Fabrication and Properties of Novel Polymer-Metal

Fabrication and Properties of

Novel Polymer-Metal Composites

Using 3D Printing by

Matthew A. Ryder

A Thesis

Submitted to the Faculty

of the

WORCESTER POLYTECHNIC INSTITUTE

in partial fulfillment of the requirements for the

Degree of Master of Science

in

Mechanical Engineering

April 19, 2017

____________________________

Professor Diana A. Lados (ME-MTE), Advisor

____________________________

Professor Germano S. Iannacchione (PH), Committee Member

____________________________

Professor Amy M. Peterson (CHE), Committee Member

____________________________

Professor Pratap M. Rao (ME), Department Representative

i

Abstract This project investigated the novel fabrication and properties of ABS - stainless steel

composites using 3D printing (fused deposition modeling – FDM). Mechanical and physical

properties of acrylonitrile butadiene styrene (ABS) - 420 stainless steel (SS) composites (with

10, 15, and 23 wt% stainless steel powder additions) were ascertained and compared to those

of the ABS. Tensile testing, dynamic mechanical analysis, modulated differential scanning

calorimetry, and scanning electron microscopy were employed to characterize all

materials/conditions. A new methodology to fabricate the composites was first developed. The

resulting materials were then extruded into composite filaments, which were further used to

print tensile specimens. Controlling printing parameters, deposition direction, and build

orientation were systematically investigated in order to optimize the process (minimize

porosity and enhance homogeneity and interlayer bonding) and improve the mechanical

properties of the resulting tensile specimens. The results demonstrate the feasibility of using

3D printing to create ABS-SS composites with increased functionality, while preserving their

mechanical properties.

ii

Acknowledgements I would to thank the following individuals and groups for their contributions to this

project. Without their tremendous help, it would not have been possible for me to complete

my thesis and to learn what I have in this whirlwind of an academic year. Special thanks to my

advisor, Professor Diana Lados (ME-MTE), for her invaluable guidance and support that ensured

the continuous progress and success of this study, Professors Germano Iannacchione (PH) and

Amy Peterson (CHE) for their insight, observations, and data analysis, Kristen Markuson (ME),

Mila Maynard (ME), and Daniel Braconnier (ME) for their assistance in material preparation,

Anthony D’Amico (CHE) and Xuejian Lyu (MTE) for their help in data acquisition, Roger Steele

(PH) and Russ Lang (CEE) for providing equipment critical to the success of this project, and Ian

Donaldson (GKN Sinter Metals) for providing the stainless steel powders utilized in the study.

iii

Table of Contents Abstract .......................................................................................................................................................... i

Acknowledgements ...................................................................................................................................... ii

List of Figures ................................................................................................................................................ v

List of Tables ................................................................................................................................................ vi

2. Literature Review ...................................................................................................................................... 1

2.1. Introduction ....................................................................................................................................... 1

2.2. Additive Manufacturing ..................................................................................................................... 1

2.3. Fused Deposition Modeling ............................................................................................................... 2

2.4. Polymer – Non-Metallic Composites ................................................................................................. 3

2.5. Polymer – Metal Composites ............................................................................................................. 4

3. Methodology ............................................................................................................................................ 6

3.1. Materials ............................................................................................................................................ 6

3.1.1. Polymer Matrix ........................................................................................................................... 6

3.1.2. Metal Powder Additive ............................................................................................................... 6

3.2. Composite Fabrication ....................................................................................................................... 8

3.2.1. Synthesis ..................................................................................................................................... 8

3.2.1.1. Dissolution Experiments ...................................................................................................... 8

3.2.1.2. Composite Drying ................................................................................................................. 8

3.2.1.3. Mass Production of Composite Solution .............................................................................. 9

3.2.1.4. Homogenizing Composite Solutions .................................................................................... 9

3.2.2. Filament Extrusion .................................................................................................................... 11

3.2.2.1. Feedstock Preparation ....................................................................................................... 11

3.2.2.2. Porosity and Re-Extrusion .................................................................................................. 11

3.2.3. Printing ...................................................................................................................................... 12

3.2.3.1. Tensile Bar Geometry ......................................................................................................... 12

3.2.3.2. Build Direction and Infill Patterns ...................................................................................... 13

3.2.4. Optimization of Print Parameters ............................................................................................. 14

3.2.4.1. First Prints .......................................................................................................................... 14

3.2.4.2. Single Bar Prints ................................................................................................................. 14

3.2.4.3. G-Code Modification for Single Bar Prints ......................................................................... 15

iv

3.2.4.4. Final Protocols .................................................................................................................... 16

3.3. Modulated Differential Scanning Calorimetry and Dynamic Mechanical Analysis.......................... 17

3.4. Tensile Testing ................................................................................................................................. 17

3.5. Fractography Studies ....................................................................................................................... 17

4. Results and Discussion ............................................................................................................................ 18

4.2. Modulated Differential Scanning Calorimetry and Dynamic Mechanical Analysis.......................... 18

4.3. Tensile Properties ............................................................................................................................ 22

4.3.1. Factory ABS and ABS ................................................................................................................. 22

4.3.1.1. Ultimate Tensile Strength of Factory ABS and ABS ............................................................ 22

4.3.1.2. Ductility of Factory ABS and ABS ........................................................................................ 22

4.3.2. ABS-SS Composites ................................................................................................................... 22

4.3.2.1. Ultimate Tensile Strength of ABS-SS Composites .............................................................. 22

4.3.2.2. Ductility of ABS-SS Composites .......................................................................................... 23

4.3.3. Comparisons between ABS and ABS-SS Composites ................................................................ 25

4.4.3.1. Ultimate Tensile Strength Comparisons between ABS and ABS-SS Composites ............... 26

4.4.3.2. Ductility Comparisons between ABS and ABS-SS Composites ........................................... 27

4.4. Fractographic Results and Analysis .................................................................................................. 28

4.4.1. Fracture Surface Comparison of Vertical and Horizontal Bars ................................................. 28

4.4.2. Raster Angle Impact on Fracture Surfaces ................................................................................ 29

4.4.3. Interfacial Debonding and Toughening Mechanisms ............................................................... 29

4.5. Other Functional Properties of the ABS-SS Composites .................................................................. 36

5. Conclusions ............................................................................................................................................. 37

6. Future Work ............................................................................................................................................ 38

Appendices ................................................................................................................................................. 39

A. Slic3r Parameters ................................................................................................................................ 39

B. Macro G-Code Editor Script ................................................................................................................ 42

References .................................................................................................................................................. 45

v

List of Figures Figure 1. Differentiation between additive manufacturing processes. [6] ........................................................ 2

Figure 2. FDM infill deposition protocol.[12] ..................................................................................................... 3

Figure 3. Acrylonitrile (C3H3N), 1,3-butadiene (C4H6), and styrene (C8H8).[14] .................................................. 6

Figure 4. Stainless steel powder characterization. .......................................................................................... 7

Figure 5. Stainless steel powder particle size distribution. .............................................................................. 7

Figure 6. Aggregation of particles in vial after mixing. .................................................................................... 8

Figure 7. Thick skin formation in vial. .............................................................................................................. 9

Figure 8. ABS-23%SS air dried sheet. ............................................................................................................. 10

Figure 9. Homogeneous distribution of stainless steel particles in ABS-23%SS sheet. ................................. 10

Figure 10. ABS-23%SS composite sheets, strips, and feedstock. ................................................................... 11

Figure 11. ASTM D638 type V tensile specimen (dim in mm). ....................................................................... 12

Figure 12. Hyrel System 30M 3D Printer........................................................................................................ 12

Figure 13. Layer cross-sections for all build direction and raster angle combinations.................................. 13

Figure 14. Five bar line (a) and pentagon (b) print layouts. ........................................................................... 14

Figure 15. Failure outside of gage length in 45/-45H tensile specimens. ...................................................... 16

Figure 16. First (a) and second (b) heating and cooling scans of ABS. ........................................................... 18

Figure 17. MDSC heating scans. ..................................................................................................................... 19

Figure 18. MDSC cooling scans. ...................................................................................................................... 20

Figure 19. Dynamic Mechanical Analysis results. .......................................................................................... 21

Figure 20. Delamination during tensile test. .................................................................................................. 23

Figure 21. Insufficient infill density. ............................................................................................................... 24

Figure 22. Maximum stress - strain curves for all materials. ......................................................................... 26

Figure 23. UTS of all build orientations and raster angles (error bars given for cases where SD > 1%). ....... 27

Figure 24. Ductility of all build orientations and raster angles (error bars given where SD > 1%). ............... 28

Figure 25. Ductility pattern on gage of fractured specimen. ......................................................................... 28

Figure 26. Incomplete particle - matrix adhesion and evidence of rubber toughening in the surrounding

polymer matrix. .............................................................................................................................................. 30

Figure 27. SEM images of Factory ABS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications. 31

Figure 28. SEM images of ABS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications. ............. 32

Figure 29. SEM images of ABS-10%SS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications. . 33



Figure 32. ABS-10%SS composite tensile bar attracted by a magnet. ........................................................... 36

vi

List of Tables Table 1. Mechanical properties of ABS and reinforced ABS ......................................................................... 5

Table 2. Stainless steel chemical composition .............................................................................................. 7

Table 3. MDSC heating scans ...................................................................................................................... 20

Table 4. MDSC cooling scans ....................................................................................................................... 20

Table 5. C-Factor values for tested materials ............................................................................................. 21

Table 6. Mechanical properties for all studied materials in all orientations .............................................. 25

Table 7. Functionalized stainless steel heating scan ................................................................................... 38

Table 8. Functionalized stainless steel cooling scan ................................................................................... 38

1

2. Literature Review 2.1. Introduction

Fused Deposition Modeling (FDM) is the most prevalent style of polymer additive

manufacturing. While the niche for FDM – 3D printed parts have traditionally been for creating

rapid prototypes, the flexibility and low-waste nature of this technology places it at the

forefront of advanced manufacturing. The largest drawback of parts produced via FDM is the

reduction in mechanical properties relative to traditionally manufactured parts. These

reductions are the result of the layer upon layer method through which additively

manufactured parts are produced. In an effort to mitigate anisotropy in mechanical properties,

studies have been performed assessing the viability of composite materials as alternatives to

pure polymers for FDM. The objective of these studies is to determine a material that can be

used for FDM that does not result in losses in mechanical properties. This would allow for the

production of parts with equivalent mechanical properties to their traditionally manufactured

counterparts. While non-metal additives have failed to provide such a solution, this study aims

to add to the sparse amount of literature focusing on the potential for metal particle additives

to enhance the mechanical properties of parts produced through FDM, while adding beneficial

functional characteristics to the materials.

2.2. Additive Manufacturing Stereolithography was developed in 1984 by Chuck Hull of 3D Systems (U.S. Patent

4575330 - March 11, 1986) and can be considered the advent of commercial additive

manufacturing (AM).[1] In the following years, the field has grown in scope to encompass a wide

array of rapid prototyping technologies. Being hailed as the 3rd Industrial Revolution,[2,3] AM

allows for the creation of not only rapidly prototyped parts, but also functional components, in

a short time span. Other benefits of AM include efficiency in material use, lack of a need for

auxiliary resources, and part flexibility. Unused material can easily be recycled with minimal

processing, and building parts from layers omits the need for fixtures, clamps, dies, or tooling,

and allows for the creation of single part assemblies.[4] AM refers to a process in which

components are digitally designed in CAD (computer aided design) programs, exported as “STL”

(stereolithography) files. These files are imported into a host software and sliced into layers

which are converted into G-Code commands. The parts are then built from an input material as

2

opposed to being machined using a subtractive process, which would result in significantly

higher waste. For example, laser material deposition (LMD) can achieve material savings of

approximately 60%, along with time savings of 30% relative to traditional 5-axis milling

processes for some aerospace components, such as bladed discs.[5] LMD also prevents the

generation of unrecyclable waste chips, instead using powders where unused powder is 95-98%

recyclable.[5]

Additive manufacturing processes can be categorized into three main groups based on

the material that is utilized, where processes are categorized into liquid-based, solid-based, and

powder-based additive manufacturing methods, as shown in Figure 1.[6] Amongst these

processes, FDM is the most commonly used polymer rapid prototyping process and was the

focus of this study.

Figure 1. Differentiation between additive manufacturing processes. [6]

2.3. Fused Deposition Modeling FDM was developed by Stratysis founder Scott Crump, U.S. Patent 5121329, and

outlined in his 1992 paper as a “single step operation” that “safely generates 3D prototypes

from CAD software data [which] reduces the new product development cycle and allows

validation of part design and production tooling.”[7] In FDM, molten filament is extruded out of

a nozzle and laid down in thin bead roads. As the print head lays down the infill in the XY plane,

adjacent extruded bead roads cool, bond to one another, and the layer solidifies (Figure 2). The

3

printer then moves to the next layer, building the part in the Z direction. A consequence of

building a part in layers is the resulting anisotropy in mechanical properties, as has been widely

reported.[8-11]

Figure 2. FDM infill deposition protocol.[12]

As compared to injection molded parts, which have no layer interfaces, parts created by

FDM have reduced mechanical properties where the degree of reduction is contingent on the

infill pattern, print orientation, print parameters, and degree of interlayer diffusion. Tensile

bars created by FDM have higher properties if they are printed with the long axis of the tensile

bar oriented horizontally (H), where layers are built parallel to the direction of the applied

tensile force, versus when the bars are printed vertically (V), and layers are built perpendicular

to the direction of the applied tensile force.[8-11] In both cases, parts created with FDM have

lower mechanical properties relative to the injection molded control samples, and optimum

print parameters result in the creation of tensile bars with only up to 73% of the ultimate

tensile strength (UTS) of an injection molded bar formed from the same material.[9]

2.4. Polymer – Non-Metallic Composites In an effort to mitigate the print orientation-induced anisotropy in the mechanical

properties of parts created with FDM, studies have been conducted to investigate material

solutions to this manufacturing problem. Many studies have attempted to improve the

properties of 3D printed materials by utilizing composite filaments, where fibers (carbon, glass,

4

jute) or ceramics (metal oxides) are the most commonly studied additives. As observed in Table

1, when ceramics are incorporated into polymer matrices, mechanical properties are typically

reduced, even at small wt% addition. The one reported exception where an additive did not

decrease the ductility of a composite material relative to ABS, a 2 wt% addition of MayaCrom

Blue, did not statistically increase the ductility, but reduced the UTS by approximately 50%.[13]

In terms of fiber reinforcements, fibers can offer improved or reduced UTS values relative to

ABS depending on the type and orientation of the reinforcement (e.g. UTS of ABS+5 wt% Jute

fibers is ~24 MPa compared to ~34 MPa for ABS[11] while the UTS of ABS+30 wt% glass fibers is

~60 MPa compared to ~40 MPa for ABS).[13] Increases in UTS using fiber reinforcements come

at the expense of ductility. However, for ABS, high fracture toughness is a critical property of

the material, and a loss in ductility reduces the applicability of ABS – fiber composites. In

addition, ceramic and fiber reinforcements increase print anisotropy (e.g. ABS shows a 48%

difference between orientations, while ABS+5 wt% SrTiO3 has a 72% difference),[11] which fails

to solve the problem initially attempting to be addressed with their addition.

2.5. Polymer – Metal Composites Metal particle additions to polymer matrices result is very low mechanical properties

due to the difficulties associated with achieving strong adhesion between the metal and the

polymer matrix. As a result, these materials are not extensively researched in the literature

with respect to seeking improvement of mechanical properties relative to ABS. Instead, metal

additives tend to be investigated in efforts to improve thermal conductivity and radiation

shielding. It has often been reported that increasing the wt% of metal particle additions in

polymer results in reduced mechanical properties of the metal composites due to void

formation caused by poor adhesion between particles and the surrounding polymer. Polymer –

metal composites do not offer increased mechanical properties (compared to ABS) in any

condition, while fiber reinforcements have been demonstrated to do so at least with respect to

improving UTS in specific fiber orientations. Table 1 highlights the effect of metal particle

additives on the mechanical properties of printed tensile specimens, relative to ABS control

samples. Despite the initial low reduction in UTS, values for UTS decrease sharply as the wt% of

metal additives is increased. In addition, the ductility also decreases with metal additions, such

as Cu or Fe particles, as shown in Table 1.

5

Table 1. Mechanical properties of ABS and reinforced ABS

Author Orientation Additive Elongation [%] UTS [MPa]

Catrell et al.[15] 24.4 +/- 0.5

Riddick et al.[16] 13.61+/-1.13

Perez et al.[17] 14.1

Torrado et al.[11] 2.08 17.73

Catrell et al.[15] 25.8 +/- 0.3

Riddick et al.[16] 19.8+/-2.22

Catrell et al.[15] 29.1 +/- 0.3

Riddick et al.[16] 25.69+/-1.75

Perez et al.[17] 28.5

Torrado et al.[11] 8.64 33.96

Chockalingam et al.[18] 33.94

Catrell et al.[15] 28.8 +/- 0.2

Riddick et al.[16] 27.77+/-.92

Ziemian et al.[19] 16.90+/-0.09

Hwang et al.[20] 8.3 45.7

5% Jute Fibers 25.9

5% TiO2 32.2

5% Thermoplastic Elastomer 24

5% JUTE FIBER 4.25 24.25

2% MayaCrom®Blue 8.86 17.31

5% TiO2 3.77 32.9

2% ZnO 6.32 20.7

5% SrTiO3 5.56 21.6

5% Al2O3 2.94 28.8

5% Jute Fibers 12.9

5% TiO2 18.4

5% Thermoplastic Elastomer 9.1

5% JUTE FIBER 1.55 8.63

2% MayaCrom®Blue 2.02 7.79

5% TiO2 1.61 16.67

2% ZnO 1.07 7.41

5% SrTiO3 1.06 5.95

5% Al2O3 1.6 12.14

10% Cu 5.4 42

30% Cu 3.3 26.5

10% Fe 6.1 43.4

30% Fe 5.1 40.6

40% Fe 4.5 36.2

Pure ABS

Non-metal Additives

45/-45 V

45/-45 H

0/90 H

0/90 V

0/90 H

0/90 V

Perez et al.[17]

Metal Additives

Perez et al.[17]

Torrado et al.[11]

Torrado et al.[11]

Hwang et al.[20]

6

3. Methodology 3.1. Materials 3.1.1. Polymer Matrix

The ABS material is a copolymer of acrylonitrile, butadiene, and styrene: (C8H8)x •

(C4H6)y • (C3H3N)z, with chemical structures expanded in Figure 3. In the literature, ABS and

polylacticacid (PLA): (C3H4O2)n are the most commonly used materials for FDM. Compared to

PLA, ABS has greater heat resistance (up to ~105°C for ABS compared to ~60°C for PLA) and

impact resistance, and has superior machinability. Applications for ABS include Lego bricks,

helmets, white water canoes and kayaks, and luggage and protective carrying cases. Due to

these advantages, ABS was selected as the polymer matrix for the ABS-Stainless steel (ABS-SS)

composites in this study. Large spools of factory manufactured ABS were used for the

optimization of the 3D printing process. ABS pellets (Raw ABS) were further used for the in-

house fabrication of ABS filaments (for comparison with Factory ABS filaments), as well as for

the matrix of fabricated ABS-SS composites.

Figure 3. Acrylonitrile (C3H3N), 1,3-butadiene (C4H6), and styrene (C8H8).[14]

3.1.2. Metal Powder Additive

Stainless Steel powders (420 stainless steel, apparent density: 2.83 g/cm3) from

Hoeganaes Corporation, provided by GKN Sinter Metals, were used for the metallic particulate

additions to the ABS polymer matrix. The non-spherical powders were chosen for the possibility

of improved bonding to the polymer matrix as compared to a spherical particle additive due to

the surface roughness of the particles resulting from their irregular morphology. The particles

were characterized with the SEM as observed in Figure 4, and the average particle dimensions

7

parallel to the long axis (~30 mm) are provided in the data sheet in Figure 5. Due to the

martensitic, high carbon composition of 420 Stainless Steel (chemical composition of this batch

given in Table 2), the resulting composite materials will have magnetic properties, which would

not be the case for an austenitic stainless steel.

Table 2. Stainless steel chemical composition

Figure 4. Stainless steel powder characterization.

Figure 5. Stainless steel powder particle size distribution.

Chemical Composition wt%

Carbon 0.29

Oxygen 0.33

Sulfur 0.010

Nitrogen 0.024

Chromium 14.0

Silicon 0.62

Nickel 0.07

Iron Bal 84.66

8

3.2. Composite Fabrication 3.2.1. Synthesis

3.2.1.1. Dissolution Experiments

In preliminary tests to reduce the viscosity of ABS through use of a solvent, small

amounts of raw ABS pellets, ranging from 1-3 g, were mixed with 1-3 ml of acetone. Acetone

(C3H6O) was selected as the solvent for this study, and due to its high vapor pressure, it readily

evaporated at room temperature, resulting in significantly lower drying times than would have

been possible with less volatile solvents. Acetone is compatible with ABS and dissolution of the

raw ABS pellets began immediately at room temperature. After letting the pellets dissolve for

24 hours, the vials were stirred and left to sit for another 24 hours. One gram of 420 stainless

steel powder was then added to these ABS – acetone solutions, and the solution was

homogenized by stirring. It was determined that at the viscosity at which the stainless steel

particles were able to be stirred into the solution and remain suspended without sinking to the

bottom, a maximum of ~23 wt% stainless steel could be added to the ABS – acetone solution.

Higher weight percentage additions of stainless steel resulted in visible aggregation, Figure 6.

Figure 6. Aggregation of particles in vial after mixing.

3.2.1.2. Composite Drying

Having determined an ideal formulation for synthesizing ABS-stainless steel solutions

(ABS-SS), drying experiments were performed, where batches of the finalized solution were

created in test tubes. When leaving the test tubes out to dry, it was observed that a thick dried

skin would form on the surface of the solution, Figure 7, preventing the acetone below this

solidified barrier from evaporating and escaping to the surrounding air. To address this

problem, batches of solution were prepared and poured out onto glass plates to dry, which

prevented acetone from being trapped under a thick dried skin in a test tube. As the surface of

the solution dried too quickly to easily pour from the test tubes, it was evident that larger

batches of less viscous solution would be needed.

9

Figure 7. Thick skin formation in vial.

3.2.1.3. Mass Production of Composite Solution

The surface of the optimized solution, where the stainless steel would remain

suspended in the ABS – Acetone solution, solidified too quickly to pour out of test tubes into

sheets for drying. Less viscous solutions that could be poured out into sheets would result in

the stainless steel settling to the bottom, though this settling would occur over the span of

minutes or hours rather than seconds. Should the solution be homogenized immediately before

being poured into sheets, the sheets of drying composite would have uniform particle

distributions. In order to homogeneously mix larger batches of solution, jars were obtained that

permitted for the mixing of approximately 100 ml of solution per batch. ABS and stainless steel

were added to acetone and allowed to sit for 2 days before being mixed as described in the

next section.

3.2.1.4. Homogenizing Composite Solutions

A drill was obtained due to its high torque output, and a 3.175 mm thick strip of 150 mm

by 25 mm aluminum was twisted into a helix and a tang snipped into the end such that it would

be able to be accepted by the drill chuck. The drill – blender was tested with a jar of water, with

a hole cut into the top to provide access to the drill and to double as a potential splash guard,

to observe the mixing pattern. When driven clockwise, the blender splashed water up onto the

walls, sides, and top of the jar. However, when driven in reverse, the drill – blender was still

10

easily able to churn the water, but with no backsplash onto the sides or top of the jar, or onto

any of the blender attachment above the waterline. The drill – blender was then tested on a jar

of unmixed ABS-SS composite, and the solution was immediately mixed into a uniform grey

color. The larger batch of solution permitted for a thin, circular sheet approximately 2 mm high

by roughly 200 mm in diameter to be poured out in under 5 seconds, with only a negligible

amount of solution drying too rapidly in the jar to be poured out. These ABS-SS composite

sheets were left to dry, and 48 hours later, the acetone had appeared to have evaporated off,

leaving small bubbles in the surface of a leathery/brittle ABS-SS composite sheet, as shown in

Figure 8.

Figure 8. ABS-23%SS air dried sheet.

It was crucial to verify the homogeneity of the ABS-SS dispersions to validate the mixing

process employed in this study. Figure 9 clearly depicts a uniform particle distribution and is

representative of the effectiveness of the mixing process in evenly distributing the

microparticles in a batch of ABS-23%SS composite. The rationale for this conclusion is that if the

maximum amount of stainless steel does not result in aggregation, reduced concentrations will

not result in aggregation either.

Figure 9. Homogeneous distribution of stainless steel particles in ABS-23%SS sheet.

11

3.2.2. Filament Extrusion

3.2.2.1. Feedstock Preparation

ABS-SS composite sheets were sliced into 1 mm thick strips by using a large paper cutter

and were subsequently cross cut into small pieces with an approximate cross-section of ~2.0 x

2.5 mm2, shown in Figure 10(a). The size of these squares correlated closely with the

dimensions of a measured raw ABS pellet, Figure 10(b), which had been specifically sized to be

accepted by the Filabot extruder used in this project. The composite feedstock was then placed

in an oven at 110°C for 24 hours with a pan of desiccant to further drive off any residual

acetone. It was observed that should this step be omitted, the composite feedstock would melt

together making extrusion impossible, and further MDSC work verified the effectiveness of this

second drying phase. The resulting ABS-SS composite feedstock pellets were removed from the

oven and extruded into filament.

(a) (b)

Figure 10. ABS-23%SS composite sheets, strips, and feedstock.

3.2.2.2. Porosity and Re-Extrusion

Attempts at spooling the extruded ABS-SS composite filament revealed that the filament

was extremely brittle, and the cross-section of the broken filament indicated that there was

porosity in the filament. Inspection under an optical microscope confirmed this conclusion, and

the extruded ABS-SS composite filament was chopped up again and reintroduced into the

extruder. The resulting filament spooled smoothly, and the cross-section showed no air

bubbles. ABS-SS composite filament was manufactured at 5, 10, 15, and 23 wt% for this study.

As the filament diameter that was able to be accepted by the printer was tested to be 1.70 +/-

0.4 mm, extruding as much filament as possible for each of the differing weight percentages

12

played a critical role in the success of this study. During the filament extrusion process, it

became evident how many variables contributed to variation in filament diameter, with the

most impactful being spooler RPM and the distance between the spooler and extruder. As the

drum spooler speed control was dial operated, and no single speed seemed to satisfy all of the

materials, repeatability in extrusion conditions were difficult to maintain, however, the final

diameters of all extruded filament was ensured to remain within the 1.70 +/- 0.4 mm

specification for consistency.

3.2.3. Printing

3.2.3.1. Tensile Bar Geometry

The largest limitations of this project were mass production of filament and cycle times

for tensile bar prints. With these constraints in mind, the ASTM D638 Type V tensile bar

(drawing included in Figure 11) was selected, as it would use a minimal amount of material. In

order to 3D print this geometry, the corresponding CAD model was built in Solidworks,

exported as an STL file, and uploaded to Repitrel (a version of Repetier Host) installed on the

Hyrel System 30M printer that was used in this study (Figure 12). In Repitrel, the STL file could

be reoriented and repositioned on the print bed, and after the STL file had been suitably placed

on the virtual print bed, the program Slic3r, which is integrated into the Repetrel interface, was

able to “slice” the STL file into layers of G-Code that are readable by the printer. Slicing

parameters are included in Appendix A.

Figure 11. ASTM D638 type V tensile specimen (dim in mm). Figure 12. Hyrel System 30M 3D Printer.

13

3.2.3.2. Build Direction and Infill Patterns

From the literature it is evident that for a single material, there are a wide array of

tensile properties that can be obtained solely by varying the raster orientation and build

direction of tensile bars (Table 1). Raster orientation refers to the infill (the area contained by

the single perimeter wall) pattern of each layer of the tensile bar, so a 45/-45 raster orientation

would mean that the first layer of infill would be laid down rotated 45° relative to the long axis

of the bar, and the second layer would be oriented -45° relative to the long axis of the bar.

Every other layer is rotated 90° so that there is no anisotropy in the bars with respect to the XY

plane, with the only anisotropy resulting from building upwards in the Z direction due to the

layers produced in FDM. Build direction refers to the direction that the long axis of the bar is

printed in, where horizontal bars are oriented with the long axis flat on the print bed, and

vertical bars are oriented with the long axis extending vertically up from the print bed. The

resulting orientations mean that horizontally oriented bars have 0 layers perpendicular to the

long axis, while vertically printed bars have all of their layers perpendicular to the long axis. This

is significant in that layer interfaces perpendicular to the long axis, the direction of the applied

force in tensile testing, are weak points in the bars. In this study, all four cases: 45/-45V and H,

0/90V and H bars were tested, and the cross-sections and infill patterns can be observed in

Figure 13.

Figure 13. Layer cross-sections for all build direction and raster angle combinations.

14

3.2.4. Optimization of Print Parameters

3.2.4.1. First Prints

Due to the large volume of prints that were needed for print optimization, a spool of

factory – made ABS was initially used, and while the parameters for ABS-SS composites prints

would likely deviate from this best print protocol, an idealized Factory ABS printing protocol

would at least provide a starting point to begin optimization of ABS-SS prints. The first batches

of Factory ABS prints were five bar prints, where five tensile bars were printed simultaneously

for each of the 4 configuration patterns tested. The bars were initially positioned in a straight

line 15 mm apart facing the front of the printer, Figure 14(a), however, after noticing that

material tended to leak out while transitioning from bar 5 to bar 1 (over a 150 mm travel), the

bars were realigned into a pentagon orientation, Figure 14(b), such that there would be a

consistent travel time between all of the bars. The resulting properties were significantly lower

than the values obtained in the literature for ABS, especially for the vertical prints, though

there was no porosity in the bars, and the printed bars were dimensionally accurate relative to

the initial imported STL file.

(a) (b)

Figure 14. Five bar line (a) and pentagon (b) print layouts.

3.2.4.2. Single Bar Prints

As ABS-SS composite production and bar optimization were simultaneously progressing

components of this study, it was determined that due to the tolerances of the extruder, under

absolutely ideal conditions rated to +/- 0.5 mm with ABS, it was impossible to extrude 4

consecutive meters of filament that maintained the 1.70 +/- 0.4 mm specification that had been

laid out based on printer capabilities. With that limitation in mind, being able to print a tensile

bar from even the smallest length of suitable filament became a focal point of print

optimization. Instead of printing in batches of five bars, where a single layer would be printed

15

for bar 1, then bar 2, etc. before returning to bar 1 to print the second layer, single bar prints

would be more consistent. Preliminary 0/90V (the condition that yielded the lowest tensile

properties) single bar prints with Factory ABS resulted in bars with low dimensional tolerance

adherence. This was particularly evident in the gage of the bar, where an oblique gage length

formed as opposed to one with a square cross-sectional area. It was observed that the layers

did not appear to be fully solidified before the following layer was printed on top of it, resulting

in large dimensional inaccuracies. With this problem in mind, a method of increasing the

elapsed time between adjacently printed layers for single bar prints was needed, to allow the

previously printed layer to cool.

3.2.4.3. G-Code Modification for Single Bar Prints

In an effort to make the single bar 0/90V prints take the same amount of time as the

0/90V five bar prints the Slic3r program was modified to include a skirt around the bar. A low 2.5

mm/s print speed for the skirt would increase the print time of a single bar print from about 19

minutes and 30 secs to 1 hour and 28 minutes. To simulate the print head moving away from the

first bar in the print sequence and laying down filament on the other 4 bars, as would be the

case in a five bar print, the G-Code was imported into Microsoft Excel where a macro was

written which added 100 mm to all of the X coordinate values in lines of G-Code including the

word “skirt.” This meant that the print head would move away from the bar before extruding

the skirt, which mimics the time that would have been spent on the other 4 bars in a five bar

print, before returning to print the next layer on the single bar. A batch of single bar 0/90V

prints were built and tested and the tensile properties were even lower than the five bar prints,

albeit while being dimensionally accurate. Interestingly, when the macro was not applied and a

skirt was allowed to be built around the bar, the tensile properties were significantly higher than

that of the five bar prints. An indication as to why this was the case revealed itself by a visual

inspection of the fracture surfaces. The clear raster pattern visible on the bars with the macro

suggested that there was not as much interlayer diffusion as there was in the bars where the

macro was not applied. With no macro, the skirt was built around the bar and the fracture

surfaces revealed no clear raster pattern with visual inspection. These observations suggested

that the close proximity of the print head to the bar resulted in increased temperatures, which

facilitated interlayer diffusion which in turn increased the tensile strength of the bar relative to

16

five bar prints, yet while still allowing enough time for the previous layer to cool to ensure

dimensional accuracy. In the interest of pursuing this breakthrough while still saving material, a

new macro was written which changed all of the “G1” instances in lines of G-Code including the

word “skirt” to “G0” which meant that the print head would be told to go to these points, but

not to extrude any filament while doing so. The macro would also be easy to modify such that

the time away from the bar could be easily tuned. There was a clear balance between

dimensional accuracy and print cooling rate, where the strongest bars were the ones that were

printed as quickly as possible, relative to time elapsed between the printing of adjacent layers,

while still maintaining dimensional accuracy. The introduction of this macro resulted in single

bar 0/90V prints that took the same amount of time to print as 0/90V five bar prints, yet used

only 1/5 of the required filament. The total elapsed print time for a 0/90V bar was further

reduced to 53 minutes, where the bar was still dimensionally accurate, and offered tensile

properties comparable to recorded values in the literature.

3.2.4.4. Final Protocols

The final protocol for the 0/90V and 45/-45V prints had now been determined, and

0/90H single bar prints were built and tested, where every single tensile result reported in this

study was obtained from a single bar print, and when observing that the interlayer diffusion

appeared to be low, print speed was increased to the maximum that the Hyrel System 30M

could manage while still maintaining dimensional accuracy, where prints of 45/-45H tensile bars

would fail in the same area outside the gage length for print speeds greater than 12.5 mm/s as

observed in Figure 15. With the final protocols for 0/90V, 45/-45V, 0/90H, and 45/-45H bars

determined, and baseline tests performed for all of these orientations for Factory ABS and ABS,

the stage was set for ABS-SS composite tests.

Figure 15. Failure outside of gage length in 45/-45H tensile specimens.

17

3.3. Modulated Differential Scanning Calorimetry and Dynamic Mechanical Analysis Modulated Differential Scanning Calorimetry (MDSC) was conducted to assess the effect of the

stainless steel additives on the thermal properties and glass transition temperature of the material.

Samples were cut from air dried processed ABS and air dried ABS-SS composites and a raw ABS pellet

was used as the reference material. Approximately 10 mg of material was used in a crimp sealed

aluminum DSC pan. Scans were performed using a TA Instruments MDSC 2920 at a ramp rate of

+/- 0.5K/minute and a modulation amplitude of 0.5K with a 60 second period, which ensured high

resolution data collection. A 30 minute dwell was included at the extrema of the test conditions, 10°C

and 250°C, to guarantee complete phase conversion. These maximum and minimum temperatures

were selected as they provide a complete window on the full range of temperatures the materials

utilized in this study would be subjected to, with the low temperature being room temperature, and

the highest temperature being 237°C during the printing process. A heat then cool scan cycle was

performed repeatedly until the scans were reproducible. This usually took 2-3 scan cycles.

Dynamic Mechanical Analysis (DMA) was conducted to analyze the viscoelastic behavior of the

ABS and ABS-SS composites. After setting the base offset for the DMA, the sample was loaded. Sample

length for all samples was 0.625 mm as determined by the separation of the grips. DMA was then

conducted from -125°C to 120°C at 2°C/min with a target amplitude of 1 mm and a 1 Hz frequency.

3.4. Tensile Testing Tensile specimens were printed in accordance with the above procedure and in

compliance with the ASTM D638 type V geometry specification. Tests were performed on an

Instron 5567A at room temperature and a constant strain rate of 1 mm/min. A minimum of three

specimens were tested for each condition, and every data point reported is a single bar print.

3.5. Fractography Studies Fracture surfaces of the tensile specimens were observed under the SEM to characterize

the mode of failure and to allow for greater insight into the effect of the stainless steel particles

on the fracture mechanics. Specimens were sliced parallel to the fracture plane with a razor

blade, and the fracture surface was attached to a sample holder with adhesive tape. It was

necessary to sputter coat the samples with gold due to the non-conductive nature of ABS, and

all samples were coated for consistency. A voltage of 5kV was utilized, and fracture surface

images were acquired at 23x, 150x, and 330x magnifications.

18

4. Results and Discussion 4.2. Modulated Differential Scanning Calorimetry and Dynamic Mechanical Analysis

The MDSC tests indicated that the drying process successfully evaporated all of the

acetone that had been used to dissolve the ABS plastic matrix. Initial MDSC scans yielded non-

reproducible data, as large, random jumps occurred during the first heating run due to the

presence of trapped acetone that had failed to evaporate. These initial runs were conducted

with specimens of processed ABS and ABS-SS composites that had been sliced after air drying,

but had yet to be dried in an oven. The confirmation of the presence of trapped acetone

validated the use of a second drying phase at 110°C, with desiccant, in an oven. Another factor

resulting in the non-reproducibility of the initial heating scan is the flowing of the sample as it

equilibrated above the glass transition temperature (Tg). Secondary heating runs yielded

reproducible data however, clearly depicting that the acetone solvent had been completely

driven off and that the material had completely equilibrated. This is indicated by the nearly

identical secondary heating and cooling runs, Figure 16(b), and lack of random steps observed

in initial heating runs, Figure 16(a). Heating and cooling scans were conducted until consecutive

heating runs were completely reproducible.

(a) (b)

Figure 16. First (a) and second (b) heating and cooling scans of ABS.

The MDSC tests also provided new insight regarding the relationships between the

thermal characteristics of the materials examined in this study, and the effect that the stainless

steel particles had on the heat capacity of the composites. The green curve, CBG, represents the

background specific heat used to determine ΔCp, where ΔCp = CP - CBG. As presented in Figure

17, the heating runs suggest that with up to a 10 wt% addition of stainless steel to the ABS

19

polymer matrix, the ΔCp step at the glass transition temperature increases, demonstrating that

the stainless steel additions resulted in the creation of a composite material with a more stable

glassy phase as compared to ABS. The curves for 5 and 10 wt% loadings of stainless steel all

follow similar trends, depicting more stable glass transitions than ABS, while the 23 wt%

loading results in a less stable glassy phase due to the smaller ΔCp step at Tg. It was is also

interesting to note that the ΔCp actually decreased at Tg initially for the ABS-23%SS composite,

indicating that there was a change in thermal conductivity of the material during the phase

change from glass to molten rubber. Relative to the processed ABS, the addition of 5 wt%

stainless steel resulted in a reduction in Tg, while additions of 10, 15, and 23 wt% resulted in an

increased Tg (values reported in Tables 3 and 4 for heating and cooling scans). Due to the

presence of the red pigment additive to Factory ABS, the glass transition temperature was

lower than that of all of the other materials tested.

The MDSC scans further allowed verification of the homogeneity of the synthesized ABS-

SS composites regarding distribution of the stainless steel particles as well, as the smooth ΔCp

vs °C curves indicate a single, sharp phase transition at Tg. Should the material be non-

homogeneous, regions would reach their respective glass transition temperatures at differing

values, resulting in multiple steps at Tg, rather than the single large step as the data indicate.

Figure 17. MDSC heating scans.

20

Table 3. MDSC heating scans

Table 4. MDSC cooling scans

Sample Mass [mg] Scan Tmin [°C] Tg [°C] Tmax [°C]

Raw ABS

Pellet11.0 H2 103.7 106.8 109.7

ABS 11.3 H2 106.3 108.4 110.8

Factory ABS 11.0 H2 97.3 101.7 105.1

ABS-5%SS 11.0 H3 103.6 106.6 110.6

ABS-10%SS 11.7 H2 106.9 109.2 110.6

ABS-15%SS 11.1 H3 105.4 109.5 110.9

ABS-23%SS 11.0 H3 108.7 109.6 112.4

HEATING

Sample Mass [mg] Scan Tmin [°C] Tg [°C] Tmax [°C]

Raw ABS

Pellet11.0 C2 103.1 107.7 110.2

ABS 11.3 C2 106.1 108.5 110.1

Factory ABS 11.0 C2 97.1 100.9 105.2

ABS-5%SS 11.0 C3 102.8 106.9 110.3

ABS-10%SS 11.7 C2 106.4 109.4 110.9

ABS-15%SS 11.0 C3 105.7 108.1 110.7

ABS-23%SS 11.0 C3 106.8 108.9 111.3

COOLING

Figure 18. MDSC cooling scans.

21

The DMA results indicate that as the temperature increased, the storage modulus

decreased. The white ABS, the ABS-5%SS, and the ABS-10%SS had very similar E’ values at

lower temperatures. The ABS-23%SS had the lowest E’ value which indicates that it had the

lowest stiffness out of all of the materials, while the Factory ABS had the highest stiffness.

Figure 19 also establishes the glass transition temperatures for the materials. It is interesting to

note that the ABS and the ABS-10%SS deviate from the Factory ABS, ABS-5%SS and ABS-23%SS.

This graph also provides data to calculate the C-factor. This number indicates how the

fillers affect the composites by comparing the moduli in the glassy and rubbery regions of the

graph. If the number is higher, this signifies that the filler is less effective The equation to

determine the C-factor is: C = (E’g / E’r)composite / (E’g / E’r). E’g and E’r are the storage

modulus values in the glassy region (90°C) and the rubbery region (120°C) respectively. The

material used as the reference material was ABS. Table 5 contains the values for the C-factor.

Table 5. C-Factor values for tested materials

Figure 19. Dynamic Mechanical Analysis results.

Material C-Factor

Factory ABS 2.37

ABS-5%SS 1.74

ABS-10%SS 0.88

ABS-23%SS 2.30

22

4.3. Tensile Properties 4.3.1. Factory ABS and ABS

4.3.1.1. Ultimate Tensile Strength of Factory ABS and ABS

Ultimate tensile strength values for Factory ABS and ABS (Table 6) were similar in all build

directions and raster orientation combinations. The average UTS values for Factory ABS and ABS

in 45/-45 vertical prints were slightly higher than were the values for 0/90 vertical prints (31.48

MPa vs 31.34 MPa for Factory ABS and 30.99 MPa vs 30.43 MPa for ABS), while in horizontal

prints the values for 45/-45 prints were higher than those for 0/90 prints for Factory ABS but not

ABS (38.75 MPa vs 37.86 MPa for Factory ABS and 40.05 MPa vs 40.47 MPa for ABS).

4.3.1.2. Ductility of Factory ABS and ABS

In a similar pattern relative to the relationships in UTS values, ductility values for Factory

ABS and ABS (Table 6) were also similar in all build directions and raster orientation

combinations with the exception of 45/-45 horizontal tensile bars. The average ductility values

for Factory ABS and ABS in 45/-45 vertical prints were slightly higher than were the values for

0/90 prints (8.82% vs 7.75% for Factory ABS and 9.24% vs 7.74% for ABS), while in horizontal

prints the values for 45/-45 prints were higher than those for 0/90 prints for Factory ABS but

not ABS (31.18% vs 29.69% for Factory ABS and 26.98% vs 28.16% for ABS). It is interesting to

note that while the UTS values for horizontal bars were ~25% higher than those for vertical bars

printed with the same conditions, the ductility for horizontal bars was roughly quadruple

compared with their vertical counterparts (with the exception of the ABS 45/-45 horizontal

print which resulted in a ductility just under triple that of its vertical counterpart).

4.3.2. ABS-SS Composites

4.3.2.1. Ultimate Tensile Strength of ABS-SS Composites

The UTS values for ABS-10%SS and ABS-15%SS composites were similar in all printing

conditions and orientations, while the UTS values for ABS-23%SS tensile bars was greatly

reduced, only offering around 60% of the UTS for vertical bars and 70% of the UTS for

horizontal bars relative to the values of the other two composites. There were no distinct

trends regarding whether the 45/-45 or 0/90 raster orientation resulted in higher UTS values

for composite tensile bars, however it was clear that horizontal bars tended to have increased

UTS values relative to vertical bars in all cases.

23

4.3.2.2. Ductility of ABS-SS Composites

The ductility of ABS-SS composites printed in the vertical direction were greatest for

ABS-10%SS composites, second greatest for ABS-23%SS composites, and the lowest for ABS-

15%SS composites in both the 45/-45 and 0/90 orientations. This trend does not correspond to

the increase in wt% additives of stainless steel to the polymer matrix, though the reasoning for

this deviation is further explained in the fractography section. For 0/90H bars, ABS-10%SS had

the greatest ductility, followed by ABS-23%SS, with ABS-15%SS offering the lowest properties of

all of the composites tested, which will be further elaborated on in both this chapter and the

fractography chapter. For 45/-45H bars, ABS-10%SS had a greater ductility value than that of

ABS-15%SS, while the average value of the ABS-10%SS bars was lower than that of the single

data point for ABS-23%SS tensile bars. As only one of the tensile bars was able to be observed

of the ABS-23%SS data set, and the maximum ductility from the ABS-10%SS composite data set

was higher than the single value for the ABS-23%SS data point, despite the set average being

lower, there is not enough evidence to make a conclusion regarding the ductility of ABS-10%SS

composites relative to ABS-23%SS composites in the 45/-45H configuration.

In the 45/-45H, ABS-23%SS tensile bars, during the tensile test, the layers of the bar

delaminated along a plane parallel to the direction of the applied tensile force as observed in

Figure 20. This indicates that for this orientation, there was extremely insufficient interlay

diffision, however, due to the orientation of the layers parallel to the applied force, the 45/-

45H, ABS-23%SS specimen that did not delaminate during tensile testing demonstrated

impressively high ductility (~29% at failure). This layer delamination induced failure was an

anomaly, only occurring in 2 bars in this single data set and none in any other sets.

Figure 20. Delamination during tensile test.

24

The ABS-15%SS tensile bars had significantly reduced ductility as compared to ABS-

10%SS and ABS-23%SS with the same print conditions, in every orientation. The reason for this

reduction in ductility can be observed in a SEM micrograph of the fracture surface, where it can

be observed that there is an increased amount of air gap space between many of the filament

bead roads. These gaps, observable in Figure 21, are a result of low filament diameter. This

suggests that the filament used for these prints, while within the 1.70 +/- 0.4 mm specification

outlined in the methodology, may have been on the lower end of the specification for more

than a short length of filament, resulting in a more sparse amount of infill to be deposited for

the horizontally printed bars.

Figure 21. Insufficient infill density.

25

4.3.3. Comparisons between ABS and ABS-SS Composites

Combined effects of material selection, build direction, and raster angle are presented in

Table 6, where all bars were printed in accordance with the 4 predetermined optimum protocols. Table 6. Mechanical properties for all studied materials in all orientations

* All but one sample delaminated during tensile testing (Figure 20).

Factory ABS ABS ABS-10%SS ABS-15%SS ABS-23%SS

Average UTS (MPa) 31.34 30.43 29.08 29.09 18.35

Average Ductility (Elongation %) 7.75 7.74 8.69 6.66 7.89

Standard Deviation UTS 0.09 0.05 0.25 0.20 1.51

Standard Deviation Ductility 0.61 0.42 0.66 0.21 0.21

Max UTS (MPa) 31.4 30.5 29.2 29.3 19.4

Max Ductility (Elongation %) 8.4 8.2 9.1 6.8 8


Average UTS (MPa) 31.48 30.99 29.54 26.26 14.76




Max UTS (MPa) 31.9 31.8 29.8 26.6 15.2

Max Ductility (Elongation %) 9.4 9.7 8.8 6.8 8.6


Average UTS (MPa) 37.86 40.47 36.79 35.63 26.51




Max UTS (MPa) 38.6 40.7 36.9 36.3 27.4

Max Ductility (Elongation %) 31.9 30.8 32.4 20.6 26.1


Average UTS (MPa) 38.75 40.05 37.22 37.12 N/A*

Average Ductility (Elongation %) 31.18 26.98 25.97 12.26 N/A*

Standard Deviation UTS 0.80 0.23 0.72 1.86 N/A*

Standard Deviation Ductility 3.28 4.99 2.73 4.33 N/A*

Max UTS (MPa) 40 40.5 38.06 39.1 24.5

Max Ductility (Elongation %) 36 34.4 29.2 16.8 29

0/90 Vertical

45/-45 Vertical

0/90 Horizontal

45/-45 Horizontal

26

(a)

(b)

(c)

(d)

Figure 22. Maximum stress - strain curves for all materials.

4.4.3.1. Ultimate Tensile Strength Comparisons between ABS and ABS-SS Composites

In all cases, horizontally printed tensile bars, with the layer direction built parallel to the

axis of the applied force, had higher measured values for UTS than their vertically printed

counterparts (Figure 23). This is because the weakest location in FDM printed specimens is the

interlayer interface, where delamination of the layers is the cause of fracture in the weaker

vertically printed tensile bars. With the layers aligned parallel to the direction of the applied force,

the horizontally printed specimens failed due to the mechanical limit of the material rather than

due to the weakness of the layer interface.

Tensile results indicate that additions of stainless steel up to 15 wt% resulted in only slightly

reduced measured values for UTS for printed tensile specimens in both the vertical and horizontal

build directions for both tested raster angle patterns, 0/90, and 45/-45. Despite insufficient layer

diffusion in the ABS-15%SS composite bars, it was interesting to note that the UTS did not suffer to

the magnitude that the ductility of these samples did. In the literature it has been reported that the

optimal print direction for generating the highest possible UTS and ductility is 45/-45H. It is

possible that the slicing parameters used to generate the G-Code, and with the narrow gauge of

27

the ASTM type D638 tensile specimen, are the reasons for the similarities in values for both the

vertical and horizontal build directions when comparing 0/90 to 45/-45 raster angles.

Figure 23. UTS of all build orientations and raster angles (error bars given for cases where SD > 1%).

4.4.3.2. Ductility Comparisons between ABS and ABS-SS Composites

In all cases, the ductility of horizontally printed tensile bars was greater than that of

vertically printed bars (Figure 24). This is due to the same layering effect as mentioned above,

where horizontal bars are built with the layers parallel to the direction of the axial force while

vertical bars are printed with the layers perpendicular to the direction of the applied force,

resulting in layer delamination in vertically printed specimens. The ABS-SS composite tensile

bars did not have significantly reduced levels of reproducibility as compared to ABS horizontally

printed bars, where the standard deviation values were all within a few percent of one another.

In all cases, horizontally printed tensile bars had greater standard deviation values. This is likely

due to the fact that for vertically printed bars, layer delamination is the overarching limiting

factor, while in horizontally printed bars there are a combination of factors that govern

whether the bar will elongate or fracture. Of the horizontally printed bars, the condition that

yielded the lowest reproducibility was the 45/-45H orientation.

Relative to ABS, additions of 10, and 23 wt% stainless steel to the ABS matrix did not

significantly reduce the measured ductility of tensile specimens. This was unforeseen as due to

the formation of voids around the metal particle additives, it was anticipated that the ductility

would be dramatically reduced. Mechanical testing yielded results that suggest otherwise,

where even up to a 23 wt% addition of stainless steel only moderately diminished the ductility

28

of the tensile specimens, even improving the ductility of the ABS-SS composite relative to ABS

in the case of a 10 wt% addition of stainless steel in the 0/90V and 0/90H tensile bars.

Figure 24. Ductility of all build orientations and raster angles (error bars given where SD > 1%).

In both 0/90 and 45/-45 raster orientations for horizontally printed tensile bars, it can

be observed that within layers, the air gap between extruded bead roads is the location at

which the tensile bar ultimately fails. However, prior to failure, the material pulls on either side

of the air gap, straining it, resulting in the white stripes visible on either side of the fractured

tensile bar in Figure 25. This behavior is observed in every material and in both orientations,

0/90 and 45/-45 for horizontally printed tensile bars.

Figure 25. Ductility pattern on gage of fractured specimen.

4.4. Fractographic Results and Analysis 4.4.1. Fracture Surface Comparison of Vertical and Horizontal Bars

Inspection of the fracture surfaces of Factory ABS and ABS vertically printed tensile bars

reveal that there are a few, clearly defined fracture planes which can be observed in Figures 27(a,c)

29

and Figures 28(a,c). As a 10 wt% addition of stainless steel is incorporated into the polymer matrix,

the number of fracture planes greatly increases for vertically printed specimens, shown in Figures

29(a,c). As SS loading increased from 15 wt% and 23 wt%, the number of fracture planes was

reduced. The resulting fracture surfaces of the ABS-15%SS and ABS-23%SS tensile specimens will be

discussed in section 4.4.3. As compared to vertically printed tensile bars, 0/90 horizontally printed

tensile bars for Factory ABS and ABS clearly showed the air gap resulting from limitations in infill

packing density, where this porosity is observed in Figures 27(e) and 28(e). With increases in wt%

additions of stainless steel, the porosity from these limitations in infill packing density remain

clearly visible in 0/90 horizontally printed tensile bars, shown in Figures 29(e), 30(e), and 31(c).

4.4.2. Raster Angle Impact on Fracture Surfaces

Fracture surfaces of vertically printed tensile bars do not show a clear variation in

morphology due to raster orientation, with the exception of ABS-15%SS specimens where, as

indicated in Figures 31(a,c) there is a distinct variation. The ABS-15%SS 0/90V tensile bar exhibited

a more ductile fracture surface as compared to its 45/-45V counterpart, which aligns with reported

ductility values in Table 6. The mechanism of fracture for the ABS-15%SS 45/-45V tensile bar will be

explored further in the following section. For horizontally printed tensile bars, raster orientating

has a much more visible effect on the fracture surface. As compared to 0/90H tensile bars where

porosity can be observed due to infill packing limitations, porosity cannot be observed in 45/-45H

bars with the same material due to the angle of viewing. This can be observed in Figures 27(a,e),

28(a,e), 29(a,e), and 30(a,e). Conversely, the only case where this is not observed is in Figures

31(a,c) where there is no discernible packing limitation induced porosity, instead, the only visible

voids are those caused from particle delamination from the surrounding matrix.

4.4.3. Interfacial Debonding and Toughening Mechanisms

The SEM studies indicate that there was incomplete adhesion of the stainless steel

additives to the ABS polymer matrix, as observed in Figure 26. This highlights one of the major

difficulties in creating polymer-metal composites – insufficient matrix-particle interfacial strength.

It is clear that while the stainless steel is not completely adhered to the surrounding matrix, the

close proximity of the matrix to the surface of the stainless steel particles suggests that the

surface roughness of the particles contributed to the ability for the particles to remain lodged in

the matrix.

30

In Figure 26 it is also observed that the ABS-23%SS composite has the look of a rubber

toughened material. The holes surrounding the particle are significantly smaller in diameter

and too uniform in shape to be a stainless steel addition. These holes represent the crack

circumventing the rubbery acrylonitrile phase, causing the rubbery phase to “pull-out” of the

surrounding matrix. Both of these instances can be classified as adhesive failures, where the

particle and the rubbery phase separated from the bulk material.

Figure 26. Incomplete particle - matrix adhesion and evidence of rubber toughening in the surrounding polymer matrix.

In Figures 30(c,d), fractographic images of a 45/-45V ABS-15%SS print, indicate that there

was incomplete layer diffusion resulting in two distinct fracture patterns. The dark region of the

image represents layer delamination where a low ductility fracture pattern can be observed, as

compared to the lighter region, where the increased deformation of the surface indicates a

more ductile fracture within the layer rather than an abrupt, brittle inter-layer delamination.

Figures 31(a,b) indicates that there is evidence of the stainless steel particles pinning the

distribution and advance of the damage. This has been inferred from the elongated traces left by

the cracks circumventing the particles on the fracture surfaces, whose directions coincide with

the layer deposition directions. This conclusion becomes obvious from the appearance of the

fracture surfaces that span consecutive layers deposited in different directions (0 and 90),

leading to traces elongated in the same respective directions. Despite poor adhesion to the

polymer matrix, the presence of the stainless steel particle additives clearly impacted the

advancement of damage, and further improvements can be achieved by increasing the adhesion

of the particle additions to the polymer matrix.

31

(a) Factory ABS 0/90V, 23x

(b) Factory ABS 0/90V, 150x

(c) Factory ABS 45/-45V, 23x

(d) Factory ABS 45/-45V, 150x

(e) Factory ABS 0/90H, 23x

(f) Factory ABS 0/90H, 150x

(g) Factory ABS 45/-45H, 23x

(h) Factory ABS 45/-45H, 150x

Figure 27. SEM images of Factory ABS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications.

32

(a) ABS 0/90V, 23x

(b) ABS 0/90V, 150x

(c) ABS 45/-45V, 23x

(d) ABS 45/-45V, 150x

(e) ABS 0/90H, 23x

(f) ABS 0/90H, 150x

(g) ABS 45/-45H, 23x

(h) ABS 45/-45H, 150x

Figure 28. SEM images of ABS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications.

33

(a) ABS-10%SS 0/90V, 23x

(b) ABS-10%SS 0/90V, 150x

(c) ABS-10%SS 45/-45V, 23x

(d) ABS-10%SS 45/-45V, 150x

(e) ABS-10%SS 0/90H, 23x

(f) ABS-10%SS 0/90H, 150x

(g) ABS-10%SS 45/-45H, 23x

(h) ABS-10%SS 45/-45H, 150x

Figure 29. SEM images of ABS-10%SS fracture surfaces at low (a,c,e,g) and high (b,d,f,h) magnifications.

34

(a) ABS-15%SS 0/90V, 23x

(b) ABS-15%SS 0/90V, 150x

(c) ABS-15%SS 45/-45V, 23x

(d) ABS-15%SS 45/-45V, 150x

(e) ABS-15%SS 0/90H, 23x

(f) ABS-15%SS 0/90H, 150x

(g) ABS-15%SS 45/-45H, 23x

(h) ABS-15%SS 45/-45H, 150x


35

(a) ABS-23%SS 0/90V, 23x

(b) ABS-23%SS 0/90V,150x

(c) ABS-23%SS 0/90H, 23x

(d) ABS-23%SS 0/90H, 150x

(e) ABS-23%SS 45/-45H, 23x

(f) ABS-23%SS 45/-45H, 150x


36

4.5. Other Functional Properties of the ABS-SS Composites The 420 stainless steel particulate additions to the ABS matrix resulted in 3D printed,

ABS polymer based tensile bars that demonstrated magnetic properties, where tensile bars of

ABS-10%SS (Figure 32), ABS-15%SS, and ABS-23%SS all attracted small, non-Neodymium bar

magnets, with the ABS-23%SS tensile bars visibly being the most responsive to the magnet. As

observed when extruding, the ABS-SS composites cool more quickly than the pure polymer.

With this in mind, coupled with the fact that elapsed time between layers was limited by the

ability of the material to rapidly cool, this suggests that for a printer capable of depositing

filament at higher rate than the Hyrel System 30M utilized in this study, the optimal print speed

would be higher for the ABS-SS composite materials relative to ABS, reducing print times for

parts.

Figure 32. ABS-10%SS composite tensile bar attracted by a magnet.

37

5. Conclusions In this study, a methodology has been determined allowing for the homogeneous

mixing of metal particulate additives to a polymer matrix, and the drying, extrusion, and

printing of these composite materials. Dispersions of stainless steel in an ABS matrix were

manufactured and the particulate additive distribution homogeneity confirmed through SEM

observations. The MDSC scans also reinforced this conclusion due to the single sharp step in

ΔCp at Tg, as opposed to a multi-step-like increase as would be the case for a non-

homogeneous material. Through use of a two phase drying process, MDSC scans confirmed

that all of the acetone solvent was evacuated from the sample, and the twice dried ABS-SS

composites were chopped into feedstock and extruded into filament, then re-extruded to

eliminate any porosity in the filament. In parallel, ASTM D638 type V tensile bars were built in

Solidworks, and imported into the host software on the Hyrel System 30M Printer used in this

study. Bar arrangement was altered from linear to a pentagon arrangement to mitigate the

print head travel time difference when moving between bars. When it was determined that

only single bar prints would be a viable option, experiments were conducted where print

parameters were varied in an attempt to have a single bar print reproduce the mechanical

properties of five bar prints. Through this optimization process, the relationship between

dimensional accuracy and interlayer diffusion was revealed. G-Code modifying macros were

utilized to maximize interlayer diffusion while maintaining dimensional accuracy, and the

resulting tensile bars printed in all build direction and raster angle combinations for Factory

ABS yielded the highest UTS and ductility values of test prints. With these four idealized

protocols, one for each build direction and raster angle combination, tensile bars were built

from Factory ABS, ABS, ABS-10%SS, ABS-15%SS, and ABS-23%SS and the mechanical properties

were subsequently obtained, and it was observed that up to a 23 wt% addition of stainless steel

only resulted in a slightly reduced. The ABS-15%SS composites demonstrated similar UTS values

relative to their ABS counterparts. Fracture surfaces observed with the SEM revealed the

effects of the stainless steel additive on the fracture mechanisms of the composites, indicating

the temporary pinning of the damage advancement upon interactions with the particles.

38

6. Future Work Further print optimization would reveal the appropriate print protocols for each ABS-SS

composite, and facilitate a comparison of mechanical properties of the strongest, most ductile

tensile bars for each material in this study, rather than using the same protocols for Factory

ABS to print tensile bars of all materials (i.e., customized printing parameters for each material

and build condition). Coupling the printing of the tensile specimens with thermal data from a

high resolution camera will assist in the optimization process. Knowing the traveling bead

temperature, thermal management optimization can be done for each condition in both ABS

and ABS-SS composites. In this way, quantitative data would be able to reinforce the qualitative

observations that the cooling rate during printing and extruding of the ABS-SS composite was

greater than that of the Factory ABS.

Future studies should also focus on the effect of functionalizing steel powders (to

enhance the adhesion with the polymer matrix) on the properties of ABS-SS composites using

silane solutions in both water and ethylene, and water and heptane. Preliminary MDSC scans of

the functionalized powder based composites reveal that there was a dramatic change in glass

transition temperature relative to non-functionalized powder based composites, as indicated

by the data shown in Tables 7 and 8.

Table 7. Functionalized stainless steel heating scan Table 8. Functionalized stainless steel cooling scan

Sample Scan Tmin [°C] Tg [°C] Tmax [°C]

ABS-

10%SSH2 106.93 109.24 110.56

ABS-

10%SS (e)H3 100.11 104.58 106.94

ABS-

10%SS (h)H3 96.52 100.77 104.2

HEATING

Sample Scan Tmin [°C] Tg [°C] Tmax [°C]

ABS-

10%SSC2 106.42 109.36 110.85

ABS-

10%SS (e)C3 98.94 104.2 106.89

ABS-

10%SS (h)C3 94.9 100.35 105.01

COOLING

39

Appendices A. Slic3r Parameters Variations between H and V that are edited after slicing: -Vertical bars have 68.85% fan. Horizontal bars have 5% fan. -Vertical bars have a skirt that is as high as the bar, Horizontal bars have a 1 layer skirt. -Vertical bars have the macro turn all skirt commands above the first layer into a "G0" command, so the print head will move there, but will not extrude any filament. -Vertical bars have a 150 mm/s skirt travel speed. ; avoid_crossing_perimeters = 1 ; bed_shape = 0x0,275x0,275x225,0x225 ; bed_temperature = 100 ; before_layer_gcode = ;announce new layer <[layer_num]>\n;---\nM756 S[layer_height]\nM790 ;execute any new layer actions\n;--- ; bridge_acceleration = 0 ; bridge_fan_speed = 27 ; brim_width = 3 ; complete_objects = 0 ; cooling = 0 ; default_acceleration = 0 ; disable_fan_first_layers = 1 ; duplicate_distance = 6 ; end_gcode = M107 T10 ; turn off fans and lasers\nM104 S0 ; turn off temperature\nM140 S0 ;turn off the hot bed.\nG91 ;\nG1 Z5.0 ; Drop bed 5mm for extra clearance \nG90 ; absolute\nG28 X0 Y0 ; home X axis\nG92 X0 Y0 ; confirm we are at zero\nM84 ; disable motors\nM30 ; End ofprogram ; extruder_clearance_height = 20 ; extruder_clearance_radius = 20 ; extruder_offset = 0x0,0x0,0x0,0x0 ; extrusion_axis = E ; extrusion_multiplier = 1,1,1 ; fan_always_on = 1 ; fan_below_layer_time = 60 ; filament_colour = #FFFFFF ; filament_diameter = 1.72,1.72,1.72 ; first_layer_acceleration = 0 ; first_layer_bed_temperature = 100 ; first_layer_extrusion_width = 0.4 ; first_layer_speed = 12.5 ; first_layer_temperature = 237,237,237 ; gcode_arcs = 0 ; gcode_comments = 1 ; gcode_flavor = reprap ; infill_acceleration = 0 ; infill_first = 0 ; layer_gcode = ; max_fan_speed = 70 ; max_print_speed = 150

40

; max_volumetric_speed = 0 ; min_fan_speed = 27 ; min_print_speed = 10 ; min_skirt_length = 0 ; notes = ; nozzle_diameter = 0.35,0.35,0.35,0.35 ; only_retract_when_crossing_perimeters = 1 ; ooze_prevention = 0 ; output_filename_format = [input_filename_base].gcode ; perimeter_acceleration = 0 ; post_process = ; pressure_advance = 0 ; resolution = 0 ; retract_before_travel = 2,2,2,2 ; retract_layer_change = 1,1,1,1 ; retract_length = 0,0,0,0 ; retract_length_toolchange = 0,0,0,0 ; retract_lift = 125,125,125,125 ; retract_restart_extra = 0,0,0,0 ; retract_restart_extra_toolchange = 0,0,0,0 ; retract_speed = 20,20,20,20 ; skirt_distance = 10 ; skirt_height = 600 ; skirts = 1 ; slowdown_below_layer_time = 2 ; spiral_vase = 0 ; standby_temperature_delta = -5 ; start_gcode = M104 T10 S[temperature]\nG21 ; use millimeters\nG90 ; absolute coordinates\nG0 Z5 ; lift head to avoid collisions\nG28 X0 Y0 ; home X and Y\nG92 X0 Y0 ; reset origin: X and Y\nG0 X0 Y0 ; move to desired origin\nG92 X0 Y0 ; reset origin: X and Y\nM83 ; relative extruder coordinates\nM109 S[temperature] ;wait for temperture to come up.\nM756 S[first_layer_height] ;set flowfor the first layer please\n\n ; temperature = 237,237,237 ; threads = 8 ; toolchange_gcode = ; travel_speed = 30 ; use_firmware_retraction = 0 ; use_relative_e_distances = 0 ; use_volumetric_e = 0 ; vibration_limit = 0 ; wipe = 0,0,0,0 ; z_offset = 0 ; dont_support_bridges = 0 ; extrusion_width = 0.4 ; first_layer_height = 0.2 ; infill_only_where_needed = 0 ; interface_shells = 0 ; layer_height = 0.2 ; raft_layers = 0

41

; seam_position = aligned ; support_material = 0 ; support_material_angle = 0 ; support_material_contact_distance = 0.2 ; support_material_enforce_layers = 0 ; support_material_extruder = 1 ; support_material_extrusion_width = 0 ; support_material_interface_extruder = 1 ; support_material_interface_layers = 3 ; support_material_interface_spacing = 0 ; support_material_interface_speed = 100% ; support_material_pattern = pillars ; support_material_spacing = 2.5 ; support_material_speed = 200 ; support_material_threshold = 0 ; xy_size_compensation = 0 ; bottom_solid_layers = 0 ; bridge_flow_ratio = 1 ; bridge_speed = 12.5 ; external_fill_pattern = rectilinear ; external_perimeter_extrusion_width = 0.4 ; external_perimeter_speed = 12.5 ; external_perimeters_first = 0 ; extra_perimeters = 1 ; fill_angle = 45 ; fill_density = 100% ; fill_pattern = rectilinear ; gap_fill_speed = 12.5 ; infill_every_layers = 1 ; infill_extruder = 1 ; infill_extrusion_width = 0.4 ; infill_overlap = 50% ; infill_speed = 12.5 ; overhangs = 1 ; perimeter_extruder = 1 ; perimeter_extrusion_width = 0.4 ; perimeter_speed = 12.5 ; perimeters = 1 ; small_perimeter_speed = 12.5 ; solid_infill_below_area = 0 ; solid_infill_every_layers = 0 ; solid_infill_extruder = 1 ; solid_infill_extrusion_width = 0.55 ; solid_infill_speed = 15 ; thin_walls = 1 ; top_infill_extrusion_width = 0.55 ; top_solid_infill_speed = 15 ; top_solid_layers = 0

42

B. Macro G-Code Editor Script

Sub Skirt_Mover_LM()

'

' Macro1 Macro

'

' Keyboard Shortcut: Ctrl+t

'

'moving skirt 100mm back in the Y direction

'

Dim i As Integer

Dim X As Integer

Dim cellcopy As String

Dim Xstring As String

Dim XstringInt As Integer

Dim XstringInt2 As Integer

Dim Xstring2 As String

Dim cellcopy2 As String

Dim cellcopy3 As String

Dim XstringE As String

Dim XstringIntE As Integer

Dim XstringInt2E As Integer

Dim Xstring2E As String

'

'fixing extrusion while moving to perimeter point

'

'L = 14, M = 14, R=13

'L= 58, M=59, R = 58

Dim cellcopyB As String

Dim cellcopyC As String

Dim cellcopyD As String

Dim cellcopyE As String

Dim cellcopyF As String

43

Dim cellcopyG As String

Dim cellcopyH As String

X = 25000

For i = 1 To X

If InStr(Cells(i, 1), "skirt") Then

cellcopy = Cells(i, 1)

Xstring = Mid(cellcopy, 14, 1)

XstringInt = CInt(Xstring)

XstringInt2 = XstringInt + 10

Xstring2 = CStr(XstringInt2)

cellcopy2 = Replace(cellcopy, Xstring, Xstring2, 14, 1)

If XstringInt2 = 0 Then

cellcopy2 = Replace(cellcopy2, Xstring2, "", 1, 1)

End If

cellcopy3 = Left(cellcopy, 13) & cellcopy2

Cells(i, 1).Value = cellcopy3

End If

'

'

'

'

If InStr(Cells(i, 1), "move to first perimeter point") And InStr(Cells(i + 1, 1), "move to first

perimeter point") And InStr(Cells(i + 2, 1), "move to first perimeter point") And InStr(Cells(i + 3,

1), "move to first perimeter point") Then

cellcopyC = Cells(i, 1)

cellcopyD = Cells(i + 4, 1)

Cells(i, 1).Value = "G0" & Right(cellcopyC, 59)

Cells(i + 1, 1).Value = ""


44


XstringE = Mid(cellcopyD, 14, 2)

XstringIntE = CInt(XstringE)

XstringInt2E = XstringIntE + 0

Xstring2E = CStr(XstringInt2E)

cellcopy2E = Replace(cellcopyD, XstringE, Xstring2E, 14, 1)

Cells(i + 4, 1).Value = "E0 " & Left(cellcopyD, 13) & cellcopy2E

cellcopyG = Cells(i + 4, 1)

XstringG = Mid(cellcopyG, 8, 3)

XstringIntG = CInt(XstringG)

XstringInt2G = XstringIntG + 5

Xstring2G = CStr(XstringInt2G)

cellcopy2G = Replace(cellcopyG, XstringG, Xstring2G, 8, 1)

Cells(i + 4, 1).Value = Left(cellcopyG, 7) & cellcopy2G

End If

If InStr(Cells(i, 1), "move to first perimeter point") And InStr(Cells(i + 1, 1), "move to first

perimeter point") And InStr(Cells(i + 2, 1), "move to first perimeter point") And InStr(Cells(i + 3,

1), "move to first perimeter point") Then

End If

Next i

End Sub

45

References

[1] T. Wohlers, “Viewpoint: History of Additive Fabrication (Part 1),” Wohlers Associates, 2008. [Online].

Available: https://wohlersassociates.com/MarApr08TCT.htm. [Accessed: 18-Apr-2017].

[2] “A third industrial revolution,” The Economist, 21-Apr-2012. [Online]. Available:

http://www.economist.com/node/21552901. [Accessed: 16-Apr-2017].

[3] M. Christopher and L. J. Ryals. The supply chain becomes the demand chain. Journal of Business

Logistics 35(1), pp. 29-35. 2014. DOI: 10.1111/jbl.12037.

[4] T. S. Srivatsan, T. S. Sudarshan and ENGnetBASE. Additive Manufacturing: Innovations, Advances, and

Applications 2016; 2015.

[5] S. Ford and M. Despeisse. Additive manufacturing and sustainability: An exploratory study of the

advantages and challenges. Journal of Cleaner Production 137pp. 1573-1587. 2016. DOI:

10.1016/j.jclepro.2016.04.150.

[6] J.P. Kruth, Material Incress Manufacturing by Rapid Prototyping Techniques, CIRP Annals -

Manufacturing Technology, Volume 40, Issue 2, 1991, Pages 603-614, ISSN 0007-8506

[7] S. S. Crump. Rapid prototyping using FDM. Modern Casting 82(4), pp. 36. 1992.

[8] A. R. Torrado Perez, D. A. Roberson and R. B. Wicker. Fracture surface analysis of 3D-printed tensile

specimens of novel ABS-based materials. Journal of Failure Analysis and Prevention 14(3), pp. 343-353.

2014. DOI: 10.1007/s11668-014-9803-9.

[9] Sung-Hoon Ahn Michael Montero Dan Odell Shad Roundy Paul K. Wright, (2002),"Anisotropic

material properties of fused deposition modeling ABS", Rapid Prototyping Journal, Vol. 8 Iss 4 pp. 248 –

257

[10] O. S. Es-Said et al. Effect of layer orientation on mechanical properties of rapid prototyped

samples. Materials and Manufacturing Processes 15(1), pp. 107-122. 2000. DOI:

10.1080/10426910008912976.

[11] A. R. Torrado et al. Characterizing the effect of additives to ABS on the mechanical property

anisotropy of specimens fabricated by material extrusion 3D printing. Additive Manufacturing 6pp. 16-

29. 2015. DOI: 10.1016/j.addma.2015.02.001.

[12] W. Wu et al. Influence of layer thickness and raster angle on the mechanical properties of 3D-

printed PEEK and a comparative mechanical study between PEEK and ABS. Materials 8(9), pp. 5834-

5846. 2015. DOI: 10.3390/ma8095271.

46

[13] “Ultimate Tensile Strength,” Tensile Property Testing of Plastics. [Online]. Available:

http://www.matweb.com/reference/tensilestrength.aspx. [Accessed: 27-Apr-2017].

[14] “Acrylonitrile butadiene styrene,” Wikipedia, 04-Apr-2017. [Online]. Available:

https://en.wikipedia.org/wiki/Acrylonitrile_butadiene_styrene. [Accessed: 16-Apr-2017]

[15] J. Cantrell et al., "Experimental characterization of the mechanical properties of 3D printed ABS and

polycarbonate parts," Conference Proceedings of the Society for Experimental Mechanics Series, vol. 3,

pp. 89-105, Jun. 2017.

[16] J. C. Riddick et al., "Fractographic analysis of tensile failure of acrylonitrile-butadiene-styrene

fabricated by fused deposition modeling," Additive Manufacturing, vol. 11, pp. 49-59. Jul. 2016.

[17] A. R. Torrado Perez et al., "Erratum to: Fracture surface analysis of 3D-Printed tensile specimens of

novel ABS-Based materials," Journal of Failure Analysis and Prevention, vol. 14, no. 4, pp. 549–549, Jul.

2014.

[18] K. Chockalingam et al., "Enhancement of Anisotropic Strength of Fused Deposited ABS Parts by

Genetic Algorithm," Materials and Manufacturing Processes, vol. 31, no. 15, pp. 2001-2010, Dec. 2016.

[19] S. Ziemian et al., "Tensile and fatigue behavior of layered acrylonitrile butadiene styrene," Rapid

Prototyping Journal, vol. 21, no. 3, pp. 270-278, 2015.

[20] S. H. Huang et al. Additive manufacturing and its societal impact: A literature review. The

International Journal of Advanced Manufacturing Technology 67(5), pp. 1191-1203. 2013. DOI:

10.1007/s00170-012-4558-5.

[21] LaMonica, Martin. "GE Will Make Jet Part with Additive Manufacturing." MIT Technology Review.

MIT Technology Review, 09 Dec. 2015. Web. 19 Mar. 2017.

[22] M. Nikzad, S. H. Masood, and I. Sbarski, "Thermo-mechanical properties of a highly filled polymeric

composites for Fused Deposition Modeling," Swinburne University of Technology, 2011.

[23] K. V. Wong and A. Hernandez, "A Review of Additive Manufacturing," Department of Mechanical

and Aerospace Engineering, University of Miami, International Scholarly Research Network, vol. 2012.

[24] Reprap.org, Content is available under GNU Free Documentation License 1.2.

[25] S. Kumar and J. P. Kruth, "Composites by rapid prototyping technology," ed: Elsevier Ltd., 2009.

47

[26] S. Kalpakjian, I. I. Of technology, and S. R. Schmid, Manufacturing Engineering and Technology 7th

Edition, 7th ed. Singapore, Singapore: Pearson Education South Asia, 2013.

[27] J. F. Rodriguez, J. P. Thomas, and J. E. Renaud, ""Mechanical behavior of acrylonitrile butadiene

styrene (ABS)

fused deposition materials. Experimental investigation," vol. 7, ed: Emerald Insight, 2001.

[28] L. Li, Q. Sun, C. Bellehumeur, and P. Gu, "Investigation of Bond Formation in FDM Process," Journal

of Manufacturing Processes.

[29] F. Rayegani and G. C. Onwubolu, "Fused deposition modelling (FDM) process parameter prediction

and optimization using group method for data handling (GMDH) and differential evolution (DE)," vol. 73,

ed. The International Journal of Advanced Manufacturing Technology, 2014, pp. 509-519.

[30] I. Durgun and R. Ertan, "Experimental investigation of FDM process for improvement of mechanical

properties and production cost," vol. 20, ed. Rapid Prototyping Journal Emerald Group Publishing

Limited, 2013, pp. 228-235.

[31] N. Sa’ude, M. S. Wahab, and M. H. I. Ibrahim, "Effect of Powder Loading and

Binder Materials on Mechanical Properties in Iron-ABS Injection Molding Process," Applied Mechanics

and Materials, vol. 315, pp. 532–586, Sep. 2016. [Online]. Available:

https://www.researchgate.net/publication/237005209. Accessed: Jan. 30, 2017.

[32] Humboldt University of Berlin, "Investigation of Polymers with Differential Scanning

Calorimetry,"2015.

[33] Brown, Eric N., Scott R. White, and Nancy R. Sottos. "Microcapsule induced toughening

in a self-healing polymer composite." Journal of Materials Science 39.5 (2004): pp. 1703-1710.

[34] G. W. Melenka, B. K. O. Cheung, J. S. Schofield, M. R. Dawson, and J. P. Carey, "Evaluation and

prediction of the tensile properties of continuous fiber-reinforced 3D printed structures," Composite

Structures, vol. 153, pp. 866–875, Oct. 2016.

[35] A. W. Fatimatuzahraa, B. Farahaina, and W. A. Y. Yusoff, "The effect of employing different raster

orientations on the mechanical properties and microstructure of Fused Deposition Modeling parts," ed:

2011 IEEE Symposium on Business, Engineering and Industrial Applications (ISBEIA), Langkawi, Malaysia,

2011.

[36] A. Garg, A. Bhattacharya, and A. Batish, "Chemical vapor treatment of ABS parts built by FDM:

Analysis of surface finish and mechanical strength," ed. The International Journal of Advanced

Manufacturing Technology: Springer-Verlag London 2016, 2016, pp. 1-17.

https://www.researchgate.net/publication/237005209

48

[37] M. Kaveh, M. Badrossamay, E. Foroozmehr, and A. H. Etefagh, "Optimization of the printing

parameters affecting dimensional accuracy and internal cavity for HIPS material used in fused deposition

modeling processes," vol. 226, ed, 2015, pp. 280-286.

[38] L. Li, Q. Sun, C. Bellehumeur, and P. Gu, "Investigation of Bond Formation in FDM Process," ed:

Journal of Manufacturing Processes.

[39] M. Montero, S. Roundy, D. Odell, S.-H. Ahn, and P. K. Wright, "Material Characterization of Fused

Deposition Modeling

(FDM) ABS by Designed Experiments," ed: Society of Manufacturing Engineers, 2001.

[40] L. Novakova-Marcincinova and J. Novak-Marcincin, "Verification of Mechanical Properties of ABS

Materials Used in FDM Rapid Prototyping Technology," vol. 8, ed: Proceedings in Manufacturing

Systems, 2013, pp. 87-92.

[41] J. F. Rodriguez, J. P. Thomas, and J. E. Renaud, "Mechanical behavior of acrylonitrile butadiene

styrene (ABS)

fused deposition materials. Experimental investigation," vol. 7, ed: Emerald Insight, 2001.

[42] Q. Sun, G. M. Rizvi, C. T. Bellehumeur, and P. Gu, "Effect of processing conditions on the bonding

quality of FDM polymer filaments," vol. 14, ed. Rapid Prototyping Journal: Emerald Insight, 2008.

[43] Stratasys. (2016). 3D Printing Solutions | Stratasys. Available: http://www.stratasys.com/

[44] H. 3D. (2016). Hyrel 3D. Available: http://www.hyrel3d.com/

[45] L. Novakova-Marcincinova and J. Novak-Marcincin, "Verification of Mechanical Properties of ABS

Materials Used in FDM Rapid Prototyping Technology," vol. 8, ed: Proceedings in Manufacturing

Systems, 2013, pp. 87-92.

[46] H. K. Garg and R. J. Singh, "Comparison of wear behavior of ABS and Nylon 6—Fe powder composite

parts prepared with fused deposition modelling," ed: Central South University Press and Springer-Verlag

Berlin Heidelberg 2015, 2015.

[47] P. J. Nuñez, A. Rivas, E. Garcia-Plaza, E. Beamud, and A. Sanz-Lobera, "Dimensional and Surface

Texture Characterization in Fused Deposition Modelling (FDM) with ABS plus," vol. 132, ed. Procedia

Engineering: Elvesier, 2015, pp. 856-863.

[48] M. Dawoud, I. Taha, and S. J. Ebeid, "Mechanical behaviour of ABS: An experimental study using

FDM and injection moulding techniques," vol. 21, ed. Journal of Manufacturing Processes: Elsevier and

Science Direct, 2016.

http://www.stratasys.com/

http://www.hyrel3d.com/

49

[49] R. Design, and Education for Additive Manufacturing Systems (DREAMS) Laboratory. (2016). Laser

Sintering. Available: http://www.me.vt.edu/dreams/laser-sintering/

[50] N. S. A. Bakar, M. R. Alkahari, and H. Boejang, "Analysis on fused deposition modelling

performance," vol. 11, ed. Journal of Zhejiang University - SCIENCE A: Zhejiang University and Springer-

Verlag Berlin Heidelberg 2010, 2010, pp. 972-977.

[51] D. Espalin, J. A. Ramirez, F. Medina, and R. Wicker, "Multi-material, multi-technology FDM:

exploring build process variations," ed. Rapid Prototyping Journal: Emerald Group Publishing Limited,

2013.

http://www.me.vt.edu/dreams/laser-sintering/

Documents

Fabrication and Properties of Novel Polymer-Metal