A critical review of ash slagging mechanisms and viscosity measurement for low-rank coal and bio-slags

Abstract

Gasification or combustion of coal and biomass is the most important form of power generation today. However, the use of coal/biomass at high temperatures has an inherent problem related to the ash generated. The formation of ash leads to a problematic phenomenon called slagging. Slagging is the accumulation of molten ash on the walls of the furnace, gasifier, or boiler and is detrimental as it reduces the heat transfer rate, and the combustion/gasification rate of unburnt carbon, causes mechanical failure, high-temperature corrosion and on occasions, superheater explosions. To improve the gasifier/combustor facility, it is very important to understand the key ash properties, slag characteristics, viscosity and critical viscosity temperature. This paper reviews the content, compositions, and melting characteristics of ashes in differently ranked coal and biomass, and discusses the formation mechanism, characteristics, and structure of slag. In particular, this paper focuses on low-rank coal and biomass that have been receiving increased attention recently. Besides, it reviews the available methodologies and formulae for slag viscosity measurement/prediction and summarizes the current limitations and potential applications. Moreover, it discusses the slagging behavior of different ranks of coal and biomass by examining the applicability of the current viscosity measurement methods to these fuels, and the viscosity prediction models and factors that affect the slag viscosity. This review shows that the existing viscosity models and slagging indices can only satisfactorily predict the viscosity and slagging propensity of high-rank coals but cannot predict the slagging propensity and slag viscosity of low-rank coal, and especially biomass ashes, even if they are limited to a particular composition only. Thus, there is a critical need for the development of an index, or a model or even a measurement method, which can predict/measure the slagging propensity and slag viscosity correctly for all low-rank coal and biomass ashes.

Key words： slag; viscosity; biomass; low-rank coal; combustion; gasification

Introduction

Fossil fuels account for approximately 85% of the primary energies generated today, and around 38% of the world’s electricity generated is derived from coal [¹,²]. However, due to the exhaustion of fossil fuel and increasing awareness about environmental issues, the appeal for renewable energy sources has been increased over the years [³]. One of the primary sources of renewable energy is biomass, as it can conceivably produce heat and power [⁴,⁵]. For heat and power generation, gasification and combustion of coal/biomass are known as the most sophisticated methods, of which the most commonly used process is the pulverised-coal fired boiler and fluidised bed or entrained bed for gasification [⁶–⁸]. For all these technologies, the ash generated is a critical issue that can lead to a variety of problematic phenomena such as slagging and fouling inside the furnace [⁹,¹⁰].

The phenomena of slagging refer to the accumulation of partially or fully molten ash on the furnace walls of a boiler/gasifier or the transmission exterior disclosed to radiant heat [¹¹–¹³]. In the case that melted or soften ash particles are not cooled down to a rigid condition when they reach the heated surface, slag is formed [¹⁴,¹⁵]. Typical ash fusion temperature or initial deformation temperature (IDT) of ash is polarized either at 1150°C to 1200°C, or at 1350°C or above. Once molten, the resultant slag can stick to comparatively cooler walls [¹⁶,¹⁷]. Because of relatively lower temperature at the tube surface or the reactor wall, most of the molten ash tends to be re-solidified [¹⁸]. However, the molten ash may not have enough time to be re-solidified in the case that the ash melting point is relatively low, the size of the furnace is too small, or the exit gas temperature is too high. Consequently, they prefer to stick to the heated surface and cause the build-up of deposits, which eventually leads to slagging [¹⁹]. From a microscopic point of view, it largely depends on properties of individual ash particles, their aerodynamic behavior, and the surface properties of existing slags [¹⁸,¹⁹] whether an ash particle can be captured by heat transfer surfaces. The capture of particles is somewhat selective, depending on which stage they fall into [¹⁸,¹⁹].

The slagging and fouling propensity of ash is directly derived from both the quantity and the quality of the inherent ash-forming elements within a feedstock [²⁰,²¹]. Both aspects are affected mainly by the nature and source of the biomass/coal adapted [²²,²³], along with the operating conditions and configuration of the combustor/gasifier [²⁴,²⁵]. Specifically, the biomasses/coals are rich in some mineral components such as alkali elements which are responsible for the formation of low-melting-point eutectics which preferentially deposit inside a combustor [²⁶–³⁰]. On the other hand, during gasification of biomass/coal, the carbonaceous matrix of biomass/coal is gasified into synthetic gas, and the mineral compounds present in the matrix mostly convert molten ash/slag under the strong reducing environment [³¹–³⁶]. The resultant liquid slag deposits either on the membrane or on the refractory wall of the gasifier [³⁷]. These liquid slags usually flow down from the downside of the gasifier and consolidate in a vessel containing heated water [³⁶]. However, a fraction of molten ash moves with the synthetic ash and deposits in the heat recovery zone and blocks it [³⁸–⁴⁰]. Viscosity is the governing factor for slag stickiness.

Viscosity determines whether or not an ash particle or its respective melt/slag sticks, which is the most crucial property affecting the slagging, fouling, and deposition within a combustor or a gasifier [⁴¹]. Viscosity of slags is a complex function of the slag composition, temperature, and oxygen partial pressure in the system. It also reflects their flow properties and their tendency to capture incoming particles [⁴²]. At low temperatures, a slag may solidify if it has a high viscosity and eventually blocks the gasifier. Thus, a higher gasifier temperature is usually required compared to the ash melting temperature to assure uninterrupted slag flow. Moreover, the slag viscosity should be less than 25 Pa·s (250 poise) for an easy flow [⁴¹]. At low temperatures, the refractory lining on the wall of a gasifier behaves as a thermal obstacle and guards the wall. However, at a higher temperature, the molten slag can penetrate or corrode the refractory lining [³⁵,⁴²]. Due to the penetration/corrosion in the refractory lining, the properties and microstructures of the refractory change rapidly, leading to the cracking of the wall [⁴²]. Hence, to improve the efficiency of a combustor or a gasifier, it is essential to understand the key ash properties such as ash fusion temperature, critical viscosity temperature, viscosity, optimum operating temperature, and slag flow characteristics.

This paper briefly reviews the state-of-the-art papers regarding the characteristics and the melting/slagging propensity of differently ranked coal and biomass ash. Both combustion and gasification are included since they both employ a high temperature, and a reducing environment is always present around the carbonecous matrix that hosts the minerals. Besides, starting from the characteristics and compositions of slag, it discusses the methodo-logies and formulas available for slag viscosity measurement/prediction and summarizes the current limitations and potential applications when these established models/formulas are extended to low-rank coal and biomasses. Moreover, it intends to touch the base of the slagging behavior of differently ranked coal and bio-slag, and elaborate on the knowledge gap and new research areas required for the study on biomass slagging.

Characterization of different ash and slag

Ash content in coal and biomass

Depending upon the formation conditions, handling process and mining techniques of coal, the ash content in coal can vary remarkably [⁴³]. Similarly, based on the plant species, growing conditions and stage of growth, the composition of biomass ash also differ remarkably from one another [⁴⁴,⁴⁵]. Figure 1 shows the relationship between the overall ash content and volatile content of different ranks of coal and biomass. As a general rule of thumb, the ash content increases with increasing volatile contents for a decrease in the coal rank and upon the change from coal to sewage sludge. However, the ash content in each coal rank varies broadly. For the highest rank anthracite coal, the ash content varies from 1 wt% to 20 wt% [⁴⁶] whereas it varies from a few percentages up to almost 40 wt% for bituminous coals. Moreover, the most substantial variation in ash content is found for the sub-bituminous coal or brown coal, from a few percentages to nearly 50 wt%, and in some extreme cases, even up to 68 wt% [⁴⁷].

The biomass ash content is interesting, demonstrating the least ash amount, although its volatile content reaches approximately 80 wt%. More interestingly, the ash content in biomass is even comparable to anthracite that is the highest rank coal. It is also comparable to the low-end of bituminous coal, sub-bituminous, and even brown coals. Moreover, the variation of ash content with biomass type is remarkable. According to Vassilev et al. [⁴⁸], the bio-ash content varies between 0.1 wt% and 46 wt% for different bio-ashes. In general, the bio-ash content follows a descending trend of animal biomass (AB)>contaminated biomass (CB)>herbaceous and agricultural straw (HAS)>herbaceous and agriculture biomass (HAB)>herbaceous and agricultural residue (HAR)>herbaceous and agricultural grass (HAG)>wood and woody biomass (WWB) [⁴⁸]. Except for AB, CB, and some other HAB, the ash generation rate in biomass was found to be much lower compared to coal. AB and CB have a large amount of ash because they are mainly comprised of chicken litter, meat and bone meal, sewage sludge, solid refuse fuel, plastic waste, etc. Additionally, due to the nutrients take-up during the growing period, HAR has more abundant ash content compared to WWB [²²,⁴⁹,⁵⁰]. Of the WWB components, wood has the lowest ash content, followed by bark and foliage [⁵¹].

Fig.1 Relationship between ash and volatile content for different rank coal and biomass (adapted from Ref. [⁴⁶] with modification. Note that d denote dry basis, and VM^daf denote volatile matter content on the dry and ash free basis).

Full size|PPT slide

Composition and characteristics of coal and biomass ash

Table 1 lists the averaged elemental compositions of typical coal samples (ashed at 815°C) and different biomass ashes groups (ashed at 550°C–600°C) based on high-temperature ash analyses. Depending on the coal rank, the compositions of coal ashes can vary remarkably. The major oxides found in coal ashes are SiO₂, Al₂O₃, SO₃, CaO, and Fe₂O₃ while the remaining oxides are below 10% in total [⁴⁶]. Usually, a higher amount of alkaline earth metals is found in the low-rank coal ashes such as lignite, brown coal, and sub-bituminous coals. The amount of alkaline earth metals can be even higher than that of iron (CaO+ MgO>Fe₂O₃) in some of the low-rank coal ashes [⁵²]. Moreover, low-rank coals are relatively rich in iron and calcium. As a result, their ashes have a lower melting temperature compared to those of other coals [⁵³]. Depending on the ash content, the amount of silica in ash also varies. Usually, high ash content refers to a dominating percentage of silica within the ash [⁵³].

Tab.1 Average elemental compositions of different biomass and coal ashes based on high-temperature ash analyses

Ash	Rank/Group	SiO2	CaO	K2O	P2O5	Al2O3	MgO	Fe2O3	SO3	Na2O	TiO2	Reference
Peat	Lowest rank (precursor of coal)	37.53	9.97	1.12	2.75	20.14	2.14	13.83	12.11	0.10	0.31	[48,62]
Lignite	Low rank coal	44.87	13.11	1.48	0.20	17.11	2.50	10.80	8.64	0.48	0.81	[48,63,64]
SBC	Medium/Low rank coal	54.74	7.05	1.67	0.08	22.86	2.14	5.30	4.07	1.09	1.00	[48,63,64]
VBC	Low rank coal	26.90	6.00	0.30	–	8.60	14.30	20.00	17.10	6.50	0.50	[65]
XJC	Low rank coal	19.13	34.34	0.68	–	8.44	9.37	7.67	13.72	5.45	0.53	[66]
BC	High rank coal	56.14	4.90	1.61	0.22	24.82	1.55	6.68	2.16	0.77	1.15	[48,63,64]
Anthracite	Highest rank coal	53.50	3.40	4.90	0.05	27.60	2.10	6.00	–	1.0	1.00	[67]
WWB	Woody biomass	22.22	43.03	10.75	3.48	5.09	6.07	3.44	2.78	2.85	0.29	[25,48,68–72]
HAB	Herbaceous and agriculture biomass	33.39	14.86	26.65	6.48	3.66	5.62	3.26	3.61	2.29	0.18	[48]
HAG	Herbaceous and agriculture biomass	46.18	11.23	24.59	6.62	1.39	4.02	0.98	3.66	1.25	0.08	[48,71–73]
HAS	Herbaceous and agriculture biomass	43.94	14.13	24.49	4.13	2.71	4.66	1.42	3.01	1.35	0.16	[23,48,71,72,74]
HAR	Herbaceous and agriculture biomass	24.47	16.58	28.25	7.27	4.90	6.62	4.84	3.80	3.05	0.22	[48,69,71,75]
AB	Animal biomass	2.90	49.04	7.67	28.17	1.69	2.75	0.35	3.91	3.50	0.02	[23,48]
CB	Contaminated biomass	35.73	18.30	3.45	3.64	15.41	3.60	9.78	3.45	1.90	4.74	[23,48,76]

Notes: SBC–sub-bituminous coal; VBC–Victorian brown coal; XJC–coal of Xinjiang Uygur Autonomous Region; BC–brown coal

More specifically, according to the content of individual elements, the coal ash can be further divided into:

1)Medium alkali content: Polish, Russian, and Colombian coal ash belong to this group. Additionally, these ashes can also be rich in calcium and iron content, and hence, have a higher slagging and fouling propensity. Higher chlorine content may also be found in this type of ash. For example, Polish coal ash is rich in chlorine content [⁵⁴,⁵⁵];

2)High calcium but low alkali content: South African coal ash is the example of this type of ash which is considered as good quality ash and usually used for blending with alkali-rich ash to reduce the slagging propensity [⁵⁶,⁵⁷];

3)Sulfur- rich coals: Due to the presence of pyrites, this type of coal, also rich in iron, has a significant slagging propensity [⁵⁷–⁵⁹];

4)High alkali, iron and calcium contents: This type of ashes has high sintering but low deposition tendencies. The brown coals, including Indonesian brown coal and Victorian brown coal, are examples of this type of ash [⁶⁰,⁶¹].

Table 1 also demonstrates the broad variation of the elemental composition of bio-ashes. Based on Table 1, Fig. 2 further qualitatively outlines the contents of the principal oxides in each different bio-ashes in a downward trend [⁴⁸]. By observing these patterns, it is clearly seen that a substantial difference exists among the order of these oxides between different biomass groups. However, a notable similarity exists among the sub-groups, especially the HAG and straw. Consequently, bio-ashes can be broadly divided into three major categories, those rich in Si/K, including HAG, straw and residue; those rich in Ca including wood and woody residue, and CB; and those rich in phosphorous (P) which is mainly AB [⁷⁰].

Fig.2 Mean contents of the principal oxides found in biomass ash in a downward trend (Note that WWB stands for wood and woody biomass; HAG for herbaceous and agricultural grass; HAS for herbaceous and agricultural straw; HAR for herbaceous and agricultural residue; AB for animal biomass, and CB for contaminated biomass.)

Full size|PPT slide

Melting propensity of bio-ash and coal ash

The three different groups for biomass ash possess different melting propensities [⁷⁰].

Those rich in Si/K usually has a low melting temperature. Again, the majority of the agricultural and herbaceous biomasses belong to this group, and K is abundant in this type of bio-ash [⁴⁶,⁷⁷].

Those rich in Ca and even potassium usually has a high melting temperature. Most of the woody biomass fuels, which are known for its low ash content, belong to this group. Few other species of biomass such as rape straw and willow with a high Ca content also belong to this group. Moreover, some of the biomasses in this group can even have a relatively large amount of phosphorous and alkali metals, which considerably decreases the ash melting temperature [⁴⁶].

The melting propensity of those rich in phosphorous (P) is quite intricate, which does not solely depend on the concentration of P, but also depends on the concentration of a few other elements including Ca, Mg, K and even Fe.

Based on the ash composition, either higher melting K-Ca/Mg phosphates or low melting K-rich phosphates may be constituted. Cereal grains are the example of comparatively high sources of P, Mg and K. Most of the animal wastes, poultry litters, and manures are the sources of high P and Ca. In the context of sewage sludge, the presence of P is very abundant. Depending on precipitant agents used during the water treatment process, iron phosphates, aluminum phosphates, and calcium phosphates are formed within the sewage sludge ash, which decrease the melting point of the entire ash considerably [⁴⁶,⁷⁸].

The melting tendencies of coal ashes are rather simple, since plenty of researches have been conducted in this area. Based on the melting point, the coal ash can be generally categorised into low-melting temperature ashes derived from low-rank brown coal or lignite with higher concentrations of Ca, Mg, S, Fe, and Na, which are rich in carbonate, oxides, sulphates, montmorillonite, sulphides, and feldspars as well [⁷⁹]; high-melting temperature ashes derived from high-rank coal including anthracite and bituminous coal with comparatively low concentrations of S, Ca, Mg, Fe, and Na, which are rich in illite, kaolinite, quartz, and rutile as well; and medium melting temperature ashes derived from sub-bituminous coal and other coals that rank between low-rank and high-rank coals [⁷⁹].

Slagging mechanism

Figure 3 portrays the alkali-induced slagging process, silicate melt-induced slagging mechanism, and bio-ash agglomeration during the combustion of biomass. Very similar phenomena take place during pulverised coal combustion [⁴⁶,⁸⁰,⁸¹]. Alkali metals are released as ultra-fine aerosols of hydroxide, chloride, sulfate, and their mixtures during the combustion process [⁸²]. The alkali elements tend to vaporise and condense into submicron fly ash particles by following a series of reactions including nucleation, adsorption, condensation, and chemical reaction upon the decrease in flue gas temperature. A portion of the resultant submicron particles can also turn into a sticky introductory slagging layer on the heat exchange tube surface through thermophoresis and turbulent diffusion. This adhesive layer fills in like a glue [⁸³,⁸⁴]. The alkali metal mist continues to concentrate and consolidate on the outside of the slag, either by structuring a sticky layer or by reacting with the SiO₂ and Fe₂O₃ contained in the fly ash [⁸⁴]. In addition, some soluble base metal vaporisers form respective eutectic mixtures at low temperatures, such as the Na₂SO₄ + NiSO₄ eutectics melting between 670°C and 883°C and KCl+ K₂SO₄ melting at 550°C [⁸⁴,⁸⁵]. In this way, coarse fly ash, with or without an adhesive self-surface layer, is deposited by inertial impaction on the outside of the initial sticky slagging layer [⁸⁶,⁸⁷]. When the underlying slagging layer loses sufficient bond to accumulate further coarse fly ash, a new sticky layer is formed from the accumulation of submicron ash particles rich in alkali metals. The accumulation of submicron ash particles and the capture of coarse ash particles lead to the growth of a multi-layered rotating framework [⁸⁷]. Different from the alkali-induced slagging, the silicate melting led slagging is mainly dependent on Si, Al, and various ash elements. When the furnace temperature exceeds the melting point of the entire ash, it undergoes disfigurement and softening and subsequently attaches by inertial impaction to the heating surface. Some reviews have shown that the IDT improves with reduced contents of K₂O, MgO, CaO, Fe₂O₃, Al₂O₃, and SiO₂ in the ash [⁸⁸]. However, a higher Si/Al ratio lessens the IDT in light of the fact that Al₂O₃ causes a significant increment in the ash fusion temperature compared to SiO₂ [⁸⁹,⁹⁰]. In the interim, the refractory minerals including quartz (SiO₂), metakaolinite (Al₂O₃·2SiO₂·2H₂O), mullite (Al₆O₁₃Si₂), and rutile (TiO₂) elevate the ash fusion temperature, while the fluxing minerals including anhydrite (CaSO₄), calcium silicate (Ca₂O₄Si), hematite (Fe₂O₃) can lessen it [⁷⁹].

Fig.3 Formation mechanism of slagging and agglomeration during biomass combustion (adapted from Ref. [²¹] with permission).

Full size|PPT slide

In the gasifier where a strong reducing environment is more beneficial for ash melting, the molten ash particles tend to initially deposit on the internal walls of the gasification chamber. Afterwards, the wall is covered by a layer of solidified slag, over which the molten slag flows under the force of gravity into a water extinguishing framework from the base of the gasifier [⁹¹]. Figure 4 illustrates the slag formation and deposition inside a gasifier. The ash particles initially cling to the outer surface of the slag and afterwards diffuse into the molten slag. In the case that the molten slag has a high viscosity, it would be solidified locally rather than flowing down, which eventually increases the thickness of the solid slag layer [⁴¹]. Consequently, this phenomenon decreases the overall gasifier efficiency and even leads to operating failure. Thus, for the operational success of a slagging gasifier, the characteristics of the molten/fluid slag layer formed between ash and solid slag are critical and must be precisely identified [⁹²].

Fig.4 Slag formation and deposition inside a gasifier (adapted from Ref. [⁹²] with permission.)

Full size|PPT slide

Characteristics of slag

Slags exhibit an ionic nature. Depending on the oxide contents in the slag, the extent of polymerisation of slags varies. As a result, the viscosities of slags are highly influenced by the extent of ions, electrostatic interactions, and the structure of the slags [⁹³]. Various components are involved in the metallurgical process of the slag system. The most common oxides found in the slag system are SiO₂, Al₂O₃, CaO, MgO, FeO, MnO, and Cr₂O₃ [⁹⁴].

Network formers such as SiO₂, P₂O_5, and B₂O₃ are responsible for high viscosities, as they possess strong and highly covalent metal-oxygen bonds. In the slag system, the alkali and alkaline earth oxides including K₂O, Na₂O, Li₂O, CaO, and MgO work as a network breaker. Besides, divalent oxides like MnO and FeO also work as a network breaker, too. Due to the addition of these network breakers in the slag, the slag is depolymerised, and consequently its viscosity decreases. The extension of depolymerisation is highly dependent on the composition and volume of the network breakers presents in the slag. Moreover, depending on the compositions of the slag, amphoteric oxides such as Fe₂O₃ and Al₂O₃ may act either as a network breaker or a former [⁹³,⁹⁵,⁹⁶].

In a slag matrix, the cations such as Na⁺, K⁺, Ca²⁺ , and Mg²⁺ create ionic non-bridging oxygen bonds (e.g., O⁻-Na⁺ bond) by breaking the covalent bridging oxygen bonds (Si-O bond) [⁹⁷]. In addition, around the non-bridging, they organize themselves in an octahedral orientation. The number of coordination can vary depending on the size of the cation. Generally, the number of coordination increases with increasing the cation size [⁹⁸]. Besides, the movement of a cation is highly dependent on its size. The larger the cation, the easier the movement [⁹⁹].

Depending on their oxidation state, transition metals such as Fe and Cr can differently affect the slag viscosity. The main factors that critically affect the oxidation state are the operating pressure, temperature, and the formation of the slag. Any modification in the above factors can strongly influence the slag viscosity [⁹⁵,¹⁰⁰].

Structure of slag

Based on various research conducted on the slag structure it is imagined that the structure of the slag is a SiO₂ based one [¹⁰¹–¹⁰⁴]. The matric SiO₂ structure in slag is a 3-dimensional array with each Si⁴⁺ consisting of four tetrahedrally organized O²⁻ s, each of which links with two Si⁴⁺ (referred to as bridging O) causing a 3-dimensional (3-D) cluster to develop as shown in Fig. 5(a). In SiO₂, these SiO₄⁴⁻ polyhedral structures are associated with a 3-D polymerised structure as illustrated in Fig. 5(b), and the oxygens are transcendent connecting ones (O^o). Cations, for example, Mg²⁺ , Ca²⁺ , and so on are likely to disintegrate Si-O bonds and de-polymerise the matrix by creating non-bridging oxygen and free oxygen (denoted O²⁻ ), which is not at all linked to Si, but linked to Na⁺ and so forth instead. Different cations, such as Ti⁴⁺ , Al³⁺_, P⁵⁺ may fit into the Si polymeric chain but still need to maintain the charge balance. For example, in the case that an Al³⁺ is consolidated into a Si⁴⁺ chain, it must have a Na⁺ (or a half of a Ca²⁺) near the Al³⁺ to maintain the nearby charge balance as portrayed in Fig. 5(c). Ti⁴⁺ was defined as a network breaker, in this case, based on the outcomes of the viscosity measurements [¹⁰¹]. Smaller cations, such as Mg²⁺, are likely to provide a wider chain length dissemination than bigger cations, such as Ba²⁺. Cations such as Fe³⁺ can behave as network breakers in small fixations; however, Al³⁺ can take part in higher concentrations in the chain along these lines. The level of polymerisation can be determined concerning the numbers of free-oxygen (NO²⁻), bridging and non-bridging (N^Oo and N^O−), and can be denoted by (NBO/T) [¹⁰²]. The term (NBO/T) refers to the ratio of non-bridging oxygen to the tetragonally bonded oxygen. The structure of slag can be described using thermodynamic quantities since thermodynamics gives a depiction of bond strengths [¹⁰³].

Fig.5 Schematic drawing (adapted from Ref. [¹⁰⁴]. Note that here pink sphere denotes non-bridging oxygen.) (a) SiO₂ structure showing 3-D characteristics; (b) silicate chain with bridging, non-bridging and free oxygen; (c) Al³⁺ incorporation in silicate chain.

Full size|PPT slide

Methodologies and formulas available for slag viscosity measurement

The knowledge related to slag viscosity plays a vital role in various types of metallurgical procedures [¹⁰⁵]. The performance of slagging is assessed by its viscosity, as it limits the volume of ash and assists it to flow out from the high-temperature process [¹⁰⁵]. The heat transfer and the performance of combustion and gasification are significantly affected by slag deposition [¹⁰⁶]. Slag viscosity and its change with temperature are crucial as they determines the degree of movement of the slag occurring at the interface [⁹³]. Over the years, various technologies have been established to measure slag viscosity. However, because of the involvement of elevated temperatures in the measuring process, only a few of them are practically applicable. The most common measuring methods are the capillary method, the rotating cylinder method, the falling body method, and the oscillating method [⁹³]. Table 2 tabulates the viscosity measuring methods applicable for slags, fluxes, and glasses [⁹³,¹⁰⁴,¹⁰⁷–¹⁰⁹].

Tab.2 Available viscosity measuring methods for slags, fluxes and glasses

Methods	Procedure	Range /(Pa∙s)	Comment
Capillary	By measuring the torque of a rotating plate; By measuring the sample height and time for parallel plate By measuring the rate of penetration in indentation	102 –1011	Impractical for high temperature
Rotating crucible	By measuring torque on static bob	10–2 –101	The exceptionally exact vertical arrangement required
Rotating bob	By measuring torque on the bob	100–102	Using flexible joint alignment problems can be resolved
Falling body	By measuring time for bob to fall (or drag) through a known distance	100.5 –105	Need a broad zone of uniform temperature
Oscillating	By measuring log decrement of the amplitude of twisting	10–4 –10–1	Applicable for depolymerised slag with low viscosity
IP	By measuring slag travel length	1.5–6	Inclination range is restricted between 9°–23°
M-IP	By measuring slag travel length	1–17.9	Restricted to an upper temperature of 1400°C

Due to its simplicity, the rotating bob method is more widely accepted. It comprises of a centrally aligned bob in a cylindrical crucible [¹⁰⁷]. By measuring the torque of the rotating bob or the rotating crucible, the viscosity can be calculated consequently. To ensure the Newtonian flow characteristics, at least two measurements are generally taken at two different rotation speeds. Although the capillary viscometer and falling sphere methods have been used, they are not widely accepted because of the complex operating process. To operate at high temperatures, the falling sphere viscometer requires a broad zone of uniform temperature. For measuring de-polymerised slag viscosities, oscillating viscometers are preferred [¹⁰⁷].

By using an inclined plane (IP), Mills et al. conducted a simple test for the measurement of slag viscosities [¹⁰⁹]. In this method, at high temperatures (1200°C –1400°C) an ash sample was first melted into slag and then cautiously discharged onto an IP to slide down. One of the drawbacks of this method is that its inclination range is restricted at 9–23 only, and its viscosity range is restricted at 1.5–6 Pa·s. Besides, due to the prior melting requirement, operation and measurements are highly risky. Moreover, the accuracy of the measurement could be easily affected by the cooling in the transferring process. In 2018, Dai et al. further extended this method to a new one, namely, the modified inclined plane (M-IP) method [¹⁰⁸]. The distinct features of this methodology include no earlier melting of ash needed, a tiny amount of ash sample required (200–300 mg), and independence on slag compositions and its nature (i.e., Newtonian or non-Newtonian). Using this methodology, Dai et al. further developed an empirical equation to predict slag viscosity. However, because of the impediment of the slagging furnace, this empirical equation is restricted to an upper temperature of 1400°C and a maximum limit of 17.9 Pa∙s.

It is noteworthy that it is challenging to measure slag viscosities at elevated temperatures from any of the aforementioned facilities. Many factors such as material, instrument, and hydrodynamics can also influence the accuracy of the measurement. Therefore, for many slags, the uncertainty in viscosity measurements is approximately ±20% [⁹³,¹⁰⁴]. Furthermore, for any further use of the experimentally determined viscosity, a crucial check and analysis of the data should be conducted before accepting these data. The erroneous and inaccurate measurements should be excluded from consideration [¹⁰⁰,¹¹⁰].

Considering the high cost of viscosity measurement equipment and complicated operating procedure, many empirical models have also been developed, most of which simply use the elemental composition of slag as the input. Table 3 lists the most extensively used slag viscosity prediction models. The first one is the Urbain model developed mainly based on the CaO-Al₂O₃-SiO₂ system. According to this model, the slag compositions can be categorised as slag formers, breakers, and amphoteric [¹¹¹]. In 2001, Kondratiev and Jak advanced this method for a modified Urbain model (M-Urb) based on different slag compositions (CaO-Al₂O₃-SiO₂-FeO) [¹¹²]. In this model, a four-component system was used instead of a three-component system. The Riboud model was deve-loped based on the SiO₂-CaO-Al₂O₃-CaF₂-Na₂O system and was able to predict slag and mold viscosities [¹¹³]. However, it is unable to differentiate the roles of different cations. A new model, namely the National Physical Laboratory (NPL) model to predict slag viscosity was developed by Mills and Sridhar in 1999 [¹¹⁴]. By using optical basicity, this model can correlate slag viscosity with slag structure [¹¹⁴,¹¹⁵]. However, its accuracy is not that high compared to the other models. In 2000, Iida et al. developed a model based on the Arrhenius type of equation and used the basicity index as a medium for describing the network structure of a slag [¹¹⁶]. Sufficient experimental data and a prior calibration are crucial for the prediction accuracy of this method [¹⁰⁴]. Relying on the Eyring-Polanyi equation, the Royal Institute of Technology in Stockholm (KTH) model was constructed [¹¹⁷], taking the Gibbs energy for activation into account [⁹³,¹¹⁸]. However, the necessity of identifying network former and breakers in the slag, and the complex interaction between anions and cations make this viscosity prediction model very complicated. Moreover, the coefficients required for calculating viscosity using this model are not available. The Mills model was developed based on the slag flowability at high temperatures and correlates slag viscosity with slag traveling length and inclination angle. However, it can only predict viscosities within a range of 1.5–6 Pa∙s and the inclination angle is restricted between 9–23 [¹⁰⁹]. The Dai model is the upgraded version of the Mill model, which correlates the slag viscosity with slag travel length, inclination angle and slag thickness. This model successfully resolved the issues with the inclination angle. An upper temperature limit of 1400°C and a viscosity limit of 17.9 Pa∙s are the drawbacks of this mode [¹⁰⁸]. In conclusion, a generic model for the prediction of slag viscosity is still missing.

Tab.3 Available slag viscosity prediction models

Models	Applicability	Correlation	Remarks	Reference
Urbain	Various	\eta =A_{\mathrm{A}}{\mathrm{e}}^{\frac{E_{\mathrm{A}}}{RT}}	Authentic for specific compositions and temperature category	[111]
Modified Urbain	Coal	\eta =AT{\mathrm{e}}^{\frac{\mathrm{1000}B}{T}}	Accurate for ash within the four component system (CaO-Al2O3-SiO2-FeO)	[112]
Riboud	Mould powder	\eta =AT{\mathrm{e}}^{\frac{B}{T}}	Unable to differentiate between different cations; accurate for SiO2-CaO-Al2O3-CaF-Na2O system	[113]
NPL	Industrial slags and mould fluxes	\eta =A{\mathrm{e}}^{\frac{B}{T}}	Optical basicity data requires for accurate prediction	[114]
Iida	Mould fluxes and metallurgical slags	\eta =A{n}_{\mathrm{0}}{\mathrm{e}}^{\frac{E}{B_{\mathrm{i}}}}	For accurate prediction basicity index value required; calibration of experimental data is necessary	[116]
KTH	Metallurgical slags	\eta ={\displaystyle \frac{h{N}_{\mathrm{A}}\rho }{M}}{A}_{\mathrm{A}}{\mathrm{e}}^{\frac{G}{RT}}	Coefficients are not available	[117]
Mills	Coal, mould powder, non-ferrous slag and blast furnace slag	{\log }_{\mathrm{10}}\mathrm{\hspace{0.17em}}\eta ={\log }_{\mathrm{10}}\mathrm{\hspace{0.17em}}A+{\displaystyle \frac{B}{T}}	Viscosity range is limited between 1.5–6 Pa∙s; inclination range is restricted between 9–23	[109]
Dai	Coal	\ln \left(\eta \right)={\displaystyle \frac{\mathrm{1}}{n}}\ln \left(\cos \beta \right)-\ln \left(v_z\right)+\ln \left\{{\displaystyle \frac{{\left(\rho g\right)}^{\frac{\mathrm{1}}{n}}{\delta }^{\frac{\mathrm{1}}{n}+\mathrm{1}}}{\left({\displaystyle \frac{\mathrm{1}}{n}}+\mathrm{2}\right)}}\right\}	Restricted to an upper temperature of 1400°C and a viscosity limit of 17.9 Pa∙s	[108]

Comparison of slagging behavior of different rank coals and bio-slags

Slagging behavior of high-rank coal slag

Over the years, much research has been conducted on the slagging behavior of high-rank coals such as anthracite and bituminous coals, due to the abundance and preferential use of these coals for combustion and gasification [¹¹⁹–¹²²]. Hurst and Pattenson conducted extensive work on slag qualities, viscosity measurement, and empirical predictions of Australian bituminous coals [¹²¹,¹²²]. To evaluate the suitability of Australian bituminous coals for use in integrated combined cycle gasification technologies, they evaluated the viscosity of 85 liquid slags from 52 Australian bituminous coal ashes and obtained four distinct empirical viscosity models at the FeO concentration of 0 wt%–2.5 wt%, 2.5 wt%–5 wt%, 5 wt%–7.5 wt%, and 7.5 wt%–10 wt%, respectively [¹²²]. The major components of Australian bituminous coal ashes are SiO₂, Al₂O₃, CaO, and FeO. Among these components, FeO is highly susceptible at high temperatures. Besides, the amount of FeO in Australian bituminous coal ashes range between 0.5 wt% and 10 wt%. Thus, Hurst and Pattenson divided these ashes into four distinctive category based on their FeO concentration (0 wt%–2.5 wt%, 2.5 wt%–5 wt%, 5 wt%–7.5 wt%, and 7.5 wt%–10 wt%) to assess their effects on slagging propensity. In addition, to compare the experimental results with different viscosity models they selected four distinctive slags, namely slag 9, slag 38, slag 59, and slag 75 which represented the ash group that contained 0 wt%–2.5 wt% FeO, 2.5 wt%–5 wt% FeO, 5 wt%–7.5 wt% FeO, and 7.5 wt%–10 wt% FeO, respectively. They performed viscosity measurements by using a rotational viscometer (Haake-1700) under reducing conditions at elevated temperatures, and empirically fitted the measured data by polynomial expressions for the four FeO concentration ranges using a modified Urbain treatment. As has been confirmed, for the first three FeO ranging up to 7.5 wt%, the value of the anticipated viscosities in descending order was the Urbain model, the synthetic SiO₂–Al₂O₃–CaO–FeO model, and the coal ash slag least square model, with the last one providing the nearest consent to the experimental data. The expected viscosity values for the 7.5 wt%–10 wt% range of FeO are generally in the same order, but the viscosity values for the synthetic model are sometimes lower than those for the least square model. This is evidenced by the examples of the accordance for different FeO contents given in Fig. 6. In addition to the above work described, Hurst and Pattenson also investigated the effect of the addition of limestone flux by examining the phase diagrams. The low iron-containing Australian coal ashes (<2.5 wt%) require fluxing as their ash composition lies in the high-temperature mullite (3Al₂O₃·2SiO₂) region and the addition of limestone flux changes the slag composition to the lower temperature anorthite (CaAl₂Si₂O₈) region in the SiO₂-Al₂O₃-CaO phase diagram. The liquidus temperatures meet the normal slag tapping range of 1400°C–1500°C, and the amount of flux is determined by the need to meet the optimum tapping and the maximum tapping viscosity values of 15 and 25 Pa∙s, respectively.

Fig.6 Comparison of viscosity models for Australian bituminous coal slag (adapted from Ref. [¹²²] with permission. Note that slag 9 has a composition of SiO₂ 48.1 wt%, Al₂O₃ 25.3 wt%, CaO 25.4 wt%, and FeO 1.2 wt%. Slag 38 consists of SiO₂ 50.1 wt%, Al₂O₃ 30.6 wt%, CaO 15.5 wt%, and FeO 3.8 wt%. Slag 59 has a composition of SiO₂ 53.9 wt%, Al₂O₃ 29.4 wt%, CaO 11.3 wt%, and FeO 5.5 wt%. Slag 75 is made up of SiO₂ 53.8 wt%, Al₂O₃ 25.6 wt%, CaO 11.5 wt%, and FeO 9.1 wt%. Model 1 denotes Urbain model; Model 2 denotes synthetic slag SAC (SiO₂-Al₂O₃-CaO) model; Model 3 denotes coal ash slag model for<2.5 wt% FeO; Model 4 denotes synthetic slag SACF (SiO₂-Al₂O₃-CaO-FeO) model for 5 wt% FeO; Model 5 denotes coal ash slag SACF model for 2.5 wt%–5 wt% FeO; Model 6 denotes coal ash slag SACF model for 5 wt%–7.5 wt% FeO; Model 7 denotes synthetic slag SACF model for 10 wt% FeO; Model 8 denotes coal ash slag SACF model for 7.5 wt%–10 wt% FeO.)

Full size|PPT slide

For coal gasification in a fixed-bed reactor, the viscosity of 5 Pa∙s at bed temperature is required, whereas the maximum viscosity for the entrained flow processes can go up to 15 Pa∙s [¹²¹]. At a tapping temperature of 1500°C, the upper limit before flux addition can be 25 Pa∙s. The temperature at or below which the viscosity of slag increase sharply or continuously is termed as the temperature of critical viscosity or T_cv [¹²⁰,¹²³]. Coal ash slags with low SiO₂/Al₂O₃ molar ratios (<1.6) may have a relatively high T_cv at 1400°C–1450°C, preventing slag tapping at 1400°C. A T_cv of around 1400°C enables lower gasification temperatures and higher cold gas efficiencies [¹²¹]. However, most of the high rank coals have high ash fusion temperatures close to or even above 1500°C, requiring the addition of flux materials, which is usually limestone (CaCO₃). Patterson and Hurst prepared 55 high-rank bituminous coal samples from 35 Australian coal deposits (from New South Wales and Queensland) and investigated the additional flux requirement for each sample [¹²¹]. The results showed that several bituminous coals did not require any additional flux for 1400°C slag tapping. For tapping at 1500°C, more than half of the coals needed a limestone amount of less than 3 wt% based on the original coal mass. In addition, such coals can be mixed with other low-melting coals to minimise the need for fluxing materials [¹²¹].

Both Li et al. [¹¹⁹] and Kong et al. [¹²⁰] investigated the slagging characteristics of anthracite coals. Coal ash slags at elevated temperatures display distinct viscosity behaviors from one another. Some slags show the classic behavior of a glassy slag with a constant increase in viscosity as the temperature drops, while others show a fast rise in viscosity when the temperature is below its liquidus temperature at which slag is fully in liquid, T_Liquidus. The slag viscosity for two Chinese anthracites, namely Zhaozhuang Coal and Datong Coal is shown in Fig. 7. The former coal slag shows a glassy slag behavior, while the latter, instead, a close-to a crystalline slag behavior. For each slag, its viscosity below T_Liquidus is strongly dependent on the quantity of the solid phase that is also affected by the cooling rate. For the glassy slag of Zhaozhuang Coal, its slag viscosity is merely affected by the cooling rate since the principal solid species mullite (Al₆Si₂O₁₃) is rarely crystalised at any cooling rate. However, for the crystalline slag of deformation temperature (DT), owing to a complete crystallization of CaAl₂Si₂O₈ at the slow heating rate, its viscosity drops considerably with increasing the cooling rate. Additionally, considering that the temperature fluctuation around the T_cv in a gasifier could lead to the blockage of the gasifier by the solidified slag, Fig. 8 shows the slag T_cv measured at various cooling rates for the two anthracites. The impact of cooling rate is clearly more pronounced for the crystalline slag of Datong Coal.

Fig.7 Slag viscosity measured at distinct cooling rates (adapted from Ref. [¹²⁰] with permission.)

Full size|PPT slide

Fig.8 For ZZ and DT anthracite coal T_cv at the different cooling rate (adapted from Ref. [¹²⁰] with permission. Note that ZZ stands for Zhaozhuang and DT for Datong.)

Full size|PPT slide

Slagging behaviors of low-rank coal slag

Compared to high-rank coal slag, the slagging behavior of low-rank coals, including lignite and brown coal, was less studied [¹²⁴–¹²⁸]. One of the most recent works on low-rank coal slagging was conducted by Wu et al. [¹²⁹] and Dai et al. [¹³⁰]. Dai et al. [¹⁰⁸] even developed a new methodology to quantify slag viscosities. They justified the necessity of developing a new methodology by comparing the measured viscosities of standard coal ash with the predicted viscosity by using different prediction models, as shown in Fig. 9(a). Using the new method, they analyzed the effects of mass, temperature, residence time, and inclination angle by conducting experiments on seven synthetic standard ash samples under a reducing environment of 1 vol% CO in nitrogen, 1100°C–1400°C, and inclination angles from 25° to 90° and different duration times of 10 to 40 min. The slag weight ranged from 0.1 g to 0.4 g. Multiple linear regression was performed to create an empirical equation based on the slag traveling length per unit mass (L′/(mm∙g⁻¹) and the inclined angle (cosb) to predict slag viscosity (h/(Pa∙s)). The results show that the traveling length of the unit mass of slag is a function of slag mass, temperature, and the residence time of slagging. The traveling length of the slag also shows a linear weight relationship within the weight range of 0.1–0.3 g at 1400°C. A linear relationship was established between the slag traveling length logarithm and the temperature reciprocal as shown in Fig. 9(b), proving that the Arrhenius type relationship can satisfactorily represent the temperature dependence of the slag traveling length that is subsequently used to calculate slag viscosity. Finally, a linear correlation was derived among the slag traveling length logarithm, the inclination angle, and the slag viscosity logarithm, as shown below in Eq. (1).

(1)\ln \mu =\mathrm{3}.\mathrm{282281cos}\beta –\mathrm{1}.\mathrm{882827ln}L\text{'}+\mathrm{7}.\mathrm{3971}\mathrm{0}\mathrm{8}.

They further justified the newly developed equation by predicting the viscosity of standard coal ashes and comparing them with measured viscosities, as shown in Fig. 9(c). Additionally, based on the M-IP method, they also investigated the slagging behavior of low-rank coals, namely the lignite in Xinjiang Uygur Autonomous Region with five different ash compositions labeled A–E, including basic ashes A and C, neutral ash B, and acidic ashes D and E [¹³⁰]. The results in Fig. 10 demonstrate the flow patterns of the five slags at three different temperatures in 1 vol% CO balanced by N₂. It is obviously observed that the neutral ash displays the greatest flowability owing to an excellent equilibrium between the components forming basic and the acidic ash within it. Based on the slag traveling length and Eq. (1), Fig. 11 shows the viscosities calculated for the five slags at 1400°C. The prediction from other metals was also included for comparison. It can be seen that these models differ widely, especially for the three ashes from A to C. The calculation of viscosity for slag B based on the M-IP method is close to three existing models, namely the Urbain, the M-Urbs, and the Riboud. For the two acidic slags D and E, the calculation from the M-IP method is very close to that of most existing models except the Iida. This is reasonable as the high-rank coal slags are acidic, and most of the existing models were just established based on acidic slags [¹³¹,¹³²]. In addition, the inapplicability of the existing models to the slags A to C could be due to the assumption of a Newtonian liquid for acidic slags [¹³³,¹³⁴]. This is the case for acidic slags and even the neutrals ones. However, for the basic slags that are prevalent for low-rank coals, the abundant basic elements within it generally undergo notable physical changes during slagging [¹³⁵]. Therefore, the slags are more similar to a non-Newtonian fluid. In this sense, the M-IP technique seems more collective and autonomous on the ash compositions. It calculates the viscosity based on traveling distance of a slag, rather than based on ash composition and any assumptions. To be specific, the M-IP methodology measures the traveling length of the whole slag regardless of its behavior as Newtonian or non-Newtonian fluid. Moreover, it considers the effects of solidification and particle precipitation.

Fig.9 For standard coal ash sample (adapted from Ref. [¹⁰⁸] with permission.)

Full size|PPT slide

Fig.10 Ash slag pictures of Xinjiang Uygur Autonomous Region at 1300°C–1400°C at the exposure time of 40 min for A–C and 2 h for D and E, for the inclined angle of 25° (adapted from Ref. [¹³⁰] with permission).

Full size|PPT slide

Fig.11 Xinjiang lignite ash viscosity of Xinjiang Uygur Autonomous Region obtained at 1400°C from different models (adapted from Ref. [¹³⁰] with permission. Note that M-IP denotes modified incline plane, Rib denotes Riboud model, Urb denotes Urbain model, Mill denotes Mill model, Iida denotes Iida model, For denotes Forsbacka model, M-Urb denotes modified Urbain model, OB denotes optical basicity model, Fact denotes Factsage, and Wu denotes Wu model).

Full size|PPT slide

Ilyushechkin and Roberts investigated the slagging behavior of Australian brown coal ashes at the temperature of 900°C–1650°C [¹²⁵]. Their evaluation of the slag microstructures showed that the brown coal ash rich in both Fe and Mg may be used in non-slagging gasifiers as the ash has a very high melting point. Similarly, the Si-rich brown coal can also be used at low temperatures in non-slagging-type gasifiers, due to its high melting point. Interestingly, brown coals that are rich in Al along with Na, Mg, Ca, Si, and Fe possess a low melting point, which were thus suggested for slag-tapping gasifiers. Only a few high-silica group slags have adequate slag viscosities at high temperatures to match the viscosity need for an entrained-bed gasifier. Some high-silica ash coals need fluxing to lower the viscosity of the slag. Slags rich in Ca, Mg, and Fe have a low-viscosity fluid with significant quantities of solids, causing a non-Newtonian flow behavior. Mixing these type of coal ashes with the coal ashes rich in Si could be a useful option to adjust the viscosity of the slag. Moreover, Ilyushechkin and Roberts also observed that the formation of liquid phase at low temperatures (900°C–1100°C) was less dependent on the thermodynamic phase equilibrium, as these temperatures are far below the ash fusion temperature (AFT). In contrast, the slag formation at high temperatures (1200°C–1600°C) is usually governed by phase equilibrium, with a few exceptions observed as well [¹²⁵].

Kondratiev and Ilyushechkin conducted a critical review on the transition of coal ash slag from Newtonian to non-Newtonian behavior [¹³⁶]. According to their study, the flow transformation from Newtonian to non-Newtonian is strongly related to the degree of crystallization. The transformation takes place at a particular concentration of crystals in the slag, which is firmly dependent on the shape and size of the solid phases as shown in Fig. 12. For instance, for spherical crystals within a slag, the transition to non-Newtonian flow can take place at about 40 vol% for the crystals. In contrast, for non-sphere-shaped crystals within a slag, the transition to non-Newtonian flow can occur at a much lower crystal content [¹³⁷]. The circumstance of non-Newtonian flow generally implies the creation of interactions between crystals. Non-Newtonian flow is known to occur primarily on the basis of the dimensional analysis equation when the Peclet or Reynolds numbers exceed those limiting values (Pe≤10³ and Re≥10⁻³) [¹³⁸]. Nevertheless, the fact that non-Newtonian behavior can be observed in a wide variety of Pe and Re in coal ash slags is confirmed by several reports [¹³⁶]. Song et al. examined the slag viscosity at different shearing rates and found Newtonian behavior for the fully molten slag [¹³⁹]. However, as the solids begin to precipitate at low temperatures, the dependency of slag viscosity on shear stress deviates from a linear trend, suggesting a non-Newtonian flow for the slag. Furthermore, according to the results for the phase, viscosity, and composition of slags (for instance, like those shown in Refs. [¹³⁹–¹⁴¹]), it is concluded that three forms of transition from Newtonian to non-Newtonian behavior can be observed in coal slags, which are the transition occurring almost at the same temperatures where solids start to precipitate, the transition occurring when solids reach a particular concentration in the slag, and the reversed transition with temperature at specific changes in slag composition followed by a second Newtonian to non-Newtonian transition. The structure and morphology of solid phases and changes in the resulting liquid phase are closely related to these transitions.

Fig.12 Viscosity as a function of the volume fraction of crystals (adapted from Ref. [¹³⁶] with permission).

Full size|PPT slide

Slagging behavior of bio-slag

There has been considerable research on the slagging behavior of biomass for co-combustion with coal [¹⁴²–¹⁴⁵], but much less on the slagging behavior of pure biomass ashes, especially in the field of bio-slag viscosity measurement. Garcia-Maraver et al. [³] conducted a critical review of the predictive coefficients of biomass ash slagging tendency. They applied different slagging and fouling index (Fu) such as silica content (SiO₂), chlorine content (Cl), basic to acidic compounds ratio (B/A), bed agglomeration index (BAI), Babcock index (Rs), ash fusibility index (AFI), Fu, slag viscosity index (Sr), softening temperature (ST), and IDT to predict the slagging propensity of various biomass ashes. However, the coefficients conventionally used to anticipate the coal ash deposition show mixed outcomes when applied to the bio-ash. Table 4 demonstrates the average slagging propensity of different biomass groups in different slagging indexes [³]. They concluded that to date, no distinctive and generic formula could unambiguously correlate the slagging and fouling propensity of all biomass fuels. To confirm the contradictory outcomes of these indexes, they further developed a relationship between the indexes and found a weak relationship between the them, as shown in Fig. 13. The closest relationship was found between BAI and Fu with 70.8% of matching. This suggests that there is a need for creating a new index to predict the slagging propensity, and the new indexing should be based on realistic outcomes from actual combustion/gasification experiences, taking into account not only the impact of the process operating conditions but also the heterogeneity and physicochemical characteristics of bio-ashes.

Tab.4 Slagging propensity of different biomass in the different slagging indexes

Fig.13 Correlations among different slagging indices (adapted from Ref. [³] with permission).

Full size|PPT slide

Näzelius et al. [¹⁴⁶] proposed a new slagging index for bio-ash by conducting a systematic evaluation of all the results achieved by burning 36 distinct biomasses in a small grate furnace, as well as the chemical analysis of the respective bottom ashes and slags. According to their slagging tendency, bio-ash can be catogirised into non-slag, minor slagging tendency, moderate slagging tendency, and intense slagging tendency. For non-slag bio-ash, the fuel compositions consist primarily of low Si and K combined with a high Ca content. Instead of forming an ash melt, these fuels tend to produce a low-viscosity carbonate melt that is not likely to form slag. For the minor slagging tendency bio-ash, it has a higher Si content that indend to form a small fraction of less-sticky silicate melt at high temperatures. The bio-ash with an additional increase in Si content is prone to carry a mild slagging tendency. The quantity of silicate melt increases within increasing the Si content. However, the viscosity of the silica melt is larger (more sticky) than the bio-ash with a minor slagging tendency. Finally, the bio-ash with large contents of both Si and K carries an intense slagging tendency, which tends to convert to sticky K-silicate at high temperatures.

Regarding the viscosity measurement of bio-slag, one of the most recent works conducted by Chen and Zhao is noteworthy [¹⁴⁷]. They investigated the slag viscosities with a broad variety of the contents of SiO₂ and K₂O in the SiO₂-K₂O-CaO scheme. The content of SiO₂ varies from 49.8 to 78.2 mol%, while that of K₂O falls in 7.7 mol%–29.5 mol%. The viscosity was measured by a rotational spindle method at 1000°C–1600°C. Figure 14 demonstrates the effect of SiO₂ content on the slag viscosity with two different K₂O/(K₂O+ CaO) ratios. Regardless of the K₂O/(KO+ CaO) ratio, a linear relationship can be established between the viscosity logarithm and the SiO₂ concentration for each temperature. This strongly suggests a negative effect of SiO₂ on the bio-slag viscosity for the concentration ranges studied.

Fig.14 Effect of SiO₂ concentrations on the viscosities (adapted from Ref. [¹⁴⁷] with permission).

Full size|PPT slide

From the empirical modeling perspective, there is still debate on whether or not the viscosities of the SiO₂-K₂O-CaO system predicted by several class models are within their predictive capacity. To predict the slag viscosity, the FactSage viscosity model [¹⁴⁸] and quasi-chemical viscosity model (QCV) [¹⁴⁹] were built based on the Qi-species and bond fractions. Another model, namely the Zhang model calculating the bio-ash viscosity by considering various oxygen ions, was also developed [¹⁵⁰]. The Lakatos model [¹⁵¹] was further established by taking into account the notion of free volume within a slag. However, the prediction of these models is inconsistent with one another. Figure 15 shows that FactSage has the best matching with the experimental readings (labeled as Present). In contrast, most of the other models overestimate the viscosity except the NPL model, whose prediction are less than the experimental readings. Clearly, these empirical models need to be tailored to fit the bio-slags. Likewise, a brand new model should be developed specifically for bio-slags. Apart from K₂O and CaO, the other unique elements such as Na, Cl, and P and their interaction with the major oxides including SiO₂, Al₂O₃, FeO/Fe₂O₃, and CaO in the bio-ash slagging propensity should also be intensively examined. Simultaneously, the measurement of bio-slag viscosity in the existing viscometers should also be extensively conducted to assist in the development of empirical models. In addition, from the fundamental research perspective, pure oxides should be mixed as multi-component blends or synthetic bio-ashes, which are then subjected to advance in situ optical measurements such as the M-IP method [¹⁰⁸] and molecular beam mass spectrometry (MBMS) [¹⁵²,¹⁵³] to explore the dynamics of bio-ash slagging and the release/loss of elements during slagging, while the final slag should also be subjected to different off-line characterization methods for a deep understanding of its physical and chemical properties.

Fig.15 Comparisons between the measured and predicted viscosities from different slag models (adapted from Ref. [¹⁴⁷] with permission. Note that experimental readings are shown as Present).

Full size|PPT slide

Conclusions

The characterization and prediction of ash slagging propensity is a topic that has been of long-term scientific interest for solid fuel combustion and gasification. However, there is still a lack of understanding about it, due to the complex nature of slag. To date, extensive research on slag has identified its formation mechanism, structure, and nature during the combustion and gasification of high-rank coals. Based on these results, many indices have been developed to predict slagging propensity. One of the most critical parameters for determining the performance of slag is its viscosity. It minimises the volume of the ash and helps it to drain out of the process rather than accumulate within the equipment. A variety of viscosity measuring equipment has thus been purposely constructed. In parallel, a number of empirical equations have also been predicted, based on various properties of the original ash within the solid fuel. This review shows that the existing viscosity models and slagging indices can only predict the viscosity and slagging propensity of high-rank coals, which are, however, unavailable upon the extension to low-rank coal ashes and bio-ashes. In particular, slagging indices and viscosity models available are unable to predict the slagging propensity and viscosity of the bio-slags. Thus, there is a critical need for an index, a model, or even a new measurement method, which can correctly predict and/or measure the slagging propensity and slag viscosity for low-rank coal ash and bio-ash. Moreover, further understanding should be developed for the thermodynamics and even kinetics underpinning the slagging of low-rank coal and bio-ash, such as the transformation of individual elements, the crystalisation chracteristics upon quenching and low cooling, as well as the phase changes.

Acknowledgment

This work was financially supported by the Australian Research Council (ARC) under its Industrial Research Training Hub (IH170100009) scheme for the joint project between Monash University and Shanghai Boiler Works Co Ltd. Md Tanvir Alam would like to thank Monash University for his Ph.D Scholarship.

References

1 Peters G P, Andrew R M, Canadell J G, Fuss S, Jackson R B, Korsbakken J I, Le Quéré C, Nakicenovic N. Key indicators to track current progress and future ambition of the Paris Agreement. Nature Climate Change, 2017, 7(2): 118–122

2 Dudley B. BP Statistical Review of World Energy 2016. 2016

3 Garcia-Maraver A, Mata-Sanchez J, Carpio M, Perez-Jimenez J A. Critical review of predictive coefficients for biomass ash deposition tendency. Journal of the Energy Institute, 2017, 90(2): 214–228

4 Fang X, Jia L. Experimental study on ash fusion characteristics of biomass. Bioresource Technology, 2012, 104: 769–774

5 Rajamma R, Ball R J, Tarelho L A C, Allen G C, Labrincha J A, Ferreira V M. Characterisation and use of biomass fly ash in cement-based materials. Journal of Hazardous Materials, 2009, 172(2–3): 1049–1060

6 Chen S, Lior N, Xiang W. Coal gasification integration with solid oxide fuel cell and chemical looping combustion for high-efficiency power generation with inherent CO₂ capture. Applied Energy, 2015, 146: 298–312

7 Fang X, Jia L, Yin L. A weighted average global process model based on two-stage kinetic scheme for biomass combustion. Biomass and Bioenergy, 2013, 48: 43–50

8 Yin C, Rosendahl L A, Kær S K. Grate-firing of biomass for heat and power production. Progress in Energy and Combustion Science, 2008, 34(6): 725–754

9 Kleinhans U, Wieland C, Frandsen F J, Spliethoff H. Ash formation and deposition in coal and biomass fired combustion systems: progress and challenges in the field of ash particle sticking and rebound behavior. Progress in Energy and Combustion Science, 2018, 68: 65–168

10 Du S, Yang H, Qiani K, Wang X, Chen H. Fusion and transformation properties of the inorganic components in biomass ash. Fuel, 2014, 117(Part B): 1281–1287

11 Li G, Li S, Huang Q, Yao Q. Fine particulate formation and ash deposition during pulverized coal combustion of high-sodium lignite in a down-fired furnace. Fuel, 2015, 143: 430–437

12 Hosseini S, Gupta R. Inorganic matter behavior during coal gasification: effect of operating conditions and particle trajectory on ash deposition and slag formation. Energy & Fuels, 2015, 29(3): 1503–1519

13 Garba M U, Ingham D B, Ma L, Degereji M U, Pourkashanian M, Williams A. Modelling of deposit formation and sintering for the co-combustion of coal with biomass. Fuel, 2013, 113: 863–872

14 Fryda L, Sobrino C, Cieplik M, van de Kamp W L. Study on ash deposition under oxyfuel combustion of coal/biomass blends. Fuel, 2010, 89(8): 1889–1902

15 Shen Z, Liang Q, Xu J, Liu H, Lin K. Study on the fragmentation behaviors of deposited particles on the molten slag surface and their effects on gasification for different coal ranks and petroleum coke. Energy & Fuels, 2018, 32(9): 9243–9254

16 Yan T, Kong L, Bai J, Bai Z, Li W. Thermomechanical analysis of coal ash fusion behavior. Chemical Engineering Science, 2016, 147: 74–82

17 Ma T, Fan C, Hao L, Li S, Song W, Lin W. Fusion characterization of biomass ash. Thermochimica Acta, 2016, 638: 1–9

18 Zhou H, Zhou B, Zhang H, Li L. Behavior of fouling deposits formed on a probe with different surface temperatures. Energy & Fuels, 2014, 28(12): 7701–7711

19 Wang G, Silva R, Azevedio J, Martins-Dias S, Costa M. Evaluation of the combustion behaviour and ash characteristics of biomass waste derived fuels, pine and coal in a drop tube furnace. Fuel, 2014, 117(Part A): 809–824

20 Khoshnaw F M. WIT Transactions on State-of-the-art in Science and Engineering. Southampton, UK: WIT Press, 2015, 85: 137–147

21 Niu Y, Tan H, Hui S. Ash-related issues during biomass combustion: alkali-induced slagging, silicate melt-induced slagging (ash fusion), agglomeration, corrosion, ash utilization, and related countermeasures. Progress in Energy and Combustion Science, 2016, 52: 1–61

22 Loo S V, Koppejan J. The Handbook of Biomass Combustion and Co-Firing. London: Earthscan Publications Ltd., 2012

23 Tortosa Masiá A A, Buhre B J P, Gupta R P, Wall T F. Characterising ash of biomass and waste. Fuel Processing Technology, 2007, 88(11–12): 1071–1081

24 Park H Y, Lee J E, Kim H H, Park S, Baek S H, Ye I, Ryu C. Thermal resistance by slagging and its relationship with ash properties for six coal blends in a commercial coal-fired boiler. Fuel, 2019, 235: 1377–1386

25 Demirbas A. Potential applications of renewable energy sources, biomass combustion problems in boiler power systems and combustion related environmental issues. Progress in Energy and Combustion Science, 2005, 31(2): 171–192

26 Pronobis M. Evaluation of the influence of biomass co-combustion on boiler furnace slagging by means of fusibility correlations. Biomass and Bioenergy, 2005, 28(4): 375–383

27 Singh A, Sharma V, Mittal S, Pandey G, Mudgal D, Gupta P. An overview of problems and solutions for components subjected to fireside of boilers. International Journal of Industrial Chemistry, 2018, 9(1): 1–15

28 Baxter L L, Miles T R, Miles T R Jr, Jenkins B M, Milne T, Dayton D, Bryers R W, Oden L L. The behavior of inorganic material in biomass-fired power boilers: field and laboratory experiences. Fuel Processing Technology, 1998, 54(1–3): 47–78

29 Szemmelveisz K, Szűcs I, Palotás Á B, Winkler L, Eddings E G. Examination of the combustion conditions of herbaceous biomass. Fuel Processing Technology, 2009, 90(6): 839–847

30 Aho M, Silvennoinen J. Preventing chlorine deposition on heat transfer surfaces with aluminium-silicon rich biomass residue and additive. Fuel, 2004, 83(10): 1299–1305

31 Knudsen J N, Jensen P A, Dam-Johansen K. Transformation and release to the gas phase of Cl, K, and S during combustion of annual biomass. Energy & Fuels, 2004, 18(5): 1385–1399

32 Guan G. Clean coal technologies in Japan: a review. Chinese Journal of Chemical Engineering, 2017, 25(6): 689–697

33 Zhuang Q, Biondi M, Yan S, Bhagat K, Vansickle R, Chen C, Tan H, Zhu Y, You W, Xia W. TRIG^TM: an advanced gasification technology to utilize low rank coals for power. Fuel, 2015, 152: 103–109

34 Falcke T J, Hoadley A F A, Brennan D J, Sinclair S E. The sustainability of clean coal technology: IGCC with/without CCS. Process Safety and Environmental Protection, 2011, 89(1): 41–52

35 Mondal P, Dang G, Garg M. Syngas production through gasification and cleanup for downstream applications—recent developments. Fuel Processing Technology, 2011, 92(8): 1395–1410

36 Wang P, Massoudi M. Slag behavior in gasifiers. Part I: influence of coal properties and gasification conditions. Energies, 2013, 6(2): 784–806

37 Kurowski M P, Spliethoff H. Deposition and slag flow modeling with SPH for a generic gasifier with different coal ashes using fusibility data. Fuel, 2016, 172: 218–227

38 Maurstad O. An Overview of Coal based Integrated Gasification Combined Cycle (IGCC) Technology. MIT LFEE 2005–002 WP. 2005

39 Duane Brooker D, Oh M S. Iron sulfide deposition during coal gasification. Fuel Processing Technology, 1995, 44(1–3): 181–190

40 Brooker D. Chemistry of deposit formation in a coal gasification syngas cooler. Fuel, 1993, 72(5): 665–670

41 Higman C, Van der Burgt M. Gasification Processes. 2nd ed. Burlington: Gulf Professional Publishing, 2008, 91–191

42 Guo Z Q, Han B Q, Dong H. Effect of coal slag on the wear rate and microstructure of the ZrO₂-bearing chromia refractories. Ceramics International, 1997, 23(6): 489–496

43 Liu G, Zhang H, Gao L, Zheng L, Peng Z. Petrological and mineralogical characterizations and chemical composition of coal ashes from power plants in Yanzhou mining district, China. Fuel Processing Technology, 2004, 85(15): 1635–1646

44 Jenkins B M, Baxter L L, Miles T R Jr, Miles T R. Combustion properties of biomass. Fuel Processing Technology, 1998, 54(1–3): 17–46

45 Sami M, Annamalai K, Wooldridge M. Co-firing of coal and biomass fuel blends. Progress in Energy and Combustion Science, 2001, 27(2): 171–214

46 Plaza P. The development of a slagging and fouling predictive methodology for large scale pulverised boilers fired with coal/biomass blends. Dissertation for the Doctoral Degree. Cardiff, UK: Cardiff University, 2013

47 van Dyk J C, Keyser M J, Coertzen M. Syngas production from South African coal sources using Sasol-Lurgi gasifiers. International Journal of Coal Geology, 2006, 65(3–4): 243–253

48 Vassilev S V, Baxter D, Andersen L K, Vassileva C G. An overview of the chemical composition of biomass. Fuel, 2010, 89(5): 913–933

49 Obernberger I, Brunner T, Bärnthaler G. Chemical properties of solid biofuels—significance and impact. Biomass and Bioenergy, 2006, 30(11): 973–982

50 Liao C, Wu C, Yani Y, Huang H. Chemical elemental characteristics of biomass fuels in China. Biomass and Bioenergy, 2004, 27(2): 119–130

51 Werkelin J, Skrifvars B J, Hupa M. Ash-forming elements in four Scandinavian wood species. Part 1: summer harvest. Biomass and Bioenergy, 2005, 29(6): 451–466

52 Hatt R M. Fireside deposits in coal-fired utility boilers. Progress in Energy and Combustion Science, 1990, 16(4): 235–241

53 Baxter L, DeSollar R. Applications of Advanced Technology to Ash-Related Problems in Boilers. New York: Springer Science & Business Media, 1996

54 Joseph G, Singer P E. Combustion Fossil Power. 4th ed. Connecticut: Combustion Engineering, Inc., 1991

55 Dick E P, Ryabov G A, Tugov A N, Soboleva A N. Comparing properties of coal ash and alternative-fuel ash. Thermal Engineering, 2007, 54(3): 231–235

56 The British Coal Utilisation Research Association (BCUR). The BCURA Coal Sample Bank: a User’s Handbook. 2002

57 Laursen K, Frandsen F J. Classification system for ash deposits based on SEM analyses. In: Gupta R P, Wall T F, Baxter L., eds. Impact of Mineral Impurities in Solid Fuel Combustion. Boston: Springer, 2002, 205–216

58 Raask E. Mineral Impurities in Coal Combustion: Behavior, Problems, and Remedial Measures. New York: Hemisphere Publishing Corporation, 1985

59 Given P H, Cronauer D C, Spackman W, Lovell H L, Davis A, Biswas B. Dependence of coal liquefaction behaviour on coal characteristics. 2. Role of petrographic composition. Fuel, 1975, 54(1): 40–49

60 Vuthaluru H B, French D. Ash chemistry and mineralogy of an Indonesian coal during combustion: part 1—drop-tube observations. Fuel Processing Technology, 2008, 89(6): 595–607

61 Vuthaluru H B, French D. Ash chemistry and mineralogy of an Indonesian coal during combustion: part II—pilot scale observations. Fuel Processing Technology, 2008, 89(6): 608–621

62 Misra M K, Ragland K W, Baker A J. Wood ash composition as a function of furnace temperature. Biomass and Bioenergy, 1993, 4(2): 103–116

63 Demirbas A. Combustion characteristics of different biomass fuels. Progress in Energy and Combustion Science, 2004, 30(2): 219–230

64 Bryers R W. Fireside slagging, fouling, and high-temperature corrosion of heat-transfer surface due to impurities in steam-raising fuels. Progress in Energy and Combustion Science, 1996, 22(1): 29–120

65 Miles T R, Miles T R Jr, Baxter L L, Bryers R W, Jenkins B M, Oden L L. Alkali deposits found in biomass power plants: a preliminary investigation of their extent and nature. Volume 1. Technical Report, USA: NREL/TP-433-8142-Vol.1, 1995

66 Moilanen A. Thermogravimetric characterisations of biomass and waste for gasification processes. VTT Publications, 2006, 607: 3–103

67 Scurlock J M O, Dayton D C, Hames B. Bamboo: an overlooked biomass resource? Biomass and Bioenergy, 2000, 19(4): 229–244

68 Thy P, Jenkins B, Grundvig S, Shiraki R, Lesher C. High temperature elemental losses and mineralogical changes in common biomass ashes. Fuel, 2006, 85(5–6): 783–795

69 Werther J, Saenger M, Hartge E U, Ogada T, Siagi Z. Combustion of agricultural residues. Progress in Energy and Combustion Science, 2000, 26(1): 1–27

70 Thy P, Lesher C, Jenkins B M. Experimental determination of high-temperature elemental losses from biomass slag. Fuel, 2000, 79(6): 693–700

71 Theis M, Skrifvars B J, Hupa M, Tran H. Fouling tendency of ash resulting from burning mixtures of biofuels. Part 1: deposition rates. Fuel, 2006, 85(7–8): 1125–1130

72 Vassilev S V, Vassileva C G. A new approach for the classification of coal fly ashes based on their origin, composition, properties, and behaviour. Fuel, 2007, 86(10–11): 1490–1512

73 Vassilev S V, Vassileva C G. A new approach for the combined chemical and mineral classification of the inorganic matter in coal. 1. Chemical and mineral classification systems. Fuel, 2009, 88(2): 235–245

74 Li C Z. Advances in the Science of Victorian Brown Coal. Elsevier Inc., 2004

75 Wu X, Zhang X, Yan K, Chen N, Zhang J, Xu X, Dai B, Zhang J, Zhang L. Ash deposition and slagging behavior of Chinese Xinjiang high-alkali coal in 3 MW_th pilot-scale combustion test. Fuel, 2016, 181: 1191–1202

76 Russell N V, Méndez L B, Wigley F, Williamson J. Ash deposition of a Spanish anthracite: effects of included and excluded mineral matter. Fuel, 2002, 81(5): 657–663

77 Andersen K H, Frandsen F J, Hansen P F B, Wieck-Hansen K, Rasmussen I, Overgaard P, Dam-Johansen K. Deposit formation in a 150 MW_e utility PF-boiler during co-combustion of coal and straw. Energy & Fuels, 2000, 14(4): 765–780

78 Beck J, Brandenstein J, Unterberger S, Hein K R G. Effects of sewage sludge and meat and bone meal co-combustion on SCR catalysts. Applied Catalysis B: Environmental, 2004, 49(1): 15–25

79 Vassilev S V, Kitano K, Takeda S, Tsurue T. Influence of mineral and chemical composition of coal ashes on their fusibility. Fuel Processing Technology, 1995, 45(1): 27–51

80 Berkowitz N. An Introduction to Coal Technology. 2nd ed. New York: Academic Press, 1994

81 Jia Y, Lighty J S. Ash particulate formation from pulverized coal under oxy-fuel combustion conditions. Environmental Science & Technology, 2012, 46(9): 5214–5221

82 Wei X, Schnell U, Hein K R. Behaviour of gaseous chlorine and alkali metals during biomass thermal utilisation. Fuel, 2005, 84(7–8): 841–848

83 Niu Y, Tan H, Wang X, Liu Z, Liu Y, Xu T. Study on deposits on the surface, upstream, and downstream of bag filters in a 12 MW biomass-fired boiler. Energy & Fuels, 2010, 24(3): 2127–2132

84 Mu L, Zhao L, Liu L, Yin H. Elemental distribution and mineralogical composition of ash deposits in a large-scale wastewater incineration plant: a case study. Industrial & Engineering Chemistry Research, 2012, 51(25): 8684–8694

85 Lindberg D, Backman R, Chartrand P, Hupa M. Towards a comprehensive thermodynamic database for ash-forming elements in biomass and waste combustion—current situation and future developments. Fuel Processing Technology, 2013, 105: 129–141

86 Garba M U, Ingham D B, Ma L, Porter R T J, Pourkashnian M, Tan H Z, Williams A. Prediction of potassium chloride sulfation and its effect on deposition in biomass-fired boilers. Energy & Fuels, 2012, 26(11): 6501–6508

87 Niu Y, Tan H, Ma L, Pourkashanian M, Liu Z, Liu Y, Wang X, Liu H, Xu T. Slagging characteristics on the superheaters of a 12 MW biomass-fired boiler. Energy & Fuels, 2010, 24(9): 5222–5227

88 Niu Y, Tan H, Wang X, Liu Z, Liu H, Liu Y, Xu T. Study on fusion characteristics of biomass ash. Bioresource Technology, 2010, 101(23): 9373–9381

89 Li Q H, Zhang Y G, Meng A H, Li L, Li G X. Study on ash fusion temperature using original and simulated biomass ashes. Fuel Processing Technology, 2013, 107: 107–112

90 Niu Y, Zhu Y, Tan H, Hui S, Jing Z, Xu W. Investigations on biomass slagging in utility boiler: criterion numbers and slagging growth mechanisms. Fuel Processing Technology, 2014, 128: 499–508

91 Seggiani M. Modelling and simulation of time varying slag flow in a Prenflo entrained-flow gasifier. Fuel, 1998, 77(14): 1611–1621

92 Song W, Tang L, Zhu X, Wu Y, Rong Y, Zhu Z, Koyama S. Fusibility and flow properties of coal ash and slag. Fuel, 2009, 88(2): 297–304

93 Seetharaman S, Mukai K, Sichen D. Viscosities of slags—an overview. Steel Research International, 2005, 76(4): 267–278

94 Ji F. Studies on viscosites of some multicomponent slags. Dissertation for Doctoral Degree. Stockholm: Royal Institute of Techology (KTH), 1998

95 Kondratiev A, Jak E, Hayes P. Predicting slag viscosities in metallurgical systems. JOM, 2002, 54(11): 41–45

96 Seetharaman S. Pertinent properties for metals and slags in continuous casting. In: The Making, Shaping and Treating of Steel. 11th ed. Warrendale, PA: Association for Iron & Steel Technology, 2003

97 Mills K C. Structure and properties of slags used in the continuous casting of steel: part 1 conventional mould powders. ISIJ International, 2016, 56(1): 1–13

98 Henderson G S. The structure of silicate melts: a glass perspective. Canadian Mineralogist, 2005, 43(6): 1921–1958

99 Mysen B, Richet P. Silicate Glasses and Melts. 2nd ed. Elsevier B.V., 2018

100 Kondratiev A, Zhao B, Raghunath S, Hayes P C, Jak E. New tools for viscosity measurement and modelling of fully liquid and partly crystallised slags. In: Proceedings—European Metallurgical Conference, 2007, 2: 953–973

101 Shankar A. Studies on high alumina blast furnace slags. Dissertation for the Doctoral Degree. Stockolm: Royal Institute of Technology (KTH), 2007

102 Mysen B O. Relationships between silicate melt structure and petrologic processes. Earth-Science Reviews, 1990, 27(4): 281–365

103 Sichen D, Bygd’en J, Seetharaman S. A model for estimation of viscosities of complex metallic and ionic melts. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 1994, 25(4): 519–525

104 Mills K. The estimation of slag properties. 2019–10–19, available at the website of pyro.co.za

105 Van Dyk J C, Benson S A, Laumb M L, Waanders B. Coal and coal ash characteristics to understand mineral transformations and slag formation. Fuel, 2009, 88(6): 1057–1063

106 Couch G. Understanding Slagging and Fouling during PF Combustion. IEACR/72. 1994

107 Kekkonen M, Oghbasilasie H, Louhenkilpi S. Viscosity models for molten slags. Technical Report, Finland: Aalto University publication series–Science+ Technology, 2012

108 Dai B, Wu X, Zhang L. Establishing a novel and yet simple methodology based on the use of modified inclined plane (M-IP) for high-temperature slag viscosity measurement. Fuel, 2018, 233: 299–308

109 Kondratiev A, Hayes P C, Jak E. Development of a quasi-chemical viscosity model for fully liquid slags in the Al₂O₃-CaO-‘FeO’-MgO-SiO₂ system. Part 1. description of the model and its application to the MgO, MgO-SiO₂, Al₂O₃-MgO and CaO-MgO sub-systems. ISIJ International, 2006, 46(3): 359–367

110 Urbain G. Viscosity estimation of slags. Steel Research, 1987, 58(3): 111–116

111 Kondratiev A, Jak E. Review of experimental data and modeling of the viscosities of fully liquid slags in the Al₂O₃-CaO-‘FeO’-SiO₂ system. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 2001, 32(6): 1015–1025

112 Riboud P V, Roux Y, Lucai L D, Gaye H. Improvement of continuous casting powders. Fachberichte Huttenpraxis Metallweiterverarbeitung, 1981, 19(8): 859–869

113 Mills K C, Sridhar S. Viscosities of ironmaking and steelmaking slags. Ironmaking & Steelmaking, 1999, 26(4): 262–268

114 Ray H, Pal S. Simple method for theoretical estimation of viscosity of oxide melts using optical basicity. Ironmaking & Steelmaking, 2004, 31(2): 125–130

115 Iida T, Sakai H, Kita Y, Shigeno K. An equation for accurate prediction of the viscosities of blast furnace type slags from chemical composition. ISIJ International, 2000, 40(Suppl): S110–S114

116 Sichen F Z, Du S, Seetharaman S. Experimental studies of the viscosities in the CaO-Fe_nO-SiO₂ slags. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 1997, 28(5): 827–834

117 Sridhar S. Estimation models for molten slag and alloy viscosities. JOM, 2002, 54(11): 46–50

118 Mills K C, Halali M, Löez H P, Kinder A, Pomfref R, Walker B. A simple test for the measurement of slag viscosities. In: Proceedings of the 5th International Conference on Molten Slags, Fluxes and Salts, Sydney, Australia, 1997: 535–542

119 Li F, Li Z, Huang J, Fang Y. Understanding mineral behaviors during anthracite fluidized-bed gasification based on slag characteristics. Applied Energy, 2014, 131: 279–287

120 Kong L, Bai J, Li W, Wen X, Li X, Bai Z, Guo Z, Li H. The internal and external factor on coal ash slag viscosity at high temperatures, Part 1: effect of cooling rate on slag viscosity, measured continuously. Fuel, 2015, 158: 968–975

121 Patterson J H, Hurst H J. Ash and slag qualities of Australian bituminous coals for use in slagging gasifiers. Fuel, 2000, 79(13): 1671–1678

122 Hurst H J, Novak F, Patterson J H. Viscosity measurements and empirical predictions for fluxed Australian bituminous coal ashes. Fuel, 1999, 78(15): 1831–1840

123 Oh M S, Brooker D D, de Paz E F, Brady J J, Decker T R. Effect of crystalline phase formation on coal slag viscosity. Fuel Processing Technology, 1995, 44(1–3): 191–199

124 Schobert H H, Streeter R C, Diehl E K. Flow properties of low-rank coal ash slags: implications for slagging gasification. Fuel, 1985, 64(11): 1611–1617

125 Ilyushechkin A Y, Roberts D. Slagging behaviour of Australian brown coals and implications for their use in gasification technologies. Fuel Processing Technology, 2016, 147: 47–56

126 Moza A K, Austin L G. A new test for characterizing the slag deposition properties of a coal ash: the sticking temperature. Journal of the Institute of Energy, 1979, 52(410): 15–16

127 Wu X, Zhang X, Dai B, Xu X, Zhang J, Zhang L. Ash deposition behaviours upon the combustion of low-rank coal blends in a 3 MW_th pilot-scale pulverised coal-fired furnace. Fuel Processing Technology, 2016, 152: 176–182

128 Li F, Fan H, Fang Y. Exploration of slagging behaviors during multistage conversion fluidized-bed (MFB) gasification of low-rank coals. Energy & Fuels, 2015, 29(12): 7816–7824

129 Wu X, Ji H, Dai B, Zhang L. Xinjiang lignite ash slagging and flowability under the weak reducing environment at 1300°C—a new method to quantify slag flow velocity and its correlation with slag properties. Fuel Processing Technology, 2018, 171: 173–182

130 Dai B, Wu X, Zhao J, Zhang L. Xinjiang lignite ash slagging and flow under the weak reducing environment at high temperatures —slag viscosity and its variation with ash type and addition of clay. Fuel, 2019, 245: 438–446

131 Amini S H, Brungs M P, Ostrovski O, Jahanshani S. Effects of additives and temperature on dissolution rate and diffusivity of lime in Al₂O₃-CaO-SiO₂ based slags. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 2006, 37(5): 773–780

132 Jones E T R, Fishburn R A. Additives to improve slag formation in steelmaking furnaces. US Patent: 3964899, 1976

133 Wu G, Yazhenskikh E, Hack K, Müller M. Viscosity model for oxide melts relevant to fuel slags. Part 2: the system SiO₂-Al₂O₃-CaO-MgO-Na₂O-K₂O. Fuel Processing Technology, 2015, 138: 520–533

134 Wu G, Yazhenskikh E, Hack K, Wosch E, Müller M. Viscosity model for oxide melts relevant to fuel slags. Part 1: pure oxides and binary systems in the system SiO₂-Al₂O₃-CaO-MgO-Na₂O-K₂O. Fuel Processing Technology, 2015, 137: 93–103

135 Benson S A, Austin L G. Crystallization in coal ash slags and its effect on slag strength. In: American Chemical Society Division of Fuel Chemistry Meeting, Toronto, Canada, 1988, 33: CONF-8806136

136 Kondratiev A, Ilyushechkin A. Flow behaviour of crystallising coal ash slags: shear viscosity, non-Newtonian flow and temperature of critical viscosity. Fuel, 2018, 224: 783–800

137 Lejeune A M, Richet P. Rheology of crystal-bearing silicate melts: an experimental study at high viscosities. Journal of Geophysical Research, 1995, 100(B3): 4215–4229

138 Mueller S, Llewellin E W, Mader H M. The rheology of suspensions of solid particles. Proceedings–Royal Society. Mathematical, Physical and Engineering Sciences, 2010, 466(2116): 1201–1228

139 Song W, Tang L, Zhu X, Wu Y, Zhu Z, Koyama S. Flow properties and rheology of slag from coal gasification. Fuel, 2010, 89(7): 1709–1715

140 Seok S H, Jung S M, Lee Y S, Min D J. Viscosity of highly basic slags. ISIJ International, 2007, 47(8): 1090–1096

141 Xuan W, Whitty K J, Guan Q, Bi D, Zhan Z, Zhang J. Influence of SiO₂/Al₂O₃ on crystallization characteristics of synthetic coal slags. Fuel, 2015, 144: 103–110

142 Stam A, Livingston W, Cremers M, Brem G. Review of models and tools for slagging and fouling prediction for biomass co-combustion. In: Workshop on High Cofiring Percentages in New Coal Fired Power Plants, Hamburg, Germany, 2009: 1–18

143 Teixeira P, Lopes H, Gulyurtlu I, Lapa N, Abelha P. Evaluation of slagging and fouling tendency during biomass co-firing with coal in a fluidized bed. Biomass and Bioenergy, 2012, 39: 192–203

144 Degereji M U, Gubba S R, Ingham D B, Ma L, Pourkashanian M, Williams A, Williamson J. Predicting the slagging potential of co-fired coal with sewage sludge and wood biomass. Fuel, 2013, 108: 550–556

145 Heinzel T, Siegle V, Spliethoff H, Hein K R G. Investigation of slagging in pulverized fuel co-combustion of biomass and coal at a pilot-scale test facility. Fuel Processing Technology, 1998, 54(1–3): 109–125

146 Näzelius I L, Fagerström J, Boman C, Boström D, Öhman M. Slagging in fixed-bed combustion of phosphorus-poor biomass: critical ash-forming processes and compositions. Energy & Fuels, 2015, 29(2): 894–908

147 Chen M, Zhao B. Viscosity measurements of the SiO₂-K₂O-CaO system relevant to biomass slags. Fuel, 2016, 180: 638–644

148 Bale C W, Bélisle E, Chartrand P, Decterov S A, Eriksson G, Hack K, Jung I H, Kang Y B, Melançon J, Pelton A D, Robelin C, Petersen S. FactSage thermochemical software and databases— recent developments. Computer Coupling of Phase Diagrams and Thermochemistry, 2009, 33(2): 295–311

149 Suzuki M, Jak E. Quasi-chemical viscosity model for fully liquid slag in the Al₂O₃-CaO-MgO-SiO₂ system. Part II: evaluation of slag viscosities. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 2013, 44(6): 1451–1465

150 Zhang G H, Chou K C. Measuring and modeling viscosity of CaO-Al₂O₃-SiO₂(-K₂O) melt. Metallurgical and Materials Transactions. B, Process Metallurgy and Materials Processing Science, 2012, 43(4): 841–848

151 Lakatos T, Johansson L G, Simmingskold B. Viscosity-temperature relations in the glass system SiO₂-Al₂O₃-Na₂O-K₂O-CaO-MgO in the composition range of technical glasses. Glass Technology, 1972, 13(3): 88–95

152 Bläsing M, Müller M. Mass spectrometric investigations on the release of inorganic species during gasification and combustion of Rhenish lignite. Fuel, 2010, 89(9): 2417–2424

153 Bläsing M, Müller M. Mass spectrometric investigations on the release of inorganic species during gasification and combustion of German hard coals. Combustion and Flame, 2010, 157(7): 1374–1381