Processing and mechanical properties of novel

biodegradable poly-lactic acid/Zn 3D printed scaffolds

for application in tissue regeneration
C. Pascual-González1,2,∗ , J. de la Vega1 , C. Thompson1,3 , J. P.
Fernández-Blázquez1 , D. Herráez-Molinero1,4 , N. Biurrun1,4 , I.
Lizarralde1,4,5 , J. Sánchez del Rı́o1,6 , C. González1,4 , J. LLorca1,4,∗∗
1
IMDEA Materials Institute, C/Eric Kandel 2, 28906 Getafe, Madrid, Spain.
2
Materials Science and Engineering Area, Rey Juan Carlos University, C/Tulipán s/n,
28933, Móstoles, Madrid, Spain.
3
Department of Material Science and Engineering, Universidad Carlos III de Madrid,
28911 - Leganés, Madrid, Spain. 4 Department of Materials Science, Polytechnic
University of Madrid/Universidad Politécnica de Madrid, E. T. S. de Ingenieros de
Caminos, 28040 - Madrid, Spain.
5
Hexcel Composites S.L., C/ Bruselas 10-16, 28983, Parla, Madrid, Spain.
6
Departamento de Ingenierı́a Eeléctrica, Electrónica Wutomática y Fı́sica Aplicada,
Universidad Politécnica de Madrid, E. T. S. de Ingenierı́a y Diseño Industrial, 28012 -
Madrid, Spain.

Abstract
The feasibility to manufacture scaffolds of poly-lactic acid reinforced with
Zn particles by fused filament fabrication is demonstrated for the first time.
Filaments of 2.85 mm in diameter of PLA reinforced with different weight
fractions of µm-sized Zn - 1 wt. % Mg alloy particles (in the range 3.5 to
17.5 wt. %) were manufactured by a double extrusion in method in which
standard extrusion is followed by a precision extrusion in a filament-maker
machine. Filaments with constant diameter, negligible porosity and a homo-
geneous reinforcement distribution were obtained for Zn weight fractions of
up to 10.5%. It was found that the presence of Zn particles led to limited
changes in the physico-chemical properties of the PLA that did not affect the
window temperature for 3D printing nor the melt flow index. Thus, porous

∗
Corresponding author: [email protected]
∗∗
Corresponding author: [email protected], [email protected]

Journal of the Mechanical Behavior of Biomedical Materials, accepted for publicationMay 20, 2022
scaffolds could be manufactured by fused filament fabrication at 190°C with
poly-lactic acid/Zn composites containing 3.5 and 7 wt. % of Zn and at 170°C
when the Zn content was 10.5 wt. % with excellent dimensional accuracy
and mechanical properties.
Keywords: Poly-lactic acid, Zinc, Biodegradable, Scaffold, composites,
Fused filament fabrication

1. Introduction
Temporary implants and scaffolds from bioresorbable materials, that can
be progressively degraded and absorbed in the human body, have tremendous
potential for tissue engineering applications [1–3]. As opposed to permanent
implants, that remain as foreign materials in the body after healing and may
require second surgeries to remove the implant because of inflammatory reac-
tions, bioresorbable implants disappear and only the natural tissue remains.
Among biodegradable polymers, poly-lactic acid (PLA) is one of the most
widely used and it has been approved by the Food and Drug Administra-
tion (FDA) of the United States for implantation in the human body for
tissue fixation devices and drug-delivery systems[4–7]. Moreover, PLA is a
thermoplastic polymer very suitable to be deposited using the fused filament
fabrication (FFF) technique [8]. Thus, patient-specific implants and scaf-
folds of PLA can be easily manufactured by additive manufacturing from
computer-assisted tomography medical images [9].
Obviously, the mechanical properties, degradation rate and degradation
products of PLA are not optimum for all biomedical applications. For in-
stance, the mechanical properties are low and the degradation times too
long (1 to 2 years) for orthopaedic applications [10]. Moreover, although
products resulting from the biodegradation of PLA are non-toxic and can
be metabolised by the body, they contain acidic components that do not
always favour tissue integration [11]. In order to overcome these limitations,
composites made up of a PLA matrix reinforced with metallic particles have
been manufactured. For instance, PLA/Ti composite filaments, suitable for
FFF, were prepared by a double thermal mixing route. The tensile and
compressive mechanical properties as well as the cytocompatibility of porous
scaffolds manufactured by FFF were enhanced when the volume fraction of
Ti particles was ≤ 10%. Nevertheless, this work did not provide evidence
that high quality filaments (with constant diameter, good particle disper-

2
sion and negligible porosity) could be manufactured by this route and, in
addition, Ti particles are inert but not biodegradable. Similar results and
limitations were reported in the case of PLA filaments reinforced with 316L
stainless steel particles [12].
Most of the investigations to reinforce PLA with metallic particles have
been focussed in Mg that is also biodegradable. Antoniac et al. [13] reported
the viability of PLA/Mg filaments (4 wt% of Mg) for FFF manufactured by
conventional extrusion while a colloidal route was developed by Ferrández
et al. [14, 15] that allows higher particle volume fractions. More recently,
PLA/Mg filaments were manufactured by a double extrusion method (a stan-
dard extrusion followed by a precision extrusion) that allows to produce high
quality filaments with constant diameter, full density and homogeneuos dis-
persion of the Mg particles [16]. They were successfully used to print face-
centered cubic scaffolds with optimum dimensional accuracy and fully-dense
struts. It should be noted, however, that the combination of PLA and Mg
particles has several limitations. Firstly, thermal degradation of PLA at the
extrusion and printing temperatures can be catalyzed by the presence of MgO
[17–19]. As a result, the molecular weight (MW), viscosity and mechanical
properties of PLA/Mg composites are reduced [20] and it is more difficult to
find the appropriate processing window during FFF [16]. Secondly, Mg par-
ticles suffer from rapid corrosion rates, especially in chloride solutions such
as body fluids, that cause the formation of hydrogen gas and a high local pH
around the implant, which may lead to necrosis of tissues [21–23]. Finally,
the increase in volume of the Mg fibers and particles due to the formation of
Mg(OH)2 leads to the cracking of the PLA matrix and to a large reduction
in the mechanical properties [24].
An interesting option to overcome the problems associated with the high
reactivity of Mg is to use another biodegradable metal such as Zn. Zn is an
essential element for the body and is involved in a high number of enzymatic
reactions in which it participates as a cofactor of more than 300 enzymes
[25]. This element is also associated with biological functions that include
anti-inflammatory and anti-proliferative properties, as well as in the pre-
vention of endothelial cell apoptosis, atherosclerosis risk reduction, among
other beneficial properties to the body [26]. Moreover, corrosion rates of
Zn are much lower than those of Mg and Zn alloys have demonstrated ex-
cellent biocompatibility [27–29]. However, to the authors’ knowledge, the
development of PLA/Zn filaments for FFF has not been previously achieved
and this is the main objective of this investigation. Dense, homogeneous

3
and printable PLA/Zn filaments with constant diameter were manufactured
containing different weight fractions of Zn particles. The influence of the Zn
particles and of the processing on the physico-chemical and mechanical prop-
erties of PLA/Zn composites was investigated. Finally, porous scaffolds were
manufactured by FFF and their microstructure and mechanical properties
were analyzed. These results show the viability of the fabrication of PLA/Zn
composite scaffolds for applications in regenerative tissue engineering.

2. Methods
2.1. Materials and filament fabrication
PLA/Zn filaments were manufactured from Ingeo Biopolymer 2003D PLA
(NatureWorks LLC, Minnesota, USA) and spherical particles of Zn - 1 wt.%
Mg (Zn-Mg1) alloy produced by gas atomization by Meotec GmbH (Aachen,
Germany). The size, morphology and element distribution in the Zn-Mg1
particles were examined by scanning electron microscopy (SEM) (Helios
NanoLab 600i, FEI) using secondary electrons as well as energy dispersive
spectrometry (EDS). The PLA pellets were dried in a conventional oven at
60°C for 24 hours and mixed with the Zn alloy powders by means of manual
agitation, ensuring that most of the metallic powders were stuck on the pellet
surface. PLA and PLA mixed with 3.5, 7, 10.5, 14 and 17.5 wt. % of Zn-Mg1
particles was used. The mixtures were subjected to standard extrusion at
180°C in a microcompounder (Xplore MC 15) that has two conical mixing
screws. The residence time was approximately 3 minutes and the rotating
speed of the screws 100 rpm. The diameter of extruded filament was not
constant (and, thus, not suitable for 3D printing by FFF). The PLA/Zn
filaments were pelletized (Granulator, Brabender) and the composite pel-
lets were used to feed a filament-maker machine that carries out a precision
extrusion (Precision 450, 3Devo, Utrecht, The Netherlands). Precision 450
filament-maker includes four heaters along the mixing screw and, in addition,
the diameter of the filament is continuously measured with an optical sensor
that also provides a feedback control to the automated spooling system. The
nominal diameter was set to 2.85 mm, the standard diameter used by the
Ultimaker S5 3D printer. The heater temperatures were optimised depend-
ing on the metallic particle content and they are indicated in Table 1. The
temperature of all heaters was initially 10% above the melting temperature
determined by DSC. Afterwards, the temperature of each heater was manu-
ally modified until a proper flow of material was attained. This process was

4
guided by the MFI results which indicated that the viscosity decreased with
increasing Zn content. The residence time was approximately 12 minutes.
More details about the filament production can be found in [16].
Material heater heater heater heater
1 2 3 4
PLA and PLA/Zn (3.5%, 7%) 180 200 180 170
PLA/Zn (10.5%, 14%) 170 190 170 160
PLA/Zn (17.5%) 160 180 160 150
Table 1: Temperature (in °C) of the heaters in the Precision 450 filament-maker during
precision extrusion of PLA/Zn composites. Zn content is expressed in wt. %.

2.2. Filament characterisation

The internal features of the composite filaments (porosity and particle dis-
tribution) were analyzed by X-ray computed tomography (XCT) in a Phoenix
Nanotom (General Electric). XCT was carried out with a W target and 0-
mode nanofocus at 50 KV with a current of 300 µA and the voxel size was
2 µm. In addition, PLA/Zn composite filaments were analyzed by differen-
tial scanning calorimetry (DSC) (Q200, TA Instruments) up to 180°C with
a heating rate of 10°C/min. Thermo-gravimetric analysis (TGA) (Q50, TA
Instruments) was used to evaluate the thermal degradation of PLA/Zn spec-
imens up to 500°C in air from the weight variation as a function of temper-
ature. The molecular weight (MW) distribution of the PLA/Zn composites
was measured by gel permeation chromatography (GPC) (GPC 2414, Wa-
ters). The system uses a Waters 2424 refractive index detector and a series
of narrow polystyrene standards with tetrahydrofuran as the mobile phase.
Melt flow index (MFI) tests of the filament after standard and precision ex-
trusion were carried out in a MFI tester (KJ-3092, Kinsgeo) with a load
of 2.16 Kg for 10 min following ASTM standard D1238. Finally, variable-
temperature X-ray scattering experiments were performed in an Empyrean
PANalytical X-Ray diffractometer with Ni-filtered Cu Kα radiation. The
heating stage is a homemade instrument using a coil heater provided by
Resistencias Industriales Maxiwatt and controlled with an Eurotherm 2416
temperature controller.
2.3. Scaffold 3D printing and characterisation
Scaffolds of 20 x 20 x 20 mm3 following a face-centered cubic (FCC) lattice
pattern were 3D printed by FFF with an Ultimaker S5 3D printer with

5
the precision extruded filaments. The struts of the lattice have a circular
section of 1.5 mm in diameter. The scaffolds were printed layer-by-layer
with a travel velocity of 5 mm/s and a nozzle of 400 µm in diameter. The
printing temperature was in the range 190°C to 210°C depending on the
PLA/Zn filament composition while the bed temperature was kept at 70°C.
The microstructure of selected scaffolds was studied using XCT. The sample
and detector positions were modified to attain a voxel size of 12 µm. In
addition, the elastic modulus and strength of the scaffolds were measured
by means of compression tests between under displacement control at cross-
head speed of 0.5 mm/min in an Instron 3384 universal mechanical testing
machine. Teflon paste was inserted between the scaffold and the compression
plates to minimize friction. The applied load was measured with a load cell
while the reduction in length was determined by means of a linear variable
differential transducer (LVDT) in contact with the compression plates.

3. Results and discussion

3.1. Porosity and particle distribution in PLA /Zn filaments
SEM images of the ZnMg1 particles are shown in Figures 1 (a) and (b).
Particles are spherical and the diameter distribution is in the range 2 µm
to 25 µm. The Zn and Mg distribution in the particles was analyzed by
EDS and the results are shown in Figures 1(c) and (d). Zn is homogeneously
distributed in the particles which show some Mg-rich regions. The solubility
of Mg in Zn is very low and the Mg-rich regions are formed by the Mg2 Zn11
eutectic phase [30].
The filaments with different weight fractions of Zn (x= 3.5%, 7%, 10.5%,
14% and 17.5%) manufactured by standard extrusion and standard extru-
sion followed by precision extrusion were analyzed by XCT. Representative
cross-sections are depicted in Figures 2(a) and (b) for standard and precision
extruded filaments, respectively. Black regions correspond to pores while
white regions stand for spherical Zn particles. In general, a homogeneous
distribution of the Zn particles in the PLA matrix is obtained after standard
or precision extrusion up to 14 wt. % of Zn. However, the concentration of
the filler Zn particles is higher near the center of the filament in the standard
extruded filaments with 17.5 wt. % Zn. Filler migration towards the center of
the filament during polymer extrusion has been previously reported [31, 32]
and it is the result of the radial forces that develop on the filler particles
during extrusion.

6
Figure 1: (a) and (b) Secondary electrons SEM images of the Zn-Mg1 powder. (c) and
(d) EDS analysis showing the distribution of Zn and Mg in the particles.

The tomograms of filaments of 3 mm in length were segmented to deter-

mine the pore volume fraction that can be found in Table 2. White regions
stand for the Zn particle while black regions are pores. Standard extruded
materials present non-constant diameter and irregular cross-section shape.
The porosity was low except for the filaments with 17.5 wt. % Zn (Fig-
ure 2(a)). Precision extruded filaments have constant diameter and circular
shape and the porosity of the filaments is practically reduced to zero after
precision extrusion when the Zn content is up to 10.5% (Fig. 2(b)). However,
the porosity increases rapidly for 14 wt. % and 17.5wt. % of Zn particles
and large pores can be observed in the filament containing 17.5 wt. % Zn in
Figure 2(b)).
The porosity of the PLA/Zn filaments is higher than that reported in
PLA/Mg filaments processed by the same route and with equivalent volume
fraction of metallic particles [16]. The higher porosity of the PLA/Zn fila-
ments may be attributed to the large content of small Zn particles (< 5µm),
which increases the contact surface between metal and polymer. This phe-

7
 Zn content Standard Precision
(wt. %) extrusion extrusion
3.5% 0.20 0
7% 0.80 0
10.5% 0.15 0.68
14% 0.16 2.23
17.5% 1.23 13.80
Table 2: Pore volume fraction (in %) in PLA/Zn filaments processed by standard and
precision extrusion.

nomenon leads to higher porosity because of the gas layers surrounding the
particles [32] and of the higher effective viscosity of suspension that results
in a higher gas volumes trapped in the melt as well as improper filling of
the gaps between adjacent particles [31]. In addition, the air trapped within
particle clusters and the limited metal flow inside them can contribute to the
formation of pores [31]. In particular, the pore volume fraction in the fila-
ments increased rapidly when the Zn content was around 14%, particularly
in the precision extruded filaments. This latter behavior may be attributed
to the particular processing conditions associated with precision extrusion.
Overall, it should be noted that standard extrusion followed by precision
extrusion is able to produce continuous, uniform and dense PLA/Zn filaments
which are suitable for FFF. Furthermore, this process is reproducible, pre-
dictable and can be easily scaled up. The porosity in the filaments increased
when the Zn content was around 14 wt. % and increased more rapidly for
the precision extruded filament because of the differences in the processing
conditions.

3.2. Physico-chemical properties of PLA /Zn filaments

DSC curves during heating at 10°C/min of the PLA/Zn filaments after
standard and precision extrusion are plotted in Figures 3(a) and (b), re-
spectively. All DSC curves exhibit first a peak corresponding to the glass
transition (Tg ), followed by an exothermic phase transition due to cold crys-
tallization (Tcc ), and finally by an endothermic transition due to polymer
melting (Tm ). The glass transition of neat PLA occurs at approximately
58 °C. The Tg of PLA/Zn composites after standard and precision extrusion
remains unaffected regardless of the metal content (dashed line at lowest tem-
peratures on Figures 3 (a) and (b)), indicating that polymer chain mobility

8
 9
Figure 2: Representative cross sections obtained by XCT of PLA-x Zn (x in wt.%) fila-
ments after (a) standard extrusion and (b) precision extrusion. x stands for the wt. % of
Zn. Filament diameter was 2.85 mm in all cases. White regions stand for the Zn particle
while black regions are pores.
is not modified by the addition of Zn particles. On the contrary, the metal
particles favour PLA crystallisation, that occurs at lower temperatures and
with narrower exothermic peaks, evidencing the nucleating effect of Zn. The
nucleation of PLA starts at ∼ 90°C and the cold crystallization peak (Tcc ) is
reached at ∼ 122°C (dashed lines in Figures 3(a) and (b)). Tcc decreases with
increasing Zn content and it reaches ∼ 114°C after standard extrusion and
∼ 110°C after precision extrusion in the filaments with 17.5 wt. % Zn. The
difference in Tcc between standard and precision extrusion may be ascribed
to the reduction of the MW and of the physical entanglements in PLA due
to the shear stress during the precision extrusion, since this second process-
ing step was carried out at higher temperatures and with longer residence
times than standard extrusion. Cold crystallisation has also been reported
to be facilitated by the addition of metal particles in PLA/Mg composites
[13]. The melting peak (Tm ) of PLA after standard and precision extrusion
is found at 150°C (dashed lines in Figures 3(a) and (b)). Tm of PLA/Zn
composites after standard extrusion remains unaffected up to 14 wt.% of Zn,
and a slight decrease (together with the emergence of a shoulder at higher
temperature) is observed in PLA-17.5 wt. % Zn. Tm decreases slightly after
precision extrusion due to the reduction in Tcc and new endothermic peaks
appear at higher temperatures in PLA/Zn composites, which can be asso-
ciated to melt/recrystallization processes (marked by red short dashed lines
on Figures 3(a) and (b)). The split in the melting peak is more noticeable
with increasing Zn content.
The weight loss as a function of temperature during the TGA in air is
plotted in Figure 4 for the PLA/Zn filaments subjected to standard and
precision extrusion. PLA degradation begins at ∼ 300°C and finishes at ∼
390°C. PLA/Zn composites start to degrade at lower temperatures (220-240
°C) and the end of the degradation also occurs at lower temperatures (275-
310 °C). Both temperatures decrease with increasing the Zn content but this
effect is minor as compared with presence of relative small amounts of Zn
particles. The residual mass increased upon Zn content, except for PLA+14
wt. %, that was slightly higher than that of PLA + 17.5 wt. %. This
difference may be attributed differences in the actual Zn content due to the
small size of the samples (between 10-15 mg).
The MW distributions of the PLA/Zn filaments after standard and preci-
sion extrusion were obtained from the GPC experiments and they are plotted
in Figures 5(a) and (b), respectively. The data from the as-received PLA pel-
lets are added in both figures. The presence of Zn particles led to an almost

10
Figure 3: DSC curves during the heating cycle of PLA/Zn filaments at a scanning rate of
10°C/min. Glass transition (Tg ), cold crystallization (Tcc ) and melting temperature (Tm )
of pure PLA are marked with black dashed lines for reference. New emerging peaks in the
PLA/Zn composites are indicated with short red dashed lines.

negligible reduction in the MW distribution of the PLA after standard ex-

trusion, which took place at 180°C with a residence time of 3 minutes. The

11
 Figure 4: Weight loss as a function of temperature during the TGA test in air.

reduction in MW was slightly more noticeable after precision extrusion, which

was carried out at a maximum temperature 200°C with a residence time of
12 minutes. Thus, the presence of Zn particles induced some scission of the
molecular chains during extrusion which increases with the Zn content, as
shown in the plot of the average MW (Mn) of PLA/Zn filaments with Zn
content in Figure 5(c). This process was slightly accelerated during preci-
sion extrusion because of the higher temperature and residence time. These
results are in contrast with those reported for PLA/Mg filaments processed
using the same strategy [16]. The average MW (Mn) of the PLA in PLA/5
wt. % Mg composites decreased more than 50% after precision extrusion at
190 °C and the differences between PLA/Zn and PLA/Mg composites have
to be attributed to the thermal degradation of the polymeric chains catalysed
by the presence of the Mg particles.
The MFI is a common test performed to investigate the melt-processability
of polymers and to determine the optimum temperature for 3D printing. Ac-
cording to the literature, the MFI of the filament should be in the range of
10 g/10 min to 30 g/10 min to be successfully printed by the FFF technique
[16, 33]. The MFI of the PLA/Zn composites at 190°C after standard and
precision extrusion is plotted in Figure 6(a). These results are the average

12
Figure 5: MW distributions of PLA/Zn composites. (a) After standard extrusion. (b)
After precision extrusion. (c) Evolution of the number average molecular weight (Mn)
with Zn content after standard and precision extrusion.

of five tests and the error bars stand for the standard deviation. The MFI of
precision extruded filaments increased slightly with the incorporation of up
to 7 wt. % of Zn and remained approximately constant (15 g/10 min) for
larger Zn contents. These results are in agreement with the limited reduc-
tion in the average MW of PLA after standard extrusion, Figure 5(c), which
did not influence significantly the viscosity of the composite. However, the
MFI after precision extrusion increased rapidly with the Zn content due to
the reduction of the average MW of PLA. As a result, composites with Zn
content above 10.5 wt. % were outside of the printability range at 190°C and
it was not possible to measure the MFI of the PLA/ 17.5 wt. % Zn compos-

13
ite because of its high fluidity. Thus, the MFI of the composites with more
than 10 wt. % of Zn were determined at 170°C and the results are plotted
in Figure 6(b). They were within the printability range at this temperature.

Figure 6: (a) MFI at 190◦ C of standard extruded and precision extruded PLA/Zn filaments
as a function of the Zn content. (b) MFI at 170◦ C of precision extruded filaments of
PLA/10.5 wt. % Zn, PLA/14 wt. % Zn and PLA/17.5 wt. % Zn. The MFI printability
range is marked by the horizontal dashed lines.

Overall, the results of the TGA tests (Figure 4) and of the MFI (Figure 6)
indicate that the presence of Zn particles accelerates the degradation of the
PLA. However, the temperature window required for FFF processing –which
is found between Tm and the temperature at which the material begins to

14
degrade (220-240 °C)– is not compromised.
The results of DSC (Figure 3) indicate that Zn particles promote the
formation of crystalline nuclei in PLA at lower temperatures. It is interest-
ing to analyse the evolution of the crystalline phase in PLA during heating
from room temperature because this strategy might be used to modify the
properties of PLA (and, thus, of the composite) in 3D printed scaffolds. The
evolution of the XRD patterns of neat PLA and PLA/Zn composites with
3.5, 7 and 10.5 wt. % of Zn from 70°C up to 180°C with a heating rate of
10 °C/min is plotted in Figure 7. At low temperatures (from 70°C to Tcc ),
PLA and PLA/Zn composite do not show diffraction peaks, in agreement
with the amorphous glass state that was also observed by DSC (Figure 3).
A reference dashed blue line is added at 120°C in Figure 7, corresponding
with the cold crystallization of PLA. This value is determined with the emer-
gence of the two main diffraction peaks, corresponding to two doublet forms
(100)/(200) and (113)/(203). As indicated by the blue arrows in Figure 3.2,
the diffraction peaks for PLA-3.5% Zn, PLA-7% Zn and PLA-10.5% Zn ap-
pear at lower temperatures than those for PLA, especially for PLA-10.5%Zn
composite. This means that lower temperatures are required for the cold
crystallization with increasing of Zn content. Moreover, a reference dashed
red line, that corresponds to the melting point of PLA, is added at 150°C in
Figure 7. As indicated by the red arrows in Figure 7, higher melting temper-
atures were found in the composites, in agreement with the disappearance of
the diffraction peaks.

Figure 7: Temperature dependence of XRD patterns of PLA and PLA/Zn composites

from 70°C up to 180°C.

15
3.3. Manufacture and mechanical properties of PLA/Zn scaffolds
Porous scaffolds of 20 x 20 x 20 mm with a FCC lattice and circular struts
of 1.5 mm in diameter were manufactured by FFF following the CAD model
in Figure 8(a) with precision extruded filaments of PLA, PLA/3.5 wt. %
Zn, PLA/7 wt. % Zn and PLA/10.5 wt. % Zn. There is not direct relation
between the porosity of the filaments and of the scaffolds. The scaffolds
were 3D printed using filaments of pure PLA, PLA+3.5% Zn, PLA+7% Zn
and PLA+10.5% Zn with porosities below 0.7%. They are shown in Figures
8(b)-(d), respectively. The printing temperatures were selected according the
MFI experiments (Figure 6): 210°C for neat PLA, 190°C for PLA/3.5 wt. %
Zn and 170°C for the composites with 7 and 10.5 wt. % of Zn. It was not
possible to print scaffolds with higher Zn content due to clogging problems.
Cross-sections of the XCT tomograms of the scaffolds made up of PLA
and PLA/Zn composites with 3.5, 7 and 10.5 wt.% Zn are shown in Figures
8(e)-(h). The figures indicate that the scaffolds have good geometrical accu-
racy and that the Zn particles were homogeneously distributed in the case
of the composites. Moreover, slightly higher internal porosity was observed
in the case of the PLA/3.5 wt. % scaffold (marked with arrows in Figure
8(f)). The addition of higher loads of Zn to the filaments reduced the internal
porosity of the scaffolds (and it was possible to print the scaffolds at lower
temperatures) probably due to the reduction in viscosity during processing
which facilitated the deposition of the material.
The nominal stress-strain curves in compression of the 3D printed scaf-
folds are plotted in Figure 9. Stress was calculated from the applied load
divided by the initial cross-section of the scaffold while the engineering strain
was obtained from the displacement measured by the LVDT divided by the
initial height. All the samples were deformed up to 10% compressive strain.
The PLA scaffolds showed continuous hardening during deformation (or, at
least, a constant flow stress), which is indicative of homogeneous deformation.
In fact, the optical micrograph of the lateral section of the PLA scaffold at
the beginning of the compression test (Figure 10(a)) and after 10% of com-
pressive strain (Figure 10(b)) show a slight barreling of the scaffold after
deformation. On the contrary, softening after the peak load was found in all
the scaffolds containing Zn particles, which is indicative of the early local-
ization of damage. This is shown in Figure 10(d), where the lateral view of
the PLA + 7 wt. % Zn scaffold is depicted after 10% compressive strain.
Damage has been localized in the first row of the scaffold lattice and vertical

16
Figure 8: FCC lattice scaffolds of PLA/Zn particles manufactured by FFF. (a) CAD
model. (b) Scaffold of PLA, (c)PLA-3.5% Zn and (d) PLA-10.5% Zn. (e)-(h) Cross-
sections of FCC scaffolds extracted from XCT tomograms. Internal pores are marked
with white arrows.

cracks (parallel to the loading direction) have developed. Thus, the presence
of Zn particles embrittles the material, leading to early crack formation.

17
 Elastic Modulus (MPa) Strength (MPa)
PLA 650 ± 32 16 ± 1
PLA- 3.5% Zn 529 ± 1 13 ± 1
PLA-7% Zn 675 ± 103 16 ± 1
PLA- 10.5% Zn 517 ± 10 13 ± 1
Table 3: Mechanical properties of the 3D printed FCC scaffolds deformed in compression.

The average value and standard deviation of the elastic modulus and of
the strength of the scaffolds without and with Zn particles are indicated in
Table 3. The mechanical properties of the PLA and PLA + 7 wt. % Zn
scaffolds were very close while those of the scaffolds with 3.5 wt. % and 10.5
wt. % Zn were slightly lower, very likely due to the porosity in the former and
to the brittleness induced by the Zn particles in the latter. Thus, PLA + Zn
particle scaffolds present similar properties to those found in PLA scaffolds
and the main objective of the Zn particle reinforcement should be to tailor
the degradation rate and the biocompatibility. Both issues will be addressed
in the future.

Figure 9: Engineering stress-strain curves in compression of 3D printed scaffolds of PLA

and PLA with 3.5, 7.0 and 10.5 wt. % of Zn.

18
Figure 10: Lateral view of the scaffolds. (a) PLA, before testing. (b) PLA, after 10%
compressive strain. (c) PLA + 7 wt. % Zn, before testing. (d) PLA + 7 wt. % Zn, after
10% compressive strain.

4. Conclusions
The feasibility to manufacture scaffolds of PLA reinforced with Zn parti-
cles by FFF has been demonstrated by the first time. To this end, filaments
of 2.85 mm in diameter of PLA reinforced with different weight fractions of
µm-sized Zn - 1 wt. % Mg alloy particles (in the range 3.5 to 17.5 wt. %) were
manufactured by a double extrusion in method in which standard extrusion
is followed by a precision extrusion in a filament-maker machine. Filaments
with constant diameter, negligible porosity and a homogeneous reinforcement
distribution were obtained for Zn weight fractions of up to 10.5%. It is found
that the presence of Zn particles led changes in the physico-chemical proper-

19
ties of the PLA. In particular, Zn particles favoured the nucleation of crystals
and reduced Tcc . Moreover, the start and end degradation temperatures and
the average MW of the PLA are reduced with the Zn content after precision
extrusion at 200°C. Nevertheless, all these changes were limited and did not
affect the window temperature for 3D printing by FFF nor led to significant
changes in the MFI. Thus, it was possible to print scaffolds with complex
shape at 190°C with PLA/Zn composites containing 3.5 and 7 wt. % of Zn
and at 170°C when the Zn content was 10.5 wt. %. Composites with higher
Zn content could not be printed because of clogging of the nozzle. The di-
mensional accuracy of FCC scaffolds of 20 x 20 x 20 mm3 was excellent and
the porosity was neglible in scaffolds with 7 wt. % of Zn content, very likely
because of the viscosity reduction that facilitates 3D printing. The mechani-
cal properties of the scaffolds containing 7 wt. % of Zn were similar to those
of the PLA scaffolds in terms of stiffness and strength, although the presence
of Zn particles led to a more brittle failure mode.

Acknowledgements
This investigation was supported by the European Unions Horizon 2020
research and innovation programme under the European Training Network
BioImpant (Development of improved bioresorbable materials for orthopaedic
and vascular implant applications), Marie Sklodowska-Curie grant agreement
813869. Additional support from the Spanish Ministry of Science and Inno-
vation through the project ADDICOMP (grant RTI2018-094435-B-C33) is
gratefully acknowledged.

References
[1] Z. Sheikh, S. Najeeb, Z. Khurshid, V. Verma, H. Rashid, M. Glogauer,
Biodegradable materials for bone repair and tissue engineering applica-
tions, Materials 8 (9) (2015) 5744–5794.
[2] C. Li, C. Guo, V. Fitzpatrick, A. Ibrahim, M. J. Zwierstra, P. Hanna,
A. Letchtig, S. J. Lin, D. L. Kaplan, Design of biodegradable, im-
plantable devices towards clinical transition, Nature Reviews Materials
5 (2020) 61–81.
[3] R. Singh, M. J. Bathaei, E. Istifa, L. Beker, A review of bioresorbable
implantable medical devices: materials, fabrication and implementation,
Advacned Healthcare Materials 9 (2020) 2000790.

20
 [4] P. A. Gunatillake, R. Adhikari, Biodegradable synthetic polymers for
tissue engineering, Eur Cell Mater 5 (1) (2003) 1–16.

[5] P. B. Maurus, C. C. Kaeding, Bioabsorbable implant material review,

Operative Techniques in Sports Medicine 12 (3) (2004) 158–160.

[6] D. da Silva, M. Kaduri, M. Poley, O. Adir, N. Krinsky, J. Shainsky-

Roitman, A. Schroeder, Biocompatibility, biodegradation and excretion
of polylactic acid (PLA) in medical implants and theranostic systems,
Chemical Engineering Journal 340 (2018) 9–14.

[7] C. Chen, G. Lv, C. Pan, M. Song, C. Wu, D. Guo, X. Wang, B. Chen,

Z. Gu, Poly (lactic acid)(PLA) based nanocompositesa novel way of
drug-releasing, Biomedical materials 2 (4) (2007) L1.

[8] B. Tyler, D. Gullotti, A. Mangraviti, T. Utsuki, H. Brem, Polylactic acid

(PLA) controlled delivery carriers for biomedical applications, Advanced
Drug Delivery Reviews 107 (2016) 163–175.

[9] F. Auricchio, S. Marconi, 3D printing: clinical applications in or-

thopaedics and traumatology, EFORT open reviews 1 (5) (2016) 121–
127.

[10] A. Agüero, M. d. C. Morcillo, L. Quiles-Carrillo, R. Balart, T. Boronat,

D. Lascano, S. Torres-Giner, O. Fenollar, Study of the influence of the
reprocessing cycles on the final properties of polylactide pieces obtained
by injection molding, Polymers 11 (12) (2019) 1908.

[11] P. U. Rokkanen, O. Böstman, E. Hirvensalo, E. A. Mäkelä, E. K. Partio,

H. Pätiälä, S. Vainionpää, K. Vihtonen, P. Törmälä, Bioabsorbable fix-
ation in orthopaedic surgery and traumatology, Biomaterials 21 (2000)
2607–2613.

[12] D. Jiang, F. Ning, Fused Filament Fabrication of Biodegradable

PLA/316L Composite Scaffolds: Effects of Metal Particle Content, Pro-
cedia Manufacturing 48 (2020) 755–762.

[13] I. Antoniac, D. Popescu, A. Zapciu, A. Antoniac, F. Miculescu,

H. Moldovan, Magnesium filled polylactic acid (PLA) material for fila-
ment based 3D printing, Materials 12 (5) (2019) 719.

21
[14] A. Ferrández-Montero, M. Lieblich, J. L. González-Carrasco, R. Be-
navente, V. Lorenzo, R. Detsch, A. Boccaccini, B. Ferrari, Development
of biocompatible and fully bioabsorbable PLA/Mg films for tissue re-
generation applications, Acta biomaterialia 98 (2019) 114–124.

[15] A. Ferrández-Montero, M. Lieblich, R. Benavente, J. L. González-

Carrasco, B. Ferrari, Study of the matrix-filler interface in PLA/Mg
composites manufactured by Material Extrusion using a colloidal feed-
stock, Additive Manufacturing 33 (2020) 101142.

[16] C. Pascual-González, C. Thompson, J. de la Vega, N. B. Churruca, J. P.

Fernández-Blázquez, I. Lizarralde, D. Herráez-Molinero, C. González,
J. LLorca, Processing and properties of PLA/Mg filaments for 3D print-
ing of scaffolds for biomedical applications, Rapid Prototyping Journal
28 (2022) 884–894.

[17] S. Feliu Jr, A. Pardo, M. Merino, A. Coy, F. Viejo, R. Arrabal, Corre-

lation between the surface chemistry and the atmospheric corrosion of
AZ31, AZ80 and AZ91D magnesium alloys, Applied Surface Science 255
(2009) 4102–4108.

[18] M. Esmaily, J. Svensson, S. Fajardo, N. Birbilis, G. Frankel, S. Virtanen,

R. Arrabal, S. Thomas, L. Johansson, Fundamentals and advances in
magnesium alloy corrosion, Progress in Materials Science 89 (2017) 92–
193.

[19] T. Motoyama, T. Tsukegi, Y. Shirai, H. Nishida, T. Endo, Effects of

MgO catalyst on depolymerization of poly-L-lactic acid to L, L-lactide,
Polymer degradation and stability 92 (2007) 1350–1358.

[20] S. C. Cifuentes, M. Lieblich, F. López, R. Benavente, J. L. González-

Carrasco, Effect of Mg content on the thermal stability and mechanical
behaviour of PLLA/Mg composites processed by hot extrusion, Materi-
als Science and Engineering: C 72 (2017) 18–25.

[21] V. D. M. Echeverry-Rendón, D. Quintero, M. C. Harmsen, F. Echev-

errı́a, Novel coatings obtained by plasma electrolytic oxidation to im-
prove the corrosion resistance of magnesium-based biodegradable im-
plants, Surface & Coatings Technology 354 (2018) 28–37.

22
[22] A. Kopp, T. Derra, M. Müther, L. Jauer, J. H. Schleifenbaum,
M. Voshage, O. Jung, R. Smeets, N. Kröger, Influence of design and
postprocessing parameters on the degradation behavior and mechanical
properties of additively manufactured magnesium scaffolds, Acta Bio-
materialia 98 (2019) 23–35.

[23] M. Li, F. Benn, T. Derra, N. Kroeger, M. Zinser, R. Smeets, J. M.

Molina-Aldareguı́a, A. Kopp, J. LLorca, Microstructure, mechanical
properties, corrosion resistance and cytocompatibility of WE43 Mg alloy
scaffolds fabricated by laser powder bed fusion for biomedical applica-
tions, Materials Science and Engineering C 119 (2021) 111623.

[24] W. Ali, M. Echeverry-Rendón, C. G. A. Kopp, J. LLorca, Strength, cor-

rosion resistance and cellular response of interfaces in bioresorbable poly-
lactic acid/Mg fiber composites for orthopedic applications, Journal of
the Mechanical Behavior of Biomedical Materials 123 (2021) 104781.

[25] C. G. Fraga, Relevance, essentiality and toxicity of trace elements in

human health, Molecular Aspects of Medicine 26 (2005) 235–244.

[26] B. Hennig, R. Slim, M. Toborek, L. W. Robertson, Linoleic acid ampli-

fies polychlorinated biphenyl-mediated dysfunction of endothelial cells,
Journal of Biochemical and Molecular Toxicology 13 (1999) 83–91.

[27] N. S. Murni, M. S. Dambatta, S. K. Yeap, G. R. A. Froemming, H. Her-

mawan, ICytotoxicity evaluation of biodegradable Zn-3Mg alloy toward
normal human osteoblast cells, Materials Science and Engineering C 49
(2015) 560–566.

[28] C. Shen, X. Liu, B. Fan, P. Lan, F. Zhou, X. Li, H. Wang, X. Xiao,

L. Li, S. Zhao, Mechanical properties, in vitro degradation behavior,
hemocompatibility and cytotoxicity evaluation of Zn-1.2 Mg alloy for
biodegradable implants, RSC Advances 6 (2016) 86410–86419.

[29] H. Gong, K. Wang, R. Strich, J. G. Zhou, In vitro biodegradation be-

havior, mechanical properties, and cytotoxicity of biodegradable Zn-Mg
alloy, Journal of Biomedical Materials Research Part B 103 (2015) 1632–
1640.

23
[30] S. Liu, G. Esteban-Manzanares, J. LLorca, First-principles analysis
of precipitation in Mg-Zn alloys, Physical Review Materials 4 (2020)
093609.

[31] J. L. Colón Quintana, T. Heckner, A. Chrupala, J. Pollock, S. Goris,

T. Osswald, Experimental study of particle migration in polymer pro-
cessing, Polymer Composites 40 (6) (2019) 2165–2177.

[32] S. Goris, T. Osswald, Process-induced fiber matrix separation in long

fiber-reinforced thermoplastics, Composites Part A: Applied Science and
Manufacturing 105 (2018) 321–333.

[33] S. Wang, L. Capoen, D. R. Dhooge, L. Cardon, Can the melt flow index
be used to predict the success of fused deposition modelling of com-
mercial poly (lactic acid) filaments into 3D printed materials?, Plastics,
Rubber and Composites 47 (1) (2018) 9–16.

24