The final version of the test mold used to test the ceramic under pressure. Lithography-based ceramic manufacturing (LCM) is one of the current techniques that can be used for AM of ceramics.
INTRODUCTION
B ACKGROUND ON A DDITIVE M ANUFACTURING
AM allows for greater design freedom, as a wider range of geometries can be produced, compared to using traditional methods that can impose limitations on the desired geometry when designing parts. Lithography-based ceramic manufacturing (LCM) is unable to print right angles, and the radii of curvature that can be produced are limited because some radii result in cracking during post-processing thermal treatment6.
B ACKGROUND ON L ITHOGRAPHY - BASED C ERAMIC M ANUFACTURING (LCM)
- Printing Process
- Post Processing
- Printing Direction
- EXPERIMENTAL PROCEDURE
The light intensity at the corners of the building platform was measured as 98.4% of the intensity at the center8. When the building platforms descend to make contact with the tub, one side of the floor platform that holds the tub drops.
P RINTING M ATERIAL
Plan to start with a simplified design to verify the process and make it easier to attribute difficult-to-manufacture geometries. The idea behind this was if a part gave good results, then the design was modified closer to the end goal by adding another design element to the geometry. At first a simplified nozzle design was used, two models were produced to check the process capabilities.
All corners and edges on all designs had small fillets on them to make them rounded. The designs worked to maintain the same wall thickness, or as close as possible, throughout the section. LithaLox alumina particles mixed with DI water being observed in an ESEM in Low Vac mode.
For this method, a line of known length, Lline, is drawn across the image and the grain boundaries that intersect the line are counted. The grain size was calculated for each line and then the average of the six calculated sizes was calculated. Lines drawn with this over alumina particles represent the layout of the six lines used to determine the average grain size via the intercept method.
P RODUCING S AMPLES AND S ELECTING THE G EOMETERY
First Design
Due to an insufficient layer of adhesive film that was applied to the platform, the parts got stuck on the construction platform. When cleaning the parts, it took longer to form the can, as the flat top, perpendicular to the wall, made it difficult to remove the excess slurry. While cleaning the test discs, some were found to have pits or voids on the surface of the disc.
The parts would also be inspected using a flashlight to shine light through the part to illuminate any cracks. While light was shining through the part, if there was a crack or defect, a difference could be seen in the light. Inspection of parts with light was best when the parts were sintered, as light is able to pass through the sintered material better.
It was harder to shine light through a part that had just been printed or decomposed, but with a bright flashlight and a dark room it would work. No cracks were visible in the parts, but there were a few issues during decomposition. Occasionally, part of the top portion of the can design, the flat top, would delaminate during the decomposition process.
Second Design
Parts are printed in the x-direction with the flared edge based on contact with the build platform. It could be printed without problems, but after the bonding process almost all the parts would crack around that area. The problem with printing flared curves on parts is that during the bonding process the part loses the binder that up until this point held the structure together.
In the decomposition fire, the parts are heated to 1100°C (2012°F), this temperature burns the organic binder but also causes the part to shrink towards the end of the firing schedule. The combination of the two results in forces being applied that the geometry cannot handle, causing the part to break in the curved region to relieve the stress. With the parts resting flat on an alumina plate, there are a few problems, including that 1) air is not allowed to flow out evenly from the inside of the part, and that 2) if the part shrinks there drag forces on the flared lip are caused. .
The weight of the part acts downward on the edge and as the part shrinks, it slides inward onto the aluminum oxide plate. The combination of these two problems causes the part to crack at least half the time. It shows the forces that act on the geometry during the parting process and cause the stress that causes the part to break.
Third Design
N ON -D ESTRUCTIVE O PTICAL I NPSECTION
Inspecting with Light
The third version of the Bullet Design that was produced. illuminate a specific part of the part and prevent the light from washing out the cracks making them invisible. When the nose was placed directly over the flashlight, the light would only be directed into the interior of the geometry.
Inspecting with Ink
The ink was able to flow into the crack near the bottom of the sample, which allows us to visually see it. Shows a sample that was thermally shocked at ΔT of 500°C and then coated with black ink.
C ORRECTING FOR P ART D EFECTS
Regulating the Temperture of the Print Environment
By increasing the temperature, the goal was to decrease the viscosity of the slurry, especially the photopolymer contained in the slurry. The decrease in surface tension allows any air bubbles that form to move more easily between the slurry. This gives a greater opportunity, if any air bubbles are formed, to escape from the cross-section of the part when the build platform comes into contact with the vessel.
To determine the temperature to which the printing envelope should be heated, several prints were made at different temperatures. A mount was 3D printed to hold all the components together and make it easier to attach the unit to the wall of the print envelope. While the heater is mounted halfway up the wall to keep it out of the way while printing, the thermometer for the unit hangs lower near the drum to make sure the slurry temperature is affected.
When the heater is placed in the center of the print envelope, the fan circulates the air inside the print envelope and keeps the entire space warm. The large cross-section was more affected by the hooks than smaller sections of the cross-section. Among the various temperatures used, 26.6 °C (80 °F) was chosen to be a good temperature for printing.
Using a Setter Plate
T ESTING S AMPLES
- Designing and Producing a Testing Jig
- Pressure Tests Early On
- Thermal Shock
- RESULTS AND DISCUSSION
The polycarbonate is attached to the inside of the metal cover with a screw, washer and nut. Since the box will be exposed to high pressure, relief holes are drilled in the back of the metal box. The small diameter holes are drilled along the bottom edge of the box to remove the pressure.
A hole is drilled in the shell so that the body of the geometry can pass through, but then catches on the flared lip. The flared washer was in contact with the underside of the flared lip to provide a seal at the bottom of the geometry. An O-ring is placed at the top of the flared lip to provide a seal at the top of the geometry.
An exploded view of the parts used to secure the nozzle during the pressure tests. The final version of the test jig used to pressure test the ceramic parts. The parts produced for this work will experience a force on the inside of the geometry that comes from gas under pressure.
C HARACTERICATION OF S AMPLES
TGA
The specimens were inserted into the test fixture and pressure tested at 3.4 MPa (500 psi), new specimens were tested at 6.9 MPa (1,000 psi).
Examining Microstructure
During this process, a vacuum is drawn and filled with argon gas to ensure that an even layer is applied to the surface of the sample. As a result, a portion of the detached uncoated sample was mounted and re-examined without gold coating. The higher potential allowed electrons to flow through the top layer of the sample and reduced the amount of charging that occurred on the surface.
While a majority of the grains are tightly packed, some porosity can be seen throughout the grains. The sintered sample contains particles that are significantly larger than the initial size of the particles. Due to the high temperature of the combustion, the grains, as seen in Figure 24, grew a large amount.
Due to the high firing temperature, 1,650°C, of the sample during the sintering process, it is not surprising that the grains have grown in size. The fracture surface of the broken samples was examined using ESEM (Figure 25, Figure 26, Figure 27) to see if there was any variation in the sample microstructure. A piece of each of the samples was examined and while looking at the crack surface, no difference could be seen as the microstructure was similar in the three samples.
P RESSURE AND T HERMAL T ESTING
SUMMARY AND CONCLUSIONS
While undergoing the thermal process to remove the organic binder, the part cracked. It was seen that the use of a curing plate, produced in the same manner as the desired part, prevented the sample from cracking during the debonding process. The part was placed on the setter plate for post-processing thermal treatment and resulted in the successful production of the part.
The combination of the solenoid valve, controlled via Arduino, with the industrial grade nitrogen tank allowed us to quickly and accurately pressurize the samples. This indicates that critical damage has occurred after the cross section of the wall has broken.
FUTURE WORK
Zhao, J.; Oh, X.; Deng, J.; Wang, Z., A model of the thermal shock resistance parameter for functional gradient ceramics. Koyama, T.; Nishiyama, A.; Niihara, K., Effect of grain morphology and grain size on the mechanical properties of Al2O3 ceramics.
