Page 149 - MSAM-4-2

Materials Science in Additive Manufacturing Bi-modal powder spreading behavior of ceramics

to understand the variation from the measured packing 3.4. Part relative density
density in the spreading direction. The difference in To understand the effect of the preferential powder
packing density was not statistically significant (p>0.350) deposition and density variation, a cube of 10 mm ×
within individual X positions; however, it was statistically 10 mm × 5 mm was printed at distinct locations on the XY
significant (p<0.001) when comparing different X plane, similar to the density coupons. The relative density
positions. There was a clear trend of packing density of the green parts was calculated using Equation 5. As a
decreasing in the spreading direction. The average packing result, the relative density followed the trend seen for the
density at the beginning of the spreading process (X ) was
1
0.6867, followed by 0.6745 in the middle (X ), and 0.6342 at powder packing density. The relative density of the parts
was higher at the beginning of the spreading direction and
2
the end of the spreading process (X ). The packing density decreased from there. The average relative density of the
3
decreases by 7.65% at the end of the spreading process printed samples measured at the start, middle, and end of
compared to the beginning. However, the packing density
remains relatively consistent along the perpendicular the powder bed along the spreading direction is 0.6724,
direction of the spreading for the given location in the 0.6611, and 0.6215, respectively. This indicates a decrease
spreading direction. The three separate locations sampled in relative density at the end of the powder bed, amounting
in the experiments yielded an average packing density of to 8.19% and 6.38% when compared to the start and
0.6647, 0.6646, and 0.6661 at Y , Y , and Y respectively. middle of the bed, respectively. Similar to the packing
1
2
3,
These values are close to the overall average packing density density, there was no significant variation perpendicular to
of 0.6651. The sampling locations and the corresponding the spreading direction for a given X location. In addition,
packing density measurements are shown in Figure 6. the average relative density measured perpendicular to
the spreading direction exhibits negligible variation, with
From the experimental results, the packing density values of 0.6516 at location Y , 0.6521 at location Y , and
1
2
shows a decreasing trend in the spreading direction, which 0.6512 at location Y . These values are close to the overall
3
follows the trend of the preferential powder deposition average packing density of 0.6517. The sampling locations
discussed in the previous section. The smaller powders and the corresponding relative density measurements are
were deposited at the beginning of the spreading process, shown in Figure 6. The relative density of the printed parts
and the larger powders dominated at the end of the lies within the range of the relative density observed for
spreading process. The packing density was highest where green parts with similar materials in previous studies. 32,77,81
the powder size distribution was dominated by smaller
powder, which decreased with the increase in powder Compared to the packing density of the part, the
size. The reason is the smaller interparticle gap between relative density of the green parts decreased across all the
smaller and larger powders. As the roller advances, regions samples, as shown in Figure 7. This can be attributed to the
increasingly dominated by coarser particles exhibit looser perturbation caused by the printing process. Binder kinetic
packing and reduced bulk density due to larger void spaces energy, which is influenced by the velocity and size of the
and reduced surface area for interparticle interactions. droplets, significantly affects the ejection of powder particles
82
This gradient is further influenced by frictional resistance during the printing process. The droplet in the BJT process
and shear-induced void during roller motion, which can can act as a projectile due to its kinetic energy, and when it
temporarily expand the powder bed at certain locations, impacts the powder bed, the momentum transfer can lead to
particularly where coarse particles dominate. However, in local powder ejection or rearrangement. This effect is more
bimodal powder, the smaller particles help to reduce the pronounced at the end of the spreading direction, where
interparticle gaps between the larger powder, resulting packing density is already reduced due to preferential coarse
in increased packing density. Increased packing density particle deposition. In addition, the surface tension and
6,7
enhances binder penetration and reduces binder spreading spreading behavior of the binder can displace fine particles
time. Furthermore, higher packing density enhances laterally or vertically, disturbing the initial packing achieved
6
40
the green density, improving the final sintered density. during spreading. Inkley et al. showed that binder droplets
31
Similar directional density gradients and segregation- can cause powder particles to be ejected from the bed,
driven heterogeneities have also been reported in other leading to increased porosity and reduced density in the
ceramic systems, including titanium–aluminum alloys final part. However, the binder introduces some mass in the
and alumina–zirconia composites, where powder bed and green part from its own component. Although the solvent
part characteristics were closely linked to initial packing portion of the binder dries off during the curing process,
structure. 58,80 It is possible to use the differential packing it leaves behind the polymeric component, which helps to
density in the same powder bed to impart different keep the shape intact. The difference between the powder
densities in the same part. bed packing density and the relative density indicates that

Volume 4 Issue 2 (2025) 10 doi: 10.36922/MSAM02510016