with thicker layers, making them more suitable for certain applications.

The wear depth investigation results under all conditions (Fig. 8c) did not show any specific trend, as the wear depth varied between high and low values. However, most of the samples had a wear depth ranging between 10 and 40 μm, which can be assessed as low. The temperature evaluation shown in Figure 8d indicated a consistent increase with the testing time, which can be attributed to the increase in friction between the surface of the sample and the counterpart in tribology test over the sliding time. However, similar temperature values were reported at the end of the tests, regardless of the printing technology and process parameters.

Measurements of the geometric structure of the surface showed significant differences between the 2D roughness parameters depending on the three additive technologies—FDM, PJM, and SLS, which are presented in Table 6 for the selected samples. The highest values of surface parameters were produced for the 3D printing technology—FDM.

This may be the result of the fact that FDM technology very well shows the method of layering as well as technological losses of the material in the place of spreading the material coming out of the nozzle (building path). For samples in the PJM (FC 720) and FDM (ABS) technology, the wear trace is visible in the central part of the sample, whereas for the PA 2200, it occurs on the outer and inner edges of the tested surface.

The measurements of the geometric structure of the surface showed significant differences in the value and size of individual irregularities depending on the additive technology used and technological parameters of 3D printing, in particular the angle of the print. The quantitative analysis carried out showed that for the PJM technology, the lowest values of the analyzed 2D roughness parameters were recorded before the start of the friction tests (average values) and their increase after the end of the process. Such characteristics may be the result of the physicochemical properties of the material in the form of a liquid polymer resin, susceptible to deformation caused by temperature increase and caused by friction forces.

Similar values of individual parameters were obtained in the case of the two remaining technologies; however, in the case of the FDM technology, these values are slightly higher than for the SLS technology. When analyzing the parameters determining the height of the roughness, namely the mean height of roughness profile—Rc and the total height of profile—Rt, a significant reduction of these parameters can be noted for the FDM and SLS technologies.

In the case of the FDM technology, the Rc parameter was in some cases reduced by as much as 149%. This may be due to the friction of the abrasive pair, which resulted in micromachining and “smoothing” of the tested surfaces. Similar conclusions may be drawn by analyzing the most popular roughness parameters, that is, the Ra parameter and the Rq parameter.

Analysis of the measured values of the Rsk parameter (skewness), which determines the nature of the unevenness of the measured surfaces in detail, leads to the conclusion that the primary surface for FDM and PJM technologies (before friction) was dominated by the presence of areas with numerous sharp peaks. However, following the specific friction cycles, a decrease in the value of the Rsk parameter was noted for both technologies.

In particular, it is visible for the PJM technology where the change in the nature of the distribution of peaks and valleys in the assessed profiles was obtained. For these samples, the Rsk parameter reached negative values, which indicates the presence of a surface morphologically similar to the plateau. This may indicate that the friction process of the friction pair flattened the sharp tops.

Therefore, smoothing these asperities (peaks and valleys) during the friction cycles caused rising the wear depth of both FDM and PJM technologies' materials. On the other hand, the opposite trend was observed in the SLS technology, which interpreted the lowest coefficient of friction obtained for this material's samples. Yu et al.³⁵ conducted a study to examine the properties of composites that were made by mixing PA12 with SiC particles using SLS.

They studied the worn surface of the composite and found that the deformed asperities on it were able to bear a more tangential load, leading to a decrease in specific wear rate, which is in good agreement with the findings of the current study.

In addition, the tests carried out showed that in the case of the PJM technology and FC 720 material, the wear process has shown a change in the nature of the distribution of unevenness, which directly translated into the values of the roughness parameters of individual profiles. Only in the case of this material, an increase in the values of roughness parameters was noted due to the resulting traces of friction and the interaction of the friction pair.

The traces of friction formed on the surface determined the formation of scratches and high irregularities, which directly translated into the measurement results. This may also be due to the fact that the values of roughness parameters before the friction process for this material were the smallest among all the additively manufactured materials tested.

During the summary of the overall results of the metrological tests, it was found that the values for all manufactured samples after the friction process reached similar values regardless of the initial distribution of the samples' unevenness, which proves the high interference of the friction process on the geometric structure of the surface and the high repeatability of the process friction for the tested additive technologies.

The wear tracks shown in Figures 9–11 can give an indication of the wear mechanisms that occurred with each printing technology's material. Scratch marks are obvious on the wear tracks of the worn surface for the FDM and PJM printed samples, unlike the SLS (PA 2200). This plenty of scratches implies the occurrence of severe abrasive wear as the dominant wear mechanism.

On the other hand, since there is significant plastic deformation on the worn surface of the SLS-printed material, it can be concluded that the plastic deformation shown on the wear track is how PA 2200's abrasion took place. In addition, the wear debris formation and containment within the wear track, which functions as a lubricating film to regulate friction, is the primary difference that can be observed in the PA 2200's worn area compared with the other technologies' materials.

Studies in the literature investigated the tribological properties of 3D-printed materials and showed a similar attitude, regarding the wear mechanisms. After analyzing the EDS, SEM, and FTIR results, Yu et al.³⁵ inferred that the wear mechanisms of PA12 reinforced with SiC particles (fabricated using SLS) were a result of abrasive and fatigue modes.

Further, Dangnan et al.²⁹ studied the wear and friction of additively manufactured 3D ABS and Verogray polymers created using PJM technology. They also observed a similar abrasive wear mechanism.

Conclusions

This study investigated the tribological properties of samples fabricated by FDM, PJM, and SLS 3D printing systems with different materials. The effect of the technological parameters (printing orientation, layer thickness, and energy density) of selected techniques was evaluated. Moreover, the surface texture of the sample models was discussed. The following conclusions may be drawn in result:

• The materials used in additive processes have much more influence on the friction coefficient than the printing parameters evaluated.

• SLS technology with higher energy density can be recommended for the industry to produce better frictional characteristics parts from PA 2200 material. In contrast, the PJM components made from FullCure 720 are not to be fit for heavy duty tribological applications due to their unstable and high friction attitude.

• Smaller layer thickness samples showed enhanced (lower) friction tendency compared with thicker layers, which makes them more recommended for certain applications.

• Most of the samples offered a wear depth ranging between 10 and 40 μm by the end of the tests, which can be considered high wear strength.

• The smallest wear depth values were reported for the PA 2200 polyamide material for SLS technology (for both applied energy density), indicating the suitability of this technology and material for wear resistance uses products more than the other materials examined.

Moving forward, there is potential to expand on this study by examining how different factors such as the speed of sliding, frequency of movement, microscopic characteristics such as interlocking or welding, and the underlying microstructure of the materials used in 3D printing impact their wear and friction properties.

Authors' Contributions

T.K.: conceptualization, methodology, software, validation, formal analysis, writing—original draft preparation, writing—review and editing, supervision, and investigation. M.M.H.: software, validation, writing—original draft preparation, writing—review and editing, and investigation. P.Z.: project administration, software, and investigation. D.G.: software, investigation. M.R.: software, investigation. W.S.: software, investigation.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The research was financed by the National Centre for Research and Developments part of the Lider XI project, number LIDER/44/0146/L-11/19/NCBR/2020, under the title, “An analysis of application possibilities of the additive technologies to rapid fabrication of casting patterns.”