The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

https://doi.org/10.1007/s00170-024-13140-7

ORIGINAL ARTICLE

Impact of process parameters on improving the performance of 3D

printed recycled polylactic acid (rPLA) components
Mohammad Raquibul Hasan1 · Ian J. Davies2 · Alokesh Pramanik2 · Michele John1 · Wahidul K. Biswas1

Received: 8 September 2023 / Accepted: 26 January 2024 / Published online: 20 February 2024
© The Author(s) 2024

Abstract
The main goal of this research was to investigate the influence of additive manufacturing (AM) printing parameters on the
mechanical properties and surface roughness of specimens fabricated using recycled polylactic acid (rPLA). In order to
achieve this goal, significant printing parameters such as layer thickness, infill density, and nozzle temperature were selected
based on prior research. A three-level L9 orthogonal array, based on the Taguchi method, was used in the experimental design.
The mechanical properties of virgin PLA and recycled PLA printed specimens were examined and compared. To facilitate
the analysis of variance (ANOVA) examination, the response data for mechanical and surface roughness parameters were
transformed to signal-to-noise (S/N) ratios. The inspected responses under consideration were the surface roughness, shore
D hardness, tensile strength, flexural strength, and impact strength. The main findings suggest that careful consideration of
the layer height is crucial for achieving optimum mechanical properties in the recycled PLA specimens. Furthermore, the
nozzle temperature also played an important factor that affected the mechanical and surface roughness properties of the 3D
printed PLA specimens. Microscopic investigation demonstrated that the number and size of voids increased significantly
when the layer thickness and temperature were low, namely, 0.1 mm and 195 ℃, respectively. Finally, the optimal combi-
nation of printing parameters for each performance characteristic was determined. Following this, a confirmation test was
performed using the preferred combination of parameters, which indicated a strong correlation with the outcomes predicted
statistically. The results obtained from this study revealed that recycled PLA exhibited mechanical properties comparable to
that of virgin PLA under certain conditions. In summary, the results of this study will serve as a valuable dataset in the field
of additive manufacturing, providing valuable insights for other researchers working with recycled PLA material.

Keywords Fused deposition modelling (FDM) · Recycled polylactic acid (recycled PLA) · Taguchi design of experiment
(DoE) · Mechanical properties

1 Introduction
* Mohammad Raquibul Hasan
[email protected] Technological advancements have a substantial impact
on the competitiveness amongst manufacturers in various
Ian J. Davies
[email protected] industry sectors due to the increasing demand for high-qual-
ity customised parts that are economical and have struc-
Alokesh Pramanik
[email protected] turally sound mechanical qualities [1]. Additive manufac-
turing (AM) technology, also known as three-dimensional
Michele John
[email protected] (3D) printing, has the potential to address the mentioned
issues due to their design flexibility, manufacturing simplic-
Wahidul K. Biswas
[email protected] ity, lower production costs, and reduction of raw material
waste [2]. These advantages make 3D printing an attrac-
1
Sustainable Engineering Group, School of Civil tive alternative to conventional manufacturing methods
and Mechanical Engineering, Curtin University, Bentley, [3–6]. There are several commercially available AM pro-
WA 6102, Australia
cesses, such as inkjet modelling (IJM), fused deposition
2
School of Civil and Mechanical Engineering, Curtin modelling (FDM), stereolithography (SLA), and laminated
University, Bentley, WA 6102, Australia

Vol.:(0123456789)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3752 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

object manufacturing (LOM). However, FDM stands out conducted to improve the mechanical properties of rPLA
as a widely utilised technology that involves the extrusion by changing the process parameters. Dey et al. [24] altered
of semi-solid thermoplastic materials through a nozzle in the infill percentage, layer thickness, printing speed, nozzle
contrast to other AM techniques that employ diverse laser temperature, build orientation, and raster angle to modify
systems, powders, and resins [7–9]. The FDM process has the mechanical properties of recycled PLA and observed an
the capability to construct complex geometries using light- increase. Atakok et al. [17] conducted an experiment using a
weight materials, resulting in a notable reduction in process- commercially recycled PLA filament to examine the impact
ing time [10, 11]. of the layer thickness, infill percentage, and infill pattern on
A range of thermoplastic polymer materials, such as the mechanical properties. The findings of the investigation
polylactic acid (PLA), polyethylene terephthalate (PET), indicated that employing a layer thickness of 0.25 mm and
acrylonitrile–butadiene styrene (ABS), polypropylene (PP), an infill percentage of 70% yielded mechanical properties
and high-density polyethylene (HDPE), are the most exten- that closely resembled those of the vPLA. Correia et al. [25]
sively documented polymers employed in the FDM process determined that by utilising the optimal combination print-
[12–14]. Amongst these polymers, PLA has attracted con- ing parameters, such as 0.2 mm layer thickness, 100% infill,
siderable interest as a bio-based and biodegradable polymer and 40 mm/s travel speed, would effectively improve the
that serves as a viable substitute for petroleum-derived plas- mechanical properties of rPLA. In their study, Tan et al. [22]
tics [15]. PLA exhibits a relatively lower molecular weight determined that the optimal printing parameters for enhanc-
and low melting point, rendering it highly compatible with ing the tensile strength and flexural strength of recycled pol-
a wide range of FDM equipment [14, 16]. Moreover, it has ylactic acid (rPLA) include a fan cooling speed of 100%, an
been observed that the mechanical properties of PLA exhibit extruder temperature of 190 °C, and a bed temperature of
a higher level of performance when compared to ABS and 50 °C. Breški et al. [26] assessed that 30% infill and 0.1 mm
PETG [17]. In addition, PLA has better recyclable properties layer thickness were most favourable for higher mechanical
than other thermoplastic polymers [18]. By considering lim- properties of rPLA. These studies showed that different infill
ited material resources and the relatively high price of virgin percentages and layer thicknesses can improve the mechani-
polymers, future recycling of PLA could become economi- cal properties of rPLA 3D printed parts. Therefore, modifi-
cally attractive. Therefore, recycling of PLA for 3D printing cation and optimisation of different printing parameters can
could be a feasible option, as it offers environmental benefits improve the overall mechanical properties [27].
and comparable mechanical characteristics [19]. Numerous researchers have attempted to investigate the
Previous research has indicated that the recycling process impact of the process parameters on the behaviour of 3D
is subject to a high shear force and temperature, which facili- printed vPLA components. Lanzotti et al. [28] conducted
tate chain scission processes. This causes a reduction in the a study to examine tensile properties of 3D printed PLA
molecular weight and viscosity, which ultimately reduces components by varying the layer thickness and build orienta-
the mechanical strength hindering the use of recycled PLA tion. The findings of this study demonstrated that a decrease
(rPLA) [3, 20]. Furthermore, the anisotropic characteristics in layer thickness resulted in the attainment of the highest
of FDM produce pores, which reduce its strength and other ultimate tensile strength. Eryildiz [29] experimentally deter-
important characteristics [21]. As a consequence, whilst mined that a flat orientation showed higher tensile proper-
the utilisation of virgin PLA (vPLA) in the context of 3D ties compared to the upright build orientation due to the
printing has been extensively documented in the academic intralayer and interlayer phenomena. Benamira et al. [30]
literature, the application of recycled PLA remains rela- also printed tensile specimens in a flat position and observed
tively uncommon. Hence, proper use of recycled materials improved tensile properties. Tsouknidas et al. [31] examined
is important towards the efficacy of recycled plastic materi- the influence of various printing speeds on the compressive
als and green manufacturing techniques. strength of the PLA samples and concluded that by lowering
According to existing research, the overall mechanical the printing speeds, maximum strength could be achieved. In
performance of printed specimens can be improved by modi- their study, Carneiro et al. [32] investigated parameters such
fying various printing parameters, such as the nozzle/extru- as the infill density, layer thickness, and raster orientation
sion temperature, layer thickness/height, nozzle diameter, and found that the layer thickness had the least impact on the
infill percentage, infill pattern, printing speed, build orien- mechanical properties. In their study, Kam et al. [33] found
tation (flat, edge, and upright), raster/infill orientation, and that the infill percentage directly influenced the tensile and
top and bottom layer thicknesses [3, 17, 22]. Therefore, it is Izod impact strengths of PLA parts. This study identified a
essential to optimise and control all these factors effectively significant positive correlation between strength and infill
within the FDM manufacturing process to fabricate com- percentage. In a study conducted by Suteja and Soesanti
ponents with higher dimensional precision and enhanced [34], it was observed that the layer thickness had a nota-
mechanical functionality [23]. Several studies have been ble impact on the tensile strength. Behzadnasab et al. [35]

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3753

observed that increasing nozzle temperature from 180 to guide the researchers and encourage the manufacturing com-
240 °C increased mechanical strength in PLA 3D printed pany to fabricate products for engineering applications using
parts. Finally, Sun et al. [36] demonstrated that the bed tem- recycled PLA. This study employs the design of experiments
perature plays a crucial role in the bonding process, conse- (DOEs) technique to investigate the impact of selected print-
quently leading to an enhancement in the tensile strength. ing parameters on the surface roughness, hardness, tensile,
Regarding the optimization of printing parameters, flexural, and impact strengths. In addition, confirmation tests
researchers have utilised a range of design of experiments were conducted to validate the optimal process parameters
(DoE) methods, including full factorial design (FFD), fuzzy obtained by the DoE technique in order to propose the most
logic (FL), response surface methodology (RSM), central appropriate printing parameter as a viable manufacturing
composite design (CCD), analysis of variance (ANOVA), protocol for future process optimisation.
and the Taguchi method [10, 21, 37, 38]. Several recently
published studies have focused on optimising 3D printing
process parameters using DoE methods. In their research, 2 Materials and methods
Chari et al. [39] examined the compressive strength and
hardness of 3D printed PLA components by using the Tagu- The main objective of this study was to investigate a pre-
chi method with varying nozzle temperature, infill, and layer ferred combination of 3D printing parameters to optimise
thickness. Heidari-Rarani et al. [37] employed the Taguchi the mechanical properties of virgin PLA and recycled PLA
DoE method to examine the mechanical properties of PLA specimens using the Taguchi method. The proposed method-
specimens by modifying infill density, print speed, and ology followed in the work has been shown in Fig. 1.
layer thickness. This study found that the impact of the infill As a first step, the experimental setup was planned and
density on the mechanical characteristics was significantly prepared, wherein the control parameters for 3D printing
greater than that of the printing speed and layer thickness. were carefully chosen. The range of the control parameters
To determine the impact of printing factors on the mechani- was selected, specific levels were assigned to each parameter,
cal strength of FDM printed specimens, Sood et al. [40] and then a Taguchi orthogonal array was chosen. After the
used CCD for DoE and utilised ANOVA to optimise print- initial stage, experiments were conducted using the Taguchi
ing settings. The Taguchi approach was used by Alafaghani method. The experimental data were statistically analysed in
et al. [41] to investigate the impact of the infill pattern and the form of average values, S/N ratios, main effect plots, and
percentage, layer thickness, and nozzle temperature on the analysis of variance (ANOVA). Finally, the optimal values
characteristics of specimens made using PLA filaments. By for each control parameter were identified, and the optimised
utilising this approach, researchers aimed to gain insights results were subsequently validated through confirmation
into the optimal combination of these parameters to enhance experiments. A comprehensive description of the materials,
the overall quality of the fabricated specimens. Therefore, printing parameter selection, testing conditions, and analysis
the selection and optimisation of the FDM parameters are employed in the production and characterisation of the sam-
essential for achieving the desired technical properties of 3D ples will be provided in the subsequent subsections.
printed components.
A review of the sources mentioned above indicates that 2.1 Materials
recent scientific investigations have focused on analysing
the influence of various factors on the mechanical strength Polylactic acid is one of the most commonly used thermo-
of vPLA printed specimens, such as layer thickness, infill plastics for fused deposition modelling due to its favourable
density, infill pattern, print speed, build orientation, and noz- printing properties and ease of use as a filament in 3D print-
zle temperature. However, the mechanical strength of rPLA ing [17]. This biodegradable, thermoplastic material origi-
still remains unexplored. Additionally, no previous research nates from cleaner sources when compared to conventional
has investigated the impact of layer height, infill density, and plastics, with less hazardous components, and can be printed
nozzle temperature on rPLA 3D printed specimens, and this at lower temperatures saving energy [4]. PLA has a melting
is considered a novelty of the present work. Thus, it is neces- point of approximately 150–160 °C, can be easily processed
sary to conduct an in-depth analysis of the effects of layer at 190–230 ℃, and has a high tensile strength of 50–60 MPa
height, infill density, and nozzle temperature on the perfor- when compared to ABS and PET [42]. For this study, virgin
mance of rPLA in order to achieve optimal outcomes and PLA and recycled PLA (recycled from residual extrusion
provide significant statistical information for future research- waste stream, which are subsequently recompounded and
ers in the field. Therefore, this study is aimed at assessing the homogenised) filaments, with 1.75 mm diameter, were pur-
mechanical properties of FDM manufactured samples utilis- chased (X3D, Perth, Western Australia) and used without
ing virgin and recycled PLA materials by varying the layer further modification, for example, no additives or thermal/
height, infill density, and nozzle temperature. This study will chemical changes. PLA filaments can absorb moisture from

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3754 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

Fig. 1  Flow diagram of the

research methods followed in
the present work

their surroundings, which can impair the quality of printed were selected for analysis. The final input parameter consid-
specimens [43]. Filaments were stored in plastic desiccators ered was the nozzle temperature (alternatively known as the
containing silica gel in order to avoid moisture adsorption printing temperature), which denotes the temperature of the
prior to their use. extruding nozzle. Three levels of nozzle temperatures were
selected at 195 ℃, 205 ℃, and 215 ℃ to analyse the strength
2.2 Selection of printing parameters of the bond between the layers. It is noted that 190–220 ℃
is the temperature range recommended by the manufacturer
The mechanical properties of FDM printed parts are known for printing. The configurations of printing parameters have
to be influenced by several printing parameters. The litera- been summarised in Table 1.
ture review noted that different printing parameters had been
taken into consideration by previous researchers to improve 2.3 Experimental design using Taguchi orthogonal
the mechanical properties of 3D printed PLA parts, includ- approach
ing layer thickness, shell, infill percentage, infill pattern,
nozzle temperature, print speed, and print orientation [13, The number of 3D printed specimens required to assess the
14, 33, 44, 45]. Therefore, based on earlier studies, parame- influence of printing parameters on the mechanical proper-
ters that directly impacted mechanical properties and surface ties and microstructure needed to be chosen. Therefore, it is
roughness, such as infill density, layer thickness, and nozzle important to organise, conduct, evaluate, and interpret tests
temperature, were chosen for this study across three levels. to assess the factors that impact the value of a parameter or
The choice of the factors and their corresponding levels set of parameters. Design of experiment is a useful tool to
was motivated by the interest in maintaining the printing improve the consistency of results whilst minimising the
quality and mechanical performance of 3D printed compo- number of experiments without loss of accuracy. The DoE
nents. The first input parameter considered was the layer methodology employs orthogonal arrays (OA) to determine
height, which measures the individual layer height deposited the optimal number of trials and their corresponding levels
by a nozzle tip. The printing time and printing quality of
a specimen could be influenced by this printing parameter
[46]. Values of 0.1 mm, 0.2 mm, and 0.3 mm were cho- Table 1  Proposed levels of 3D printing parameters for the experimental
sen as being typical values for the layer height. Second, the design
infill density indicates the degree to which the printed part Factors Level 1 Level 2 Level 3
is solid or hollow, where a 0% infill is fully hollow, and a
Layer height (mm) 0.1 0.2 0.3
100% is fully solid. Printing time, part cost, and mechanical
Infill density (%) 60 80 100
strength are known to generally increase with infill density;
Nozzle temperature (℃) 195 205 215
for this reason, three levels of infill, 60%, 80%, and 100%,

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3755

[47]. Numerous types of DoE have been described in the parameters were selected (Sect. 2.2), therefore, considering
literature, including full factorial design (FFD), central com- the software settings of the 3D printer, the other printing
posite design (CCD), definitive screening design (DSD), parameters (as listed in Table 3) for the operation of the
Box-Behnken design (BBD), and the Taguchi method [48]. 3D printing instrument were kept constant during the print-
In this work, Taguchi method was selected to design the ing operation. In this study, a rectilinear infill structure was
OA due to its effectiveness and ability to minimise the num- selected because it presents a high yield on mechanical prop-
ber of experiments performed [49]. Based on the selected erties and its ability to provide strong support for solid layers
parameters, an L9-OA was created using the MINITAB 21 with varying bottom and top thicknesses compared to other
software. The Taguchi DoE using L9-OA for the sample infill structures such as honeycomb and triangular [33]. The
preparation is presented in Table 2. The array was utilised sliced 3D model was then transferred to an FDM 3D printer
as a base for every material (vPLA and rPLA) when the test through its USB or SD card connector. To ensure printing
rounds were planned. accuracy and precision, and to provide a better possibility of

2.4 Fabrication of test samples by 3D printing

Table 3  Fixed printing parameters and their corresponding values
Once the L9 specifications of the specimens (Table 2) Parameter Specific parameter Values (constant)
were determined for the different printing parameters, the
specimens were printed according to these specifications. Layer Shell thickness 2 mm
A Raise 3D E2 FDM printer with a nozzle diameter of First layer height 0.25 mm
0.4 mm was used to manufacture the specimens according First layer flowrate 100%
to the L9 orthogonal array. This printer uses the ideaMaker First layer solid fill pattern type Lines
slicer software to slice the 3D model (STL file) and modifies Extruder Extrusion width 0.4 mm
the 3D printing process parameters, such as the infill type, Retraction speed 40 mm/s
percentage, layer thickness, heated bed temperature, and Infill Infill pattern Rectilinear
fan speed. After designing the test parts using Solidworks Infill angle − 45°/45°
computer aided design (CAD) programme, the models were Infill flowrate 100%
transformed into standard tessellation language (STL) files Infill overlap 15%
with linear tolerance of 0.05 mm and angular tolerance of Solid fill Bottom solid fill layers 5
1° ASTM D638 standard [50]. These files were utilised to Top solid fill layers 5
orient the model for the building process and mathemati- Bottom surface solid fill pattern Line
type
cal slicing, which allowed for layer-by-layer printing by the
Top surface solid fill pattern type Line
FDM machine. Following this, the 3D model was converted
Speed Printing speed 50 mm/s
into printing instruction code (G-code) using the ideaMaker
Infill speed 50 mm/s
slicer programme. This software featured various options
Support Support structure None
for modifying process parameter settings, such as infill
Platform addition Brim
type, calculating the tool path, and regulating the heated
Temperature Heat bed temperature 45 ℃
bed temperature and fan speed. As the influencing printing

Table 2  Experimental layout Experimental run Coded matrix Un-coded matrix

based on Taguchi L9 OA design
A B C Layer height Infill (%) Nozzle
(mm) temperature
(℃)

1 1 1 1 0.1 60 195
2 1 2 2 0.1 80 205
3 1 3 3 0.1 100 215
4 2 1 2 0.2 60 205
5 2 2 3 0.2 80 215
6 2 3 1 0.2 100 195
7 3 1 3 0.3 60 215
8 3 2 1 0.3 80 195
9 3 3 2 0.3 100 205

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3756 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

predicting the effects of parameters [30], all samples were resulting in a total of 45 hardness measurements for each
printed horizontally on the platform, aligned with the print experimental run.
head axis, whilst the vertical orientation was determined by Surface roughness (SR) is utilised as a technical criterion
the sample thickness. Prior to conducting every experiment, for mechanical components to quantify the existence of micro-
the build plate was cleaned to ensure that it was free from irregularities on the surface texture [55]. This property is com-
adhesives and that the process was unaffected [51]. monly employed as a technical specification for mechanical
products because of its impact on aesthetic appearance and
2.5 Characterisation of 3D‑printed specimens critical role in ensuring the structural precision of a component
[56]. In order to evaluate the influence of printing parameters
Following the 3D printing operation, all samples were kept on surface roughness, a Mitutoyo surface roughness tester
in a sealed packet, and silica gel was used to avoid moisture (model SJ-210) was utilised in this study. This compact and
absorption. For all the printed samples, tests were carried portable device offers a maximum resolution of 0.001 µm. The
out at room temperature (typically 23 °C) according to the stylus moves over the sample surface at a predetermined meas-
ASTM-D618-21 [52] standard. The mechanical properties uring velocity and distance in order to identify any irregulari-
and microstructural testing procedures followed have been ties present on the surface of the workpiece [57]. Roughness
described below. was measured on the largest surface of the tensile specimens
(Fig. 3). The arithmetic mean roughness ( Ra) and root mean
2.5.1 Determination of hardness and surface roughness square roughness ( Rq ) were quantitatively measured at the
micro-metre (μm) level.
The hardness of a material is a fundamental attribute that sig- The parameter Ra represents the mean absolute deviation
nifies its ability to withstand localised deformations, such as of roughness irregularities from the mean line over a certain
penetration or indentation on its surface. [53]. The hardness of length of sampling [58]. The numerical representation of the
the printed specimens was assessed using a specialised shore D arithmetic average height parameter is shown by Eq. 1 [58]:
hardness testing device (Sauter HBD 100–0) that is designed n
to test the hardness of rigid plastics, thermoplastics, and hard 1 ∑| |
Ra = y (1)
rubber materials. The ASTM-D2240-15 [54] standard, which n i=1 | i |
is commonly employed for assessing the durometer hardness
of polymers, was employed as the metric in this test. Accord- where n represents the total number of samples and yi repre-
ing to this standard, the thickness of the test specimen should sents the deviation from the sample mean line.
be at least 6.0 mm (0.24 in). Therefore, the dimensions of the The root mean square roughness ( Rq ) is another signifi-
hardness test specimens were manufactured with a thickness cant parameter for characterising surface roughness, which
greater than 6 mm ( 40mm × 40mm × 8mm), as illustrated indicates the standard deviation of the distribution of surface
in Fig. 2. Five pieces of 3D printed test block were prepared heights. The digital representation of this parameter can be
for each experimental condition in an orthogonal array. Nine mathematically expressed using Eq. 2 [58]:
different surface hardness measurements with 5 mm spaces √
√ n
√1 ∑
between each point were collected from each test specimen, Rq = √ |y | (2)
n i=1 | i |

where n represents the total number of samples and yi repre-

sents the deviation from the sample mean line.
Since waviness, peaks, and valleys affect the surface qual-
ity, five points were chosen on the top side of the tensile speci-
men to analyse the surface texture with a measuring speed of
0.5 mm/s (Fig. 3), and their mean values were tabulated [59,
60]. Five samples printed from each experimental run were
analysed using a tracing length of 5.6 mm, an evaluation length

Fig. 3  Five-point selections from the top surface for surface rough-
Fig. 2  Schematic diagram of the hardness test specimen ness measurements

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3757

of 4.0 mm, and a cut-off wavelength of 0.8 mm to achieve high Microsoft Excel, and the ultimate tensile strength (UTS)
accuracy and precision, according to ASTM-D7127-17 [61] was evaluated from these data. The elastic modulus of each
[59]. The roughness of the printed surface was determined specimen was evaluated according to the initial slope of its
vertically to the feed mark [62]. The representative value of stress/strain curve using Microsoft Excel’s ‘Trend Line’
each specimen was determined by calculating the arithmetic function. For this purpose, a line diagram was plotted using
mean of five measurements. strain values within the 0.0005 < ε < 0.0025 range. Several
standard mechanical property evaluations were conducted,
2.5.2 Determination of mechanical properties including the yield strength, ultimate tensile strength (UTS),
work at UTS, and work at fracture, and the average results
The mechanical properties of the printed parts were obtained were tabulated to obtain the arithmetic mean.
from the tensile, three-point bending (flexural), and impact The flexural strength (3-point bending) refers to the
tests. Following ASTM standards, the standard number of ability of a material to resist bending forces that act per-
isotropic test specimens per sample was a minimum of five. pendicular to its longitudinal axis. The 3-point bending
Hence, for all three tests, five samples were printed for each test specimens (strip-like samples) having dimensions of
experimental run. Therefore, 45 specimens were printed for 127mm × 12.7mm × 3.2mm (as shown in Fig. 5) were pre-
each material (vPLA and rPLA), and in total, 90 specimens pared and manufactured according to ASTM-D790-17 [64]
were printed for each type of mechanical test (tensile, three- standard. This test was conducted using Shimadzu Auto-
point bending, and impact). The average data were reported graph AGS-X equipment with a 10 kN cell mounted with a
based on the testing results of at least five samples for each 3-point bending apparatus, and a displacement rate of 3 mm/
material batch, along with the calculated standard deviations min and a span of 51.2 mm (span to depth ratio 16:1) were
to evaluate the test reproducibility. employed for the test. The instantaneous data for the load
Tensile tests were conducted using dog-bone samples that and displacement were obtained using the Trapezium-X
were fabricated following the ASTM-D638-22 [50] Type IV software.
specifications, with dimensions of 115mm × 19mm × 3.2mm In this study, the flexural strength was determined by cal-
(as shown in Fig. 4). All tests were performed using a Shi- culating the nominal stress in the central span, which was
madzu Autograph AGS-X tensile testing machine equipped obtained using the maximum load value, according to Eq. 3
with a 50 kN load cell capacity. This model has an exception- [64],
ally high level of precision, with a measurement accuracy
3PL
within ± 0.5% of the indicated test force. Prior to testing the 𝜎= (3)
2bd2
sample with the instrument, the non-shift wedge-type grips
were adjusted to a range of 0 to 7 mm and calibrated accord- where P is the load at a given point on the load–deflection
ing to the ASTM-E4 [63] standard. This instrument uses curve, L is the support span length, b is the width, and d is
TRAPEZIUM-X software to manage the test process and the thickness of the specimen.
thereby pre-process the data acquired from this machine. For The flexural modulus was determined using Eq. 4, as
the testing operation, at first, sample data such as thickness specified in the ASTM-D790-17 [64] standard. This equa-
(mm), breadth (mm), and gauge length (mm) were entered tion was derived through linear regression analysis of the
into the software programme. After that, the samples were load–displacement curves, considering the linear segment
clamped in the machine and loaded at a crosshead speed of within the range where the correlation coefficient exceeded
5 mm/min until failure occurred. The computer software was 95%.
used to record the force and displacement values. These val-
ues were used to generate force–displacement curves using L3 m
E= (4)
4bd3
where m represents the slope of the tangent line to the initial
linear segment of the load–deflection curve.

Fig. 4  Schematic diagram of the tensile test specimen Fig. 5  Schematic diagram of the flexural test specimen

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3758 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

The Charpy impact test, alternatively referred to as the the previous roughness. The final polishing stage com-
Charpy V-notch test, is a universally accepted method for menced with using diamond suspensions of 6 μm, followed
evaluating the fracture behaviour of materials under high by suspensions of 3 μm and 1 μm. Throughout the process,
strain rates. This indicates the amount of energy absorbed a continuous water flow was maintained. The specimens
by the material during the fracture [65]. Charpy impact tests were adequately cleaned with water prior to use at higher
were conducted to evaluate the fracture behaviour of the grit levels. This process aided in the removal of any accu-
rPLA-printed specimens. The test specimens were manu- mulated abrasive dust, which facilitated clear visualisation
factured using dimensions of 127 mm × 12.7 mm × 6.2 mm of the surface. After completion of the polishing process,
being notched at a 45° angle in accordance to the ASTM- the specimens were examined under an optical microscope
D6110-18 [66] standard (Fig. 6). A Zwick 5102 (model (OM, OLYMPUS BX51M, Japan) equipped with a portable
D-7900) impact tester with a 6.5 J hammer was used to test standalone colour camera. The acquired images were cap-
the fabricated specimens. The impact strength was deter- tured at 10 × magnifications by using magnifying lenses. For
mined using the energy required to break the specimen and fractographic analysis, a Pro-MicroScan microscope (model
the depth under the notch of the specimen. DCM 900) equipped with an eyepiece camera manufactured
by the Oplenic Corporation, China, was used to examine the
2.5.3 Determination of surface morphology fractured cross-sectional structure of the mechanically tested
samples. This examination was primarily concerned with
The surface morphology of polymeric materials is a sig- assessing the surface properties associated with the print-
nificant characteristic that is originated from their chemical ing quality. A few scratches and voids were observed in the
structure and the methods used in their production. These worn area, which may be due to sliding during mechanical
characteristics directly affect the final surface properties of property testing.
the workpiece [67]. Optical microscopy inspects samples
volumetrically and seeks to provide a preliminary percep- 2.6 Assessment using the signal‑to‑noise ratio
tion of the structural features exhibited by a specimen [68].
The focus of this examination is to assess the surface char- Performance characteristics are fundamental quality features
acteristics in relation to printing quality. To examine the that determine a product’s ability to meet specified design
surface morphology of the printed samples after mechanical requirements. Thus, a product of superior quality will exhibit
testing, at first, the samples were cut into sections and then consistent performance over its entire lifespan and under
embedded within acrylic resin and powder in self-curing various operating conditions [70]. The Taguchi method
polyacrylic cylinders with a dimension of 2 cm width and employs a loss function to quantify the difference between
3 cm height to expose a flat 4 × 6 mm window of surface the observed and desired values of the performance char-
for microstructure observations. The test parts (top surface) acteristics, which minimises performance variability. This
were prepared through a sequential polishing process uti- outcome is then converted into a statistical metric called the
lising silicon carbide (SiC) paper with grit sizes of P400, signal-to-noise ratio ( S∕N) ratio to determine the optimal
P800, P1000, P1200, and P4000. Grit sizes determine the level for each control factor and assess the statistical sig-
particle count in paper according to the Federation of Euro- nificance of their impact on the response variable, in other
pean Producers of Abrasives (FEPA) system, which is the words, to measure the process variability [10, 47]. The term
same as the ISO 6344 standard. SiC paper with a higher ‘noise’ refers to the impact of each factor on each operation,
‘P’ value is distinguished by the presence of smaller par- whilst the term ‘signal’ denotes the response to changes in
ticles within the paper [69]. Prior to starting the manual each operating variable [71].
polishing of the specimens, the surface was marked with an The S/N ratio is a metric that considers both the average
ink-felt tip permanent marker. The polishing process was value and variability of a given quality feature. The specific
performed until the ink was completely removed from the formula for calculating the S/N ratio depends on the criteria
surface, indicating that the current grit size had replaced used to assess the quality features that need to be investi-
gated [48]. As a consequence, when the process is optimised
in terms of S/N ratio, it ensures that the resulting optimal
process conditions are robust and stable, indicating minimal
process variation [72]. Hence, the S/N ratio was employed
in the current analysis to determine the most favourable and
optimal process conditions for the layer height, infill density,
and nozzle temperature. The study of the performance char-
acteristics involves the utilisation of three distinct types of
Fig. 6  Schematic diagram of the Charpy impact test specimen S/N ratio loss functions: larger-the-better, nominal-the-best,

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3759

and smaller-the-better. In the context of smaller-the-better for each sample is presented in the results section, based on
conditions, these characteristics typically represent unfa- an orthogonal experimental design. Finally, the influencing
vourable outputs. In contrast, for the larger-the-better con- parameters were determined from the calculated S∕N ratios,
dition, the characteristics are generally favourable [73]. response tables, and main effects plots.

2.7 Analysis of variance (ANOVA)

2.6.1 Processing of data for S/N value
After statistically examining the S∕N ratio, an analysis
At first, all output performance responses (data from the of variance (ANOVA) was performed to determine the
experiment) were analysed according to the Taguchi L9 accuracy of the predicted model and to investigate which
orthogonal array-based design. In this study, the ‘larger the parameter significantly affected the quality characteristics.
better’ and ‘smaller the better’ type S∕N ratio responses ANOVA separates the variation in the dataset into two
were used to evaluate the properties of the printed specimens parts: between-group and within-group [77, 78]. Therefore,
through the FDM process. ANOVA primarily examines the difference between groups
This study focuses on maximising the mechanical proper- in relation to the variation within each group [79]. ANOVA
ties, including tensile strength, three-point bending, impact is a statistical method used to optimise process parameters
strength, and hardness. Therefore, the criterion ‘larger-the- by analysing a set of experimental results and categorising
better’ was chosen for these quality characteristics. The prin- them according to a common variable or parameter, as well
ciple of ‘larger-the-better’ was employed in situations where as an objective function or response. This analysis offers
the purpose was to maximise the values of quality character- valuable insights into the relevance of the observed vari-
istics [74]. The larger-the-better S∕N ratio was calculated by ance as well as the connection between system parameters
employing Eq. 5 [75]. and their corresponding responses [76]. F-test and p-test
n were employed in the ANOVA to assess the extent to which
S 1∑ 1
ratio = (−10) × log10 ( ) (5) each factor contributed to the overall variation [76]. The
N n i=1 y2i
F-test was used to determine the parameters that exerted a
where n is the number of trials, i is the experiment number, statistically significant impact on the mechanical properties.
and yi is the response in each experiment. In addition, the p value estimated the significance of the
The purpose of the smaller-the-better condition is to results or the extent to which a parameter affects the objec-
minimise the assessment of the quality characteristics to tive function.
the smallest possible value, ideally zero, a target, or ideal
value. Therefore, the principle of ‘smaller-the-better’ was 2.7.1 Calculation of the F‑ratio
employed to minimise the surface roughness of specimens
produced using vPLA and rPLA. The smaller-the-better S∕N F-ratio is applied in statistical analysis to determine the sig-
ratio was computed by Eq. 6 [75]. nificant impact of process parameters on product charac-
teristics, which leads to a conclusive outcome [78]. A large
S 1∑ 2
n
F-ratio indicates a higher level of variance between groups
ratio = (−10) × log10 ( y) (6) compared to within groups, thereby enhancing the likeli-
N n i=1 i
hood of rejecting the null hypothesis, which suggests that
where n is the number of trials, i is the experiment number, all population means are identical. Generally, the higher
and yi is the response in each experiment. the F-ratio, the greater the influence of process parameter
Once the S∕N ratios were evaluated, the average S∕N modification on the performance characteristics [73]. For
ratios of the parameters at specific levels were examined. the case of F > 4, it typically indicates that modifications to
After that, the parameter exhibiting the largest difference in the design parameters can have a significant influence on the
arithmetic values between its highest mean S∕N ratio and finished product’s characteristics [78].
its lowest mean S∕N ratio was determined to observe the
major impact on the outcome [76]. The Taguchi approach 2.7.2 Calculation of p value
states that when the S∕N ratio is maximum, process vari-
ability is minimal. The greater the S∕N ratio value, the bet- In order to determine the statistical significance of the dis-
ter the outcome since it ensures the maximum quality with parity between group means, it is crucial to consider the
the least variability [73]. Therefore, the experimental values p value, which is associated with the F-ratio [77]. The p
with the highest mean S∕N ratio were considered as the opti- value is a statistical measure that is used instead of critical
mal operating values for each parameter. The influence of values to determine the minimum level of significance for
each process parameter on the S/N ratio at different levels rejecting the null hypothesis. A smaller p value indicates a

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3760 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

higher level of support for the alternative hypothesis [80]. printed specimens. In this section, a comprehensive analysis
The p value is commonly utilised with a threshold of 0.05, of the results obtained from the experimental testing of the
which corresponds to a significance level of 5%. For exam- specimens has been presented with the aim of evaluating
ple, in a case where the calculated p value of a test statistic their material properties.
is less than 0.05, it would be appropriate to reject the null
hypothesis [80]. 3.1 Surface roughness

2.8 DoE confirmation test To gain additional insight into the differences between the
printed virgin and recycled materials, an analysis of the sur-
Once the optimal level of the design parameters had been face roughness was conducted following printing (as shown
selected, the final step was conducted to predict and verify in Fig. 7). The FDM process typically results in a relatively
the improvement of the quality characteristics using the opti- rough surface, with the extent of this roughness being influ-
mal levels of the design parameters. The obtained results enced by the printing quality [82]. In order to achieve a high
could subsequently be compared to the theoretically pre- level of surface smoothness, it is common practise to subject
dicted optimal conditions based on the desired combina- a product to different processing procedures [3]. Therefore,
tion of factors and their respective levels [81]. At first, the the experiment aimed to achieve maximum surface smooth-
predicted S∕N ratio, which was expected for the optimal ness and determine whether the printing parameters under
combination of parameters, was determined. After com- investigation have an impact on the surface roughness of
puting the predicted S∕N ratio, a confirmation experiment the specimens.
was conducted to validate whether the optimal conditions The measurement and recording of the surface roughness
achieved the expected response. In engineering experiments, for each printed part have been documented in Fig. 7. The
it is expected that the level of agreement between the pre- findings presented in Fig. 7 (a) indicate that the parameter
dicted values and confirmation results should fall within the settings used in experimental run 2 (LH 0.1 mm, I 100%,
range of ± 5%, indicating a confidence interval of greater and NT 205 ℃) yielded the lowest average Ra values for
than 95% [81]. This comparison evaluated the agreement both vPLA and rPLA specimens, namely, 5.87 μm and
between the obtained and expected parameters. 6.12 μm, respectively. This was followed by experimental
run 3, which resulted in an average Ra value of 6.22 μm and
2.9 Statistical analysis 7.33 μm, and experimental run 3, which yielded average
Ra values of 6.70 μm and 7.72 μm, for vPLA and rPLA,
Microsoft Excel (Microsoft Corporation) was used for data respectively. The rPLA specimens exhibited higher Ra and
handling. Values obtained from the material property char- Rq values in comparison to the vPLA specimens, suggesting
acterisation tests are presented as mean ± standard deviation. the presence of rougher surfaces. As recycled PLA is known
Minitab 21 software was used to analyse the S/N ratio and to exhibit a lower molecular weight due to chain scission,
ANOVA. the recycled material would be expected to exhibit a lower
viscosity at the same extrusion temperature, leading to the
formation of a surface with increased roughness [62]. Based
3 Result and discussion on the data presented in Fig. 7, it can be observed that there
is a positive correlation between the nozzle temperature and
The primary objective of the present investigation was to layer thickness with the surface roughness for both types
examine the most favourable printing parameters for the of materials. Specifically, at a lower nozzle temperature
fabrication of 3D printed components utilising two differ- (195 °C) and higher nozzle temperature (215 °C), the sur-
ent materials, namely, vPLA and rPLA. The present study face roughness exhibits an increase. The surface was rough-
employed a Taguchi orthogonal array experimental design to est at 195 °C and slightly less rough at 215 °C, whereas at
conduct the experiments. The specimens were manufactured 205 °C, the surface roughness was lower. Maidin et al. [62]
using a fused deposition modelling 3D printer. Following and Mani et al. [83] observed that for vPLA, the surface
the manufacturing process, a comprehensive evaluation of roughness decreased with an increase in nozzle tempera-
mechanical properties was conducted. The properties under ture. It should be noted that the present study used different
investigation included surface roughness, hardness, tensile nozzle temperatures and layer thicknesses, which showed a
strength, flexural strength, and impact strength. The perfor- different result but a similar trend, as in previous research
mance of the rPLA specimens was compared/benchmarked on surface roughness. This may be attributed to the tem-
with 3D printed vPLA specimens using the same process- perature of the nozzle, which significantly affects the raw
ing parameters. These parameters were carefully assessed material melting process. Incomplete melting due to lower
to gain insights into the performance characteristics of the than optimum temperature would produce a rough surface.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3761

Fig. 7  Surface roughness

properties for the vPLA and
rPLA printed specimens: a Ra
and b Rq. *LH, layer height; I,
infill percentage; NT, nozzle
temperature

On the other hand, the reduction in the thickness of the lay- run 9, which indicated hardness result of 81.68 and 76.78
ers has a direct impact on the surface roughness, resulting for vPLA and rPLA, respectively. In contrast to this, the
in decreased visibility of the layers and a smoother surface lowest result was obtained for experimental run 1, which
texture for the printed object. Ayrilmis [84] also observed an showed hardness of 79.53 and 73.92 for vPLA and rPLA,
increased roughness with increasing printing layer thickness. respectively. The findings of this study demonstrate a clear
As the thickness of the layers increases, the surface of the positive correlation between nozzle temperature and infill
printed part exhibits enhanced sharpness, thereby rendering percentage and the resulting increase in hardness, which is
the individual layers more noticeable [85]. consistent with the results evaluated by Mani et al. [83] and
Maguluri et al. [87] for vPLA. According to Şirin et al. [88],
3.2 Hardness test an increase in the infill percentage resulted in a reduction in
void formation, leading to an overall increase in hardness.
The hardness test is a widely employed method in the field Another reason is that an increase in the infill percentage
of plastics to assess the ability of a material to withstand leads to an expansion in the cross-sectional area of the mate-
concentrated loads that result in significant plastic deforma- rial, thereby contributing to the development of a more rigid
tion. This test is both rigorous and informative and provides internal structure [89]. The results also demonstrated the
valuable insights into the properties of the material under influence of printing temperature on the hardness properties.
inspection [86]. The findings of the experiments conducted A reduction in the printing temperature led to a correspond-
on both materials have been summarised in Fig. 8. The ing decrease in the hardness value, the cause of which was
vPLA-printed specimens exhibited marginally higher hard- attributed to poor layer-to-layer bonding and insufficient
ness, whereas the rPLA polymers exhibited lower hardness. melting when printing at low temperatures, which resulted
The highest shore D hardness was observed for experimental in increased porosity and decreased hardness. The decrease

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3762 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

Fig. 8  Shore D hardness

values for the vPLA and rPLA
specimens. *LH, layer height;
I, infill percentage; NT, nozzle
temperature

in the rate of hardness change with increasing printing tem- stress–strain curves of vPLA and rPLA components manu-
perature was attributed to the elevated temperature during factured through 3D printing, which included all experimen-
the printing process and the enhanced adhesion between tal conditions. It is evident that the vPLA material exhibited
the fibres, resulting in a more robust material combination. higher values of tensile strength and yield strength in com-
Additionally, a reduction in the number of air gaps can also parison to rPLA. The stress–strain relationships exhibited by
contribute to this phenomenon [87]. the vPLA and rPLA specimens demonstrated a nearly linear
correlation as the strain increased. Once the stress reached
3.3 Tensile test its maximum point, it remained relatively constant as the
strain continued to rise. The examination of the outcomes
The output variables of elastic modulus, ultimate tensile revealed that both the vPLA and rPLA specimens exhibited
strength, yield strength, fracture strength, work at ultimate ductile fracture behaviour. The data clearly indicate that the
tensile strength, and work at fracture have been observed virgin PLA material exhibited a higher tensile strength than
by adjusting the input processing variables, as indicated the recycled PLA material.
in Table 2. Five specimens were produced for each Tagu- Tables 4 and 5 summarise the average values of the vPLA
chi experiment and were chosen for tensile examination to and rPLA tensile properties, including the elastic modulus,
reduce the impact of differences in the specimens on the ultimate tensile strength, yield strength, fracture strength, and
results of the experiment. The findings from these five strain at fracture, for each specimen designed by the Taguchi
specimens demonstrated a consistent level of repeatability. method. The results showed that the standard deviation of
Hence, a single stress–strain curve is illustrated, represent- all the measurements was relatively small, indicating good
ing a set of five specimens. Figures 9 and 10 illustrate the repeatability within the experimental data. When comparing

Fig. 9  Tensile stress vs. tensile

strain curves for vPLA printed
specimens. *LH, layer height;
I, infill percentage; NT, nozzle
temperature

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3763

Fig. 10  Tensile stress vs. tensile

strain curves for rPLA printed
specimens. *LH, layer height;
I, infill percentage; NT, nozzle
temperature

Table 4  Tensile properties of the vPLA printed specimens

Experimental run Yield strength Ultimate ten- Strain at UTS Fracture Strain at Elastic Work until Work until
sile strength strength fracture modulus UTS fracture
(UTS)
(MPa) (MPa) (%) (MPa) (%) (GPa) (kJ/m2) (kJ/m2)

Run 1 (LH 0.1, I 60, NT 195) 33.51 ± 0.871 36.36 ± 0.289 2.27 ± 0.047 31.87 ± 0.619 3.78 ± 0.266 2.26 ± 0.296 32.08 ± 0.770 65.07 ± 1.412
Run 2 (LH 0.1, I 80, NT 205) 38.12 ± 0.931 45.30 ± 0.328 2.48 ± 0.025 42.98 ± 0.389 3.84 ± 0.229 2.71 ± 0.042 40.34 ± 0.340 85.32 ± 0.995
Run 3 (LH 0.1, I 100, NT 215) 47.97 ± 0.954 52.81 ± 0.630 2.50 ± 0.082 49.37 ± 0.829 3.70 ± 0.696 3.13 ± 0.053 43.52 ± 0.944 100.02 ± 1.607
Run 4 (LH 0.2, I 60, NT 205) 50.17 ± 1.124 53.47 ± 0.710 2.49 ± 0.066 50.53 ± 0.951 4.05 ± 0.167 3.10 ± 0.072 43.11 ± 0.632 102.04 ± 0.890
Run 5 (LH 0.2, I 80, NT 215) 50.89 ± 0.991 54.04 ± 0.429 2.51 ± 0.031 51.33 ± 0.659 5.36 ± 0.684 3.11 ± 0.036 47.56 ± 0.871 130.97 ± 1.437
Run 6 (LH 0.2, I 100, NT 195) 48.41 ± 1.202 52.68 ± 0.652 2.41 ± 0.041 47.68 ± 1.012 5.04 ± 0.688 3.08 ± 0.038 42.25 ± 0.205 120.07 ± 2.014
Run 7 (LH 0.3, I 60, NT 215) 53.56 ± 1.141 56.85 ± 0.269 2.48 ± 0.021 52.53 ± 0.475 4.74 ± 0.442 3.19 ± 0.014 45.31 ± 0.537 131.32 ± 1.018
Run 8 (LH 0.3, I 80, NT 195) 50.61 ± 1.123 51.37 ± 0.548 2.37 ± 0.044 50.23 ± 0.613 4.31 ± 0.188 3.04 ± 0.035 41.56 ± 0.759 115.90 ± 1.752
Run 9 (LH 0.3, I 100, NT 205) 51.26 ± 0.984 54.84 ± 0.698 2.45 ± 0.069 51.08 ± 0.286 5.30 ± 0.339 3.10 ± 0.038 46.73 ± 0.409 125.63 ± 1.921

*LH, layer height; I, infill percentage; NT, nozzle temperature

the vPLA- and rPLA-printed specimens, it was observed that, 39.90 MPa, which is 25.22% lower than the correspond-
in all of the experimental investigations, the vPLA specimens ing value of 56.85 MPa for the vPLA printed specimen.
exhibited higher tensile characteristics than the rPLA-printed The experimental results indicated that the second high-
samples, as depicted in Figs. 9 and 10. In addition, the findings est recorded value was 54.84 MPa for experimental run 9,
indicated that the samples exhibited better tensile properties which utilised a layer height of 0.3 mm, 100% infill, and a
as the layer thickness and temperature increased. In previous nozzle temperature of 205 °C. Similarly, the third highest
studies utilising recycled polylactic acid, the same declining recorded value was 54.04 MPa for experimental run 5, which
pattern in the tensile properties was noted [3, 15, 17, 22, 90]. employed a layer height of 0.2 mm, 100% infill, and nozzle
In the context of experimental run 7, it was observed that temperature of 215 °C. In the case of the rPLA-printed spec-
specimens printed with a layer height of 0.3 mm, infill den- imens, the reduction in the values was 28.85% and 29.09%
sity of 80%, and nozzle temperature of 215 °C exhibited for the respective parameters. Conversely, a decrease in the
the highest tensile strength. This outcome was consistent thickness of each layer and temperature corresponded to a
for both vPLA and rPLA materials. Based on the data pre- decline in the tensile strength. To provide an example, the
sented in Table 4 and 5, it can be observed that the tensile results obtained from experimental run 1 revealed subop-
strength of rPLA in experimental run 7 was recorded as timal results, with vPLA and rPLA demonstrating tensile
strength values of 36.06 MPa and 28.81 MPa, respectively.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3764 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

The lower printing temperature resulted in incomplete melt-

87.34 ± 0.53
74.43 ± 1.27
97.65 ± 1.75
109.63 ± 2.10
135.98 ± 1.89
120.07 ± 2.28
142.24 ± 2.29
120.70 ± 2.20
136.67 ± 2.49
Work until ing, which in turn produced weak bonding between layers
fracture
and loss of internal structure, resulting in easily fractured
(kJ/m2)
specimens [89].
A noteworthy finding was observed in relation to the strain at
fracture, wherein rPLA exhibited a superior outcome compared
Work until UTS

26.77 ± 0.682
29.98 ± 0.331
24.48 ± 0.862
29.37 ± 0.774
34.63 ± 0.730
31.89 ± 0.521
40.44 ± 0.507
30.74 ± 0.461
34.66 ± 0.722
with vPLA, with increases ranging from 14.64 to 68.83%. This
ultimately indicated that rPLA demonstrated a more favour-
(kJ/m2)

able work of fracture. In the case of both the vPLA and rPLA
samples, it was observed that experimental run 7 exhibited bet-
ter fracture performance, including strain at fracture, fracture
strength, and work done until fracture, compared to the remain-
Elastic modulus

ing samples. Experimental runs 5 and 9 exhibited comparable

1.99 ± 0.031
2.19 ± 0.157
2.32 ± 0.018
2.42 ± 0.016
2.59 ± 0.007
2.38 ± 0.043
2.66 ± 0.031
2.51 ± 0.033
2.62 ± 0.019
findings to experimental run 7 in terms of the work done until
(GPa)

fracture. It is interesting to observe that both the tensile strength

and fracture strain were simultaneously improved as the nozzle
temperature increased. This phenomenon can be attributed to
Strain at fracture

the improved bond quality between the layers deposited at the

5.65 ± 0.558
4.34 ± 0.230
5.15 ± 1.001
6.50 ± 1.796
6.91 ± 0.361
6.71 ± 0.792
7.10 ± 0.769
5.94 ± 1.271
6.92 ± 0.558

interface [10].
Previous research indicated that reducing the infill
(%)

density leads to a decrease in the contact and adhesion

between layers, resulting in a reduction in the mechani-
Fracture strength

cal properties [89]. However, the findings of the present

23.84 ± 0.919
25.77 ± 0.901
26.41 ± 0.863
29.93 ± 0.645
33.84 ± 0.392
30.54 ± 0.732
34.56 ± 0.992
33.16 ± 1.357
28.42 ± 0.462

study indicate that the infill density did not have a sig-
nificant impact when the nozzle temperature and layer
(MPa)

thickness were suitably high, as observed in experimental

run 7. The rationale behind this phenomenon may lie in
the reduction of the number of layers during printing with
a higher layer thickness, resulting in a decrease in the
Strain at UTS

2.06 ± 0.053
2.01 ± 0.019
2.00 ± 0.030
2.21 ± 0.031
2.17 ± 0.031
2.24 ± 0.023
2.36 ± 0.027
2.06 ± 0.031
2.16 ± 0.021

distortion effect, consequently leading to an increase in

strength. Moreover, an increase in the thickness of each
layer facilitates the movement of the partially melted
(%)

substance within the expanded gaps between the printed

lines, consequently leading to a notable enhancement in
Ultimate tensile
strength (UTS)

28.81 ± 0.227
31.20 ± 0.116
32.11 ± 0.509
35.23 ± 0.450
38.36 ± 0.429
35.86 ± 0.563
39.90 ± 0.479
37.54 ± 0.636
38.47 ± 0.338

the structural integrity [91]. An additional factor that

could contribute to this phenomenon is the utilisation of
(MPa)

a higher printing temperature, which increases the fluidity

*LH, layer height; I, infill percentage; NT, nozzle temperature

of the material and subsequently enhances the interfacial

Table 5  Tensile properties of the rPLA printed specimens

bond strength between successive layers. Furthermore,

Yield strength

the reduction in interchain interaction and decrease in

22.78 ± 0.442
25.02 ± 0.738
25.84 ± 0.696
31.01 ± 0.458
32.80 ± 0.437
29.34 ± 0.436
34.12 ± 0.648
31.65 ± 0.394
33.25 ± 0.723

the number of air gaps would be expected to contribute

to an increase in the tensile strength of the material, as
(MPa)

demonstrated by Hsueh et al. [89].

3.4 Flexural test
Run 3 (LH 0.1, I 100, NT 215)

Run 6 (LH 0.2, I 100, NT 195)

Run 9 (LH 0.3, I 100, NT 205)

Run 1 (LH 0.1, I 60, NT 195)
Run 2 (LH 0.1, I 80, NT 205)

Run 4 (LH 0.2, I 60, NT 205)

Run 5 (LH 0.2, I 80, NT 215)

Run 7 (LH 0.3, I 60, NT 215)

Run 8 (LH 0.3, I 80, NT 195)

The experimental values of flexural properties have been

presented in Table 6 and 7. These tables provide a com-
prehensive overview of the flexural performances of the
Experimental run

tested specimens. In addition, the stress vs. strain diagrams

for the flexural tests have been depicted in Figs. 11 and 12
and visually illustrate the relationship between stress and
strain, thereby offering valuable insight into the mechanical

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3765

Table 6  Test results for vPLA flexural specimens

Experimental run Yield strength Flexural strength Flexural strain Elastic modulus Work until
flexural
strength
MPa MPa % GPa kJ/m2

Run 1 (LH 0.1, I 60, NT 195) 60.25 ± 0.25 64.78 ± 2.21 4.78 ± 0.21 2.31 ± 1.20 28.74 ± 1.41
Run 2 (LH 0.1, I 80, NT 205) 66.44 ± 0.58 71.15 ± 1.21 5.11 ± 0.85 2.56 ± 1.24 31.11 ± 2.20
Run 3 (LH 0.1, I 100, NT 215) 73.85 ± 1.20 79.98 ± 1.48 5.23 ± 0.98 2.94 ± 1.24 35.86 ± 2.01
Run 4 (LH 0.2, I 60, NT 205) 69.45 ± 1.07 82.75 ± 2.01 6.60 ± 0.67 2.91 ± 1.83 39.47 ± 1.91
Run 5 (LH 0.2, I 80, NT 215) 74.11 ± 0.87 87.39 ± 1.24 6.92 ± 0.70 2.91 ± 1.51 40.87 ± 1.01
Run 6 (LH 0.2, I 100, NT 195) 70.72 ± 1.23 82.53 ± 1.09 7.90 ± 1.24 2.89 ± 1.81 45.83 ± 1.46
Run 7 (LH 0.3, I 60, NT 215) 77.34 ± 0.89 92.44 ± 2.01 6.85 ± 1.20 3.01 ± 0.95 49.96 ± 1.47
Run 8 (LH 0.3, I 80, NT 195) 71.48 ± 1.87 86.11 ± 2.42 8.11 ± 0.51 2.85 ± 0.89 46.07 ± 2.03
Run 9 (LH 0.3, I 100, NT 205) 76.60 ± 1.74 90.26 ± 1.74 6.50 ± 0.81 2.95 ± 1.01 46.35 ± 1.58

*LH, layer height; I, infill percentage; NT, nozzle temperature

behaviour of the specimens under flexural loading. The flex- characteristics was documented [3, 17, 22]. This phenom-
ural stress–strain diagrams were determined using Eqs. 4 and enon can be attributed to the reduced mechanical properties
5, respectively. Work until flexural strength was determined of recycled PLA owing to its low molecular weight [93].
from the area under the bending load–displacement curve According to the findings presented in Table 6 and
until the point of maximum stress. During loading, the frac- 7, it is evident that experimental run 7 yielded the high-
ture strength of the flexural tests could not be determined, est value for all the flexural properties. Specifically, this
because the samples did not completely fracture up to the condition demonstrated a maximum flexural strength of
testing limit. All tests were terminated before the specimens 92.44 MPa for vPLA and 84.74 MPa for rPLA. This result
reached their fracture point because they came into contact may be attributed to the fact that experimental run 7 was
with the grips prior to fracture. This was primarily attrib- printed with a high nozzle temperature and high layer
uted to the excessive ductility of the polymers; therefore, height, which resulted in improved layer adhesion and
the measurement of strain at fracture was not achieved [3, minimal void formation; as a consequence, the specimens
92]. The findings presented in this study showed compara- exhibited excellent flexural properties in comparison to
tively poorer mechanical properties exhibited by the speci- other conditions [94]. In contrast to this, the findings of
mens printed using rPLA in comparison to those printed experimental run 1 yielded unsatisfactory outcomes for
using vPLA. In previous investigations employing rPLA, a both vPLA and rPLA with respective flexural strengths
comparable finding related to a decreasing trend in flexural of 64.78 MPa and 46.78 MPa. The stress and strain curves

Table 7  Test results for rPLA flexural specimens

Experimental run Yield strength Flexural strength Flexural strain Elastic modulus Work until
flexural
strength
MPa MPa % GPa kJ/m2

Run 1 (LH 0.1, I 60, NT 195) 41.50 ± 1.23 46.78 ± 1.83 8.75 ± 2.21 1.86 ± 2.41 71.93 ± 1.10
Run 2 (LH 0.1, I 80, NT 205) 43.15 ± 1.74 52.31 ± 1.94 8.95 ± 2.85 2.05 ± 1.08 73.44 ± 0.95
Run 3 (LH 0.1, I 100, NT 215) 53.02 ± 0.84 65.54 ± 0.89 9.20 ± 1.27 2.13 ± 1.30 79.01 ± 2.01
Run 4 (LH 0.2, I 60, NT 205) 63.40 ± 1.01 75.84 ± 1.38 7.71 ± 1.38 2.35 ± 1.49 84.48 ± 1.28
Run 5 (LH 0.2, I 80, NT 215) 68.35 ± 0.51 82.40 ± 2.01 8.55 ± 1.78 2.40 ± 1.74 92.66 ± 1.87
Run 6 (LH 0.2, I 100, NT 195) 60.33 ± 1.21 74.22 ± 1.89 8.50 ± 2.12 2.19 ± 2.41 82.50 ± 1.91
Run 7 (LH 0.3, I 60, NT 215) 72.06 ± 1.87 84.74 ± 2.24 9.90 ± 1.45 2.52 ± 1.34 93.45 ± 2.41
Run 8 (LH 0.3, I 80, NT 195) 66.34 ± 0.98 79.70 ± 1.56 8.70 ± 1.81 2.36 ± 2.29 85.94 ± 1.01
Run 9 (LH 0.3, I 100, NT 205) 67.33 ± 1.23 81.44 ± 2.03 8.77 ± 1.78 2.48 ± 2.10 86.36 ± 1.58

*LH, layer height; I, infill percentage; NT, nozzle temperature

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3766 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

for vPLA (Fig. 11) showed that experimental run 7 pos- The rationale behind this phenomenon was that elonga-
sessed maximum strength, but that the strain capacity tion of thermoplastic polymers is directly influenced by
was not maximum. Experimental run 8 exhibited lower the recycling process. As previously mentioned, the ther-
flexural strength in comparison with experimental run 7, mal degradation process during recycling induces chain
but it demonstrated greater ductility because of its better scissoring, which subsequently causes a reduction in the
strain-absorbing capacity. This was attributed to the dom- strength of the intermolecular bonds within the polymers.
ination of additional porosity within the sample due to the Hence, the polymers demonstrated a tendency to undergo
lower melting temperature, which resulted in increased elongation [95].
deflection [1]. The stress–strain curves, as illustrated in
Fig. 12, provide valuable insights into the mechanical 3.5 Impact strength
properties of rPLA. Notably, analysis of the stress and
strain curves revealed that experimental run 7 exhibited In relation to the influence of the printing parameters on the
the highest levels of flexural stress and strain amongst impact strength, Fig. 13 presents the average Charpy impact
all of the samples tested. In comparing the two curves, values, which exhibit a direct correlation with the infill per-
it became apparent that the strain displayed by rPLA is centage and layer thickness. The study conducted by Kamaal
greater in magnitude when compared to that of vPLA. et al. [96] found that there was no significant difference in

Fig. 11  Flexural stress vs.

flexural strain curves for vPLA
specimens. *LH, layer height;
I, infill percentage; NT, nozzle
temperature

Fig. 12  Flexural stress vs.

flexural strain curves for rPLA
specimens. *LH, layer height;
I, infill percentage; NT, nozzle
temperature

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3767

impact strength at low infill percentages. However, this study polymer within the expanded gaps between the printed lines,
observed that when the infill percentage reached 100%, and consequently leading to a notable enhancement in structural
the layer height was 0.3 mm, there was an increase in impact integrity. However, when the layer thickness and infill were
strength. This can be attributed to the reduction in voids high, but the nozzle temperature was low (run 9), the presence
within the printed material, as well as the increased amount of pores increased compared to run 7, but it was still very small
of material available to absorb the impact energy [97]. The compared to run 1 (shown in Fig. 14c). In contrast to this, it
outcome of the impact tests indicated that rPLA exhibited an was observed from Fig. 14 that, although the infill and layer
impact strength that is comparable to that of vPLA. The impact thickness increased, if the nozzle temperature was low then
strength of the components made from recycled PLA exhibited the presence of large voids was noted (Fig. 14d). This was
only a marginally reduced value (ranging from 4.06 to 11.08%) attributed to the low fluidity of the material, which ultimately
when compared to the components made from virgin PLA. increased the size and quantity of pores. For the case of rPLA,
Experimental run 9 exhibited the highest average impact value the same type of observations was made. In this case, large
for both the vPLA and rPLA specimens. In contrast to this, the voids were observed for the same experimental conditions
sample with the lowest average value was observed in experi- as had been noted for vPLA, i.e., Fig. 15a–d. An in-depth
mental run 1 when the layer thickness was reduced, with this investigation utilising microscopy tools was conducted to
being consistent with Atakok et al. [17], who found a decrease identify the underlying factors contributing to the observed
in impact strength as the layer height decreased, with this being reduction in mechanical strength caused by variations in the
attributed to the reduced resistance to crack propagation. nozzle temperature. The findings indicate that the inadequate
flow of material, stemming from a lower nozzle temperature
3.6 Fractography studies during the deposition process, leads to the formation of gaps
and voids within the layers. Although the infill density, as
The experimental tests revealed that the nature of fracture for specified in the slicing software, was set to 100%, the presence
the vPLA and rPLA specimens was ductile, which was due of voids was observed [98]. Figure 16a–c shows that with
to stretching and reorientation of the material that resulted decreasing nozzle temperature, the voids between layers
in deformation. The fracture surfaces of the 3D printed PLA increased owing to the poor adhesion between layers, which
samples exhibited a ductile type of fracture. The fracture ultimately decreased the strength of the rPLA samples. In
surfaces were analysed using an optical microscope with conclusion, this analysis showed a positive correlation between
results being presented in Figs. 14 and 15. It can be seen from layer thickness and nozzle temperature, i.e., even for the same
Fig. 14 that a large number of pores were present when the nominal layer thickness, increasing the nozzle temperature led
layer thickness was low, even though the infill amount and to a larger actual layer thickness. Thus, the nozzle temperature
temperature were high (run 1) as shown in Fig. 14a, which and layer thickness must be selected appropriately to improve
resulted in poor bonding of the material. When the layer the microstructure of the specimen. Overall, a good correlation
thickness and nozzle temperature were high (run 7), a smaller was noted between the optical microscopic studies and the
number of pores were present, as shown in Fig. 14b, with this resulting experimental data.
being mainly attributed to the increase in partially melted

Fig. 13  Impact strength experi-

mental results for vPLA and
rPLA specimens. *LH, layer
height; I, infill percentage; NT,
nozzle temperature

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3768 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

a) Optical microscopy image of vPLA (LH: 0.1, I:100, NT: 215) b) Optical microscopy image of vPLA (LH: 0.3, I: 60, NT: 215)

c) Optical microscopy image of vPLA (LH: 0.3, I:100, NT: 205) d)Optical microscopy image of vPLA (LH: 0.2, I:100, NT: 195)

Fig. 14  Optical micrographs of fractured surfaces of vPLA specimens: a run 3, b run 7, c run 9, and d run 6

a) Optical microscopy image of rPLA (LH: 0.1, I:100, NT: 215) b) Optical microscopy image of rPLA (LH: 0.3, I:60, NT: 215)

c)Optical microscopy image of rPLA (LH: 0.3, I:100, NT: 205) d)Optical microscopy image of rPLA (LH:0.2, I:100, NT: 195)

Fig. 15  Optical micrographs of fracture surfaces for rPLA specimens: a run 3, b run 7, c run 9, and d run 6

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3769

3.7 Optimisation of process parameters increases during the printing process, the melt viscosity of
PLA decreases, thereby leading to improved interlayer adhe-
The optimisation of process parameters within the FDM 3D sion [99]. As shown in Fig. 17, the optimum combination
printer with respect to hardness, tensile strength, and sur- of parameters was found to be A3B3C3, with a layer thick-
face roughness was carried out using Taguchi analysis and ness of 0.3 mm, infill of 100%, and a nozzle temperature of
the Minitab 21 software. The S/N ratios for tensile, flexural, 215 °C.
impact strength, hardness, and surface roughness were eval- Figure 18 illustrates the S/N ratio and main effects plot
uated with ‘larger the better’ criteria for tensile properties, for flexural strength (FS) for the larger-the-better criteria.
flexural properties, impact strength and hardness, and ‘lower The FS exhibited an almost identical parametric response to
the better’ criteria for surface roughness. The optimal levels that of the tensile strength. Here, it was determined that the
were determined by utilising the average S/N ratios for each flexural characteristics exhibited maximum values when the
response at each level. A higher value of S/N ratio indicates layer thickness, infill density, and nozzle temperature were
superior quality characteristics [23]. Therefore, to maxim- adjusted to high levels. Therefore, for the maximum flexural
ise the mechanical strength, the optimal parameters were strength response, 0.3 mm layer thickness, 100% infill, and
selected by choosing the highest S/N ratio for each factor. 215 °C nozzle temperature were determined to be the opti-
Figure 17 shows a comparison of the main effects plot, mal combination of parameters. The reasoning behind this
illustrating the S/N ratios for both the vPLA and rPLA ten- combination was the same as already discussed for the case
sile properties, as well as the FDM printing parameters. The of tensile strength.
figure reveals that layer thickness was the most significant From a dynamic application standpoint, maximising the
factor affecting tensile properties, and the temperature of the required impact strength would be important [21]. In the
nozzle head also had a substantial impact on strain at UTS. current study, the resulting Charpy impact strengths were
It is anticipated that the melt viscosity of PLA is highly examined under the condition of larger the better, as illus-
influenced by temperature, and as the nozzle temperature trated in Fig. 19. Analysis of the S/N ratio response plots

a) Optical microscopy image of rPLA (LH: 0.3, I: 80, NT: 195) b) Optical microscopy image of rPLA (LH: 0.3, I: 100, NT: 205)

c)Optical microscopy image of rPLA (LH: 0.3, I: 60, NT: 215)

Fig. 16  Micrographs of the cross-sectional area of printed samples: a run 8, b run 9, and c run 7

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3770 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

Fig. 17  Main effects plot showing S/N ratios for tensile properties of vPLA and rPLA specimens: a yield strength, b UTS, c strain at UTS, d
fracture strength, e strain at fracture, f elastic modulus, g work until UTS, and h work until fracture

revealed that the layer height and infill significantly affected height. Furthermore, the main effects plot demonstrated that
the impact strength of the vPLA and rPLA-printed speci- low density had a statistically significant impact on surface
mens. However, it was observed that nozzle temperature roughness. Therefore, the most favourable arrangement of
did not significantly influence the impact properties, such processing parameters for minimising surface roughness was
as had been the case for TS and FS. Therefore, 0.3 mm layer determined as follows: a layer thickness of 0.1 mm, an infill
thickness, 100% infill, and 205 °C nozzle temperature were percentage of 60%, and a nozzle temperature of 205 °C for
determined to be the optimal combination to achieve high vPLA; a layer thickness of 0.1 mm, an infill percentage of
impact strength. In particular, the interlayer bonding exhib- 60%, and a nozzle temperature of 205 °C for rPLA.
ited poor performance at 195 °C but improved when the tem- Figure 21 presents the S/N ratios and mean plots (using
perature was raised to 205 °C. However, a further increase in the ‘larger the better’ criterion) for shore D hardness of
the temperature resulted in increased fluidity of the molten the 3D printed specimens. Overall, it was found that the
plastic, leading to a decrease in viscosity and the formation specimen hardness increased by increasing the infill den-
of voids. Consequently, this had a slight negative impact on sity, layer height, and nozzle temperature for vPLA. The
the overall performance in terms of impact strength [97]. reason behind this trend was ascribed to the reduction in
The experimental findings were examined in order to porosity, which would be expected to improve the hard-
determine the optimal printing parameters for minimising ness [21]. However, a layer thickness of 0.2 mm exhibited
surface roughness. The analysis was conducted based on the highest hardness for rPLA and, therefore, the optimum
the S/N ratios, employing the ‘smaller the better’ criteria combinations were found to be layer height of 0.3 mm, infill
and presented in Fig. 20. The investigation revealed that of 100%, and nozzle temperature of 215 ℃ for vPLA and
the primary factor contributing to the reduction in the sur- 0.2 mm layer height, 100% infill, and 215 ℃ nozzle tem-
face roughness of the specimens was the reduction in layer perature for rPLA.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3771

Fig. 18  Main effects plot of S/N ratios for flexural properties of vPLA and rPLA specimens: a yield strength, b flexural strength, c flexural strain,
d elastic modulus, and e work until flexural strength

3.8 ANOVA analysis are thus considered as experimental errors [100]. A low

residual error was observed for each property that exhib-
An ANOVA study was conducted with a confidence level ited good control over the processing of components. The
of 95% to examine the relative contribution of the param- main purpose of the percentage contribution is to assess
eters on the individual responses of the tensile strength, the extent to which different parameters influence mechan-
flexural strength, impact strength, shore D hardness, and ical properties. In other words, a higher percentage value
surface roughness properties of both virgin PLA and recy- indicates a stronger influence of a particular parameter on
cled PLA materials. Table 8 shows the ANOVA results, the response variable [101]. The findings indicated that
where the rows labelled as ‘Residual Error’ refer to the the layer thickness had a higher percentage contribution to
errors that arise from uncontrollable factors, specifically the parameters examined (as shown in Table 8) compared
noise, which are not accounted for in the experiment and to the infill and nozzle temperatures. Therefore, the major

Fig. 19  Main effects plot of S/N ratios for impact strengths of a vPLA and b rPLA specimens

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3772 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

Fig. 20  Main effects plot of S/N ratios for the surface roughness of vPLA and rPLA specimens: a Ra and b Rq

determinant for the properties investigated in this work within the range of 60–100% investigated in this work).
was found to be the layer thickness. The second factor that From the ANOVA results, it was determined that overall
exerted an influence on the system was the nozzle temper- optimum tensile and flexural strength could be achieved
ature. In contrast to this, the influence of the infill percent- by selecting high layer thickness, high nozzle temperature,
age on the outcome was found to be negligible (at least and low infill density.

Fig. 21  Main effects plot of S/N ratios for shore D hardness of (a) vPLA and (b) rPLA specimens

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3773

Table 8  ANOVA responses for S/N ratios of various properties examined in this work
Response Source DF Seq SS Adj SS Adj MS F p % contribution

Ultimate tensile strength (vPLA) LH 2 3.09 3.09 1.54 130.35 0.008 68.04
Infill 2 0.19 0.19 0.09 8.17 0.109 4.27
NT 2 1.23 1.23 0.61 52.05 0.019 27.17
Residual error 2 0.023 0.02 0.01 0.52
Total 8 4.55 100.00
Ultimate tensile strength (rPLA) LH 2 2.94 2.94 1.47 273.29 0.004 73.48
Infill 2 0.14 0.14 0.07 13.04 0.071 3.51
NT 2 0.91 0.91 0.45 84.58 0.012 22.74
Residual error 2 0.01 0.01 0.00 0.27
Total 8 4.01 100.00
Flexural strength (vPLA) LH 2 5.97 5.97 2.98 39.68 0.025 74.55
Infill 2 0.43 0.43 0.22 2.86 0.259 5.38
NT 2 1.46 1.46 0.73 9.68 0.094 18.19
Residual error 2 0.15 0.15 0.08 1.88
Total 8 8.00 100.00
Flexural strength (rPLA) LH 2 23.50 23.50 11.75 23.61 0.041 80.39
Infill 2 1.42 1.42 0.71 1.43 0.412 4.86
NT 2 3.31 3.31 1.65 3.33 0.231 11.35
Residual error 2 0.99 0.99 0.49 3.40
Total 8 29.23 100.00
Impact strength (vPLA) LH 2 2.35 2.35 1.17 10.88 0.084 47.72
Infill 2 1.72 1.72 0.86 7.96 0.112 34.91
NT 2 0.64 0.64 0.32 2.96 0.252 12.99
Residual error 2 0.22 0.22 0.11 4.38
Total 8 4.92 100.00
Impact strength (rPLA) LH 2 1.44 1.44 0.72 4.16 0.194 35.98
Infill 2 1.62 1.62 0.81 4.68 0.176 40.50
NT 2 0.60 0.60 0.30 1.72 0.368 14.86
Residual error 2 0.35 0.35 0.17 8.65
Total 8 4.01 100.00
Shore D hardness (vPLA) LH 2 0.003 0.003 0.001763 2.46 0.289 32.11
Infill 2 0.004 0.004 0.0021 2.93 0.254 38.24
NT 2 0.001 0.001 0.000911 1.27 0.44 16.59
Residual error 2 0.001 0.001 0.000716 13.05
Total 8 0.011 100.00
Shore D hardness (rPLA) LH 2 0.028 0.028 0.014 28.77 0.034 68.94
Infill 2 0.008 0.008 0.004 8.13 0.11 19.49
NT 2 0.003 0.003 0.001 3.83 0.207 9.17
Residual error 2 0.001 0.001 0.0004 2.40
Total 8 0.041 100.00
Surface roughness Ra (vPLA) LH 2 47.42 47.42 23.71 85.37 0.012 82.09
Infill 2 1.63 1.63 0.81 2.95 0.253 2.83
NT 2 8.15 8.15 4.07 14.68 0.064 14.11
Residual error 2 0.55 0.55 0.27 0.96
Total 8 57.76 100.00
Surface roughness Ra (rPLA) LH 2 43.64 43.64 21.82 44.58 0.022 75.99
Infill 2 0.85 0.85 0.42 0.87 0.536 1.48
NT 2 11.96 11.96 5.98 12.22 0.076 20.83
Residual error 2 0.98 0.98 0.49 1.70
Total 8 57.43 100.00

*DF represents the degree of freedom; Seq SS represents the sequential sum of squares; Adj SS represents the adjusted sum of squares; Adj MS
represents the adjusted mean square; F represents the ratio of explained variance to unexplained variance; p represents the probability of obtain-
ing F value

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3774 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

3.9 DoE confirmation test

Description
The optimal parameters resulting in the most favourable

100
215
0.3
mechanical properties for the vPLA and rPLA speci-

Description
mens have been summarised in Table 9. A confirmation
Level
rPLA
experiment was conducted to validate the potential of

205
0.1
60
3
3
3
the optimised parameter conditions determined by Tagu-
chi analysis to improve the mechanical properties. The
Description

test was necessary to validate the conditions suggested

by the Taguchi analysis, which resulted in the optimal
100
205
0.3

properties. The predicted results for the optimal condi-

Impact Strength

tions were determined solely by considering the opti-

mum performance factors. Confirmation test results for
several mechanical properties, such as tensile strength,
vPLA
Level

Level
rPLA

flexural strength, impact strength, shore D hardness,

and surface roughness, were obtained using the optimal
3
3
2

1
2
2

combination of factors, as listed in Table 9. The obtained

Description

Description

results for these properties were found to be 61.25 MPa

Surface Roughness

and 45.97 MPa for tensile strength, 104.05 MPa and

100
215

205

89.87 MPa for flexural strength, 5.61 and 5.37 kJ/m2 for
0.3

0.1
60

impact strength, 82.30 and 79.60 for shore D hardness, and

1.97 μm and 2.28 μm for surface roughness of vPLA and
vPLA
Level
rPLA

Level

rPLA specimens, respectively. The results demonstrated

3
3
3

1
1
2
Table 9  Summary of the optimum parameters required to attain the highest possible mechanical characteristics

that the samples produced using the optimal combination

of factors exhibited higher levels of accuracy than those
Description

generated by the L9 orthogonal array. Therefore, the S/N

Flexural Strength

ratio of the selected combination of factors was employed

100
215
0.3

Description

for S/Nconfirmation.
The optimal combination of factors was determined by
vPLA
Level

calculating the predicted values. The comparison between

100
215
0.3
3
3
3

the confirmation test data and the predicted results is pre-

sented in Table 10. It is widely recognised in the engi-
Description

neering community that achieving a 95% confidence

interval is a fundamental principle in engineering tests
100
215
0.3

in order to ensure high-quality products and testing valid-

ity [102]. Therefore, the CI for both the confirmed and
Level
rPLA

Level
rPLA

predicted outcomes were determined. For all the attrib-

utes listed in Table 10, the findings showed a signifi-
3
3
3

3
3
3

cant degree of overlap between the CI obtained from the

Description

Description

confirmation tests and predicted outcomes. A significant

observation was made for the surface roughness property,
Shore D hardness
Tensile Strength

as the experimental value showed a comparatively better

100
215

100
215
0.3

0.3

result than the predicted value. The primary reason for

this was the improvement in interlayer adhesion when the
vPLA

vPLA
Level

Level

rPLA samples were printed using an optimal combination

3
3
3

3
3
3

of printing parameters [17, 97]. This result suggests that

Nozzle Temperature (°C)

Nozzle Temperature (°C)

the selected combination of factors in terms of material

preparation and processing factors was acceptable and
Layer Height (mm)

Layer Height (mm)

could be considered as optimal conditions during manu-

facturing. The ANOVA results suggested that the signifi-
cant factors could be accurately predicted and combined
Infill (%)

Infill (%)
Factors

Factors

to achieve good reproducibility.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3775

Table 10  Confirmation test Output response Prediction Confirmation

summary based on optimum
factors S∕N predicted C.I predicted S∕N confirmation C.I confirmation

Tensile strength vPLA 35.80 ± 2.54 35.81 ± 0.21

rPLA 32.21 ± 2.07 33.11 ± 1.21
Flexural strength vPLA 39.74 ± 6.48 40.34 ± 1.45
rPLA 39.43 ± 8.44 39.07 ± 1.47
Impact strength vPLA 14.75 ± 0.31 14.97 ± 0.12
rPLA 14.12 ± 0.26 14.60 ± 0.13
Shore D hardness vPLA 38.21 ± 0.96 38.23 ± 1.41
rPLA 37.62 ± 1.14 38.07 ± 1.12
Surface roughness vPLA − 14.11 ± 2.12 − 5.90 ± 0.17
rPLA − 15.35 ± 2.37 − 7.19 ± 0.14

4 Conclusions roughness being 0.1 mm, 205 ℃, and 80%, respectively.

The results revealed that for a high layer height, infill
The present study employed vPLA and rPLA filaments amount, and nozzle temperature, the bonding between
to investigate the influence of the layer thickness, infill layers increased, which also had a beneficial effect on the
percentage, and nozzle temperature on 3D printed mechanical strength. Microscopic investigations showed
specimen surface roughness, hardness, tensile, flexural, that a high nozzle temperature and layer height resulted in
and Charpy impact properties. This study used the a stronger microstructure. In contrast to this, the existence
Taguchi L9 OA method to design the experiments, of voids was observed when the temperature was low
whilst the obtained results were additionally analysed by (195 ℃), causing a reduction in mechanical properties.
ANOVA to determine the optimal printing parameters Finally, the predicted responses for individual parameters
for maximising the mechanical strength of recycled were verified through experimental confirmation. These
PLA specimens. The experimental results indicated that additional experiments showed that using the optimally
vPLA possessed overall stronger mechanical properties determined printing parameters, rPLA achieved a high
compared to rPLA at all the selected printing factors. The tensile strength of 45.97 MPa, which was comparable to
rPLA specimen properties exhibited lower overall results the tensile strength of vPLA when printed with 0.1 mm
than the vPLA for the same layer thickness and nozzle layer thickness, 215 °C nozzle temperature, and 80% infill
temperature, but the former with a high layer thickness density. Therefore, the experimental investigation proved
and high nozzle temperature could attain almost the same that rPLA could be a feasible environmentally friendly
property as the latter. The main effect plots show that the alternative material for virgin PLA owing to the similarity
tensile and flexural strengths exhibited the highest values of its mechanical properties under optimum processing
when the layer thickness and nozzle temperature were conditions. Since this study evaluated the optimal
high. ANOVA analysis also indicated that layer thickness printing parameters for the layer thickness, infill, and
had the highest significance amongst the input variables, nozzle temperature of recycled PLA specimens, future
with p values below 0.05, for all responses. In addition, research could focus on choosing other important printing
the temperature of the nozzle also played a significant parameters, such as printing speed, infill angle for the
influence in enhancing the tensile characteristics. In evaluation of mechanical properties, and determining
contrast, the infill density had an insignificant influence the fatigue properties of vPLA and rPLA specimens.
on the range of values encountered in this study The investigation and minimisation of the residual
(60–100%). The optimum printing parameters for FDM stress could be an interesting field for future research.
using vPLA and rPLA filament based on ANOVA results In addition, investigating the use of blends of recycled
for tensile, flexural, impact, and shore D hardness tests and virgin PLA may be useful in 3D printing in order to
were 0.3 mm layer thickness, 215 ℃ nozzle temperature, evaluate the resulting mechanical properties and attendant
and 100% infill density with the values for surface sustainability benefits.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3776 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

Acknowledgements The authors are grateful to the technical staff of 6. Pulok MKH, Rahman MS, Chakravarty UK (2021) Crack
the School of Civil and Mechanical Engineering, Curtin University, Propagation and Fracture Toughness of Additively Manu-
Australia, for their valuable technical support. factured Polymers. Proceedings of the ASME 2021 Interna-
tional Mechanical Engineering Congress and Exposition Vol-
Author contribution Experimentation, data curation, and writing the ume 4: Advances in Aerospace Technology. Virtual, Online.
original draft: M.R.H; conceptualisation and methodology: all authors; November 1-5, 2021. ASME. https://​doi.​org/​10.​1115/​imece​
writing, review, and editing: all authors. 2021-​72061
7. Matsuzaki R, Ueda M, Namiki M, Jeong T-K, Asahara H,
Funding Open Access funding enabled and organized by CAUL and Horiguchi K, Nakamura T, Todoroki A, Hirano Y (2016)
its Member Institutions. Three-dimensional printing of continuous-fiber composites
by in-nozzle impregnation. Sci Rep 6:23058. https://​doi.​org/​
Data availability All data generated or analysed during this study, 10.​1038/​srep2​3058
along with associated materials, will be made available at Curtin 8. Jayawardane H, Davies IJ, Gamage JR, John M, Biswas WK
Research Data Collection. (2022) Investigating the ‘techno-eco-efficiency’ performance
of pump impellers: metal 3D printing vs. CNC machining. Int
Declarations J Adv Manuf Technol 121:6811–6836. https://​doi.​org/​10.​1007/​
s00170-​022-​09748-2
Ethics approval and consent to participate The article followed 9. Uddin MS, Sidek MFR, Faizal MA, Ghomashchi R, Pramanik
the guidelines of the Committee on Publication Ethics (COPE) and A (2017) Evaluating mechanical properties and failure mech-
involved no studies on human or animal subjects. anisms of fused deposition modeling acrylonitrile butadiene
styrene parts. J Manuf Sci Eng 139. https://​doi.​org/​10.​1115/1.​
Consent for publication All authors read and approved the final manuscript. 40367​13
10. Maguluri N, Suresh G, Rao KV (2023) Assessing the effect of
Competing interests The authors declare no competing interests. FDM processing parameters on mechanical properties of PLA
parts using Taguchi method. J Thermoplast Compos Mater
36:1472–1488. https://​doi.​org/​10.​1177/​08927​05721​10530​36
Open Access This article is licensed under a Creative Commons Attri- 11. Jayawardane H, Davies IJ, Leadbeater G, John M, Biswas WK
bution 4.0 International License, which permits use, sharing, adapta- (2021) ‘Techno-eco-efficiency’ performance of 3D printed
tion, distribution and reproduction in any medium or format, as long impellers: an application of life cycle assessment. Int J Sustain
as you give appropriate credit to the original author(s) and the source, Manuf 5. https://​doi.​org/​10.​1504/​ijsm.​2021.​116871
provide a link to the Creative Commons licence, and indicate if changes 12. Badia JD, Ribes-Greus A (2016) Mechanical recycling of poly-
were made. The images or other third party material in this article are lactide, upgrading trends and combination of valorization tech-
included in the article’s Creative Commons licence, unless indicated niques. Eur Polymer J 84:22–39. https://​doi.​org/​10.​1016/j.​eurpo​
otherwise in a credit line to the material. If material is not included in lymj.​2016.​09.​005
the article’s Creative Commons licence and your intended use is not 13. Sabil M, Prabhakar DAP (2022) Optimisation of extrusion tem-
permitted by statutory regulation or exceeds the permitted use, you will perature and infill density of PLA material by using L16 orthogo-
need to obtain permission directly from the copyright holder. To view a nal array. Aust J Mech Eng 1–17. https://​doi.​org/​10.​1080/​14484​
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 846.​2022.​20730​22
14. Hikmat M, Rostam S, Ahmed YM (2021) Investigation of tensile
property-based Taguchi method of PLA parts fabricated by FDM
3D printing technology. Results Eng 11:100264. https://​doi.​org/​
10.​1016/j.​rineng.​2021.​100264
References 15. Jayawardane H, Davies IJ, Gamage JR, John M, Biswas WK
(2023) Additive manufacturing of recycled plastics: a ‘techno-
1. Abdullah Aloyaydi B, Sivasankaran S, Rizk Ammar H (2019) eco-efficiency’ assessment. Int J Adv Manuf Technol 126:1471–
Influence of infill density on microstructure and flexural behav- 1496. https://​doi.​org/​10.​1007/​s00170-​023-​11169-8
ior of 3D printed PLA thermoplastic parts processed by fusion 16. Torres J, Cole M, Owji A, Demastry Z, Gordon AP (2016)
deposition modeling. AIMS Mater Sci 6:1033–1048. https://​doi.​ An approach for mechanical property optimization of fused
org/​10.​3934/​mater​sci.​2019.6.​1033 deposition modeling with polylactic acid via design of experi-
2. Jayawardane H, Davies IJ, Gamage JR, John M, Biswas WK ments. Rapid Prototyp J 22:387–404. https://​doi.​org/​10.​1108/​
(2023) Sustainability perspectives – a review of additive and sub- rpj-​07-​2014-​0083
tractive manufacturing. Sustain Manuf Service Econ 2. https://​ 17. Atakok G, Kam M, Koc HB (2022) Tensile, three-point bending
doi.​org/​10.​1016/j.​smse.​2023.​100015 and impact strength of 3D printed parts using PLA and recycled
3. Pinho AC, Amaro AM, Piedade AP (2020) 3D printing goes greener: PLA filaments: a statistical investigation. J Market Res 18:1542–
study of the properties of post-consumer recycled polymers for the 1554. https://​doi.​org/​10.​1016/j.​jmrt.​2022.​03.​013
manufacturing of engineering components. Waste Manag 118:426– 18. Taib N-AAB, Rahman MR, Huda D, Kuok KK, Hamdan S, Bakri
434. https://​doi.​org/​10.​1016/j.​wasman.​2020.​09.​003 MKB, Julaihi MRMB, Khan A (2022) A review on poly lactic
4. Farazin A, Mohammadimehr M (2021) Effect of different acid (PLA) as a biodegradable polymer. Polym Bull 80:1179–
parameters on the tensile properties of printed polylactic acid 1213. https://​doi.​org/​10.​1007/​s00289-​022-​04160-y
samples by FDM: experimental design tested with MDs simu- 19. Zhao XG, Hwang K-J, Lee D, Kim T, Kim N (2018) Enhanced
lation. Int J Adv Manuf Technol 118:103–118. https://​doi.​org/​ mechanical properties of self-polymerized polydopamine-
10.​1007/​s00170-​021-​07330-w coated recycled PLA filament used in 3D printing. Appl Surf
5. Geyer R, Jambeck JR, Law KL (2017) Production, use, and fate Sci 441:381–387. https://​doi.​org/​10.​1016/j.​apsusc.​2018.​01.​257
of all plastics ever made. Sci Adv 3:e1700782. https://​doi.​org/​ 20. Cruz Sanchez FA, Boudaoud H, Hoppe S, Camargo M (2017)
10.​1126/​sciadv.​17007​82 Polymer recycling in an open-source additive manufacturing

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3777

context: mechanical issues. Addit Manuf 17:87–105. https://​doi.​ PLA printed parts. Rapid Prototyp J 26:381–389. https://​doi.​
org/​10.​1016/j.​addma.​2017.​05.​013 org/​10.​1108/​rpj-​02-​2019-​0048
21. Singh S, Singh M, Prakash C, Gupta MK, Mia M, Singh R (2019) 36. Sun Q, Rizvi GM, Bellehumeur CT, Gu P (2008) Effect of pro-
Optimization and reliability analysis to improve surface quality cessing conditions on the bonding quality of FDM polymer fila-
and mechanical characteristics of heat-treated fused filament fab- ments. Rapid Prototyp J 14:72–80. https://d​ oi.o​ rg/1​ 0.1​ 108/1​ 3552​
ricated parts. Int J Adv Manuf Technol 102:1521–1536. https://​ 54081​08620​28
doi.​org/​10.​1007/​s00170-​018-​03276-8 37. Heidari-Rarani M, Ezati N, Sadeghi P, Badrossamay MR (2020)
22. Tan WS, Tanoto YY, Jonoadji N, Christian AA (2021) The effect Optimization of FDM process parameters for tensile properties
of cooling and temperature in 3D printing process with fused of polylactic acid specimens using Taguchi design of experiment
deposition modelling technology on the mechanical properties method. J Thermoplast Compos Mater 35:2435–2452. https://​
with polylactic acid recycled material. Int Rev Mech Eng 15. doi.​org/​10.​1177/​08927​05720​964560
https://​doi.​org/​10.​15866/​ireme.​v15i12.​21573 38. Maazinejad B, Mohammadnia O, Ali GAM, Makhlouf ASH,
23. Ahmad MN, Ishak MR, Mohammad Taha M, Mustapha F, Nadagouda MN, Sillanpää M, Asiri AM, Agarwal S, Gupta VK,
Leman Z, Anak Lukista DD, Irianto Ghazali I (2022) Application Sadegh H (2020) Taguchi L9 (34) orthogonal array study based
of Taguchi method to optimize the parameter of fused deposition on methylene blue removal by single-walled carbon nanotubes-
modeling (FDM) using oil palm fiber reinforced thermoplastic amine: adsorption optimization using the experimental design
composites. Polymers (Basel) 14. https://​doi.​org/​10.​3390/​polym​ method, kinetics, equilibrium and thermodynamics. J Mol Liq
14112​140 298. https://​doi.​org/​10.​1016/j.​molliq.​2019.​112001
24. Dey A, Hoffman D, Yodo N (2020) Optimizing multiple process 39. Chari VS, Venkatesh P, Krupashankar K, Dinesh V (2018) Effect
parameters in fused deposition modeling with particle swarm of processing parameters on FDM process. AIP Conference Pro-
optimization. Int J Interact Des Manuf 14:393–405. https://​doi.​ ceedings. AIP Publishing. https://​doi.​org/​10.​1063/1.​50296​37
org/​10.​1007/​s12008-​019-​00637-9 40. Sood AK, Ohdar RK, Mahapatra SS (2012) Experimental investi-
25. Correia C, Gomes TEP, Gonçalves I, Neto V (2022) Reprocess- gation and empirical modelling of FDM process for compressive
ability of PLA through chain extension for fused filament fabrica- strength improvement. J Adv Res 3:81–90. https://​doi.​org/​10.​
tion. J Manuf Mater Process 6:26. https://d​ oi.o​ rg/1​ 0.3​ 390/j​ mmp6​ 1016/j.​jare.​2011.​05.​001
010026 41. Aa A, Qattawi A, Alrawi B, Guzman A (2017) Experimental
26. Breški T, Hentschel L, Godec D, Đuretek I (2021) Suitability of optimization of fused deposition modelling processing param-
recycled PLA filament application in fused filament fabrication eters: a design-for-manufacturing approach. Procedia Manuf
process. Tehnički glasnik 15:491–497. https://​doi.​org/​10.​31803/​ 10:791–803. https://​doi.​org/​10.​1016/j.​promfg.​2017.​07.​079
tg-​20210​80512​0621 42. Freeland B, McCarthy E, Balakrishnan R, Fahy S, Boland A,
27. Subramaniyan M, Karuppan S, Eswaran P, Appusamy A, Naveen Rochfort KD, Dabros M, Marti R, Kelleher SM, Gaughran J
Shankar A (2021) State of art on fusion deposition modeling (2022) A review of polylactic acid as a replacement material for
machines process parameter optimization on composite mate- single-use laboratory components. Materials 15. https://​doi.​org/​
rials. Mater Today: Proc 45:820–827. https://​doi.​org/​10.​1016/j.​ 10.​3390/​ma150​92989
matpr.​2020.​02.​865 43. Auffray L, Gouge P-A, Hattali L (2022) Design of experiment
28. Lanzotti A, Grasso M, Staiano G, Martorelli M (2015) The analysis on tensile properties of PLA samples produced by fused
impact of process parameters on mechanical properties of parts filament fabrication. Int J Adv Manuf Technol 118:4123–4137.
fabricated in PLA with an open-source 3-D printer. Rapid Pro- https://​doi.​org/​10.​1007/​s00170-​021-​08216-7
totyp J 21:604–617. https://​doi.​org/​10.​1108/​rpj-​09-​2014-​0135 44. Lokesh N, Praveena BA, Sudheer Reddy J, Vasu VK, Vijaykumar
29. Eryildiz M (2021) Effect of build orientation on mechanical S (2022) Evaluation on effect of printing process parameter through
behaviour and build time of FDM 3D-printed PLA parts: an Taguchi approach on mechanical properties of 3D printed PLA speci-
experimental investigation. Eur Mech Sci 5:116–120. https://​ mens using FDM at constant printing temperature. Mater Today:
doi.​org/​10.​26701/​ems.​881254 Proc 52:1288–1293. https://​doi.​org/​10.​1016/j.​matpr.​2021.​11.​054
30. Benamira M, Benhassine N, Ayad A, Dekhane A (2023) Inves- 45. Hanon MM, Dobos J, Zsidai L (2021) The influence of 3D
tigation of printing parameters effects on mechanical and failure printing process parameters on the mechanical performance of
properties of 3D printed PLA. Eng Fail Anal 148. https://d​ oi.o​ rg/​ PLA polymer and its correlation with hardness. Procedia Manuf
10.​1016/j.​engfa​ilanal.​2023.​107218 54:244–249. https://​doi.​org/​10.​1016/j.​promfg.​2021.​07.​038
31. Tsouknidas A, Pantazopoulos M, Katsoulis I, Fasnakis D, Maro- 46. Park K, Kim G, No H, Jeon HW, Kremer GEO (2020) Identifica-
poulos S, Michailidis N (2016) Impact absorption capacity of tion of optimal process parameter settings based on manufactur-
3D-printed components fabricated by fused deposition model- ing performance for fused filament fabrication of CFR-PEEK.
ling. Mater Des 102:41–44. https://​doi.​org/​10.​1016/j.​matdes.​ Appl Sci 10. https://​doi.​org/​10.​3390/​app10​134630
2016.​03.​154 47. Oloyede CT, Jekayinfa SO, Alade AO, Ogunkunle O, Laseinde
32. Carneiro OS, Silva AF, Gomes R (2015) Fused deposition mod- OT, Adebayo AO, Abdulkareem AI, Smaisim GF, Fattah IMR
eling with polypropylene. Mater Des 83:768–776. https://d​ oi.o​ rg/​ (2023) Synthesis of biobased composite heterogeneous catalyst
10.​1016/j.​matdes.​2015.​06.​053 for biodiesel production using simplex lattice design mixture:
33. Kam M, İpekci A, Şengül Ö (2021) Taguchi optimization of optimization process by Taguchi method. Energies 16. https://​
fused deposition modeling process parameters on mechanical doi.​org/​10.​3390/​en160​52197
characteristics of PLA+ filament material. Scientia Iranica. 48. Mrabti IE, Bouziane K, Touache A, Hakimi AE, Chamat A,
https://​doi.​org/​10.​24200/​SCI.​2021.​57012.​5020 Daya A (2022) Effect of process parameters on the deep draw-
34. Suteja TJ, Soesanti A (2020) Mechanical properties of 3D printed ing formability of aluminum and advanced high-strength steel
polylactic acid product for various infill design parameters: a square cups. Int J Adv Manuf Technol 124:1827–1842. https://​
review. J Phys: Conf Ser 1569. https://​doi.​org/​10.​1088/​1742-​ doi.​org/​10.​1007/​s00170-​022-​10616-2
6596/​1569/4/​042010 49. Boswell B, Islam MN, Davies IJ, Pramanik A (2015) Effect of
35. Behzadnasab M, Yousefi AA, Ebrahimibagha D, Nasiri F (2019) machining parameters on the surface finish of a metal matrix
Effects of processing conditions on mechanical properties of

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

3778 The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779

composite under dry cutting conditions. Proc Inst Mech Eng Part William Andrew Publishing. pp. 165–172. https://​doi.​org/​10.​
B: J Eng Manuf 231:913–923. https://​doi.​org/​10.​1177/​09544​05415​ 1016/​B978-0-​323-​44371-5.​00019-0
583776 66. ASTM-D6110-18 (2018) Standard test method for determining
50. ASTM-D638-22 (2022) Standard test method for tensile proper- the charpy impact resistance of notched specimens of plastics.
ties of plastics. ASTM International. West Conshohocken, PA, ASTM International. West Conshohocken, PA, United States.
United States. https://​doi.​org/​10.​1520/​D0638-​22 https://​doi.​org/​10.​1520/​d6110-​18
51. Ramesh M, Panneerselvam K (2020) PLA-Based Material Design 67. Popelka A, Zavahir S, Habib S (2020) Morphology analysis. In:
and Investigation of Its Properties by FDM. In: Shunmugam MS, AlMaadeed MAA, Ponnamma D, Carignano MA (eds) Polymer
Kanthababu M (eds) Advances in Additive Manufacturing and Science and Innovative Applications. Elsevier. pp. 21–68. https://​
Joining. Lecture Notes on Multidisciplinary Industrial Engineer- doi.​org/​10.​1016/​b978-0-​12-​816808-​0.​00002-0
ing. Springer, Singapore. pp. 229–241. https://​doi.​org/​10.​1007/​ 68. Vidakis N, David C, Petousis M, Sagris D, Mountakis N (2023)
978-​981-​32-​9433-2_​20 Optimization of key quality indicators in material extrusion 3D
52. ASTM-D618-21 (2021) Standard practice for conditioning printing of acrylonitrile butadiene styrene: the impact of criti-
plastics for testing. ASTM International. West Conshohocken, cal process control parameters on the surface roughness, dimen-
PA, United States. https://​doi.​org/​10.​1520/​d0618-​21 sional accuracy, and porosity. Mater Today Commun 34. https://​
53. Kim G, Barocio E, Tsutsui W, Wang P, Dubikovsky S, Pipes RB, doi.​org/​10.​1016/j.​mtcomm.​2022.​105171
Sterkenburg R (2021) Enhancing surface characteristics of addi- 69. Kim YJA (2014) The Effect of Surface Roughness on Biofilm
tively manufactured fiber reinforced thermoplastic mold using Adhesion to Clear Heat-cured Poly (methyl Methacrylate)
thermoset coating with ceramic particles. Surf Coat Technol 422. used for Ocular Prostheses. Master of Health Sciences, University
https://​doi.​org/​10.​1016/j.​surfc​oat.​2021.​127536 of Otago. [Online]. Available: http://​hdl.​handle.​net/​10523/​4792
54. ASTM-D2240-15 (2021) Standard test method for rubber prop- 70. Kackar RN (1989) Taguchi’s quality philosophy: analysis and
erty-durometer hardness. ASTM International. West Consho- commentary. In: Dehnad K (ed) Quality Control, Robust Design,
hocken, PA, United States. https://​doi.​org/​10.​1520/​d2240-​15r21 and the Taguchi Method. Springer, US Boston, MA, pp 3–21
55. Chand R, Sharma VS, Trehan R, Gupta MK, Sarikaya M (2023) 71. Peace GS (1993) Taguchi methods: a hands-on approach. Addi-
Investigating the dimensional accuracy and surface roughness for son Wesley Publishing Company
3D printed parts using a multi-jet printer. J Mater Eng Perform 72. Phadke MS, Kackar RN, Speeney DV, Grieco MJ (1989) Off-line
32:1145–1159. https://​doi.​org/​10.​1007/​s11665-​022-​07153-0 quality control in integrated circuit fabrication using experimen-
56. Bouzaglou O, Golan O, Lachman N (2023) Process design and tal design. In: Dehnad K (ed) Quality Control, Robust Design,
parameters interaction in material extrusion 3D printing: a review. and the Taguchi Method. Springer, US Boston, MA, pp 99–141
Polymers (Basel) 15. https://​doi.​org/​10.​3390/​polym​15102​280 73. Dehnad K (2012) Quality control, robust design, and the Taguchi
57. Mishra P, Sood S, Bharadwaj V, Aggarwal A, Khanna P (2023) method Springer New York, NY
Parametric modeling and optimization of dimensional error and 74. Tamizharasan T, Senthilkumar N, Selvakumar V, Dinesh S
surface roughness of fused deposition modeling printed polyeth- (2019) Taguchi’s methodology of optimizing turning parameters
ylene terephthalate glycol parts. Polymers (Basel) 15. https://d​ oi.​ over chip thickness ratio in machining P/M AMMC. SN Appl Sci
org/​10.​3390/​polym​15030​546 1. https://​doi.​org/​10.​1007/​s42452-​019-​0170-8
58. Gadelmawla ES, Koura MM, Maksoud TMA, Elewa IM, Soli- 75. Ross PJ (1988) Taguchi techniques for quality engineering:
man HH (2002) Roughness parameters. J Mater Process Technol loss function, orthogonal experiments, parameter and toler-
123:133–145. https://​doi.​org/​10.​1016/​s0924-​0136(02)​00060-2 ance design. McGraw-Hill
59. Lalegani Dezaki M, Ariffin MKAM, Serjouei A, Zolfagharian 76. Chen W-H, Carrera Uribe M, Kwon EE, Lin K-YA, Park Y-K,
A, Hatami S, Bodaghi M (2021) Influence of infill patterns gen- Ding L, Saw LH (2022) A comprehensive review of thermo-
erated by CAD and FDM 3D printer on surface roughness and electric generation optimization by statistical approach: Taguchi
tensile strength properties. Appl Sci 11:7272. https://​doi.​org/​10.​ method, analysis of variance (ANOVA), and response surface
3390/​app11​167272 methodology (RSM). Renew Sustain Energy Rev 169. https://​
60. Fly D, Açan M (2013) 3D Printed Internal Structure: Influence on doi.​org/​10.​1016/j.​rser.​2022.​112917
Tensile Strength. 2013 North Midwest Section Meeting. Fargo, 77. Anderson MJ (2001) A new method for non-parametric multi-
North Dakota. October 17-18, 2013. ASEE. https://​doi.​org/​10.​ variate analysis of variance. Austral Ecol 26:32–46. https://​doi.​
18260/1-​2-​1139-​36234 org/​10.​1111/j.​1442-​9993.​2001.​01070.​pp.x
61. ASTM-D7127-17 (2017) Standard test method for measure- 78. Yang WH, Tarng YS (1998) Design optimization of cutting
ment of surface roughness of abrasive blast cleaned metal sur- parameters for turning operations based on the Taguchi method.
faces using a portable stylus instrument. ASTM International. J Mater Process Technol 84:122–129. https://​doi.​org/​10.​1016/​
West Conshohocken, PA, United States. https://​doi.​org/​10.​1520/​ s0924-​0136(98)​00079-x
d7127-​17 79. Rizk TH (2023) Chapter 31 - Analysis of variance. In: Eltorai
62. Maidin S, Fadani I, Md. Nor Hayati N, Albaluooshi H, (2022) AEM, Liu T, Chand R, Kalva SP (eds) Translational Interven-
Application of Taguchi method to optimize fused deposition tional Radiology. Academic Press. pp. 149–152. https://​doi.​org/​
modeling process parameters for surface roughness. J Teknol 10.​1016/​B978-0-​12-​823026-​8.​00024-9
84:29–37. https://​doi.​org/​10.​11113/​jurna​ltekn​ologi.​v84.​18430 80. Greenwald AG, Gonzalez R, Harris RJ, Guthrie D (1996) Effect
63. ASTM-E4 (2021) Standard practices for force calibration and sizes and p values: what should be reported and what should be
verification of testing machines. ASTM International. West Con- replicated? Psychophysiology 33:175–183. https://​doi.​org/​10.​
shohocken, PA, United States. https://​doi.​org/​10.​1520/​e0004-​21 1111/j.​1469-​8986.​1996.​tb021​21.x
64. ASTM-D790-17 (2017) Standard test methods for flexural prop- 81. Belavendram N (1995) Quality by design: taguchi techniques for
erties of unreinforced and reinforced plastics and electrical insu- industrial experimentation. Prentice Hall
lating materials. ASTM International. West Conshohocken, PA, 82. Bikas H, Stavropoulos P, Chryssolouris G (2015) Additive manu-
United States. https://​doi.​org/​10.​1520/​D0790-​17 facturing methods and modelling approaches: a critical review.
65. Chandrasekaran C (2017) 19 - Testing of Rubber Lining. Int J Adv Manuf Technol 83:389–405. https://​doi.​org/​10.​1007/​
In: Chandrasekaran C (ed) Anticorrosive Rubber Lining. s00170-​015-​7576-2

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

The International Journal of Advanced Manufacturing Technology (2024) 131:3751–3779 3779

83. Mani M, Karthikeyan AG, Kalaiselvan K, Muthusamy P, Muru- structure on the crystalline forms (alpha’ and alpha) and mechan-
ganandhan P (2022) Optimization of FDM 3-D printer process ical properties of wet spinning fibres. Polymers (Basel) 9. https://​
parameters for surface roughness and mechanical properties doi.​org/​10.​3390/​polym​90100​18
using PLA material. Mater Today: Proc 66:1926–1931. https://​ 94. Kumar S, Singh R, Singh TP, Batish A (2019) Flexural, pull-
doi.​org/​10.​1016/j.​matpr.​2022.​05.​422 out, and fractured surface characterization for multi-material 3D
84. Ayrilmis N (2018) Effect of layer thickness on surface properties printed functionally graded prototype. J Compos Mater 54:2087–
of 3D printed materials produced from wood flour/PLA filament. 2099. https://​doi.​org/​10.​1177/​00219​98319​892067
Polym Testing 71:163–166. https://​doi.o​ rg/1​ 0.​1016/j.​polyme​ rtes​ 95. Thoden van Velzen EU, Chu S, Alvarado Chacon F, Brouwer
ting.​2018.​09.​009 MT, Molenveld K (2020) The impact of impurities on the
85. Shirmohammadi M, Goushchi SJ, Keshtiban PM (2021) Opti- mechanical properties of recycled polyethylene. Packag Technol
mization of 3D printing process parameters to minimize surface Sci 34:219–228. https://​doi.​org/​10.​1002/​pts.​2551
roughness with hybrid artificial neural network model and parti- 96. Kamaal M, Anas M, Rastogi H, Bhardwaj N, Rahaman A (2020)
cle swarm algorithm. Prog Addit Manuf 6:199–215. https://​doi.​ Effect of FDM process parameters on mechanical properties of
org/​10.​1007/​s40964-​021-​00166-6 3D-printed carbon fibre–PLA composite. Prog Addit Manuf
86. Zeng YS, Hsueh MH, Lai CJ, Hsiao TC, Pan CY, Huang WC, 6:63–69. https://​doi.​org/​10.​1007/​s40964-​020-​00145-3
Chang CH, Wang SH (2022) An investigation on the hardness of 97. Syrlybayev D, Zharylkassyn B, Seisekulova A, Akhmetov M,
polylactic acid parts fabricated via fused deposition modeling. Perveen A, Talamona D (2021) Optimisation of strength proper-
Polymers (Basel) 14. https://​doi.​org/​10.​3390/​polym​14142​789 ties of FDM printed parts-a critical review. Polymers (Basel) 13.
87. Maguluri N, Suresh G, Guntur SR (2022) Effect of printing https://​doi.​org/​10.​3390/​polym​13101​587
parameters on the hardness of 3D printed poly-lactic acid parts 98. Tao Y, Kong F, Li Z, Zhang J, Zhao X, Yin Q, Xing D, Li P
using DOE approach. IOP Conf Ser: Mater Sci Eng 1248. https://​ (2021) A review on voids of 3D printed parts by fused filament
doi.​org/​10.​1088/​1757-​899x/​1248/1/​012004 fabrication. J Market Res 15:4860–4879. https://​doi.​org/​10.​
88. Şirin Ş, Aslan E, Akincioğlu G (2022) Effects of 3D-printed PLA 1016/j.​jmrt.​2021.​10.​108
material with different filling densities on coefficient of friction 99. Akhoundi B, Nabipour M, Hajami F, Shakoori D (2020) An
performance. Rapid Prototyp J 29:157–165. https://​doi.​org/​10.​ experimental study of nozzle temperature and heat treatment
1108/​rpj-​03-​2022-​0081 (annealing) effects on mechanical properties of high-tempera-
89. Hsueh MH, Lai CJ, Liu KY, Chung CF, Wang SH, Pan CY, ture polylactic acid in fused deposition modeling. Polym Eng
Huang WC, Hsieh CH, Zeng YS (2021) Effects of printing tem- Sci 60:979–987. https://​doi.​org/​10.​1002/​pen.​25353
perature and filling percentage on the mechanical behavior of 100. Shahavi MH, Hosseini M, Jahanshahi M, Meyer RL, Darzi GN
fused deposition molding technology components for 3d printing. (2015) Clove oil nanoemulsion as an effective antibacterial agent:
Polymers (Basel) 13. https://​doi.​org/​10.​3390/​polym​13172​910 Taguchi optimization method. Desalin Water Treat 57:18379–
90. Lanzotti A, Martorelli M, Maietta S, Gerbino S, Penta F, Gloria 18390. https://​doi.​org/​10.​1080/​19443​994.​2015.​10928​93
A (2019) A comparison between mechanical properties of speci- 101. Fazita RN, Johary N, Khalil HPSA, Norazli N, Azniwati A,
mens 3D printed with virgin and recycled PLA. Procedia CIRP Haafiz MKM (2021) Parameter optimization via the Taguchi
79:143–146. https://​doi.​org/​10.​1016/j.​procir.​2019.​02.​030 method to improve the mechanical properties of bamboo particle
91. Dave HK, Prajapati AR, Rajpurohit SR, Patadiya NH, Raval reinforced polylactic acid composites. BioResources 16:1914–
HK (2020) Open hole tensile testing of 3D printed parts using 1939. https://​doi.​org/​10.​15376/​biores.​16.1.​1914-​1939
in-house fabricated PLA filament. Rapid Prototyp J 26:21–31. 102. Salam H, Dong Y, Davies IJ, Pramanik A (2018) Identification
https://​doi.​org/​10.​1108/​rpj-​01-​2019-​0003 of preferred combination of factors in manufacturing bioepoxy/
92. Dong Y, Milentis J, Pramanik A (2018) Additive manufacturing clay nanocomposites. Adv Compos Mater 27:511–530. https://​
of mechanical testing samples based on virgin poly (lactic acid) doi.​org/​10.​1080/​09243​046.​2018.​14801​47
(PLA) and PLA/wood fibre composites. Adv Manuf 6:71–82.
https://​doi.​org/​10.​1007/​s40436-​018-​0211-3 Publisher's Note Springer Nature remains neutral with regard to
93. Puchalski M, Kwolek S, Szparaga G, Chrzanowski M, Krucin- jurisdictional claims in published maps and institutional affiliations.
ska I (2017) Investigation of the influence of PLA molecular

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]