applied sciences Article
Rapid Prototyping of Efficient Electromagnetic Interference Shielding Polymer Composites via Fused Deposition Modeling Luiz Gustavo Ecco 1 , Sithiprumnea Dul 2 , Débora Pereira Schmitz 1 , Guilherme Mariz de Oliveira Barra 1 , Bluma Guenther Soares 3 , Luca Fambri 2 Alessandro Pegoretti 2, * 1
2 3
*
and
Department of Mechanical Engineering, Federal University of Santa Catarina—UFSC, Florianópolis 88040-900, SC, Brazil;
[email protected] (L.G.E.);
[email protected] (D.P.S.);
[email protected] (G.M.d.O.B.) Department of Industrial Engineering, University of Trento, Trento 38100, Italy;
[email protected] (S.D.);
[email protected] (L.F.) Department of Metallurgic and Materials Engineering, Federal University of Rio de Janeiro, Rio de Janeiro 21941-909, RJ, Brazil;
[email protected] (B.G.S.) Correspondence:
[email protected]; Tel.: +39-0461-282452
Received: 17 November 2018; Accepted: 18 December 2018; Published: 22 December 2018
Featured Application: The potential applications of the developed materials are in the field of the protection of electronic devices from electromagnetic interferences. Moreover, the proven possibility to use additive manufacturing widely extends the complexity of the geometries. Abstract: Acrylonitrile–butadiene–styrene (ABS) filled with 6 wt.% of multi-walled carbon nanotubes and graphene nanoplatelets was extruded in filaments and additively manufactured via fused deposition modeling (FDM). The electrical conductivity and electromagnetic interference shielding efficiency (EMI SE) in the frequency range between 8.2 and 12.4 GHz of the resulting 3D samples were assessed. For comparison purposes, compression molded samples of the same composition were investigated. Electrical conductivity of about 10−4 S·cm−1 and attenuations of the incident EM wave near 99.9% were achieved for the 3D components loaded with multi-walled carbon nanotubes, almost similar to the correspondent compression molded samples. Transmission electron microscopy (TEM) images of ABS composite filaments show that graphene nanoplatelets were oriented along the polymer flow whereas multi-walled carbon nanotubes were randomly distributed after the extrusion process. The electrical conductivity and electromagnetic interference (EMI) shielding properties of compression molded and FDM manufactured samples were compared and discussed in terms of type of fillers and processing parameters adopted in the FDM process, such as building directions and printing patterns. In view of the experimental findings, the role of the FDM processing parameters were found to play a major role in the development of components with enhanced EMI shielding efficiency. Keywords: 3D printing; carbon nanotubes; polymer-matrix composites (PMCs); fused deposition modeling; electromagnetic interference
1. Introduction The term “electromagnetic interference” (EMI) refers to an unwanted disturbance occurring in electronic instruments and devices. Technological problems attributed to EMI date back to the 1930s [1], however, in the modern digital era, due to the extensive global development of novel electronic and Appl. Sci. 2019, 9, 37; doi:10.3390/app9010037
www.mdpi.com/journal/applsci
Appl. Sci. 2019, 9, 37
2 of 19
telecommunications technologies, EMI has become a major concern [2]. In Europe, a directive destined to the electrical and electronic engineering industries regulating the electromagnetic compatibility (EMC) was released in 2014. The aim is to assure that electro-electronic apparatuses do not cause, or are not affected by EMI [3]. Consequently, there is the need to develop effective and practical EMI shielding materials specifically designed for such applications [4]. In recent years, considerable research efforts have been conducted for the development of novel electrically conductive polymer composites with enhanced EMI shielding efficiency. When compared to traditional metallic materials, polymer composites have the advantage of lightness, ease of processing/shaping, and of not suffering from corrosion processes. Various polymer composite formulations, i.e., different filler types and polymer matrixes, as well as mechanisms towards electromagnetic attenuation have been recently reviewed [5–8]. The most common processing routes for the development of electrically conductive polymer composites for EMI shielding applications are solvent casting [9–12] and melt-mixing followed by compression-molding [13–16]. Nevertheless, solvent casting remains limited to lab-scale. Moreover, it is sometimes difficult to shape polymer composites into complex forms via solvent casting or compression molding. Recent advances in additive manufacturing (AM) technologies have opened new possibilities for rapid prototyping of electrically conductive polymer composites for EMI shielding [17]. The so-called fused deposition modeling appears as an emerging technology that is considered as one of the most common techniques for rapid prototyping of polymeric components. In fused deposition modeling (FDM), 3D manufactured components are grown layer-by-layer from polymeric filaments [18,19]. On the other hand, obtaining extruded filaments of electrically conductive polymer composites suitable for FDM is not a simple task. In fact, modifications of the rheological behavior due to the addition of fillers may complicate the extrusion process of the filaments and the subsequent FDM step [19]. Nevertheless, the successful preparation of electrically conductive filaments filled with carbon nanotubes (CNT) and their use for producing electrically conductive samples by FDM has been recently reported [20]. In this manuscript, acrylonitrile–butadiene–styrene copolymer filled with graphene nanoplatelets (GNP) or multi-walled CNT at a fixed weight fraction of 6 wt.% was extruded in filaments and rapid prototyped via fused deposition modeling. It is important to notice that the concentration of 6 wt.% was carefully chosen based on previous studies [20,21]. In fact, the comparison between graphene and carbon nanotube composites indicated that a filler content of 6 wt.% is a good compromise to maintain an acceptable processability and to obtain positive effects on the physical properties of acrylonitrile–butadiene–styrene (ABS) composites [21]; additionally, ABS/CNT at 6 wt.% of loading has been successfully used in FDM [20]. In this new research, the EMI shielding efficiency of 3D manufactured components of ABS/GNP and ABS/CNT was also compared to that of compression-molded samples. Moreover, the role of the FDM printing patterns on the electrical and EMI shielding properties was also investigated. 2. Materials and Methods 2.1. Materials The matrix was an acrylonitrile–butadiene–styrene copolymer (ABS, trade name Sinkral® F322) from Versalis S.p.A (Mantova, Italy) with a density of 1.04 /g/cm3 . The following nanofillers were selected:
•
•
Multi-walled carbon nanotubes (MWCNT, NanocylTM NC7000, Belgium) with surface area of 250 to 300 m2 ·g−1 ; density of 2.15 ± 0.03 g·cm−3 ; carbon purity of 90%; average diameter and length of 9.5 nm and 1.5 µm, respectively; graphene nanoplatelets (xGnP, type M-5, East Lansing, MI, USA) with surface area of 120 to 150 m2 ·g−1 ; density of 2.06 ± 0.03 g·cm−3 ]; carbon purity of 99.5%; average particle diameter of 5 µm and particle thickness of 6 to 8 nm [20,22].
Appl. Sci. 2019, 9, 37
3 of 19
Appl. Sci. 2018, 8, x
3 of 19
2.2. Preparation of 2.2. Preparation of the the Polymer Polymer Composites Composites Figure Figure 11 details details the the various various experimental experimental procedures procedures for for sample sample fabrication, fabrication, starting starting from from melt melt compounding of ABS and carbonaceous fillers followed by compression molding, filament compounding of ABS and carbonaceous fillers followed by compression molding, filament extrusion extrusion and and FDM FDM prototyping. prototyping. The The various various testing testing techniques techniques used used in in the the work work are are indicated indicated in in Figure Figure1. 1.
Figure 1. Scheme of theofmanufacturing processes and characterization tests.tests. Figure 1. Figure 1. Scheme the manufacturing processes and characterization
2.2.1. 2.2.1. Compounding Compounding A A fixed fixed weight weight fraction fraction of of 66 wt.% wt.% of of MWCNT MWCNT and and xGnP xGnP was was melt melt compounded compounded with with ABS ABS using using TM , PolylabTM Rheomix). The temperature in the mixing an internal mixer (Thermo Scientific Haake an internal mixer (Thermo Scientific HaakeTM, PolylabTM Rheomix). The temperature in the mixing ◦ C, the rotor speed was 90 rpm and the mixing time was 15 min. For a comparison, chamber chamber was was 190 190 °C, the rotor speed was 90 rpm and the mixing time was 15 min. For a comparison, the The polymer polymer composites composites are the ABS ABS matrix matrix was was processed processed under under similar similar conditions. conditions. The are hereafter hereafter labeled for the labeled as as follows: follows: ABS ABS for for the the neat neatpolymer, polymer,ABS/GNP ABS/GNP and and ABS/CNT ABS/CNT for the composites composites loaded loaded with xGnP and MWCNT, respectively. After compounding, the resulting materials were reduced with xGnP and MWCNT, respectively. After compounding, the resulting materials were reduced to to millimeter-sized 0.6 mm mm were millimeter-sizedparticles particles(average (averageparticle particlesize sizeofof2.1 2.1±± 0.6 were obtained obtained with with D10 D10 == 1.5 1.5 mm mm and and D90 2.9 mm) mm) using using aa low-speed low-speed granulator granulator (Piovan, (Piovan, model: model: RN RN 166) 166) for for either either compression-molding compression-molding D90 = = 2.9 or extrusion processes. or extrusion processes. 2.2.2. Filament Extrusion 2.2.2. Filament Extrusion Extruded filaments with a diameter of 1.75 ± 0.10 mm were prepared using a Thermo Haake Extruded filaments with a diameter of 1.75 ± 0.10 mm were prepared using a Thermo Haake PTW16 intermeshing co-rotating twin-screw extruder (screw diameter = 16 mm; L/D ratio = 25; rod PTW16 intermeshing co-rotating twin-screw extruder (screw diameter = 16 mm; L/D ratio = 25; rod die diameter 1.80 mm, Thermo Electron, Karlsruhe, Germany). The processing temperature from die diameter 1.80 mm, Thermo Electron, Karlsruhe, Germany). The processing temperature from feeding zone (zone 1) to the rod die (zone 5) was set respectively to 180, 205, 210, 215 and 220 ◦ C, and feeding zone (zone 1) to the rod die (zone 5) was set respectively to 180, 205, 210, 215 and 220 °C, and the screw rotation speed was fixed at 5 rpm. The resulting filaments were used as feedstock materials the screw rotation speed was fixed at 5 rpm. The resulting filaments were used as feedstock materials for FDM. for FDM. 2.2.3. Preparation of the Samples for electromagnetic interference shielding efficiency (EMI SE) 2.2.3. Preparation of the Samples for electromagnetic interference shielding efficiency (EMI SE) Tests—Compression Molded Tests—Compression Molded Compression molded (CM) specimens were prepared using a hot hydraulic press (Carver ◦ C under Compression (CM) specimens were prepared using hydraulic press Laboratory) at 190 molded a pressure of 3.9 MPa for 10 min anda ahot cooling rate of 20 ◦(Carver C/min Laboratory) at 190 °C under a pressure of 3.9 MPa for 10 min and a cooling rate of 20 °C/min to obtain to obtain square plaques with dimensions 120 × 120 × 2.0 mm, which were later cut into specimens of square dimensions 120 × 120 × 2.0 mm, which were later cut into specimens of 45 × 45 × 45 × 45plaques × 2 mmwith for testing. 2 mm for testing. 2.2.4. Preparation of the 3D samples for EMI SE tests—FDM 2.2.4. Preparation of the 3D samples for EMI SE tests—FDM 3D manufactured specimens were obtained by a Sharebot Next Generation FDM machine 3D manufactured specimens were obtained a Sharebot Generation FDM (Sharebot NG, Nibionno, Italy). The open source by software Slic3r Next was used to control andmachine design (Sharebot NG, Nibionno, Italy). The open source software Slic3r was used to control and design the manufacturing process. 3D plaques with dimensions 45 × 45 × 2.0 mm were grown along three different directions: Horizontal alternate (H45), horizontal concentric (HC) and perpendicular
Appl. Sci. 2019, 9, 37
4 of 19
Appl. 2018, 8, x the Sci. manufacturing
of 19 process. 3D plaques with dimensions 45 × 45 × 2.0 mm were grown 4along three different directions: Horizontal alternate (H45), horizontal concentric (HC) and perpendicular concentric concentric (PC), (PC), as as schematized schematized in in Figure Figure 2. 2. The ThePC PCand andHC HCcomponents componentshad hadtheir their layers layersbuilt built along along concentric lines from the outer to the inner direction, and all the layers were identically deposited concentric lines from the outer to the inner direction, and all the layers were identically depositedon on the the top top of of the the previous previous one. one. The The H45 H45 growing growing direction direction was was built built alternating alternating the the orientation orientation of of each each layer layerresulting resulting in inaanetwork networkstructure. structure.The The3D 3Dcomponents componentswere weremanufactured manufacturedat atnozzle nozzletemperature temperature of 250 °C for ABS and ABS/GNP and 280 °C for ABS/CNT. In fact, due to high viscosity ◦ ◦ of 250 C for ABS and ABS/GNP and 280 C for ABS/CNT. In fact, due to high viscosityof ofABS/CNT, ABS/CNT, composites were 3D-printed at 280 °C in order to guarantee a regular extrusion, and to ◦ composites were 3D-printed at 280 C in order to guarantee a regular extrusion, and to avoid avoid the the clogging ◦ C. clogging of of the the nozzle nozzle of of the the 3D 3D printer printer machine machine as as observed observed during during processing processing at at 250 250 and and 260 260 °C.
Figure 2. 2. Scheme Schemeofofthe the fused fused deposition deposition modeling modeling (FDM) (FDM) 3D 3D manufactured manufactured samples samples for for Figure electromagnetic interference shielding efficiency (EMI SE) tests. electromagnetic interference shielding efficiency (EMI SE) tests.
2.3. Testing 2.3. Testing 2.3.1. Rheological Analysis 2.3.1. Rheological Analysis Rheological properties of neat ABS and the nanocomposites were determined using an oscillatory Rheological of neat ABS nanocomposites were parallel determined an rheometer, modelproperties Hybrid Discovery HR1 and from the TA Instrument Inc., with plate using geometry oscillatory rheometer, Hybrid Discovery HR1 fromofTA parallel plate (25 mm). The analysis model was performed at the temperatures 250Instrument and 280 ◦ CInc., in a with frequency range of geometry (25inmm). The analysis was performed at the temperatures of 250 and 280 °C in a frequency 0.1–100 Hz, the linear viscoelastic regime under a 0.1% applied deformation. range of 0.1–100 Hz, in the linear viscoelastic regime under a 0.1% applied deformation.
2.3.2. Transmission Electron Microscopy
Appl. Sci. 2019, 9, 37
5 of 19
Appl. Sci. 2018, 8, x 2.3.2. Transmission Electron Microscopy
5 of 19
The The microstructure microstructure of of nanocomposites nanocomposites from from compression compression molding molding (CM) (CM) and and extruded extruded filaments filaments ® EM 400 T (Philips, was observed by transmission electron microscopy (TEM), using a Philips ® was observed by transmission electron microscopy (TEM), using a Philips EM 400 T (Philips, Amsterdam, Amsterdam, The The Netherlands) Netherlands) transmission transmission electron electron microscope. microscope. Ultrathin Ultrathin specimens specimens of of CM CM and and filaments, having a trapezoidal shape with dimensions of about 200 × 150 µm, and were cut using filaments, having a trapezoidal shape with dimensions of about 200 × 150 m, and were cut using aa Leica Leica EM EM UC7 UC7 ultramicrotome ultramicrotome equipped equipped with with aa diamond diamond knife. knife. The The filament filament specimens specimens were were cut cut into into thin thin slices slices perpendicular perpendicular and and parallel parallel to to the the flow flow direction direction (see (see Figure Figure 3). 3). The The thin thin specimens specimens were were deposited deposited on on aa copper copper grid grid mesh mesh covered covered by byamorphous amorphous holey holeycarbon carbonfilm. film.
Figure 3. Schematic of the perpendicular and parallel section in a filament for transmission electron Figure 3. Schematic of the perpendicular and parallel section in a filament for transmission electron microscopy (TEM) observation. microscopy (TEM) observation.
2.3.3. Electrical Conductivity Measurements 2.3.3. Electrical Conductivity Measurements The electrical conductivity in the direct current (d.c.) of the investigated samples was assessed electrical method conductivity the direct current (d.c.) of the investigated samples was assessed via a The two-contacts usingina Metrohm-Autolab potentiostat, model PGSTAT302N operated in via a two-contacts using a6517A Metrohm-Autolab potentiostat, model PGSTAT302N operated in galvanostatic modemethod and a Keithley electrometer connected to Keithley 8009 test fixture (Keithley ◦ galvanostatic mode and a Keithley 6517A electrometer connected to Keithley 8009 test fixture Instruments). Measurements were conducted at room temperature (23 ± 1 C) and the electrical (Keithley Instruments). Measurements at room temperature (23 ± 1 °C) and the conductivity values represent an averagewere of at conducted least five repetitions. electrical conductivity values represent an average of at least five repetitions. 2.3.4. Electromagnetic Interference and Dielectric Constant Measurements 2.3.4. Electromagnetic Interference and Dielectric Constant Measurements EMI shielding efficiency of resulting polymer composites was tested in the frequency between 8.2 and 12.4 usingefficiency an Agilent PNA series network analyzer (N5230C Agilentbetween PNA-L, EMIGHz shielding ofTechnology resulting polymer composites was tested in the frequency Agilent, Santa Clara, CA, USA) with a rectangular waveguide with cross-section of 23 × 10PNAmm. 8.2 and 12.4 GHz using an Agilent Technology PNA series network analyzer (N5230C Agilent An incidentSanta EM wave frequency range was applied and both the×transmitted L, Agilent, Clara,within USA) the withspecific a rectangular waveguide with cross-section of 23 10 mm. An and reflected waveswithin were collected. Thefrequency instrumentation measures theand magnitude phase of and the incident EM wave the specific range was applied both the and transmitted scattering parameters the sample uponmeasures the incident whose and power is denoted reflected waves were(S-parameters) collected. Theofinstrumentation the wave magnitude phase of the as Pi. The power of the(S-parameters) transmitted and reflected Pt and Pr, power respectively. The scattering parameters of the samplewaves upon are the denoted incident as wave whose is denoted attenuation of incident waves isand expressed in waves decibels and itascan as givenThe in as Pi. The power of the EM transmitted reflected are(dB) denoted Pt be andexpressed Pr, respectively. Equation (1).of incident EM waves is expressed in decibels attenuation (dB) and it can be expressed as given in Pt Equation 1. EMI SE = 10 log (1) Pi Pt The reflected and transmitted the shielding material can be mathematically ( ) EMI SE =waves 10 logthrough (1) P i represented by the complex S-parameters named S11, S22, S12 and S21 which are correlated to the reflectance and the transmittance, respectively, according to Equations (2) and (3): through The reflected and transmitted waves the shielding material can be mathematically Pt 2 2 2 T = = S12 = |S21 (2) | represented by the complex S-parametersPnamed S11,| S22, S12| and S21 which are correlated to the i reflectance and the transmittance, respectively, according to Equation 2 and 3: 2 Pr (3) |2 = |S22|2 PtR2 = Pi 2 = |S11 T = ( ) = |S12| = |S21|2 (2) Pi
Appl. Sci. 2019, 9, 37
6 of 19
The absorbance (A), reflectivity (R), and transmittance (T) are correlated as in Equation (4): A + R + T = 1
(4)
Considering the propagation of an electromagnetic wave through an electrically conductive specimen, the tested specimen can either absorb (SEA) or reflect (SER) the incident electromagnetic waves. This can be estimated according to Equations (5) and (6). h i SER = 10 log 1−(S11)2 (" SEA = 10 log
1−
1−(S11)2 − (S21)2
(5) #)
1−(S11)2
(6)
The overall EMI SE is thus given by both SER and SEA as follows. EMI SE = SER + SEA
(7)
The dielectric constant and the loss factor of the samples were also determined using the Agilent 85071 software (Agilent, Santa Clara, USA). The software uses the Nicolson–Ross–Weir (NRW) algorithm and the S-parameters to calculate these values. Measurements were conducted at 23 ± 1 ◦ C, the reported spectra are representative of three replicated acquisitions for each specimen. The thickness of all the specimens was 2.0 mm. 3. Results and Discussion 3.1. Rheological Behavior Rheology tests can provide useful information on the processability of the polymer and nanocomposites in the molten state. The complex viscosity at two different temperatures obtained from oscillatory rheometer is reported in Figure 4. At both 250 and 280 ◦ C, ABS exhibited a pseudo-Newtonian behavior with a viscosity plateau at low frequency. The presence of nanofillers increased the viscosity in the low frequency region for both temperatures. A significant increase in viscosity was observed by the addition of carbon nanotubes. ABS/GNP nanocomposite showed a viscosity smaller than that of neat ABS at high frequencies for 250 ◦ C. This behavior could be explained by the fact that the particle-particle interaction was not pronounced, and viscosity of ABS/GNP nanocomposites was decreased by a lubricating effect played by graphene nanoplatelets [23]. Similar behavior was also observed in a prior study on ABS composites containing carbon black (1–3 wt.%) or carbon nanotubes (up to 1 wt.%) [17]. Additionally, details about the melt flow index values of ABS, ABS/GNP and ABS/CNT at the same temperature were reported in our previous paper [21].
Appl. Sci. 2019, 9, 37
7 of 19
Appl. Sci.2018, 2018, Appl. Sci. 8,8, xx
7 7ofof 1919 6 6 1010
ABS ABS ABS/GNP ABS/GNP ABS/CNT ABS/CNT
o o
Complex viscosity (Pa.s) Complex viscosity (Pa.s)
250CC 250 5 5 1010
ABS ABS ABS/GNP ABS/GNP ABS/CNT ABS/CNT
5 5 1010
4 4 1010
4 4 1010
3 3 1010
3 3 1010
2 2 1010
2 2 1010
1 1 1010 -1 -1 1010
o oC 280 280 C
Complex viscosity (Pa.s) Complex viscosity (Pa.s)
6 6 1010
0 0 1 1 1010 1010 Frequency (Hz) Frequency (Hz)
2 2 1010
1 1 1010 -1 -1 1010
0 0 1 1 1010 1010 Frequency (Hz) Frequency (Hz)
2 2 1010
◦ C; Figure Complex viscosity function frequency temperature of 250 and 280 Figure4. Complexviscosity viscosityasas asaaafunction functionofof offrequency frequencyatatattemperature temperatureof of250 250and and280 280°C; °C; Figure 4.4.Complex Acrylonitrile–butadiene–styrene; ABS/graphene nanoplatelets (GNP); ABS/carbon nanotubes (CNT) Acrylonitrile–butadiene–styrene;ABS/graphene ABS/graphenenanoplatelets nanoplatelets(GNP); (GNP);ABS/carbon ABS/carbonnanotubes nanotubes(CNT) (CNT) Acrylonitrile–butadiene–styrene;
The effect ofofgraphene graphene and carbon nanotubes on the storage modulus and loss modulus vs. Theeffect effectof grapheneand andcarbon carbonnanotubes nanotubeson onthe thestorage storagemodulus modulusand andloss lossmodulus modulusvs. vs. The frequency of the ABS can be observed in Figure 5. Both storage and loss of moduli increased frequency of the ABS can be observed in Figure 5. Both storage and loss of moduli increased frequency of the ABS can be observed in Figure 5. Both storage and loss of moduli increased significantly with graphene and The dependence ofofstorage significantlywith withgraphene grapheneand andcarbon carbonnanotubes nanotubesat lowfrequencies. frequencies.The Thedependence dependenceof storage significantly carbon nanotubes atatlow low frequencies. storage modulus on frequency (slope of G’ vs. frequency at low frequencies) decreased with respect to modulus on frequency (slope of G’ vs. frequency at low frequencies) decreased with respect the modulus on frequency (slope of G’ vs. frequency at low frequencies) decreased with respect totothe the viscosity of materials for both temperatures. For instance, the slope of neat ABS and its viscosity of materials for both temperatures. For instance, the slope of neat ABS and viscosity of materials for both temperatures. For instance, the slope of neat ABS and itsits nanocomposite decreased, The storage modulus nanocomposite nanocompositedecreased, decreased,respectively. respectively.The Thestorage storagemodulus modulusof ABS/CNTnanocomposite nanocompositewas was nanocomposite respectively. ofofABS/CNT ABS/CNT was almost independent of frequency. This pseudo-solid-like behavior is attributed to the formation of almost independent of frequency. This pseudo-solid-like behavior is attributed to the formation almost independent of frequency. This pseudo-solid-like behavior is attributed to the formation ofofaa a space-filling space-fillinginterconnected interconnectednetwork networkof nanofillers[23]. [23]. space-filling interconnected network ofofnanofillers nanofillers [23]. A typical response for a polymer melt is that elastic atat typicalresponse responsefor fora apolymer polymermelt meltisisthat thatelastic elasticbehavior behavior(storage (storagemodulus) modulus)dominates dominatesat AAtypical behavior (storage modulus) dominates high frequencies while viscous behavior (loss modulus) dominates atatlow low frequencies. The cross-over highfrequencies frequencieswhile whileviscous viscousbehavior behavior(loss (lossmodulus) modulus)dominates dominatesat lowfrequencies. frequencies.The Thecross-over cross-over high point between storage and loss modulus can be defined depending on the molecular weight and pointbetween betweenstorage storageand andloss lossmodulus moduluscan canbebedefined defineddepending dependingon onthe themolecular molecularweight weightand and point molecular weight distribution of some linear polymers. Our result showed that neat ABS seemed molecular weight distribution of some linear polymers. Our result showed that neat ABS seemed molecular weight distribution of some linear polymers. Our result showed that neat ABS seemed toto ◦ and at higher frequencies for tohave have a cross-over point a frequency of about 100 Hz at 250 cross-over point frequency about 100Hz Hzatat 250 °C andC higherfrequencies frequenciesfor for280 280°C. °C. have a across-over point atata aat frequency ofofabout 100 250 °C and atathigher ◦ 280 C. ABS/GNP nanocomposite was likely to have cross-over point at frequencies of 100 Hz for both ABS/GNPnanocomposite nanocompositewas waslikely likelytotohave havecross-over cross-overpoint pointatatfrequencies frequenciesofof100 100Hz Hzfor forboth both ABS/GNP temperatures. No cross-over point for ABS/CNT suggested the presence of a rheological percolated temperatures. No cross-over point for ABS/CNT suggested the presence of a rheological percolated temperatures. No cross-over point for ABS/CNT suggested the presence of a rheological percolated network, and the effect ofofincreasing increasing temperature significantly reduced the storage modulus and loss network,and andthe theeffect effectof increasingtemperature temperaturesignificantly significantlyreduced reducedthe thestorage storagemodulus modulusand andloss loss network, modulus of the nanocomposite. modulusofofthe thenanocomposite. nanocomposite. modulus
5 5 1010
4 4 1010
4 4 1010
3 3 1010
3 3 1010
G' (Pa) G' (Pa)
G'' (Pa) G'' (Pa)
5 5 1010
2 2 1010
1 1 1010
-1 -1 1010
0 0 1 1 1010 1010 Frequency(Hz) (Hz) Frequency
ABS ABS 2 2 10 ABS/GNP 10 ABS/GNP ABS/CNT ABS/CNT 1 1 1010 2 2 10 10
6 6 1010
6 6 1010
280oCoC 280
5 5 1010
5 5 1010
4 4 1010
4 4 1010
3 3 1010
3 3 1010
2 2 1010
ABS 2 2 ABS 1010 ABS/GNP ABS/GNP ABS/CNT ABS/CNT 1 1 1010 2 2 10 10
G'' (Pa) G'' (Pa)
6 6 1010
250oCoC 250
G' (Pa) G' (Pa)
6 6 1010
1 1 1010
-1 -1 1010
0 0 1 1 1010 1010 Frequency (Hz) Frequency (Hz)
Figure5.5.Storage Storagemodulus modulus(full (fullsymbols) symbols)and andloss loss modulus(open (open symbols) ABS, ABS/GNPand and modulus symbols) of ABS, ABS/GNP Figure symbols) and loss modulus (open symbols) ofof ABS, ABS/GNP ◦ ABS/CNTnanocomposites nanocompositesat 250and and280 280°C. °C. ABS/CNT nanocomposites atat250 250 and 280 C. ABS/CNT
AnalogousCole-Cole Cole-Coleplots plotsare areused usedtotodescribe describethe theviscoelastic viscoelasticproperties propertiesofofmaterials materialswith witha a Analogous relaxation time distribution such as heterogeneous polymeric systems [24,25]. The real and imaginary relaxation time distribution such as heterogeneous polymeric systems [24,25]. The real and imaginary componentsofofviscoelastic viscoelasticproperties propertiesare arepresented. presented.InInthis thisstudy, study,we weused usedCole-Cole Cole-Coleplots plotstoto components
Appl. Sci. 2019, 9, 37
8 of 19
Analogous Cole-Cole plots are used to describe the viscoelastic properties of materials with a relaxation time distribution such as heterogeneous polymeric systems [24,25]. The real and imaginary Appl. Sci. 2018, 8, x 8 of 19 components of viscoelastic properties are presented. In this study, we used Cole-Cole plots to investigate temperature-induced nanocomposites and thethe results are investigate temperature-induced changes changesininthe themicrostructure microstructureofof nanocomposites and results ◦ C) which were the two selected reported in Figure 6 at two different temperatures (e.g., 250 and 280 are reported in Figure 6 at two different temperatures (e.g., 250 and 280 °C) which were the two temperatures for the 3D process.process. It can be clearly observed that atthat 250at◦ C, ABS selected temperatures for printing the 3D printing It can be clearly observed 250the °C,neat the neat presents a single relaxation arc. Due to a relatively higher amount of nanofillers, the linear region ABS presents a single relaxation arc. Due to a relatively higher amount of nanofillers, the linear region was evident evident for and ABS/CNT ABS/CNTnanocomposites. nanocomposites.This Thissuggests suggestsaa restraint restraint of of long-range long-range was for ABS/GNP ABS/GNP and motion of polymer chains. In addition, our results do not show a significant effect of temperature in motion of polymer chains. In addition, our results do not show a significant effect of temperature in Cole-Cole plots. Cole-Cole plots. 106
106 ABS ABS/GNP ABS/CNT
250 C
105
104
'' (Pa.s)
'' (Pa.s)
105
o
103 102 101 101
280oC
ABS ABS/GNP ABS/CNT
104
103
102
102
103 '(Pa.s)
104
105
101 101
102
103
104
105
'(Pa.s)
Figure (’) of of ABS, ABS, ABS/GNP ABS/GNP and Figure 6. 6. Cole-Cole Cole-Cole plots plots of of imaginary imaginary viscosity viscosity (’’) (η”) versus real viscosity (η’) and ◦ C. ABS/CNT ABS/CNTnanocomposites nanocompositesatat250 250and and280 280°C.
From the the data data of the complex viscosity viscosity at at different different temperatures temperatures and and frequencies, frequencies, an activation activation From energy (E (Eact ) for the viscous flow of the investigated samples can be evaluated by using an Arrhenius energy ) for the viscous flow of the investigated samples can be evaluated by using an Arrhenius act type equation equation (Equation (Equation 6), (6)),asas previouslyreported reported[26,27]. [26,27]. type previously Eact ' )( η 0)= log log (ηo ) E−act log(η = log(ηo ) − 2.303 2.303 RT RT
(8)
(8)
where η0 is a pre-exponential factor, T is the absolute temperature for the rheological test and R is the universal constant of 8.314factor, J/molK. of intercept η0 formally where 0 isgas a pre-exponential T isThe thevalue absolute temperature for the represents rheologicalcomplex test andviscosity R is the at infinite temperature. universal gas constant of 8.314 J/molK. The value of intercept 0 formally represents complex Table reportstemperature. the values of Eact for neat ABS and its nanocomposites at different frequencies. viscosity at1infinite The average activation forEthe viscous flowand of neat ABS was foundat todifferent be 107 kJ/mol. With Table 1 reports the energy values of act for neat ABS its nanocomposites frequencies. the addition of nanofillers, the decrease of the activation energy suggests that both the investigated The average activation energy for the viscous flow of neat ABS was found to be 107 kJ/mol. With the carbonaceous nanofillers the flow of polymer a pseudo-Newtonian In the addition of nanofillers, thefavor decrease of the activationchains energyinsuggests that both the zone. investigated presence of graphene, due to the possible lubricating effect of this lamellar filler, a minimum of the the carbonaceous nanofillers favor the flow of polymer chains in a pseudo-Newtonian zone. In activationofenergy can be Moreover, lower wettability with the polymer andof more presence graphene, dueobserved. to the possible lubricating effect of this lamellar filler, a matrix minimum the particle-particle may occur [25]. Itlower is also interesting to note that the activation activation energyinteractions can be observed. Moreover, wettability with the polymer matrix andenergy more evaluated in theinteractions melt flow process follows trend with 94,that andthe 117activation kJ/mol, for ABS, particle-particle may occur [25].anIt opposite is also interesting to87, note energy ABS/GNPinand respectively as shown in ourtrend previous this latter for case, the evaluated theABS/CNT, melt flow process follows an opposite withpaper 87, 94,[21]. andIn 117 kJ/mol, ABS, difference and could be attributed to the higher shear with respect to [21]. the rheological plate-plate ABS/GNP ABS/CNT, respectively as shown in rate our previous paper In this latter case, the experiments, andbe to the consequent pseudoplastic of respect polymertomelt extrusion where difference could attributed to the higher shearbehavior rate with the during rheological plate-plate both GNP and CNT played the direct role of effectively reduction themelt viscosity in extrusion the molten flow. experiments, and to the consequent pseudoplastic behavior of polymer during where It is also interesting to underline that the data in Table 1 show the tendency of ABS to decrease both GNP and CNT played the direct role of effectively reduction the viscosity in the molten flow.the It activation energy at frequencies, contrast to ABS/GNP and ABS/CNT. is also interesting tohigher underline that the in data in Table 1 show the tendency of ABS to decrease the activation energy at higher frequencies, in contrast to ABS/GNP and ABS/CNT.
Appl. Sci. 2018, 2019, 8, 9, x37
9 of 19
Table 1. Activation energy (Eact ) in kJ/mol at various frequencies at temperature 250 and 280 ◦ C.
Table 1. Activation energy (Eact) in kJ/mol at various frequencies at temperature 250 and 280 °C. Samples ABS ABS/GNP ABS/CNT
Samples 0.1 Hz 0.1 Hz ABS 83.1 83.1 ABS/GNP 11.5 11.5 ABS/CNT 71.6 71.6
1 Hz 1 Hz 130.7 130.7 16.9 16.9 60.5 60.5
100 Hz 10 Hz Average 10 Hz 100 Hz 107 ± 23 122.5 90.4 122.5 90.4 30 ± 19 45.3 47.8 45.3 47.8 65 ± 5 66.5 61.7 66.5 61.7
Average 107 ± 23 30 ± 19 65 ± 5
3.2. Transmission Transmission Electron 3.2. Electron Microscopy Microscopy Transmission Transmission electron electron microscopy microscopy was was used used to to evaluate evaluate the the quality quality of of the the dispersion dispersion and and orientation of nanofillers in different types of processes (i.e., compression molding and orientation of nanofillers in different types of processes (i.e., compression molding and extrusion). extrusion). In In Figure 7,7,TEM micrographs of compression molded samples are presented. An evident dispersion Figure TEM micrographs of compression molded samples are presented. Angood evident good of grapheneofand carbon nanotubes be observed no detectable In aggregates. addition, both dispersion graphene and carboncan nanotubes can with be observed with aggregates. no detectable In graphene and carbon nanotubes seemed to be randomly oriented in the ABS matrix. In the ABS/GNP addition, both graphene and carbon nanotubes seemed to be randomly oriented in the ABS matrix. nanocomposite presence of wrinkled graphene flakesgraphene can be observed, while in ABS/CNT In the ABS/GNPthe nanocomposite the presence of wrinkled flakes can be observed, whilethe in nanotubes the entangled in interconnected structures can structures be detected. ABS/CNT nanotubes entangled in interconnected can be detected.
(a)
(b)
(c)
(d)
molded (CM) nanocomposites: (a,b)(a,b) ABS/GNP and (c,d) Figure 7. TEM TEMmicrographs micrographsofofcompression compression molded (CM) nanocomposites: ABS/GNP and (c,d) ABS/CNT. ABS/CNT .
The The surface surface of of nanocomposite nanocomposite filaments filaments was was investigated investigated in in two two directions directions (i.e., (i.e., perpendicular perpendicular and parallel) to flow polymer direction in order to determine the orientation of nanofillers. Similar to and parallel) to flow polymer direction in order to determine the orientation of nanofillers. Similar to the compression compression molding molding process, process, nanofillers nanofillers without without visible visible agglomerated agglomerated structures structures were were observed observed the (see and 9). 9). Figure Figure 88 presents presents the nanocomposite filament, filament, and and graphene graphene flakes flakes (see Figures Figure 88 and the ABS/GNP ABS/GNP nanocomposite were were oriented oriented along along polymer polymer flow flow in in the the extrusion extrusion direction. direction.
Appl. Sci. 2019, 9, 37 Appl. Appl. Sci. Sci. 2018, 2018, 8, 8, xx
10 of 19 10 10 of of 19 19
(a) (a)
(b) (b)
Figure 8. 8. TEM of ABS/GNP ABS/GNP Figure TEM micrograph micrograph of nanocompositefilament: filament: (a) (a)Perpendicular Perpendicular and and (b) (b) parallel. parallel. ABS/GNP nanocomposite nanocomposite filament: (a) Perpendicular and (b) parallel.
(a) (a)
(b) (b)
Figure TEM micrograph nanocompositefilament: filament: (a) (a)Perpendicular Perpendicular and and (b) (b) parallel. parallel. Figure 9. of ABS/CNT Figure 9. 9. TEM TEM micrograph micrograph of of ABS/CNT ABS/CNT nanocomposite nanocomposite filament: (a) Perpendicular and (b) parallel.
On the other other hand, nanotubes nanotubes in the the ABS matrix matrix were randomly randomly oriented from from the extrusion extrusion On On the the other hand, hand, nanotubes in in the ABS ABS matrix were were randomly oriented oriented from the the extrusion process (see Figure 9b). Additional TEM images of the composites are given in the Supplementary process process (see (see Figure Figure 9b). 9b). Additional Additional TEM TEM images images of of the the composites composites are are given given in in the the Supplementary Supplementary Materials and and they they better better illustrate illustrate the the orientation orientation and and random random distribution distribution of of the the particles in in the Materials Materials and they better illustrate the orientation and random distribution of the particles particles in the the ABS/GNP and ABS/CNT composites, respectively (see Figure S1). ABS/GNP ABS/GNP and and ABS/CNT ABS/CNT composites, composites, respectively respectively (see (see Figure Figure S1). S1). 3.3. Electrical Conductivity 3.3. 3.3. Electrical Electrical Conductivity Conductivity Table 2 summarizes the electrical conductivity and the apparent density of the compression Table Table 22 summarizes summarizes the the electrical electrical conductivity conductivity and and the the apparent apparent density density of of the the compression compression molded (CM) and 3D manufactured (FDM) samples. The apparent density was estimated from direct molded (CM) and 3D manufactured (FDM) samples. The apparent density was estimated from molded (CM) and 3D manufactured (FDM) samples. The apparent density was estimated from direct direct 3 measurement of weight square plaques and volume with size of 45 × 45 × 2 mm . 3 measurement measurement of of weight weight square square plaques plaques and and volume volume with with size size of of 45 45 ×× 45 45 ×× 22 mm mm3.. Table 2. Electrical conductivity and apparent density of ABS and its composites. Table Table 2. 2. Electrical Electrical conductivity conductivity and and apparent apparent density density of of ABS ABS and and its its composites. composites. Samples
Samples Samples
CM CM CMFDM-PC ABS FDM-PC FDM-PC FDM-HC ABS ABS FDM-HC FDM-H45 FDM-HC CM FDM-H45 FDM-H45 FDM-PC CM ABS/GNP CM FDM-HC FDM-PC FDM-PC FDM-H45 ABS/GNP ABS/GNP FDM-HC FDM-HC CM FDM-H45 FDM-PC FDM-H45 ABS/CNT CM FDM-HC CM FDM-H45 FDM-PC
ABS/CNT ABS/CNT
FDM-PC FDM-HC FDM-HC FDM-H45 FDM-H45
Electrical Conductivity (S·cm−1 )
Electrical ·cm-1-1) Electrical Conductivity Conductivity −(S (S 16 cm ) (3.06 ± 4.02) × 10-16 -16 (3.06 ± 4.02) ×× 10 (3.06 (4.93 ±±4.02) 2.89) ×10 10−16 -16 (4.93 −16 (4.93 ±±2.89) 2.89) ××10 10 (2.63 ± 3.21) × 10-16 −16 -16 (2.63 (2.67± ±3.21) 3.30) × × 10-16 (2.63 ± 3.21) × 10 10 −14 (1.55 ±3.30) 3.67)× 10-16 -16 (2.67± (2.67± 3.30) ××10 10 −15 (3.00 ± 4.08) × 10 -14 (1.55 (1.55 ±± 3.67) 3.67) ×× 10 10-14 −15 (2.34 ± 4.34) × 10-15 -15 (3.00 ±± 4.08) ×× 10 (3.00 4.08) −15 (1.89 ± 3.98) ×10 10-15 -15 (2.34 ± 4.34) × 10 (2.34 ± 4.34) × 10 (6.97 ± 1.01) × 10−3 -15 (1.89 (9.17± 0.89)× 10-15−4 (1.89 ±±3.98) 3.98) ××10 10 -3−4 (6.97 (2.34 ± 1.11) × 10-3 (6.97 ±±1.01) 1.01) ××10 10 −4 -4 (1.68 ± 1.21) × 10 (9.17 (9.17 ±± 0.89) 0.89) ×× 10 10-4 -4 (2.34 (2.34 ±± 1.11) 1.11) ×× 10 10-4 (1.68 (1.68 ±± 1.21) 1.21) ×× 10 10-4-4
Apparent Density (g/cm3 )
Apparent Apparent density density (g/cm (g/cm33)) 1.05 ± 0.02 1.05 ±± 0.02 0.87 1.05 ± 0.07 0.02 ±± 0.07 0.87 0.07 0.930.87 ± 0.03 ±± 0.03 0.950.93 ± 0.02 0.93 0.03 1.070.95 ± 0.01 ±± 0.02 0.95 0.02 0.891.07 ± 0.02 1.07 ±± 0.01 0.01 0.97 ± 0.02 0.89 ±± 0.02 0.89 0.98 ± 0.02 0.02 ±± 0.02 0.97 0.02 1.070.97 ± 0.01 ±± 0.02 0.870.98 ± 0.03 0.98 0.02 ±± 0.01 0.921.07 ± 0.03 1.07 0.01 0.940.87 ± 0.02 ±± 0.03 0.87 0.03 0.92 0.92 ±± 0.03 0.03 0.94 0.94 ±± 0.02 0.02
Appl. Sci. 2019, 9, 37 Appl. Sci. 2018, 8, x
11 of 19 11 of 19
The presence presence of of MWCNT MWCNT considerably considerably modified modified the the electrical electrical conductivity conductivity of ofthe theABS/CNT ABS/CNT The 3cm −1 10 −4-1 S 1 were achieved −3 − -1 and -4 S· S· cm were for the polymer composites. Electrical Electricalconductivities conductivitiesofofabout about1010 S·cm and 10 ·cm−achieved for the composites manufactured via compression molding andand FDM, respectively. polymer composites manufactured via compression molding FDM, respectively.This This observation created anan efficient conductive network across the suggests that the the addition additionofof66wt.% wt.%ofofMWCNT MWCNT created efficient conductive network across ABSABS matrix offering sufficient mobility to tothe of the matrix offering sufficient mobility thecharge chargecarriers. carriers.Differently, Differently,the the same same amount of modify the the electrical electrical conductivity conductivity of of ABS ABS matrix. matrix. Both ABS ABS and and graphene nanoplatelets did not modify ABS/GNP nanocomposites the ABS/GNP nanocomposites were non-conductive materials with electrical conductivity values in the − 16 − 14 − 1 −16 −14 −1 range between between 10 10 and of the fabrication route. range and10 10 S·cm S·cm, independent , independent of the fabrication route. In the the case case of of GNP-based GNP-based nanocomposites, nanocomposites, TEM images (see Figure 8) of of extruded extruded filaments filaments In that graphene graphene nanoplatelets were were oriented oriented along along the the polymer polymer flow flow direction. direction. The The contacts contacts indicate that created between between graphene graphenenanoplatelets nanoplateletswere werenot notenough enoughto topromote promoteaapercolative percolativepath pathin inABS/GNP ABS/GNP created composites resulting resulting in in aa non-conductive non-conductive composite. composite. From From Figure Figure 9, 9, it can be observed that MWCNTs composites MWCNTs were not not oriented oriented during during the the extrusion extrusion of of the the filaments. filaments. In In fact, fact, this this filler filler remained remained well-distributed well-distributed were across the the ABS ABS matrix matrix thus thus creating creating an an efficient efficient conductive conductive percolated percolated network network for for carrying carrying charges charges across across the the material. material. across 3.4. Electromagnetic Interference Interference Shielding Shielding Efficiency Efficiency The in Figure 10 refer the total electromagnetic interference interference shielding effectiveness The graphs graphsgiven given in Figure 10torefer to the total electromagnetic shielding of ABS and its composites prepared via compressed molding and FDM along the three effectiveness of ABS and its composites prepared via compressed molding and FDM along different the three growing describeddescribed in Figure 1. different directions growing directions in Figure 1. -50 CM ABS/CNT
-40
EMI SE (dB)
-50
-30
-20
-20
-10 ABS/GNP
-10
ABS
-50
9
10
11
12
7
13
-40
-40
-30
-30
7
8
9
10
11
10
11
Frequency (GHz)
12
13
13
ABS/GNP ABS
0 9
12
ABS/CNT
-10
ABS/GNP ABS
0
8
FDM-H45
-20
ABS/CNT
-10
ABS/GNP ABS
-50
FDM-HC
-20
ABS/CNT
0
8
EMI SE (dB)
-40
-30
0
FDM-PC
7
8
9
10
11
12
13
Frequency (GHz)
Figure 10. 10. Total Total EMI EMI shielding shielding effectiveness effectiveness of of ABS ABS and and composites: composites: Compression Compression molded molded (upper (upper left left Figure graph), FDM-PC FDM-PC (upper (upper right right graph), FDM-HC FDM-HC (lower (lower left graph) and FDM-H45 (lower right graph). graph), The orientations orientations PC, PC, HC HC and and H45 H45 are aredescribed describedin inFigure Figure2.2. The
The EMI SE SEisisclearly clearlya afunction functionofof both filler type as well as fabrication the fabrication route. The total EMI both thethe filler type as well as the route. The The higher SE was achieved by ABS/CNT the ABS/CNT independent the fabrication Onother the higher EMIEMI SE was achieved by the independent of theoffabrication route.route. On the
hand, the highest EMI SE values were achieved for the samples manufactured via compression molding whereas 3D-printed parts attenuated the incident EM waves in the following order: FDM-
Appl. Sci. 2019, 9, 37
12 of 19
Appl. Sci. 2018, 8, x
12 of 19
other hand, the highest EMI SE values were achieved for the samples manufactured via compression PC > FDM-HC ≈ FDM-H45, independently from the composite formulation. For example, the total molding whereas 3D-printed parts attenuated the incident EM waves in the following order: FDM-PC EMI SE of ABS/CNT composite processed via compression molding was around 45 dB, whereas the > FDM-HC ≈ FDM-H45, independently from the composite formulation. For example, the total EMI SE 3D-printed specimens of the same composite showed attenuation near 25 dB (PC direction) and 15 of ABS/CNT composite processed via compression molding was around 45 dB, whereas the 3D-printed dB (HC and H45 direction). Differently, lower attenuations were observed for the ABS/GNP, specimens of the same composite showed attenuation near 25 dB (PC direction) and 15 dB (HC and approximately 5 dB, independently from the fabrication route. Considering that the EMI SE of H45 direction). Differently, lower attenuations were observed for the ABS/GNP, approximately 5 dB, electrically conductive polymer composites is improved from the increase in electrical conductivity independently from the fabrication route. Considering that the EMI SE of electrically conductive [1,5,28], the higher values of EMI SE achieved for ABS/CNT composites with respect to other polymer composites is improved from the increase in electrical conductivity [1,5,28], the higher values formulations are reasonable. Figure 11 compares the amount of the incident radiation attenuated by of EMI SE achieved for ABS/CNT composites with respect to other formulations are reasonable. absorption and reflection by both ABS/GNP and ABS/CNT composites prepared by CM and FDM. Figure 11 compares the amount of the incident radiation attenuated by absorption and reflection by The whole absorption and reflection spectra of tested samples are given in the Supplementary both ABS/GNP and ABS/CNT composites prepared by CM and FDM. The whole absorption and Materials (see Figures S2 and S3). reflection spectra of tested samples are given in the Supplementary Materials (see Figures S2 and S3). 50
SE Reflection SE Absorption
EMI SE mean (-dB)
40
30
20
10
0 C
C
C
M FD
-P M
FD
-H M
FD
45
-H M
ABS/GNP
C
C -P M
M FD
C 45 -H -H M M FD FD
ABS/CNT
Figure 11. Shielding effectiveness by absorption and reflection of ABS composites. Figure 11. Shielding effectiveness by absorption and reflection of ABS composites.
It is worthwhile to note that ABS/CNT attenuated the incident EM waves mostly by absorption It is worthwhile to note that ABS/CNT attenuated the incident EMlow waves bythe absorption (approximately by 80%). For ABS/GNP composites instead, despite the totalmostly EMI SE, incident (approximately For ABS/GNP composites instead,in despite the low total EMI SE, the incident EM was shieldedby by80%). absorption and reflection mechanisms similar amounts. EM was reflection mechanisms CM shielded samples by areabsorption assumed toand have a negligible quantityinofsimilar defectsamounts. such as voids or holes, since CM samples are assumed toduring have a the negligible quantity process. of defectsInsuch as voids holes, since the samples are well-compacted manufacturing contrast, the or deposition of the samples are well-compacted the manufacturing process. In contrast, the inside deposition of layer-by-layer in FDM generates aduring microstructure with several material/air interfaces the 3D layer-by-layer in FDM generates a microstructure with several material/air interfaces inside the 3D components. Moreover, in the case of 45 × 45 × 2 mm samples, less material was used to produce components. in fact, the case of 45 × 45molded × 2 mm specimens samples, less was an used to produce the the specimensMoreover, via FDM. In compression arematerial denser, with apparent density 3 (Table specimens via FDM. 2). In fact, compression molded specimens denser, with an apparent of of 1.06 g/cm 3D-printed parts manufactured alongare H45 and HC directions haddensity a density 3 3 3 1.06 g/cm (Table 0.95 2). 3D-printed parts along manufactured along H45 and HC density of of approximately g/cm whereas the PC direction a density of directions 0.88 g/cm had wasa measured. 3 whereas along the PC direction a density of 0.88 g/cm3 was measured. This approximately 0.95 as g/cm This indicates that, expected, a higher porosity is generated by the FDM process. The different indicates that, can as expected, a higher is SE generated by the manufactured FDM process. via TheFDM different microstructure therefore explain theporosity lower EMI for components with microstructure canmain therefore explainbetween the lower EMI SE for components via FDM respect to CM. The differences samples obtained by CM ormanufactured FDM are represented bywith the respect to CM. main differences between CM or FDMporosity, are represented by distribution andThe dispersion of the fillers besidessamples the fact obtained that, due by to their higher 3D-printed the distribution and dispersion of the fillers besides the fact that, due to their higher porosity, 3Dcomponents offer less material to interact with the incident EM waves. printed less material to interact with the incident EM waves. It iscomponents worthwhileoffer noticing that 3D-printed ABS/CNT tested along the PC direction reached a It is worthwhile noticing that 3D-printed ABS/CNT tested alongEM thewave PC direction reached a total total EMI SE near 25 dB representing an attenuation of the incident higher than 99.9% [8]. EMImagnitudes SE near 25 dB representing anwere attenuation of the incident EM wave than 99.9% The The of the attenuation approximately 20 dB lower than higher the respective CM [8]. sample magnitudes of the attenuation were approximately 20 dB lower than the respective CM sample (~ 45 dB) and 10 dB higher than the 3D manufactured samples along the HC and H45 directions (~ 15 dB).
Appl. Sci. 2019, 9, 37
13 of 19
(~45 dB) and 10 dB higher than the 3D manufactured samples along the HC and H45 directions (~15 dB). Interestingly, the lowest apparent density was measured for FDM-PC components and the electrical conductivities of ABS/CNT composites were just an order of magnitude different; near 10−3 S·cm−1 and 10−4 S·cm−1 for the ABS/CNT composites processed via CM and FDM, respectively. Table 3 presents a comparative overview on various polymer nanocomposites prepared for EMI SE investigations.
Appl. Sci. 2019, 9, 37
14 of 19
Table 3. Comparative view of EMI SE of polymer nanocomposites investigated by various authors. Reference
Polymer Matrix
Filler
[29]
Acrylonitrile–butadiene–styrene (ABS)
[30]
[31]
[32] [33] [34] [35]
Thickness
Preparation Method
carbon black (CB) and carbon nanotubes (CNTs)
2 mm
Melt mixing followed by compression molding
44 dB 87 dB 9 dB 34 dB
ultrahigh molecular weight polyethylene (UHMWPE)
carbon nanotubes (CNTs)
1 mm
wet mixing followed by compression molding
50 dB
polylactic acid (PLA)
carbon nanotubes (CNTs)
0.4 mm
3D printable conductive scaffold microstructures
55 dB 30 dB
Acrylonitrile–butadiene–styrene (ABS) Acrylonitrile–butadiene–styrene (ABS) polycarbonate (PC) Acrylonitrile–butadiene–styrene (ABS)
multiwall carbon nanotubes (MWCNTs) multiwall carbon nanotubes (MWCNTs) GNP type M multiwall carbon nanotubes (MWCNTs)
2.4mm 2.8 mm
dry powder mixing followed by hot compression
2.2 mm
Injection molding
3.2 mm
Injection molding Solution mixing following by compression molding
1 dB 40 dB 39 dB in the Ku band (12.4–18 GHz) 26.3 dB at 800 mHz
1.1 mm
EMI SE in the X-Band Frequency Range
~30dB
5 wt.% ABS/CNT 15 wt.% ABS/CNT 5 wt.% ABS/CB 15 wt.% ABS/CB 10 wt.% UHMWPE/CNT 30 wt.% PLA/CNT solid 30 wt.% PLA/CNT scaffold 0.05 wt.% ABS/CNT 5 wt.% ABS/CNT 10wt.% ABS/CNT 15wt.% PC/GNP 10wt.%ABS/CNT
Appl. Sci. 2019, 9, 37 Appl. Sci. 2018, 8, x
15 of 19 1 of 19
To To better understand understand EMI EMI shielding shielding properties, properties, Figures Figures 12 12 and and 13 13 show show the the variation variation of of the the real real 0 ) and (dielectric andthe theimaginary imaginary(dielectric (dielectricloss, loss,ε00ε’’) parts complex dielectric function (dielectric constant, εε’) ) parts ofof thethe complex dielectric function (or (or permittivity) for investigated the investigated composites the X-band microwave frequency range. The permittivity) for the composites in the in X-band microwave frequency range. The addition addition wt.% of MWCNT considerably the dielectric of ABS 11). (Figure 11). an In of 6 wt.%ofof6 MWCNT considerably modifiedmodified the dielectric constantconstant of ABS (Figure In fact, fact, an improved value was measured for the ABS/CNT CM and specimens FDM-PC specimens improved ε0 value ε’ was measured for the ABS/CNT and bothand CM both and FDM-PC exhibited, exhibited, the frequency dependence, comparable behavior. The ε’ ispresence due to the despite thedespite frequency dependence, comparable ε0 behavior.ε’The increase of increase ε0 is due of to the of presence of mobile charge and related of the in component the charges mobile charge carriers andcarriers related to the abilitytoofthe theability component transport in thetransport charges from a point from a point to across another one across the component. to another one the component. 70 60
ε'
CM
FDM-PC
60
ABS/CNT
50
50 ABS/CNT
40
40
30
30
20
20
10 0 70
10
ABS/GNP ABS
8
ε'
70
0 9
10
11
12
13
FDM-HC
8
60
50
50
40
40
30 ABS/CNT
30
20
20
0
10
11
12
13
FDM-H45
ABS/CNT
10 ABS/GNP
ABS/GNP ABS
8
9
70
60
10
ABS/GNP ABS
0 9
10
11
Frequency (GHz)
12
13
ABS
8
9
10
11
12
13
Frequency (GHz)
Figure 12. 12. Dielectric Dielectricconstant constantof ofABS ABS and and its its composites composites in in the the X-band X-band frequency frequency range; range; FDM-PC; FDM-PC; Figure FDM-HC; FDM-H45. The orientations PC, HC and H45 are described in Figure 2. FDM-HC; FDM-H45. The orientations PC, HC and H45 are described in Figure 2.
As far far as as the the dielectric composites processed processed via via As dielectric loss loss (Figure (Figure 12) 12) is is concerned, concerned, ABS/CNT ABS/CNT composites 00 values over the X-band frequency range. The 3D-printed compression molding exhibited the highest ε compression molding exhibited the highest ε’’ values over the X-band frequency range. The 3Dspecimens showedshowed improved dielectric loss values in thisinorder: FDM-PC > FDM-H45 ≈ FDM-HC. printed specimens improved dielectric loss values this order: FDM-PC > FDM-H45 ≈ FDM00 are related to the dissipation of the energy from the incident EM waves which are The values of ε HC. The values of ε’’ are related to the dissipation of the energy from the incident EM waves which absorbed by by thethe material and converted into proven, the the are absorbed material and converted intoheat heat[22]. [22].Although Although not not experimentally experimentally proven, higher quantities quantities of of material/air material/airinterfaces interfacescreated createdininthe theFDM-PC FDM-PCcomponents componentscould couldbe be the the reason reason higher for their better ability of dissipating the energy from the incident EM waves. In fact, higher multiple for their better ability of dissipating the energy from the incident EM waves. In fact, higher multiple internal interactions interactions might efficiencies forfor FDM-PC internal might have have occurred, occurred,justifying justifyingthe thehigher higherEMI EMIshielding shielding efficiencies FDMcomponents even though they have a lower amount of material to interact with the EM radiation. PC components even though they have a lower amount of material to interact with the EM radiation.
Appl. Sci. 2019, 9, 37 Appl. Sci. 2018, 8, x
160 140
16 of 19 2 of 19
160
CM
120
120
100
ε''
80
60
60
40
40
20 ABS/GNP 0 ABS
ε''
100
ABS/CNT
80
160
8
9
10
11
12
ABS/CNT
20 ABS/GNP 0 ABS 13 8 9 160
140 FDM-HC
140
120
120
100
100
80
80
60
60
40 20
FDM-PC
140
10
11
12
13
FDM-H45
40 ABS/GNP
0 ABS 8
ABS/CNT
20 0
9
10
11
Frequency (GHz)
12
13
ABS/GNP
ABS/CNT
ABS
8
9
10
11
12
13
Frequency (GHz)
Figure Figure 13. 13. Dielectric Dielectricloss loss of ofABS ABS and and its its composites composites in in the the X-band X-band frequency frequency range. range.
In In summary, summary, MWCNT is the most efficient carbon-based filler when an electrically conductive conductive polymer polymer composite composite with with enhanced enhanced EMI EMI shielding shielding purposes purposes is desired. In In fact, fact, low low quantities quantities of of MWCNT MWCNT are are required required to to improve the mechanical, mechanical, electrical electrical and and EMI EMI properties properties of of ABS ABS based based composites composites without without compromising compromising the rheological properties [17–20]. ABS ABS polymer polymer composites composites containing carbonblack black or xGnP eventually reach EMI than SE higher than –20 dB containing carbon or xGnP couldcould eventually reach total EMItotal SE higher –20 dB corresponding corresponding attenuations >99.9%, by simply increasing the content of these fillers. However, to attenuationsto >99.9%, by simply increasing the content of these fillers. However, processing such processing such via FDM would duerheological to changes on the rheological formulations via formulations FDM would be challenging duebe to challenging changes on the behavior. In addition, behavior. In addition, a recent study has shown thecarbon advantages of using nanotubes over the a recent study has shown the advantages of using nanotubes overcarbon the traditional aluminum traditional aluminum sheets the development composites for EMI shielding sheets in the development of in polymer compositesof forpolymer EMI shielding purposes; energetic andpurposes; economic energetic and economictosavings are attributed to the polymer composites based on carbon nanotubes savings are attributed the polymer composites based on carbon nanotubes [36]. [36]. 4. Conclusions 4. Conclusions Multi-walled carbon nanotubes, and graphene nanoplatelets were added to an ABS polymer matrix at a fixed concentration of 6 wt.%. of resulting polymer composites were Multi-walled carbon nanotubes, and Extruded graphenefilaments nanoplatelets were added to an ABS polymer obtained used as feedstockoffor fusedExtruded deposition modeling. 3D specimens were manufactured matrix at aand fixed concentration 6 wt.%. filaments of resulting polymer composites were in different conductivity the electromagnetic interference obtained andprinting used as orientations. feedstock for The fusedelectrical deposition modeling. and 3D specimens were manufactured in shielding efficiency of resulting 3D printed samples were assessed and compared with those of different printing orientations. The electrical conductivity and the electromagnetic interference compressed molded of specimens theprinted same composition. As assessed expected,and compression molded samples shielding efficiency resultingof3D samples were compared with those of exhibited higher values, due to the continuity of material and the virtual absence of voids. compressed molded specimens of the same composition. As expected, compression molded samples TEMhigher images of the extruded filamentsof showed that the graphene nanoplatelets, exhibited values, due to the continuity material and virtual absence of voids.during the preparation of filaments, oriented along showed the polymer direction, whereas multi-walled TEM images of the were extruded filaments that flow graphene nanoplatelets, during the carbon nanotubes remained randomly dispersed in the polymer matrix. Polymer composites preparation of filaments, were oriented along the polymer flow direction, whereas multi-walled loaded nanotubes with multi-walled nanotubes showed electrical values and EMI carbon remainedcarbon randomly dispersed in thehigher polymer matrix.conductivity Polymer composites loaded shielding effectiveness. with multi-walled carbon nanotubes showed higher electrical conductivity values and EMI shielding effectiveness. The additive manufacturing method called fused deposition modeling has great potential for producing 3D printed components with high electromagnetic interference shielding efficiency. In
Appl. Sci. 2019, 9, 37
17 of 19
The additive manufacturing method called fused deposition modeling has great potential for producing 3D printed components with high electromagnetic interference shielding efficiency. In fact, polymer composites based on ABS multi-walled carbon nanotube are able to attenuate 99.9% of the incident electromagnetic wave when 6 wt.% of MWCNT in the composite formulation are added. Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/2076-3417/9/1/ 37/s1. Figure S1: TEM micrograph of filaments in parallel: ABS/GNP (a) and ABS/CNT (b) nanocomposites, Figure S2: EMI shielding by absorption of ABS and composites: compression moulded (upper left graph), FDM-PC (upper right graph), FDM-HC (lower left graph) and FDM-H45 (lower right graph), Figure S3: EMI shielding by reflection of ABS and composites: compression molded (upper left graph), FDM-PC (upper right graph), FDM-HC (lower left graph) and FDM-H45 (lower right graph). Author Contributions: L.G.E., S.D., L.F. and A.P. conceived and designed the experiments; L.G.E., S.D., D.P.S., G.M.d.O.B. and B.G.S., performed the experiments; L.G.E., S.D., D.P.S., G.M.d.O.B., B.G.S., L.F., A.P. analyzed the data and wrote the paper. Funding: This research was funded by the Brazilian research foundation CNPq “Conselho Nacional de Desenvolvimento Científico e Tecnológico”. Acknowledgments: Authors are thankful to Versalis S.p.A. (Mantova, Italy) for donating ABS pellets for this work. Authors also wish to thank Sharebot S.r.l. (Nibionno, LC, Italy) for providing the prototype of the HT Ne×t Generation desktop 3D-printer. One author (S.D.) gratefully acknowledges financial support from AREAS+ EU Project of Erasmus Mundus Action 2 Programme. Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2.
3.
4. 5.
6. 7. 8. 9.
10. 11. 12. 13.
Tong, X.C. Advanced Materials for Electromagnetic Interference Shielding, 1st ed.; CRC Press: Boca Raton, FL, USA, 2009. Jda Silva, P.S.; Soares, B.G.; Livi, S.; Barra, G.M.O. Phosphonium—Based ionic liquid as dispersing agent for MWCNT in melt-mixing polystyrene blends: Rheology, electrical properties and EMI shielding effectiveness. Mater. Chem. Phys. 2017, 189, 162–168. [CrossRef] European Parliament, European Council, Directive 2014/30/EU: Electromagnetic Compatibility (Recast), 29.3. 2014, pp. 79–106. Available online: http://eur-lex.europa.eu/legal-content/ES/TXT/?uri=CELEX: 32014L0030 (accessed on 20 December 2018). Chung, D.D.L. Carbon materials for structural self-sensing, electromagnetic shielding and thermal interfacing. Carbon 2012, 50, 3342–3353. [CrossRef] Thomassin, J.M.; Jérôme, C.; Pardoen, T.; Bailly, C.; Huynen, I.; Detrembleur, C. Polymer/carbon based composites as electromagnetic interference (EMI) shielding materials. Mater. Sci. Eng. R Rep. 2013, 74, 211–232. [CrossRef] Pawar, S.P.; Biswas, S.; Kar, G.P.; Bose, S. High frequency millimetre wave absorbers derived from polymeric nanocomposites. Polymer 2015, 84, 398–419. [CrossRef] Pang, H.; Xu, L.; Yan, D.X.; Li, Z.M. Conductive polymer composites with segregated structures. Prog. Polym. Sci. 2014, 39, 1908–1933. [CrossRef] Al-Saleh, M.H.; Sundararaj, U. A review of vapor grown carbon nanofiber/polymer conductive composites. Carbon 2009, 47, 2–22. [CrossRef] Kar, G.P.; Biswas, S.; Rohini, R.; Bose, S. Tailoring the dispersion of multiwall carbon nanotubes in co-continuous PVDF/ABS blends to design materials with enhanced electromagnetic interference shielding. J. Mater. Chem. A 2015, 3, 7974–7985. [CrossRef] Saini, P.; Choudhary, V.; Singh, B.P.; Mathur, R.B.; Dhawan, S.K. Enhanced microwave absorption behavior of polyaniline-CNT/polystyrene blend in 12.4–18.0 GHz range. Synth. Met. 2011, 161, 1522–1526. [CrossRef] Joseph, N.; Janardhanan, C.; Sebastian, M.T. Electromagnetic interference shielding properties of butyl rubber-single walled carbon nanotube composites. Compos. Sci. Technol. 2014, 101, 139–144. [CrossRef] Joseph, N.; Varghese, J.; Sebastian, M.T. Graphite reinforced polyvinylidene fluoride composites an efficient and sustainable solution for electromagnetic pollution. Compos. Part B Eng. 2017, 123, 271–278. [CrossRef] Li, L.; Li, S.; Shao, Y.; Dou, R.; Yin, B.; Yang, M. PVDF/PS/HDPE/MWCNTs/Fe3 O4 nanocomposites: Effective and lightweight electromagnetic interference shielding material through the synergetic effect of MWCNTs and Fe3 O4 nanoparticles. Curr. Appl. Phys. 2018, 18, 388–396. [CrossRef]
Appl. Sci. 2019, 9, 37
14. 15. 16. 17.
18. 19. 20. 21.
22.
23. 24.
25. 26. 27. 28. 29.
30. 31.
32.
33.
34.
18 of 19
Huang, C.Y.; Wu, C.C. The EMI shielding effectiveness of PC/ABS/nickel-coated-carbon-fibre composites. Eur. Polym. J. 2000, 36, 2729–2737. [CrossRef] Huang, C.-Y.; Pai, J.-F. Studies on processing parameters and thermal stability of ENCF/ABS composites for EMI shielding. J. Appl. Polym. Sci. 1997, 63, 115–123. [CrossRef] Rostami, A.; Masoomi, M.; Fayazi, M.J.; Vahdati, M. Role of multiwalled carbon nanotubes (MWCNTs) on rheological, thermal and electrical properties of PC/ABS blend. RSC Adv. 2015, 5, 32880–32890. [CrossRef] Schmitz, D.P.; Ecco, L.G.; Dul, S.; Pereira, E.C.L.; Soares, B.G.; Barra, G.M.O.; Pegoretti, A. Electromagnetic interference shielding effectiveness of ABS carbon-based composites manufactured via fused deposition modelling. Mater. Today Commun. 2018, 15. [CrossRef] Dorigato, A.; Moretti, V.; Dul, S.; Unterberger, S.H.; Pegoretti, A. Electrically conductive nanocomposites for fused deposition modelling. Synth. Met. 2017, 226, 7–14. [CrossRef] Dul, S.; Fambri, L.; Pegoretti, A. Fused deposition modelling with ABS-graphene nanocomposites. Compos. Part A Appl. Sci. Manuf. 2016, 85, 181–191. [CrossRef] Dul, S.; Pegoretti, A.; Fambri, L. Filaments Production and Fused Deposition Modelling of ABS/Carbon Nanotubes Composites. Nanomaterials 2018, 8, 49. [CrossRef] Dul, S.; Pegoretti, A.; Fambri, L. Effects of the Nanofillers on Physical Properties of Acrylonitrile-Butadiene-Styrene Nanocomposites: Comparison of Graphene Nanoplatelets and Multiwall Carbon Nanotubes. Nanomaterials 2018, 8, 674. [CrossRef] [PubMed] Kuester, S.; Demarquette, N.R.; Ferreira, J.C.; Soares, B.G.; Barra, G.M.O. Hybrid nanocomposites of thermoplastic elastomer and carbon nanoadditives for electromagnetic shielding. Eur. Polym. J. 2017, 88, 328–339. [CrossRef] Heuzey, A.G.M.; Ton-that, P.J.C.M. Morphological and rheological properties of PET/clay. Nanocomposites 2013, 59–74. [CrossRef] Wu, D.; Wu, L.; Sun, Y.; Zhang, M. Rheological Properties and Crystallization Behavior of Multi-Walled Carbon Nanotube/Poly(ε-caprolactone) Composites. J. Polym. Sci. Part B Polym. Phys. 2007, 3137–3147. [CrossRef] Abbasi, S.; Carreau, P.J.; Derdouri, A.; Moan, M. Rheological properties and percolation in suspensions of multiwalled carbon nanotubes in polycarbonate. Rheol. Acta 2009, 48, 943–959. [CrossRef] Fambri, L.; Bragagna, S.; Migliaresi, C. Biodegradable Fibers of Poly-L, DL-lactide 70/30 Produced by Melt Spinning. Macromol. Symp. 2006, 20–25. [CrossRef] MacDermott, C.P.; Shenoy, A.V. Selecting Thermoplastics for Engineering Applications; Second edition; Marcel Dekker, Inc.: New York, NY, USA, 1997. Yang, Y.; Gupta, M.C.; Dudley, K.L. Towards cost-efficient EMI shielding materials using carbon nanostructure-based nanocomposites. Nanotechnology 2007, 18, 345701. [CrossRef] Schmitz, D.P.; Silva, T.I.; Ramoa, S.D.A.S.; Barra, G.M.O.; Pegoretti, A.; Soares, B.G. Hybrid composites of ABS with carbonaceous fillers for electromagnetic shielding applications. J. Appl. Polym. Sci. 2018, 135. [CrossRef] Al-Saleh, M.H. Influence of conductive network structure on the EMI shielding and electrical percolation of carbon nanotube/polymer nanocomposites. Synth. Met. 2015, 205, 78–84. [CrossRef] Chizari, K.; Arjmand, M.; Liu, Z.; Sundararaj, U.; Therriault, D. Three-dimensional printing of highly conductive polymer nanocomposites for EMI shielding applications. Mater. Today Commun. 2017, 11, 112–118. [CrossRef] VSachdev, K.; Sharma, S.K.; Tomar, M.; Gupta, V.; Tandona, R.P. EMI shielding of MWCNT/ABS nanocomposites in contrast to graphite/ABS composites and MWCNT/PS nanocomposites. RSC Adv. 2016, 6, 45049–45058. [CrossRef] Jyoti, J.; Basu, S.; Pratap, B.; Dhakate, S.R. Superior mechanical and electrical properties of multiwall carbon nanotube reinforced acrylonitrile butadiene styrene high performance composites. Compos. Part B Eng. 2015, 83, 58–65. [CrossRef] King, J.A.; Pisani, W.A.; Klimek-mcdonald, D.R.; Perger, W.F.; Odegard, G.M. Shielding effectiveness of carbon-filled polycarbonate composites. J. Appl. Polym. Sci. 2015, 132. [CrossRef]
Appl. Sci. 2019, 9, 37
35. 36.
19 of 19
Al-Saleh, M.H.; Saadeh, W.H.; Sundararaj, U. EMI shielding effectiveness of carbon based nanostructured polymeric materials: A comparative study. Carbon 2013, 60, 146–156. [CrossRef] Pourzahedi, L.; Zhai, P.; Isaacs, J.A.A.; Eckelman, M.J.J. Life cycle energy benefits of carbon nanotubes for electromagnetic interference (EMI) shielding applications. J. Clean. Prod. 2017, 142, 1971–1978. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
