*2.2. TMA Experiments*

TMA experiments are typically conducted to investigate the melting behavior of ashes from solid fuels [18,26]. This method is known to improve the reproducibility of experiments by removing the e ffects of operators that may be present in conventional ash fusion tests such as ASTM D1857. This method provides two important types of data that can be used to identify the slagging propensity of a fuel. The first data type is the penetration tendency of the ram into the sample, which is expressed as a function of temperature. The second data type is the penetration rate according to temperature, which is expressed as "peak". The peak denotes the maximum penetration rate for a given temperature change. This peak indicates the temperature range within which the rapid melting of the ash sample occurs, and it is derived from the first derivative of the penetration trace [27].

Figure 1 shows a schematic of the TMA used in this study. The TMA apparatus is composed of a heating chamber and penetrating rod, as well as a ram, crucible, linear variable di fferential transformer (LVDT), and thermocouple. The experimental procedure used in this study was as follows:

**Figure 1.** Schematic of the thermomechanical analyzer (TMA) apparatus.

Approximately 200 mg of an ash sample was placed in a crucible and the top of the sample was flattened with a jig by applying a constant pressure of 260 kPa. The sample was arranged in the prepared sample assembly by setting the pressure between the ram and the flattened sample surface at 30 kPa with a load of 60 g. The sample was then placed in a furnace. The sample was heated in a high-purity N2 atmosphere. The furnace was first heated from room temperature to 600 ◦C at 50 ◦C/min, and then slowly heated at 5 ◦C/min from 600 ◦C to 1600 ◦C.

As shown in Figure 1, the ram sinks as the ash melts during the heating process, and the melted ash flows between the ram and the empty space in the bottom of the crucible when the ash is fully melted. The sinking displacement of the ram is measured with a laser displacement meter. This displacement indicates the height change of the ash sample from room temperature to a specific temperature and is converted to % shrinkage to evaluate the penetration of each sample according to temperature. The % shrinkage is used to identify the ash fusibility, which can be used as a predictor of ash deposition.

In addition, the peak at a specific temperature is determined from the first derivative of the % shrinkage. Generally, the peak distribution of a solid fuel shows more than two peaks, with each peak indicating rapid melting, i.e., melting acceleration intensity [28]. In this study, the peaks obtained from the shrinkage curves were di fferentiated at intervals of 10 min (or 50 °C) to prevent data scattering.

Gupta et al. [28] noted that temperatures of T25%, T50%, T75%, and T90% exhibit representative melting characteristics among the shrinkage values measured by the TMA in the melting of ash. As the label suggests, T25% refers to the temperature at which the ash sample reaches 25% shrinkage. In this case, the 25% (±15%) liquid phase appears by the softening and sintering of ash. Specifically, it can be regarded as the cause of the initial ash deposition growth because the ash particles are

sticky. At T50%, the ash has shrunk by 50% and approximately 60% (±15%) melting occurs. At T75%, over 80% melting occurs, and this temperature is judged to denote complete melting. It is known that boilers must operate at temperatures lower than T75% to allow the discharge of fly ash without it melting. Finally, T90% is regarded as the final stage of melting (liquid phase > 90%). This temperature represents the slag flow (or fluid flow) characteristics of the ash.
