INTRODUCTION
Solid Freeform Fabrication (SFF) is a special class of machine technology that
rapidly produce models from three-dimensional (3D) data using an additive approach
to form the physical models (Wohlers, 2004). There are
numerous terms to describe SFF including Rapid Prototyping (RP), Layered Manufacturing
(LM) and Desktop Manufacturing (DM). The techniques are all based on the principle
of creating 3D components directly from Computer-Aided Design (CAD) in two-dimensional
profiles on layer-by-layer process without using moulds or tools as used in
conventional manufacturing techniques (Kruth et al.,
1998). The SFF processes have been used to produce physical components for
various purposes such as patterns for prototyping, fit/assembly components and
also functional models. Currently, there are a variety of SFF techniques available
such as Stereo Lithography Apparatus (SLA), Selective Laser Sintering (SLS),
Fused Deposition Modeling (FDM) and Three-Dimensional Printing (3DP). In recent
years 3DP came to the foreground as a very competitive process in terms of cost
and speed compared to others SFF technique. 3DP is a unique technique based
powder that prints complex 3D for rapid prototype purpose (Sachs
et al., 1992; Yin et al., 2007; Seitz
et al., 2005) and this technology originally developed by the Massachusetts
Institute of Technology (MIT) (Noorani, 2006). These technology
refers to a range of techniques characterized by the method of delivering build
material or build adhesive a series of nozzle that are translated across the
build platform. The basic build process is the laying down of a layer of powder
0.1 to 0.25 mm of thickness Fig. 1. As the powder supports
the part, no support structure is required therefore allowing complex parts
to be built (Noorani, 2006).
|
| Fig. 1: |
Schematic Diagram of 3DP Process (WTEC,
1997) |
Although 3DP has demonstrated the high capability of fabricating parts of a variety of shapes, it is currently limited in the choice of materials due to the restricted capabilities of the existing binder and the cost of the raw material is high. This study aims to introduce the potential of Palm Oil based-Fly Ash (POFA), mixed with maltodextrin and Poly Vinyl Alcohol (PVA) as an alternative material for 3DP in order to reduce cost for new product development.
MATERIALS AND METHODS
Preparation of raw materials: The 3DP process uses powder as raw material. Therefore, powder preparation procedure for the process need to be established and must be suitable for the RP process. Raw materials used in this study were POFA which is collected from a palm oil mill factory owned by Kluang Mill Plantation Sdn Bhd, located at Kluang, Johor, with an average particle size of 0.3-600 μm as shown in Fig. 2. Materials used as adhesive binder were maltodextrin and PVA. The average particle size of these materials are 0-32 and 200 μm, respectively.
The raw POFA and PVA were grinded and sieved to get the suitable particle size range before it can be used with the 3DP. There were three stages involved in the powder preparation as illustrated in Fig. 3.
Auto Mortar Grinder Machine model FRITSCH was used to pulverize the particles
of the POFA and PVA before undergoing the sieving process. The particle size
of POFA and PVA used in this study is between 0 ~ 180 μm. Sieving process
was done to filter the size of the POFA and PVA particles. The Vibratory Sieve-Shaker
“Analysette 3” machine was used to sieve both materials. Figure
4a and b shows the Scanning Electron Microscopy (SEM)
images of the POFA and PVA after the pulverizing process.
The preparation of the mixture of POFA, maltodextrin and PVA by a weight ratio of 7:2:1 was done by mechanical blending through the ball-milling process which was carried out the labkorea, ball-milling machine. The mixing process lasts for 8 h at a constant speed of 300 rpm to achieve the optimum dispersion quality of the mixture.
3D Printing process: The Z310 3DP machine was used to produce samples
of the tested materials. By using default setting parameters, the test samples
were fabricated based on ISO 178:2001 for flexural test samples. The samples
were also used for dimensional accuracy and surface quality measurements. In
addition, ZP102-based samples were also printed to be made as a benchmark for
comparison purposes. The specimens were then air-blown to remove the unbound
powders followed by a post-treatment using the Z-max solution to enhance the
strength of the samples.
|
| Fig. 2: |
SEM images of (a) POFA, (b) maltodextrin and (c) PVA |
RESULTS AND DISCUSSION
Testing of mechanical properties: Mechanical testing was performed on
a universal testing machine (AG-1 Shimadzu) equipped with a 10 kN load cell
at room temperature. Flexural test was done by using three-point bending method
with loading rate at 1.9 mm min-1.
Figure 5 presents the maximum stress values for flexural
test of the fabricated samples. It can be seen that the maximum stress value
of POFA is 34.52 MPa which is more than three-times higher than ZP102 materials
(11.5 MPa). From it can be seen that there are voids present in both samples,
with the POFA showing the less number of voids. Therefore, this condition proves
POFA samples are denser in comparison to the ZP102 sample. It is also noticed
from (Fig. 6b) that the ZP102 contained more porosity which
may lead the sample to become brittle and weak affecting the overall mechanical
properties of the sample. In addition, the bimodal powder-system between the
three materials of POFA, maltodextrin and PVA also improve the strength of the
product, particularly in flexural strength.
Dimensional accuracy: Mitutoyo digital calliper was used in determining
the dimensions of the test sample. For each sample parts, 3 reading were taken
which is at x-axes (length), y-axes (width) and z-axes (thickness).
|
| Fig. 3: |
Process of powder preparation |
|
| Fig. 4: |
SEM images of the (A) POFA and (B) PVA after the pulverizing
process |
The dimensional accuracy results were analyzed according to the average value
of the collected data and calculated using the Eq. 1:
D0 is the input graphical dimension while D1 is the measured
dimension (Patirupanusaraa et al., 2008) The
results were presented in Table 1 and Fig. 7.
From the results obtained, there was no correlation in all 3 axes; x, y and
z-axis for POFA material as the x and y-axes were showing shrinkage, while the
z-axes data shows the POFA sample exceeded 10.58% of dimensional error due to
the characteristic of the adhesive material used in this study. The adhesive
binder which was the altodextrin and the PVA were water-soluble, able to dissolve
in liquid which could flow through the cavities among POFA particles and possibly
spread outside the intended area of thickness affecting the overall dimensional
accuracy.
|
| Fig. 5: |
Maximum stress of flexural strength |
|
| Fig. 6: |
SEM images of fracture surface of (a) POFA and (b) ZP102 material |
| Table 1: |
Average Value of dimensional accuracy error |
 |
|
|
| Fig. 7: |
Dimensional error of ZP102 and POFA |
In the case of ZP102 materials, it was observed that the dimension has expansion
in all axes with average error 0.96, 4.23 and 0.5%, respectively for x, y and
z-axes.
Surface quality: Surface quality measurement was conducted using Mitutoyo
SJ 400 machine with determined perpendicularly to the direction of production.
Measurement was performed at three places which were at the beginning, in the
middle and the end of the test sample. The best surface quality was determined
by measuring the samples with the lowest surface roughness. Figure
8 shows the results of the measurement. Based on analysis data, it is proved
that ZP102-based samples comprise better surface quality at all sides if compared
to POFA-based samples. The average value of the POFA composite and ZP102 in
the surface roughness measurement is 14.78 and 12.52 μm, respectively.
|
| Fig. 8: |
Roughness, Ra of POFA and ZP102 |
|
| Fig. 9: |
SEM Images of (a) POFA and (b) ZP102 |
From the SEM images shown in Fig. 9, it can be seen that
the ZP102 mixture contained smaller particle sizes compared to the POFA mixture.
It is reasonably known that smaller particles convey smoother surfaces. Therefore,
in this case, ZP102 samples present better surfaces compared to POFA samples
in terms of quality and smoothness.
CONCLUSION
The existing material available for 3DP applications is too expensive to be
consumed as the main raw material especially for study purposes. Therefore developments
and researches made in order to replace the existing material, partly or totally,
must always be welcomed with the hope that the SFF technology will be continually
developed and offer cheaper alternatives for educational purposes and small
business. In line with this, the study is conducted with the use of POFA, as
the main potential replacement to the current ZP102 material, together with
maltodextrin and PVA mixture operating as the powder-binder material for the
3DP. From the results, it was concluded that:
| • |
POFA mixed with maltodextrin and PVA can be successfully direct
fabricated by 3DP machine using distilled water as the binding liquid |
| • |
The POFA composite can improve flexural strength of 3DP material |
Further researches are needed, particularly in the material preparation, to achieve better dimensional accuracy and surface quality of the POFA composite-based products.
ACKNOWLEDGMENT
The authors would like to thank University Tun Hussein Onn Malaysia for technically supporting this research. Authors would also like to acknowledge Ministry of Higher Education, Malaysia for financially support this research under the Fundamental Research Grant Scheme (FRGS).
Subscribe Today
