**1. Introduction**

The burning of fossil fuels is a major cause of climate change owing to the massive greenhouse gas (GHG) emissions it entails. Many countries signed the Paris Agreement at the UN Climate Change Conference in 2015 with the aim of reducing GHG emissions. Consequently, these countries are developing alternative and renewable energy sources for fossil fuel replacement [1]. Biomass continues to grow in importance worldwide as a renewable and CO2-neutral energy source that may help to diversify renewable energy sources for energy production [2].

The South Korean governmen<sup>t</sup> has announced its intention to reduce greenhouse gas emissions by 37% before 2030 [3]. To achieve this target, the South Korean governmen<sup>t</sup> has recommended the use of new and renewable energies for power generation, such as biomass, by the relevant companies at the national level through the renewable portfolio standard (RPS). Furthermore, the governmen<sup>t</sup> plans to increase the supply of new and renewable energy fuels from 3% in 2015 to 10% by 2023 [4].

Biomass co-combustion or co-firing with coal is a potential strategy to achieve GHG abatement and reduce toxic emissions. Many previous studies have demonstrated that biomass co-firing can reduce net CO2 emission because of the carbon neutrality of biomass, as well as decreasing NOx and SOx production [5,6]. This technology is favorable for many power plants because the fuel can be easily used in existing powdered coal-powered boilers without entailing any environmental or economic concerns [7,8].

Although biomass co-combustion and co-firing have the advantages of simple equipment configuration, low cost, and enhanced combustion performance, biomass fuels usually contain high levels of alkali and alkaline metals, particularly Na and K, which increase slagging and fouling in biomass-fired boilers [9].

For example, Baxter [10] studied ash deposition and slagging/fouling during the combustion of coal and biomass, and postulated a mechanistic model and ash deposition characteristics for biomass combustion. According to his results, the ash deposition phenomenon is associated with the combustion conditions and the type of inorganic material used for fuel co-combustion. He found that ash deposition by biomass peaks during initial combustion and then gradually decreases.

Pronobis [11,12] investigated the effects of co-firing a medium-level fouling coal with three types of biomasses on surface fouling in the convection region of a furnace. He revealed that the properties of the biomass fuel affect both the operation variables and the efficiency of the boiler. Furthermore, he co-fired two types of bituminous coals with different slagging tendencies and four types of biomasses (straw, wood, dried sewage sludge, and bone meal) and examined the slagging effect in the furnace in terms of the correlation between the properties and fusibility of the produced ashes. The results showed that co-firing increases the risk of slagging at the fireside in furnaces.

Theis et al. [13] co-fired peat and two biomass types (bark and straw) using an entrained-flow reactor and compared the resulting ash deposition with that of single-fired peat. They reported that the ash deposition rate does not increase, even when the bark and straw fuels are co-fired with peat at levels of 30 wt% and 70 wt%, respectively.

Savolainen [14] researched the slagging behavior of boilers by measuring and monitoring the soot-blowing frequency and attemperation of water flow. She reported that slagging and fouling in the furnace were maintained at normal levels despite co-firing with biomass.

Abreu et al. [15] co-combusted bituminous coals with two types of biomasses (pine sawdust and olive stones) at 10–50 wt% (according to calorific value) and examined the ash deposition rate. Co-combustion sawdust with a low alkali content led to a lower ash deposition rate than that observed for single-firing of coal. In contrast, co-combustion with olive stones with high K contents resulted in a higher ash deposition rate than that for single coal. They posited that this difference is caused by the different adhesion tendencies on the deposition surface and attributed this effect to the ash compositions of the biomasses.

In addition to the above studies, numerous studies have been performed to understand and explain the phenomenon of ash deposition in biomass co-combustion. This includes the development of empirical indicators and several experimental methods for determining ash melting temperatures [16–18]. Many investigations using thermomechanical analyzers (TMAs), drop tube furnaces (DTF), pilot plants, and full-scale boiler trials have been performed [19–21].

These previous studies have demonstrated that slagging and fouling propensities vary, particularly in response to type of blended fuel used. Recently, the use of oil-palm empty fruit bunches (EFBs) as biomass has been attempted in power plants in Korea for economic reasons. Thus, optimal operation parameters for the application of EFBs in power plants must be derived. Oil palms are widely cultivated in tropical Asia, especially in Malaysia, Indonesia, and Thailand. Considering this abundant and

CO2-neutral fuel resource, oil-palm EFBs, as byproducts of the crude palm-oil milling process, represent one of the most promising energy resources [22,23].

There is much less literature available on the ash deposition behavior of EFBs compared to that on wood pellets (WPs) [24,25]. Thus, the goal of this study was to perform a detailed investigation into ash deposition during the co-combustion of pulverized coal with EFBs and WPs in pulverized form.

A bituminous coal from Australia with a relatively low slagging propensity was chosen as a representative fuel that is commonly used in Korean thermal power plants. The alterations in the extents and mechanisms of coal slagging were investigated during co-combustion with the two biomass types (WP and EFB). The ash melting characteristics and deposition rates were analyzed using a TMA and a DTF. In addition, X-ray fluorescence (XRF) analysis was performed to examine the chemical compositions of the ashes and derive their empirical prediction indices.

#### **2. Materials and Methods**
