Prashantha 2016
Prashantha 2016
To cite this article: K. Prashantha & F. Roger (2017) Multifunctional properties of 3D printed
poly(lactic acid)/graphene nanocomposites by fused deposition modeling, Journal of
Macromolecular Science, Part A, 54:1, 24-29, DOI: 10.1080/10601325.2017.1250311
Article views: 11
Download by: [Monash University Library] Date: 07 December 2016, At: 14:01
JOURNAL OF MACROMOLECULAR SCIENCE, PART A: PURE AND APPLIED CHEMISTRY
2017, VOL. 54, NO. 1, 24–29
http://dx.doi.org/10.1080/10601325.2017.1250311
CONTACT K. Prashantha [email protected] Mines Douai, Department of Polymers and Composites Technology and Mechanical Engineer-
ing, 941 rue Charles Bourseul, CS 10838, F-59508 Douai Cedex, France.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/lmsa.
© 2017 Taylor & Francis Group, LLC
JOURNAL OF MACROMOLECULAR SCIENCE, PART A: PURE AND APPLIED CHEMISTRY 25
Inc., USA. Before used, this filament has an electrical resistivity of 2.3.3. Dielectric characterization
15 ohm-cm determined using a cubic centimeter of material. The dielectric characteristics of the materials were measured by
using impedance bridge (HP Agilent 4284A LCR meter, Agilent
Technologies, USA). The accuracy of the dielectric measure-
2.2. Fabrication of the PLA and graphene/PLA 3D printed ment was confirmed by measuring the permittivity and tan d of
samples a standard polytetrafluoroethylene sample provided by the
3D printed samples are manufactured using a Makerbot Repli- manufacturers. For each filler concentration, at least five sam-
cator 2 FDM printer with a nozzle diameter of 0.4 mm at ples were tested. All the measurements were performed at 25 C
210 C. Because of fillers, it is recommended to use 0.4 mm to and 50% RH.
avoid extruder blockage and to remove filament from the noz-
zle promptly after printing. Printing bed temperature was kept 2.3.3. Electromagnetic induction shielding efficiency (EMI SE)
at 60 C and the distance between nozzle and bed surface was measurements
fixed at 0.2 mm. As schematically represented in Figure 1, using Electromagnetic induction shielding measurements in terms of
commercially available 10 wt% graphene filled PLA filaments, reflection coefficient (S11) and the transmission coefficient
dumbbell and parallelepiped specimens were built-up along (S21) were measured by using a Vector Network Analyzer
horizontal orientations X is the direction of filament deposition (VNA) type (E8362B Agilent technologies working from
and Z is the direction of the overlapping layers. Specimens are 10 MHz to 20 GHz). The rectangular samples (25131.2 mm)
printed with two surface layer shells, the infill is set to 100% of printed nanocomposite were inserted into a standard coaxial
and the layer height is fixed to 200 microns. Infill pattern is lin- sample holder which matches the internal dimensions of X-
ear with a stacking alternating between ¡45 and 45 . Infill band (8.2–12.4 GHz) waveguide and they have been placed
and outlines filaments are printed at 30 mm/s. between the two ports of the network analyzer.
The rectangular samples of 25131.2 mm and 41030 mm
dimension were printed for EMI and DMA characterization. 2.3.4. Dynamic mechanical analysis (DMA)
Specimens for resistivity and tensile tests samples were printed The viscoelastic behavior was studied in tension by dynamic
according to ASTM D449 6-87 and ASTM D790-03 standards, mechanical analysis (DMAC150, MetraviB, France). Dynamic
respectively. strain sweep was first performed to determine the linear visco-
elastic range of the materials. The tests were then performed at
a strain amplitude of 2% and a frequency of 1 Hz. Data were
2.3. Characterization methods collected from ¡50 to 125 C at a scanning rate of 3 C.min¡1.
Rectangular DMA specimens (nominal dimensions, 4 § 0.210
2.3.1. Morphology
§ .530 mm3) were cut from 3D printed samples.
A Rectangular Parallele piped printed sample with a section of
28mm has been analyzed by X-ray micro-computed tomogra-
phy (tungsten filament and Varian detector). Voxel size is 2.3.5. Mechanical behavior testing
6.9 micrometers cube and 3D reconstruction was done using The mechanical properties were evaluated from 3D printed
public domain ImageJ software. Also, scanning electron micro- specimens. Tensile strength and elongation at break were mea-
scope (SEM) (S-4300SE/N, Hitachi, Japan) was used to study sured at a crosshead rate of 5 mm.min¡1 and modulus at
the microstructure and nanofiller dispersion of PLA/GNP 1 mm.min¡1, using a tensile machine (Lloyd LR 50 K, USA) at
nanocomposites. The voltage of acceleration was 5kV. Before 25 C and 50% RH according to ISO 527 standard. All the
SEM recording, the samples were fractured by immersing into reported values were calculated as averages over five samples
liquid nitrogen and subsequently coated with thin layer gold. for tensile and flexural test and ten specimens for impact test.
Figure 2. X-ray microcomputed tomography (A) and SEM image (B) of PLA/Graphene nanocomposite sample.
higher in parallelepiped specimens (about 0.3-0.6 secs) because aligned along deposited direction of the filaments leading
its width is higher. The shorter time, the better the interaction higher conductivity.
and inter-joining between contiguous filaments, because the fil-
ament temperature is higher and closer to the proceding draw-
3.3. Dielectric properties
ing, determining a higher quality of filament bonding between
the layers. The dependence of the dielectric properties of the neat PLA and
The homogeneous dispersion of graphene in the PLA matrix nanocomposite as a function of frequency was investigated.
is one of the most important criteria to achieve optimal proper- Figures 3(a) and (b) display the dielectric permittivivity (i.e.
ties of nanocomposites. In the case of GNPs, additional inter dielectric constant) ɛ’, and the dielectric loss tangent, tan d, of
planar p–p interactions make it even more challenging to dis- the 3D printed neat PLA and PLA/graphene nanocomposite. It
perse them in the polymer matrices. Moreover, their hydropho- is evident from the figure that, the ɛ’ and tan d values increase
bic nature is not favorable to their dispersion within the substantially after adding graphene as the filler and the nano-
hydrophilic PLA matrix. Nevertheless, Figure 2B shows that composites showed the slightly higher frequency dependence
GNPs are well dispersed in the PLA matrix after even after, up of dielectric constant than that of PLA. The higher real permit-
to a GNPs concentration of 10 wt%. GNP’s high aspect ratio tivity in nanocomposites is attributed to the polar characteris-
and its relatively high population density facilitated it to form tics of fillers and thus increases the number of charge carriers
an interconnected filler network throughout the PLA matrix in the PLA matrix. The increase in dielectric constant can be
during the 3D fabrication. attributed to the homogenous dispersion and compatibility of
filler in the PLA matrix.
The dielectric loss factor (tan d) of neat PLA and the PLA/
3.2. Electrical resistivity
graphene nanocomposites under applied sweep frequencies is
The influence of FDM processing on volume resistivity of PLA/ shown in Figure 3(b). For the PLA matrix and PLA containing
graphene nanocomposites was studied. Before the processing, graphene, the loss factors of the samples exhibited initial
this filament had an electrical resistivity of 15 ohm-cm deter- increase at lower frequency followed by a decrease above
mined using a cubic centimeter of material. Surprisingly, vol- 110 Hz. The initial increase in loss factor at low frequency can
ume resistivity of 3D printed parts perpendicular to layers was be attributed to dipole polarization (10). As the frequency
found to be 25 ohm-cm and the measured volume resistivity of increased further, the dipole polarization effects reduced, and
3D printed parts through layers (along Z axis) is about the value of the loss factor declined accordingly. More signifi-
102 ohm-cm. This change in resistivity is certainly related to cantly, the values of the loss factors of all samples are small
the direction of the graphene nanoplatelets are preferentially compared with other related reports (11) and in the range of
JOURNAL OF MACROMOLECULAR SCIENCE, PART A: PURE AND APPLIED CHEMISTRY 27
Figure 4. EMI SE for neat PLA and PLA/graphene composites in the X-band range,
8–12 GHz.
3.4. Dynamic mechanical analysis Table 1. Tensile properties of FDM processed PLA and nanocomposite.
Dynamic mechanical thermograms of PLA and its nanocompo- Samples Modulus (MPa) Strength (MPa) Strain (mm/mm)
site after 3D-printing are shown in Figures 5. The storage mod- PLA 1827 31.6 0.026
ulus (E') (Figure 5a) and glass transition temperature from loss PLA/10 wt% Graphene 2454 40.2 0.023
tangent (tan d) (Figure 5b) as a function of temperature are
measured. It was observed that, addition of graphene to PLA
matrix, substantially increases the storage modulus of the FDM
3.5. Tensile properties
parts by about 20-30% with respect to the neat PLA below the
Tg. Increase in the storage modulus values after Tg is due to The mechanical properties such as elastic modulus (E), tensile
the cold crystallization of PLA, the effect of graphene nanofiller strength (s b) and strain at break (eb) of 3D-printed neat PLA
is manifestly more evident even above Tg. In fact, the storage and PLA nanocomposite are summarized in Table 1 and
modulus of the nanocomposite materials at 100 C is more than Figure 6. In general, it can be noted how the presence of gra-
that that of neat PLA samples, thus reveals that, added graphe- phene nanoplatelets promotes a remarkable increase of the
nes nucleates the cold crystallization in PLA matrix, thereby elastic modulus (by 30%) and ultimate tensile strength (by
enhancing the positive stiffening effect of graphene nanoplate- 27%) of the PLA matrix, but slightly decreases its strain at
lets in the molten state. In Figure 5b, the tan d peak observed at break values. The reduction of ultimate properties could be
about 66 C and 70 C is related to the glass transition tempera- attributed to a presence of some voids between the deposition
ture (Tg) of the neat PLA and PLA/graphene nanocomposites, filaments as documented by the X-Ray computed tomography
respectively. This increase of Tg values by about 4 C in nano- observations of Figure 2.
composite is due to the restriction of motion of macromole- As it clearly emerges from Figure 6 that 3D-printed samples
cules due to the orientation of polymer chains during extrusion along horizontal direction are characterized by low or no com-
in FDM printed parts. paction pressure leading to a 3D-printed specimens. PLA
presents a 1.8 GPa Young modulus and tensile strength of
31.6 MPa. When compared with PLA, nanocomposites reveal a
considerable increase on modulus (by 34%) and the strength
increases by 29%, but at the expense of ductility.
The observed tensile strength and modulus values of neat
PLA and that of nanocomposites is lower than the values
reported for injection molded bulk samples (17, 18). This is
attributed to the architecture of the printed samples and is also
due to the presence of voids in the microstructure which leads
to a lower effective cross section in the sample. Therefore, it still
remains a challenge to improve the quality and dimensional
stability of the FDM samples. Upon addition of 10 wt% of gra-
phene, the elastic modulus improves by about 30% compared
to unfilled PLA. In fact, the present samples are horizontally
built samples which are said to exhibit the highest elastic mod-
ulus and ultimate strength, compared to vertical and perpen-
dicular parts. This is due to the fact that the direction of the
Figure 5. Dynamic mechanical properties (a) storage modulus (E’) and (b) loss tan- Figure 6. Typical stress-strain curve for FDM processed PLA and PLA/graphene
gent (tan d), of neat PLA and nanocomposite samples as a function of temperature. nanocomposites.
JOURNAL OF MACROMOLECULAR SCIENCE, PART A: PURE AND APPLIED CHEMISTRY 29