For the primary hypothesis, there were two parts to explore the densification of glass particles as a function of viscosity. The first aspect was the understanding of the densification of glass particles as a function of temperature.
Introduction to Sintering
Processes and Mechanisms
D = Do(−RTQ) (1) where T is the temperature, R is the gas constant, Q is the activation energy required for an atom to jump from one place to another, and both Do and Q are constants for a particular system .3 Defining diffusion as a thermally activated phenomenon is a valuable concept for both crystalline and noncrystalline systems, even though temperature dependence and other factors vary from system to system.
Solid State Sintering
Liquid Phase Sintering
The solid grains in the liquid matrix are effectively dissolved and transported through the liquid phase. The rate of rearrangement that occurs during LPS depends on the degree of wettability and can be effectively explained by capillary forces.
Sintering Glass
Viscous Flow
Liquid phase sintering (LPS) and supersolidus liquid phase sintering (SLPS) are both dependent on viscous flow, but not exclusively. solid phase. Other models for viscous flow include Absolute-Rate Theory, where viscous flow is described as a rate process dominated by transitions to high-energy states, and Excess Entropy Theory, where a decrease in configurational entropy (decreasing temperatures) leads to deformation difficulties. .12,13 Both these theories and related models express viscous flow in terms of viscosity and temperature.
Crystalline and Non-Crystalline Materials
The glass transition, more practically referred to as the glass transition region, is not a phase transformation. Although it is common for many materials to undergo phase transitions over a range of temperatures, the glass transition is different.
Glass Viscosity
The structural randomness of glass over long distances allows interstitial displacements that increase in frequency and magnitude with increasing temperature.15,16 This nonlinear relationship with temperature gives way to a glass transition temperature (Tg) that is unique to glassy systems. The glass transition region is when a liquid leaves the subcooled equilibrium line (S.C.E.L.) on cooling and becomes a non-equilibrium system, which upon further cooling becomes a glass.17,18 Because glassy systems have characteristic glass transition regions, all glass systems have equilibrium contributions at high temperatures and non-equilibrium contributions at low temperatures. The most popular methods are beam bending viscometry (ɳ Pa∙s), parallel plate viscometry (ɳ Pa∙s), fiber elongation (ɳ Pa∙s) and cylinder rotation viscometry (ɳ ~ 1-2 Pa∙s).24 Beam bending viscometry (BBV) is has become increasingly popular because samples are relatively easy to prepare and BBV can be performed at temperatures below Littleton's softening point.25,26 For the purpose of this work, BBV was selected not only for the simplicity of sample preparation, but also because it covers the viscosity range around examine the glass transition temperature.
Log viscosity-temperature curve depicting characteristic temperatures and respective experimental methods for soda lime silicate glass (modified from Shelby).13,22. At low temperatures, a glass is perfectly isostructural, and its viscosity can be described in Arrhenius form.18 The idea of the brittleness parameter was first proposed by Angell who devised a way to classify liquid systems as either strong or brittle. .19 Angell noted that for certain composition families of glasses, general patterns of behavior became apparent at and above the glass transition temperature (Tg). Cp near Tg.13,17 It is worth noting that this change in heat capacity is unique to the glass transition region.
Curves were generated using Avramov-Milchev (AM), Vogel-Fulcher-Tammann (VFT), and MYEGA to show that regardless of the fit used, they all agree on the glass transition region. Viscosity model equation using log-viscosity-temperature relationship curve for soda lime silica (SLS) container glass (calculations by Oistad22).
Densification Investigation
- Property Characterization
- Surface Area Impact
- Sintering
- Hot Stage Microscopy
For the second sintering test, glass powders were dry pressed into pucks (3 grams at 21 MPa or 3,000 psi), quartered and sintered with a residence time of two hours. For the second compaction test, the gradient oven was closed on the loading side and the hood was secured with two type K thermocouples. The pressed samples were placed between the two thermocouples during baking to ensure a more accurate sample temperature.
After the oven was allowed time to equilibrate to the chosen set point, the samples were introduced to temperature. At higher temperatures, the samples were baked on platinum plates to prevent possible chemical interaction with the glass. The porcelain crucibles could only withstand temperatures up to 1150°C and the maximum operating temperature for the low temperature gradient furnaces was 1200°C. Therefore, fused silica samples were fired in a bottom loading oven.
The fused silica samples were dry pressed and fired in a bed of thick alumina at a temperature range of 1350–1475°C. A hot phase microscope (Misura ODHT, Expert Systems, Modena, Italy) was used to determine the glass transition (Tg) and sintering temperature (Ts) targets for densification investigation.
Viscosity
Beam Bending Viscometry
BBV samples were cut from the bulk of the glass using a diamond table saw for rough cuts, and then a low-speed diamond saw for precision cuts.
SciGlass Calculations
Glass compositions in weight percentage (oxide content less than 0.15 weight percentage was not included: Fe2O3, TiO2, P2O5, MnO, Cr2O3).
Densification Definition
Hot Stage Microscopy
Target temperatures were determined by first limiting the temperature range and second by averaging apparent Tg and Ts values given by each cycle. Uncertainties are due to variation between target temperatures determined for each cycle and 11 cycles) for a test run. A comparison of these calculated HSM values and those measured experimentally is discussed in the next section.
Example of HSM data analysis for Frit 3110: Example for a 5-cycle value, five temperatures and corresponding surface variance values used to calculate the running slope are detailed. Glass transition and sintering temperatures are calculated from HSM data using run slope and the INDEX function in Excel. Examples of HSM images for each glass frit (heated at 20K/min to 1200°C, unless otherwise noted).
Microstructures
However, HSM temperatures were still valuable as target temperatures, as the measured densification temperatures and associated sintering viscosities fell between calculated Tg and Ts, but slightly closer to Tg, as shown in Figure 16. Alkali borosilicate glasses tend to phase separate and form flux-B2O3- and SiO2-rich phases, which can occur when a glass is heat-treated to temperatures within the metastable liquid immiscibility region.21,32 Backscattered electron imaging via SEM has "splotching" revealed. Phase separation can lead to discrepancies between measured and calculated sintering temperature, the assigned sinter viscosity and the actual viscosity of the glass.
Sintering Viscosity
The average sintering temperatures for the glasses in Group I and Group II are given in Table XII and Table XIII, respectively. Still, since glass is a non-equilibrium system, any correlations between composition and sintering viscosity must be critically analyzed, especially since several external factors can affect sintering temperatures. Many studies have shown how the degree of water absorption into and adsorption onto glass powders varies depending on the glass chemistry.15,21,33 Water interacts with the surface of a glass particle forming hydroxyl groups, causing the glass to become mobile alkali and divalent lose ions.
These changes in the surface chemistry (or substrate, depending on the amount of time allowed for the interaction) can alter the viscosity-temperature curve of a glass.33,34 It is interesting to note that thermogravimetric analysis (TGA) for one glass single water loss between 100-600°C, but the data was generally inconclusive due to sensitivity limits. It has been shown that chemically bound water impurities tend to lower the glass transition temperature.21,33,34 The difference in sintering viscosities between Group I and Group II may be related to preferential water interactions. The difference between sintering temperatures calculated using HSM data and experimentally determined sintering temperatures also indicates a water adsorption problem.
Large-scale image shows the V-T curves associated with one standard deviation of the average sinter viscosity for each group. Impact of (a) silica content, (b) boron oxide content and (c) alumina content on the sintered viscosity of the glass system.
Surface Area Impact
34;Measurement of Viscosity of Glass Between Softening Point and Annealing Range (Approximately 108 Pa∙s to About 1013 Pa∙s) by Beam Bending (Metric),” ASTM Designation C 1350M-96. Olevsky, “Current Research Directions” at Onset of the Next Century of Sintering Science and Technology,” J. Throughout this work, the discussion of the most important factor affecting densification of glass is the characteristic viscosity.
In addition to initial sintering trials of different wood times, a design of experiments (DOE) was conducted to determine a necessary or appropriate amount of time to perform successful isothermal sintering of glass frit in a custom tube furnace. The DOE looked at three factors that affect heat dissipation: depth, position and equilibration time (see Figures 33 & 34.) The position factor was found to be insignificant - the temperature a sample would experience either near the top of the D- tube (a short sample) or near the top of the tube furnace (a long sample), was relatively the same. The depth factor was found to be significant - the temperature of the furnace varied non-linearly with the distance from the front mouth of the furnace.
Overall, time was not found to be a significant factor during the viscous sintering of glass frit, as has been shown by others.36 However, this DOE showed that the furnace setup required a minimum of 45 minutes at temperature to stabilize. It is interesting to note that there were no observable microstructural differences between glass frit sintered at a given temperature for 60 minutes compared to 2 hours, the only difference being the amount of glass frit sintered in each sample, partial and complete, respectively. a) Pareto chart showing that depth (distance from the front of the furnace) and furnace equilibration time are significant factors, and (b) a box plot showing that temperature uncertainties at higher temperatures in the hot zone decrease dramatically.
