Experimental Investigation of the Effect of NiCrTi Coating on the Ash Condensation Characteristics of High-Alkali Coals

Abstract

To investigate the effect of NiCrTi coating on the ash condensation characteristics of high-alkali coal in Xinjiang South Mine, we first built an experimental rig for high-alkali-coal flue gas condensation and carried out experimental research on high-alkali-coal flue gas condensation. Physicochemical characterization of the initial layer of the ash deposit (initial deposit) condensation products was also carried out using XRD, SEM, and EDX. Finally, the priority of products generated on the surface of NiCrTi coating and the three-phase diagram of Na₂O-SiO₂-Al₂O₃ were analyzed by using FactSage 8.3 thermodynamic software. The results show that the condensation products in the initial deposits layer of 15CrMo alloy contain other sodium salts, such as sodium feldspar (NaAlSi₃O₈), NaCl, and Na₂SO₄, and that other protective oxides, such as Cr₂O₃, NiCr₂O₄, and TiO₂, are formed on the surface of the NiCrTi coating. At the same time, the condensation experiment allows the fouling phase to be divided into four parts. Secondly, it was found that the densely flaky particles on the surface of NiCrTi coatings not only have excellent anti-fouling properties but also can effectively inhibit the penetration of other elements such as S. Finally, the reaction priority of protective oxides on NiCrTi coatings was calculated by FactSage 8.3 and found to have the following order: NiCr₂O₄ > Cr₂O₃ > TiO₂. The results of this paper provide theoretical support for the development of anti-staining NiCrTi coatings.

1. Introduction

The Xinjiang Zhundong Coalfield has estimated reserves of about 390 Gt [1]. Zhundong coal is an excellent power coal because of its low ash, low sulfur content, high volatile matter, high combustion rate, ease of ignite, ease of mining, and medium to high calorific value [2]. However, a large amount of sodium salt steam generated in a boiler burning Xinjiang Zhundong high-alkali coal will condense on the heating surface to form viscous initial deposits [3,4], which in turn will lead to serious problems such as fouling and ash corrosion on the heating surface of the boiler, and finally may cause pipe bursts as well as unplanned shutdowns [5,6,7].

Currently, pollution prevention and control techniques for high-alkali-coal-fired boilers include active and passive prevention and control techniques. Among them, active prevention and control techniques include enlarged hearth and combustion optimization; passive prevention and control techniques include fuel modification (blending of low-sodium coal, kaolin clay, and water washing), coating protection, and intensive soot blowing [8,9,10]. However, enlarging of the furnace chamber and intensifying soot blowing have been close to the limit and still can not achieve the efficient use of high-alkali coal, and enlarging of the furnace chamber and excessive soot blowing are not conducive to the safe operation of a unit under deep peaking. The admixture of low-sodium coal or kaolin causes problems such as increased ash content, severe wear and tear on the heating surface of the boiler, and poor flexibility. Therefore, the coating protection technology has attracted much attention, and it can be directly applied to the heating surface of the boiler to prevent the corrosive substances in the flue gas from penetrating the interior of the material and to provide a guarantee for the safe operation of the boiler combustion of high-alkali coal. An extensive reading of the literature on boiler power plants reveals that the main technique used in coating spraying is the supersonic flame spraying (HVOF) technique and that about three-quarters of the coatings are nickel-based [11]. Muthu et al. [12] prepared NiCrAlY and Cr₃C₂-25% NiCr coatings using the HVOF technique. They showed that the Cr₃C₂-25% NiCr coatings exhibited less weight gain than the NiCrAlY coatings after 60 cycles of cyclic hot corrosion in a high-temperature environment at 750 °C. Senthilkumar et al. [13,14] used the HVOF technique to prepare nanostructured and conventionally structured NiCr coatings. The results showed that nanostructured NiCr coatings have a more homogeneous and denser microstructure and have high erosion resistance at all impact angles. Therefore, the new generation of coatings needs to have a combination of corrosion resistance, fouling resistance, and fatigue wear resistance.

Bryers [15,16] found that boiler heat exchanger fouling grows in the order of initial deposits, transitional deposit layers, and outer layers in a fouling deposit, in which the initial deposits are mainly formed by the volatile ash containing sodium salts through the vapor-phase diffusion condensation and the thermal migration and electrophoretic deposition of tiny particles, in which the vapor-phase diffusion condensation of alkali metal salts plays an important role in the growth of the initial deposits. In the initial stage of fouling formation, due to the big temperature difference between the heated surface and the flue gas, high-temperature gaseous alkali metal and alkaline earth metal compounds will collide with the low-temperature heated surface, condense, and finally interact with each other to produce a low-temperature eutectic effect, which strengthens the adhesion of the microscopic particles and impedes the transfer of heat to the initial deposits [17]. Many factors affect fouling, such as mineral composition [18], reaction atmosphere [19], and reaction temperature [20,21]. Guo et al. [22] found that the salts generated from the combination of alkali metals and chlorine tend to condense as microparticles on the heat exchanger tubes, which will elevate the rate of ash deposition on the walls. Wang et al. [23] found that condensation and deposition of sodium sulfate can be promoted at flue gas temperatures of 630–872 °C. A similar conclusion was obtained by Dai et al. [24], showing that gaseous Na or Na₂O released from the combustion of highly alkaline coal tends to condense in the flue gas temperature range of 600–900 °C. Wu et al. [25] found that the condensation content of alkali metal sodium vapors in the flue gas is elevated as the pipe wall temperature is lowered from 800 °C to 600 °C and that sodium oxides condensed on the pipe wall react with sulfur dioxide, alumina, and iron oxides in the flue gas to form a variety of low-melting-point molten salts, such as sodium chloride, sodium sulfate, sodium complex sulfate, and sodium metabisulfite. Zhang et al. [26] also explored the formation rate of the fouling phase using digital imaging techniques and found that the formation rate of the fouling phase on the surface of a nickel-plated probe decreased from 0.060/min to 0.035/min as compared to a pristine probe. Although a series of research results have been achieved at home and abroad, the current research mainly focuses on the initial deposits within the fouling stage. In order to improve the anti-fouling performance of the coating, further research is needed to study the effect of the coating on the condensation characteristics of high-alkali-coal ash in Xinjiang South Mine and then to analyze the physicochemical properties of the condensation products in the initial deposits.

To investigate the effect of NiCrTi coating on the condensation characteristics of high-alkali coal ash in Xinjiang South Mine, this paper carries out an experimental study of flue gas condensation in a two-temperature-zone horizontal tube furnace. First, the crystalline phase composition of the condensation products was analyzed by XRD. Meanwhile, the formation time of the initial deposits was established by the flue gas condensation experiment. Second, SEM-EDX was used to analyze the physicochemical properties of the initial deposit condensation product on the NiCiTi coating. Finally, the chemical reaction priorities of the products generated on the surface of NiCrTi coatings were analyzed by FactSage 8.3. The findings of this paper provide theoretical support for the development of anti-fouling NiCrTi coatings for boilers burning high-alkali coal.

2. Experimental Section

2.1. Sample Preparation

The coal samples, which were high-alkali coals from the South Mine in Xinjiang, were air-dried, ground, and sieved to obtain particle sizes less than 74 μm. All coal samples needed to be dried in an oven at 105 °C for two hours before the experiment. The industrial, elemental, and ash compositional analyses of the coal samples are shown in

Table 1. Properties of high-alkali coal.

Proximate Analysis (wt.%)
Mar	7.63
Aar	2.99
Var	20.37
FCar	47.93
Ultimate analysis (wt.%, dry basis)
Car	54.51
Har	2.87
Oar	10.08
Nar	0.44
Sar	0.41
Clar	0.062
Qnet.ar(kJ/kg)	19.55
Ash composition (wt.%)
SiO2	6.38
Al2O3	5.97
Fe2O3	6.74
CaO	22.5
MgO	12.35
TiO2	0.27
SO3	22.63
K2O	0.19
Na2O	11.88

Each item represents the average of three measurements with a relative error of ±2%.

. From this table, it can be seen that Xinjiang South Mine Coal has highly volatile matter as a good power coal. However, due to the high Na₂O content found in the ash analysis of the coal from Xinjiang South Mine, it can cause severe fouling during combustion.

Before the NiCrTi coating was sprayed, the 15CrMo alloy was ultrasonically cleaned in deionized water and acetone for 10 min to remove the adsorbents such as surface oils and greases. Then, the surface of the substrate was sandblasted to increase the surface roughness and remove the surface oxides. Commercially available NiCrTi alloy powder (58Ni-41Cr-1Ti) was sprayed on 15CrMo alloy (0.95Cr-0.55Mn-0.25Si-0.16C-0.5Mo residual Fe) with dimensions of 20 × 20 × 4 mm using the HVOF technique. The thickness of NiCrTi coating was 400 μm, and the average value of Vickers hardness was 480.2 HV0.1 with a standard deviation of 10.7 HV0.1 (Table S1). The specific process parameters of the HVOF technique are shown in

Table 2. Processing parameters for HVOF spraying.

Parameter	Value
Oxygen flow rate (L/min)	810
Kerosene flow rate (L/min)	0.45
Carrier gas flow rate (N2) (L/min)	10.5
Powder delivery (g/min)	55
Spray distance (mm)	345
Linear velocity (mm/s)	500

. The instrument used is HP/HVOF-JP8000 (Praxair, Inc., Danbury, CT, USA). The spraying technology for the coatings was from Shaoxing Spray Micronano Technology Co., Ltd. (Shaoxing, China).

2.2. Experimental Setup

The experiments were carried out in a two-temperature-zone horizontal tube furnace, as shown in Figure 1. The system consists of an air intake system, a flow control system, a coal-fired furnace, a flue gas condensing furnace, and an exhaust system in which the solution is a sodium hydroxide solution. The length of the quartz tube is 200 cm, the inner diameter is 50 mm, the length of the constant temperature heating of each temperature zone is 150 mm, the maximum working temperature is 1200 °C, the rate of temperature increase is 10 °C/min, and the furnace temperature is measured with a K-type thermocouple with an accuracy of 1 °C.

In order to focus on exploring the effect of sodium salts in high-alkali coals on the growth characteristics of the initial deposits, it was necessary to control the Ca content in the flue gas, so the temperature of the coal-fired furnace was set at 1000 °C [27,28]. During the combustion process, sodium salts are released mainly in the form of NaCl, Na₂SO₄, sodium atoms, organic sodium, and silicate (aluminate) of sodium, which condenses on the heat exchanger surface, so the temperatures of the flue gas condensing furnace (TIANJIN ZHONGHUAN FURNACE Corp, Tianjin, China) were set to 600, 700, and 800 °C, to simulate the real environment of the heating surface [29]. The test samples were selected from boiler superheater tubing 15CrMo alloy [30] and NiCrTi-coated steel plate [14], both with dimensions of 20 × 20 × 4 mm. Before the experiment, the test samples were degreased and cleaned in an ethanol solution using ultrasonic agitation (Milton-Roy, Lakeland, FL, USA), dried, and then placed in a flue gas condensing oven (TIANJIN ZHONGHUAN FURNACE Corp, Tianjin, China) for testing. After the six-hour test, the samples were removed [31]. The crucible was cleaned, washed, and dried.

2.3. Condensation Experiments

This experiment maintained a coal sample feed rate of 20 g/h (10 g of pulverized coal was placed in the coal-fired furnace every 30 min to ensure a continuous supply of flue gas). The flow rate was set to 2 L/min via a mass flow meter to ensure an excess air factor of 1.2 [31]. 15CrMo alloy and NiCrTi-coated steel plates were selected as test materials and placed in the flue gas for 1–8 h and weighed in 1 h gradients.

$$\mathsf{\Delta }M=M_t-M_o$$

(1)

$$\mathsf{\Delta }{M}_f={\displaystyle \frac{\mathsf{\Delta }M}{S}}$$

(2)

Since the mass change during the growth phase of the initial deposits is very weak, the mass change per unit area per unit time was used to more accurately explore the end time of the growth of the initial deposits. The mass of the sample before the experiment per unit time is M_o/g; the mass of the sample after the experiment per unit time is M_t/g; the change in mass per unit time is ΔM/g; the external surface area of the steel plate is S/cm²; and the change in mass per unit time per unit area is ΔM_f/(g/cm²). The equations for calculating the change in mass per unit area per unit time are shown in Equations (1) and (2).

2.4. Methods of Analysis

An X-ray fluorescence spectrometer (XRF, Model FP-6500, Jasco, Tokyo, Japan) was used to determine the chemical composition of coal ash samples from Xinjiang South Mine. The crystalline phase composition of the sample surface was characterized by an X-ray diffractometer (XRD, HITACHI STA7300, HITACHI, Tokyo, Japan). A scanning electron microscope (SEM, Zeiss Supra55 VP, Carl Zeiss AG, Oberkochen, Germany) and an energy-dispersive X-ray spectrometer (EDX, Bruker X-Flash SDD 5010, Bruker, Berlin, Germany) were used to analyze the micro-morphology and elemental distribution on the surface and cross-section of the samples. Each experiment was repeated 3 times in parallel and 3 times for some of the tests to minimize measurement errors.

As an auxiliary tool, Gibbs free energy change (ΔG) and enthalpy change (ΔH) of chemical reactions of different substances were also investigated using Factsage 8.3 (GTT-Technologies, Herzogenrath, Germany) thermodynamic equilibrium software, selecting FactPS and FToxide databases.

3. Results and Discussion

3.1. Effect of Temperature on Flue Gas Condensation Products

As shown in Figure 2a, the main crystalline phase compositions on the surface of 15CrMo alloy after the condensation experiment were analyzed by XRD. As the temperature was increased from 600 °C to 700 °C, the Fe₃O₄ phase disappeared, and the content of the Fe₂O₃ phase increased, probably due to the oxidation reaction (3). NaCl is found at all three temperatures, which is related to the release of sodium chloride in reactions (4–5) [31]. Complex sodium salts, such as sodium feldspar (NaAlSi₃O₈), are produced at 700 and 800 °C. When the temperature was raised from 700 °C to 800 °C, the peak of Na₂SO₄ disappeared, and Na₆Ca(SO₄)₄ was generated.

4Fe₃O₄ + O₂ → 6Fe₂O₃

(3)

$${\mathrm{N}\mathrm{a}}^{+}+{\mathrm{C}\mathrm{l}}^{-}\stackrel{{600 }^{°}\mathrm{C}}{\to }\mathrm{N}\mathrm{a}\mathrm{C}\mathrm{l}$$

(4)

$$\mathrm{N}\mathrm{a}\mathrm{C}\mathrm{l}(\mathrm{s})\stackrel{800-{1000 }^{°}\mathrm{C}}{\leftrightarrow }\mathrm{N}\mathrm{a}\mathrm{C}\mathrm{l}(\mathrm{g})$$

(5)

As shown in Figure 2b, the main crystalline phases of the NiCrTi coating after the condensation experiment were elucidated by XRD analysis. As the temperature increases from 600 °C to 800 °C, Ni₃Fe gradually decomposes into Ni ions, and finally, NiCr₂O₄ is produced, which is related to reaction (6) and R1-2 in

Table 3. Main reactions between the flue gas and sample surface at 1000 °C.

Major Products	Reaction	Delta H/(KJ/mol)	Delta G/(KJ/mol)	Reaction Number
NiCr2O4	NiO + Cr2O3 → NiCr2O4	−10.17	−11.68	R1
	Ni + 2Cr + 2O2 → NiCr2O4	−1355.67	−937.53	R2
TiO2	Ti + O2 → TiO2	−940.97	−714.76	R3
Cr2O3	4Cr + 3O2 → 2Cr2O3	−1112.67	−799.76	R4
GaSO4	CaO + SO3 → GaSO4	−378.64	−173.28	R5
	2CaO + 2SO2 + O2 → 2GaSO4	−475.13	−153.14	R6
CaSiO3	CaO + SiO2 → CaSiO3	−92.71	−87.72	R7
	CaCO3 + SiO2 → CaSiO3 + CO2	71.79	−108.06	R8
	CaSO4 + SiO2 → CaSiO3 + SO3	285.93	85.56	R9
Fe2SiO4	2FeO + SiO2 → Fe2SiO4	−44.68	−21.20	R10
	2Fe2O3 + 2SiO2 → 2Fe2SiO4 + O2	234.80	110.31	R11
	2Fe3O4 + 3SiO2 → 3Fe2SiO4 + O2	153.76	86.96	R12
NaAlSi3O8	Na2O + Al2O3 + 5SiO2 → 2NaAlSi3O8	−170.83	−193.11	R13
	Na2SO4+ Al2O3 + 6SiO2 → 2NaAlSi3O8 + SO3	103.50	12.08	R14
	2NaCl + Al2O3 + 6SiO2 + H2O → 2NaAlSi3O8 + 2HCl	58.26	−19.02	R15

[31]. Trace amounts of CaSiO₃ were also detected at 800 °C. As the temperature increases from 700 °C to 800 °C, the proportion of Cr₂O₃ content decreases, and the oxides that play a protective role during this period are mainly NiCr₂O₄, and these oxides can improve the anti-fouling performance very well.

Ni₃Fe → 3Ni²⁺ + Fe⁶⁺

(6)

The highest content of sodium salts was found in the flue gas environment at 700 °C, and the optimum condensation temperature interval for sodium salts was (670 ± 20) °C [32]. Therefore, the condensation experiments were analyzed at 700 °C in a flue gas environment.

3.2. Condensation Experiment

Changes in quality are induced as the staining phase progresses, and the change is mainly due to phenomena such as gas-phase condensation deposition, corrosion, mineral growth, and particle collision. As the fouling phase progresses, the quality changes accordingly and shows a pattern [33].

The results of mass change per unit area per unit time of 15CrMo alloy at 700 °C are shown in Figure 3. It can be found that the mass change per unit area is not a horizontal line with time, which indicates that there will be regular changes in the mass at different growth stages [4,16,26,31]. The first stage occurs at 0–1 h, with a large rate of mass growth per unit of time, which is because the surface of 15CrMo alloy is rapidly oxidized to generate oxides, such as Fe₂O₃, Fe₃O₄, and FeO, in the beginning stage. The second phase occurs from 1 to 5 h and shows a general trend of slow decrease, which may be due to the slowing down of the oxidation rate as the experiment progresses. There will be a gas-phase condensation deposition of sodium salts. The third stage occurs at 5–6 h with an upward trend. Although the oxidation rate decreases further with time, the slope increases due to the continuation of gas-phase condensation deposition during the growth phase of the initial deposits, resulting in elevated viscosity, which traps large particles during the transition to the transitional deposit layers. The fourth stage occurs after six hours and shows a decreasing trend, which may be because the oxidation of the steel plate will be further inhibited by the growth of the transitional deposit layers [34]. It can be roughly inferred that the transition time from the initial deposits to the transitional deposit layers within the surface of 15CrMo alloy is 5–6 h. Therefore, 0–6 h is the growth stage of the initial deposits within the surface of 15CrMo alloy. Similarly, the transition stage from the initial deposits to the transitional deposit layers within the NiCrTi coating also occurs within 5–6 h intervals. Therefore, the growth stage of the initial deposits within the surface of the NiCrTi coating is also 0–6 h.

The mass change of both the 15CrMo alloy and NiCrTi coating constantly trends upward with time, and the mass change per unit area per unit time of the 15CrMo alloy is always greater than that of the NiCrTi-coated steel plate. This is because NiCrTi coatings generate protective oxides such as Cr₂O₃, NiCr₂O₄, and TiO₂ on the surface to enhance fouling resistance.

3.3. Analysis of the Surface and Cross-Section of NiCrTi Coatings After Condensation Experiments

Figure 4 shows the microstructure of the condensation products on the nicrti coating, and the samples were subsequently cast in epoxy resin to perform cross-sectional analysis, as shown in Figure 5.

Figure 4a shows that the NiCrTi coating before the experiment was mainly composed of some smooth spherical particles (40–50 μm), but after the experiment, two microstructures appeared. As shown in Figure 4b, the NiCrTi coating becomes smooth from the original uneven spherical particles after the condensation experiment. The mass ratio of the main elements in the smooth part is shown in Figure 4d, which can be presumed to be the main substances of Na₂SO₄ and Cr₂O₃ in combination with the XRD analysis. This may be due to the high Na content at some locations and the low-melting-point molten salt generated on the surface of NiCrTi coatings. As the temperature increases and the viscosity decreases, better fluidity results, making the coating surface smooth [35]. A thin chromium oxide layer is also formed on the NiCrTi coating, which in turn improves the fouling resistance of the boiler heat exchanger tubes.

As shown in Figure 4c, compact flaky particles are also found on the NiCrTi coatings, and the major elemental mass ratio is shown in Figure 4d, which can be combined with the XRD analysis to speculate that the major substances are NiCr₂O₄, TiO₂, Cr₂O₃, and NaCl. Due to the low Na content and the formation of protective oxides such as NiCr₂O₄, TiO₂, and Cr₂O₃ on the surface, phenomena such as surface smoothness do not occur, and the dense lamellar particles formed on the surface of the coating can provide good protection for the internal substrate [34].

As shown in Figure 5, a typical microstructure of the NiCrTi coating cross-section can be observed. Cracks were not detected at the cross-section, indicating that the substrate has a good bond strength with the NiCrTi coating. The content of the Fe element increases from the coating to the steel plate, and the opposite is true for the O element, indicating that the Fe element tends to diffuse outward, while the O element tends to diffuse inward. Ti in the NiCrTi coating is not a simple component addition; its performance on the coating has great improvement, and the addition of titanium can promote the stability of the coating and improve its bonding with the substrate [36]. From the distribution of elements such as Ni, Cr, Ti, and O in the cross-section, it can be found that the dense protective oxide film has good anti-fouling properties.

3.4. Factsage Simulations

Table 3 shows the main chemical reactions of 15CrMo alloy and NiCrTi coating surface products. ΔG and ΔH for a chemical reaction can be calculated using FactSage 8.3. ΔG allows you to determine whether the reaction can proceed spontaneously, and ΔH allows you to determine whether the reaction is absorbing or exothermic. In general, reactions with lower ΔG and ΔH are easier to perform. Therefore, two independent chemical reactions can be compared to determine the reaction priority by comparing ΔG and ΔH, which in turn can be corroborated with the XRD analysis results.

Reactions R1–4 represent the protective oxides generated on the surface of NiCrTi coatings, which can effectively inhibit the penetration of aggressive elements in the flue gas. By comparing the magnitude of the chemical reactions’ ΔG and ΔH on the surface of NiCrTi coatings, it can be hypothesized that the reaction priority of the protective oxidizing substances is NiCr₂O₄ > Cr₂O₃ > TiO₂, and this result can also be verified by the results of XRD analyses in Figure 2b. The reaction priority for generating the same calcium-based product is CaO > CaCO₃ > CaSO₄. For example, the reaction priority for generating CaSiO₃ is R7 > R8 > R9. Similarly, the reaction priority for generating the same iron-based product is FeO > Fe₃O₄ > Fe₂O₃. Finally, it can be found that the reaction priority for generating the same sodium-based product is Na₂O > NaCl > Na₂SO₄ by comparing R13–15 [37].

The three-phase diagram of Na₂O-SiO₂-Al₂O₃ is given in Figure 6. In this phase diagram, coal ash from the South Mine in Xinjiang is located at the coordinate points (0.281, 0.206, and 0.513), and the three values refer to the relative mass contents of normalized Na₂O, SiO₂, and Al₂O₃, respectively. We can observe a variety of complex sodium salts in the three-phase diagram, and the generation of a large amount of sodium feldspar (NaAlSi₃O₈) is detected in the XRD analysis of Figure 2a, which is further supported by FactSage calculations. The various minerals react with each other at certain temperatures, creating a low-temperature eutectic effect that increases the risk of fouling. Compared to the results of the XRD analysis, the simulations can still be used to demonstrate trends, although the non-equilibrium nature of the experiments leads to significant differences from the simulations.

Figure 7 represents the reaction process of 15CrMo alloy and NiCrTi coating surface products. The ions in the upper part of the sample come from the flue gas. On the surface of the sample are ions from the sample itself and newly generated compounds. It can be found that NiCr₂O₄ and Cr₂O₃ are generated on the NiCrTi coating, and these oxides can well enhance its anti-fouling properties. Due to the high content of alkali metal oxides, especially Na₂O, in the ash analysis of the Zhundong coal, NaCl can be found on both 15CrMo alloy and NiCrTi coatings at 700 °C, and sodic feldspar (NaAlSi₃O₈) was also detected on the 15CrMo alloy. Due to the heating temperature of the coal-fired furnace of 1000 °C, some calcium salts are released into the flue gas, resulting in trace amounts of hard gypsum (CaSO₄) detectable on the surface of the 15CrMo alloy and NiCrTi coatings. Hard gypsum is a fluxing mineral that acts as a binder, and hard gypsum and wollastonite (CaSiO₃) will condense to form low-melting-point eutectic crystals on the surface of the samples, which in turn enhances the viscosity of the initial deposits, which will continue to adsorb large particles, thus facilitating the fouling process [25]. Although the experimental environment is partially different from real engineering applications, the experimental results can still provide informative value for real engineering applications.

4. Conclusions

In order to investigate the effect of NiCrTi coating on the condensation characteristics of high-alkali-coal ash in Xinjiang South Mine, this paper carried out an experimental study of flue gas condensation in a two-temperature-zone horizontal tube furnace. The crystalline phase compositions of the condensation products were analyzed, the time of formation of the initial deposits was established, and the physicochemical properties of the initial deposits’ condensation products were also analyzed. Finally, the reaction priorities of the generated products were calculated using FactSage 8.3, which led to the following main conclusions:

(1) An analysis of the condensation products’ crystalline phase composition revealed that the NiCrTi coating’s protective oxides were mainly Cr₂O₃ and NiCr₂O₄, and trace amounts of sodium salts were also detected. In addition to a large number of iron oxides, other sodium salts such as sodium feldspar (NaAlSi₃O₈), Na₆Ca(SO₄)₄, NaCl, and Na₂SO₄ were detected on the surface of the 15CrMo alloy.

(2) The flue gas condensation experiments allow the fouling stage to be divided into four parts, and the transition time from the initial deposits to the transitional deposit layers on the surface of both the 15CrMo alloy and NiCrTi coatings was determined to be six hours at 700 °C.

(3) The protective oxide film on the NiCrTi coating is mainly composed of compact flaky particles, which have excellent anti-fouling properties and can also effectively inhibit the erosion of other elements, such as S.

(4) The ΔG and ΔH of the chemical reactions were calculated by FactSage 8.3, and it was found that the reaction priority of the protective oxides on the surface of NiCrTi coatings was NiCr₂O₄ > Cr₂O₃ > TiO₂. The production of large amounts of sodium feldspar (NaAlSi₃O₈) can also be found in the three-phase diagram of Na₂O-SiO₂-Al₂O₃.