- Email: [email protected]
Effect of water vapor on coal ash slag viscosity under gasification condition
Fuel 237 (2019) 18–27
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Effect of water vapor on coal ash slag viscosity under gasification condition a,b
a
a,⁎
Xi Cao , Lingxue Kong , Jin Bai , Zefeng Ge Wen Lia a b
a,b
a,b
, Chong He
a
T
a
, Huaizhu Li , Zongqing Bai ,
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China University of Chinese Academy of Sciences, Beijing 100049, PR China
A R T I C LE I N FO
A B S T R A C T
Keywords: Entrained flow gasification Slag viscosity Water vapor Slag structure
The entrained flow gasification process employs a high temperature, high pressure slagging gasifier, in which the viscosity of the slag plays a key role in determining operating conditions. Many studies have focused on the viscosity of slag under the reducing atmosphere, especially CO + CO2 and CO + H2. As an important component in the gasification syngas, the effect of water vapor on the slag viscosity temperature properties is not yet clear. In order to keep the stable operation of the gasifier, the effect of water vapor on the viscosity temperature properties of the slag were studied in this work. Ash fusion temperatures (AFTs) decrease with the increasing water vapor proportion. Although mineral species do not change obviously, the content of SiO2 in the ash decreases, and more amorphous substances forms when water vapor is added. The slag viscosity and temperature of critical viscosity (TCV) also decrease with the introduction of water vapor, while the effect is not obvious due to the low solubility at ambient pressure. It is confirmed that water vapor weakens the melts network structure by breaking [Si–O–Si] bonds. Meanwhile, the network modifier, [AlO6]9−, increases with the increasing water vapor proportion. Besides, water vapor inhibits the growth of the crystal. As its proportion increases, the average particle size of crystals decreases, leading to the decrease of the slag viscosity and TCV. The results provide a theoretical basis for the effect of water vapor on the slag viscosity and will be benefit for the operation of entrained flow gasification, especially for the coal water slurry gasification.
1. Introduction Coal, as the main energy material, plays a dominant role in China’s energy consumption and is an important guarantee for the rapid and stable development of the economy [1,2]. Coal gasification offers one of the most versatile and clean ways to convert coal into electricity, hydrogen, and other valuable energy products [3]. It can not only reduce the emission of harmful gases, but also can significantly increase the energy efficiency of coal. According to flow mechanics in the gasifier, coal gasification technologies are divided into fixed-bed (e.g., Lurgi and UGI), fluidized-bed (e.g., U-gas and KBR,) and entrained flow bed gasification (e.g., Shell, GE and OMB) [4,5]. In recent years, entrained flow gasification has become a predominant gasification technology in coal gasification due to its high throughput and feedstock flexibility. The entrained flow gasifiers usually operate under a high temperature (usually higher than 1300 °C) and a high pressure [6]. Entrained flow gasifiers are slagging gasifiers. Under this condition, the evolution behavior of inorganic substances (minerals, ash and slag) in the coal at high temperature, especially the slag tapping, is a key factor
⁎
for long-term running of an entrained flow gasifier. An appropriate slag viscosity property was required for the steady and reliable discharge of slag [7,8]. The high slag viscosity could cause slag blockage, while a low slag viscosity will result in rapid refractory wear [9]. According to the feedstock, entrained flow gasification includes pulverized coal gasification process (dry feed) and coal water slurry gasification process (slurry feed) [10]. For the pulverized coal process which features a membrane wall (such as the Shell gasifier), slag viscosity-temperature behavior is the key parameter to guide the smooth operation of the gasifier. It is generally accepted that slag viscosity should be 2.5–25 Pa·s at the slagging temperature when the temperature is between 1300 and 1500 °C [11]. At the same time, in order to prevent the rapid increase of slag viscosity caused by temperature fluctuations and the corresponding slagging problems in the gasifier, the slagging temperature should be higher than TCV. Besides, the slag should be a glassy slag, of which the viscosity increases continuously as the temperature decreases. For the coal-water slurry gasifier (such as GE and OMB) [12], refractory lined are used for heat insulation, and the thermal resistance is mainly concentrated on the refractory brick layer. Basically, the slag viscosity is
Corresponding author. E-mail address: [email protected] (J. Bai).
https://doi.org/10.1016/j.fuel.2018.09.137 Received 31 July 2018; Received in revised form 12 September 2018; Accepted 25 September 2018 0016-2361/ © 2018 Published by Elsevier Ltd.
Fuel 237 (2019) 18–27
X. Cao et al.
required in the range of 10–25 Pa·s under normal operation to ensure smooth slagging. The slag viscosity-temperature characteristics are affected mainly by the chemical compositions of the coal slag. The other factors affecting slag viscosity-temperature characteristics are temperature, atmosphere and operating parameters of the gasifier [13]. According to the network theory of molten slag at high temperature, the coal ash chemical compositions are divided into three categories: network formers, network modifiers and amphoterics [14]. The network formers (e.g. Si4+ and Ti4+), which become the polymers, increase the slag viscosity; the network modifiers (e.g. Na+, K+, Mg2+, Ca2+ and Fe2+) destroy the polymers, decreasing the slag viscosity; amphoterics (e.g. Al3+ and Fe3+), which can play as network formers or network modifiers in different system [15]. Atmosphere is also an important factor, affecting the slag viscosity. It mainly attributes to reduction of the oxidation state of iron under high temperature. Under reducing atmosphere, a part of the iron that would normally be Fe3+ is reduced to Fe2+. Fe3+ tends to enhance the three-dimensional structure of the melt and increases the slag viscosity [16]. However, Fe2+ is likely to disrupt the connectivity of network by providing non-bridging oxygen (NBO), lowering the slag viscosity. Except for the difference of the feedstock and inner wall for pulverized coal gasifier and coal water slurry gasifier, the compositions of the syngas also varied a lot. For most of the gasifier, the main components of the syngas are CO, CO2, H2 and H2O, while the proportion of each component is different. For example, in Shell gasifier, the ratio of CO, CO2, H2 and H2O in syngas is 62.2%, 2.3%, 31.6% and 4.0%, respectively. In Texaco gasifier, the percentage of CO, CO2, H2 and H2O is 35.4%, 15.5%, 34.4% and 14.7%, respectively [17]. In OMB gasifier, the syngas consists of 37.68% CO, 11.83% CO2, 26.54% H2 and 23.95% H2O [18]. It can be concluded that water vapor is an important component in the gasification syngas. According to statistics of the syngas in the coal-water slurry gasification, the water vapor content in the gasifier and the slagging outlet is about 10–20%. Therefore, it is necessary to investigate the influence of water vapor on the slag viscosity behavior. In this work, a typical coal in the coal-water slurry gasification was selected. The slag viscosity-temperature characteristics under different atmosphere (without and with different proportions of water vapor) were studied. X-Ray diffraction (XRD) was used to analyze the mineralogical compositions of the ash at high temperatures. The slag structures were characterized by Nuclear Magnetic Resonance Spectrometer (NMR) and Fourier Transform Infrared Spectroscopy (FTIR). Besides, solid phases in the slag during cooling were investigated by Scanning Electron Microscope and Energy Dispersive X-ray Spectroscopy (SEMEDX).
Table 2 Chemical compositions of YZ ash (wt%). SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
K2O
Na2O
TiO2
P2O5
15.08
8.55
22.30
33.59
2.37
12.49
0.05
0.79
0.35
0.02
Fig. 1. A schematic diagram of water vapor generator.
according to GB/T212-2008 and GB/T476-2001, and the results are listed in Table 1. The coal ash was prepared at 815 °C in a muffle furnace based on GB/T 1574-2007. X-ray fluorescence (XRF) (Bruker S8 Tiger, Germany) was used to characterize the chemical compositions of the ash, as listed in Table 2. The coal ash was rich in calcium oxide and iron oxide, but the contents of silicon oxide and aluminum oxide were low.
2.2. Water vapor generator A small fixed bed reactor which combined with a micro water pump (AP0010, Sanotac, China) was used to generate water vapor, and a schematic diagram is given in Fig. 1. The reactor was heated by a heating belt and the temperature was kept at 150 °C to ensure that water can be completely vaporized. A K-type thermocouple and a digital temperature controller (XMTD-2001, 0–399 °C) were used to control the temperature. The flow rate of water was accurately controlled by the micro metering pump which provided a flow rate from 0.001 ml/min to 10 ml/min with an accuracy of ± 0.5% [19]. Because the composition of the reducing atmosphere at various temperatures would be affected by possible water gas shift reaction (Eq. (1)), an argon gas (Ar) was selected to carry the water vapor. The total flow rate of the mixture gas was 700 ml/min, and the proportion of water vapor was 10%, 15% and 20%, respectively. In this work, ideal gas equation (pV = nRT) was used to calculate the amount of liquid water.
2. Experimental 2.1. Sample A bituminous coal in coal water slurry gasification, Yanzhou coal (denoted as YZ coal, Shandong province, North China), was selected in this work. The coal sample was crushed and ground to less than 75 μm. The proximate and ultimate analyses of the coal were performed
(1)
CO + H2 O→ CO2 + H2
Table 1 Proximate and ultimate analyses of YZ coal. Proximate analysis (wt%, ad)
Ultimate analysis (wt%, ad)
Moisture
Ash
Volatile
Fixed Carbon
Carbon
Hydrogen
Oxygena
Nitrogen
Sulfur
5.47
5.46
34.63
54.44
74.46
4.45
8.82
0.91
0.43
ad: air dry base. a By difference. 19
Fuel 237 (2019) 18–27
X. Cao et al.
2.3. Ash fusion temperature test
2.6. Characterization of quenched slags
The AFTs of YZ coal ash were measured under different atmosphere (without and with different proportion of water vapor) by the Carbolite CAF 1600 ash fusion point meter (Carbolite Gero Limited, Britain). According to the American standard ASTM D1857-04, the prepared ash cone was heated to 1560 °C at 8 °C/min. The four characteristic temperatures, deformational temperature (DT), softening temperature (ST), hemispherical temperature (HT), and flow temperature (FT), were determined based on the specific shapes of ash cone. The AFT test was repeated several times, and it showed a good reproducibility of results under the same conditions.
2.6.1. XRD analysis In order to identify the mineral compositions of YZ coal ash at high temperatures, an X-ray powder diffract meter (BRUKER D2, Germany) was applied by using Cu Kα radiation (40 kV, 100 mA). The samples were scanned with the step size of 0.02° over a 2θ range of 5–80°. Besides, SIROQUANT software was carried out to quantify the content of minerals in the slag [23]. The software used the full-profile Rietveld method to refine the shape of a calculated XRD pattern against the profile of a measured pattern. A crystalline zinc oxide (ZnO) as the spiking material was added into the samples to determine the content of crystalline minerals and amorphous.
2.4. Slag viscosity measurement 2.6.2. NMR The silicon and aluminum nuclei were obtained using Bruker Avance III 600 MHz Wide Bore spectrometer (14.1T), which was supplied by Kratos Analytical Inc. (Spring Valley, NY). All samples were measured in powdered form. Spinning frequency was 8 and 13 KHz, respectively. The chemical shift of 29Si and 27A1 nuclei were referenced to TMS (tetramethylsilane) and Al(NO3)3 solution as an external reference, respectively. For 29Si NMR, due to the high iron content of slag, no signal was detected in this study. 27A1 spectra were taken at 77.48 MHz for single pulse experiments.
The viscosities of YZ slag were measured by a Theta high-temperature rotating viscometer (Theta Industries, Inc., USA) under different atmosphere (without and with different proportions of water vapor). The ash samples were firstly melted in an electric furnace at a temperature 100 °C higher than the liquidus temperature (Tliq) which was calculated by FactSage to get a homogeneous slag. Tliq of coal ash is above the temperature at which the last solid phase disappears. Then the pre-melted slag was crushed to less than 3 mm for the viscosity measurement. The start temperature for test was higher than Tliq to make sure the sample totally melted. The maximum temperature of the furnace viscometer can be up to 1680 °C. The alumina rotors and cylinder crucibles were used during the process. A standard reference material 717A glass was used to calibrate the parameters of the rotor and crucible [20]. The procedure was as following: the furnace was heated to 1400 °C at 10 °C/min, and kept for 10 min, then heated to 1600 °C at 5 °C/min and hold for 10 min to be thermally in equilibrium. Then the spindle was lowered into the melt. After reaching equilibrium, the temperature was lowered linearly at 3 °C/min. The viscosity and temperature were recorded continuously at an interval of 0.1 °C, and the viscosity temperature curve was obtained. The spindle was raised out from the slag immediately when the viscosity was over 300 Pa·s or a significant solidification of the melt occurred [21]. Then the slag was cooled to room temperature for further characterization. In order to verify the results of the experiment, the slag viscosity measurement was repeated at least three times. The repeatability of the slag viscosity results is higher than 90%.
2.6.3. FTIR A Bruker Vertex 70 Fourier transform infrared spectrometer (Bruker, Germany) was used to investigate the structure of silicate melts under different atmosphere. The KBr drifts technique was used in the experiment. The powder sample was mixed with KBr with the ratio of 1:200 (weigh ratio). The finely grinded mixture was pressed into a thin tablet under 12 MPa. The spectra were recorded over the wave number range of 400–4000 cm−1 with a resolution of 0.4 cm−1. 2.6.4. SEM-EDX A JSM-7001F SEM was employed to characterize the solid phase in slag during cooling under different atmosphere. The morphology of the slag was observed in back-scattering mode (BSE), and EDX was used to analyze the chemical compositions of the solid phases. 3. Results and discussion 3.1. AFTS of YZ coal ash
2.5. Preparation of quenched slags
The AFTs of coal ash were measured under Ar, Ar + 10% water vapor, Ar + 15% water vapor, Ar + 20% water vapor, as shown in Fig. 2. The AFTs decreased slightly with increasing the proportion of
A series of quenched slags were prepared at 900–1400 °C under different atmosphere (Ar, Ar + 10% water vapor, Ar + 20% water vapor) in a horizontal tube furnace to analyze the mineral transformation during the heating. The maximum temperature of the furnace can reach 1650 °C, and the precision of temperature was ± 1 °C. When the temperature reached 800 °C and maintained for 10 min, the water vapor was input. At low temperature, water will condense, which will result in damage of corundum tube. When the furnace was heated to target temperature, the sample was pushed into the furnace and hold for 20 min. To maintain the slag composition and slag structure at high temperatures, the sample was pulled out and immersed into the ice water immediately [22]. Then the experimental gases were switched to Ar by stopping the water pump and the furnace was cooled to room temperature. Quenched slags at high temperatures were also prepared in the horizontal furnace to explore the slag structure and solid phases during cooling. The heating operation was the same as mentioned above. To ensure the totally melt of the sample, the slag was heated to 1550 °C and hold for 10 min. Then the temperature lowered at 3 °C/min to the target temperature. Afterwards, the sample was taken out and quenched in the ice water. All quenched slags were treated for further test.
Fig. 2. AFTs measured under Ar with water vapor. 20
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 3. Mineral compositions of the ashes under Ar with water vapor. 1-CaCO3 (calcite); 2-Fe2O3 (hematite); 3-SiO2 (quartz); 4-CaSO4 (anhydrite); 5-Fe3O4 (magnetite); 6-Ca2Al2SiO7 (gehlenite).
3.2. Effect of the water vapor on mineral evolution
water vapor, while its effect on DT was more obvious than that on ST, HT, and FT. For example, ST, HT and FT were 1280 °C, 1284 °C and 1290 °C under Ar, and they dropped to 1273 °C, 1277 °C and 1279 °C under Ar + 20% water vapor, respectively. When the proportion of water vapor was 10%, DT was 1270 °C, and it was 1272 °C under Ar. However, it decreased by 25 °C from 1272 °C to 1247 °C when the proportion of water vapor was 20%. The results demonstrate that water vapor is favor of decreasing the AFTs of the coal ash. Furthermore, DT was closely related to the formation of the initial liquid phase in the coal ash [24]. The decrease in DT implies that the increasing water vapor lowers the formation temperature of initial liquid phase.
Fig. 3 shows the mineral transformation of the coal ash under different atmosphere in the range of 900–1400 °C. Although the ash samples were prepared under different atmosphere, mineral species were similar with each other at the same temperature. The major minerals at 900 °C and 1000 °C were calcite (CaCO3), quartz (SiO2), anhydrite (CaSO4), and hematite (Fe2O3). According to the coal ash compositions, the content of CaO is 33.59%, most of the calcium in the ash existed in the form of CaCO3 and CaSO4. As the temperature increased to 1100 °C, the minerals in coal ash were mainly SiO2, CaSO4, Fe3O4 (magnetite) and Ca2Al2SiO7. Ca2Al2SiO7 was formed owing to 21
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 4. Minerals contents in the ashes under Ar with water vapor at different temperatures.
reaction between SiO2, Al2O3 and CaO. CaCO3 disappeared because it decomposed to CaO and CO2. At 1200 °C, the main minerals were Fe3O4 and Ca2Al2SiO7. SiO2 and CaSO4 disappeared gradually due to the formation of Ca2Al2SiO7. When the temperature reached to 1300 °C and 1400 °C, gehlenite and magnetite were the main minerals in the coal ash. Therefore, the main reactions can be listed as follows:
The contents of minerals in the coal ash under Ar with water vapor were further analyzed by Siroquant software. From Fig. 4, the content of SiO2 decreases slightly with the increasing water vapor proportion. At 900 °C, the SiO2 content decreased from 4.3% to 3.8% and 3.1% when the proportion of water vapor increases from 0% to 10% and 20%. The melting temperature of SiO2 was 1610 °C, and the high content of SiO2 can lead to a higher AFTs. This is a reason for the decrease of the AFTs with the addition of water vapor. The CaCO3 content dropped by 8.9% from 37.6% to 28.7%, while the content of CaSO4 raised by 2.1% from 6.5% to 8.6% under Ar and Ar + 20% water vapor, respectively. The opposite tendency was presented between CaCO3 and
CaCO3 → CaO + CO2
CaSO4 → CaO + SO3 Al2O3 + SiO2 + 2CaO → Ca2Al2SiO7 22
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 5. (a) Slag viscosity-temperature curves under Ar with water vapor. (b) slag viscosity at above 1460 °C; (c) TCV of the slag under Ar with water vapor.
It has been widely accepted that the coal ash slag was silicate melt as glass and magmas. Water molecule can enter into the slag structure as an –OH and destroy the Si–O–Si bonds, leading to depolymerization of the melt [26]. The depolymerization is attributed to the decrease of the slag viscosity. Therefore, the effect of water vapor on slag viscosity is similar with NBO which breaks the network structure, reducing the slag viscosity [27,28].
CaSO4, because CaO which is from the decomposition of CaCO3 reacted with SO3 to form CaSO4 which had a high decomposition temperature. Subsequently, CaSO4 will decompose to form Ca2Al2SiO7. Meanwhile, the content of Ca2Al2SiO7 decreased and amorphous substance increased with the increasing water vapor proportion. For example, at 1200 °C, the content of Ca2Al2SiO7 were 11.9%, 10.3%, and 8.0% under Ar, Ar + 10% water vapor, Ar + 20% water vapor. The content of amorphous substance was 62.7%, 65.7%, and 70.3%. The similar results were observed at other temperatures except for the ashes at 1400 °C, at which temperature the content of Ca2Al2SiO7 and amorphous substance remained a constant. The decrease of crystalline substances content and the increasing amorphous substance improve the proportion of liquid phase in the slag [25]. Therefore, a larger fraction of liquid phase in the slag accounts for the decrease of AFTs.
H2 O + Si−O−Si → (−Si−OH HO−Si−) The small effect of water vapor on slag viscosity and TCV was mainly due to the low solubility of water at ambient pressure. From previous studies, the solubility of water at ambient pressure is under 1 wt% [29]. According to the relative research in magmas systems, the influence of water was significant compared with that on coal ash slag. The fact is that the effect observed in magmas was at a high pressure, even over 1GPa. Under a high pressure, the water solubility can be several hundred orders of magnitude [30] and the influence would be much more pronounced. In fact, the real pressure in an entrained flow gasifier (e.g., 8.5 MPa, for a slurry-fed gasifier) is lower than that in the magmas system which was happened in the deep earth in nature [31]. Therefore, it can be speculated that the effect of water vapor on the viscosity of molten slag should be greater under gasification conditions than that at ambient pressure.
3.3. Viscosity temperature behavior of slags The viscosity results of the YZ slag under Ar, Ar + 10% water vapor and Ar + 20% water vapor were measured, as shown in Fig. 5(a–c). From the viscosity temperature curves (Fig. 5(a)), the slag exhibited the behavior of the crystalline slag under Ar with different proportion of water vapor. The viscosity rises rapidly when temperature is lower than TCV. Meanwhile, the slag viscosity decreased with the increasing proportion of water vapor, as shown in Fig. 5(b). At 1465 °C, the slag viscosity was 2.47 Pa·s under Ar. However, it dropped to 1.62 Pa·s and 0.69 Pa·s when the proportion of water vapor was 10% and 20%. Besides, the TCV decreased from 1430 °C to 1370 °C and 1350 °C when the proportion of water vapor increased from 0% to 10% and 20%, as shown in Fig. 5(c). These demonstrate that water vapor lowers slag viscosity and TCV of the slag, but the influence was slight.
3.4. Effect of water vapor on slag structure at high temperature 3.4.1. Effect of water vapor on Si–O As shown in Fig. 6, the characteristic bands of Si–O symmetric stretching are between the wavenumbers 2000–500 cm−1. The bands at about 800 cm−1 and 960 cm−1 were assigned to the asymmetric and 23
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 8. 27Al NMR of spectra under Ar with water vapor. (a) Ar; (b) Ar + 10% water vapor; (c) Ar + 20% water vapor. Fig. 6. FTIR spectra under different atmosphere. (a) Ar; (b) Ar + 10% water vapor; (c) Ar + 20% water vapor.
degree of polymerization. As shown in Fig. 7(b), it is noted that R increased slightly from 3.83% to 3.90% (0.13%) when the proportion of water vapor increased from 0% to 10%, but it increased drastically to 5.72% when the proportion of water vapor was 20%. It demonstrates that water vapor breaks the silicate network structure by decreasing the polymerization degree of melt. However, the effect of water vapor on the slag structure was small in this study. Previous studies [35,36] have found that compared with a high polymerized melt (like acidic melts), water causes a smaller change for a low depolymerized melt (usually basic melts) even at high temperature and high pressure. Because the ash used in this work is a basic slag for the high content of calcium and iron, it has a low polymerization degree. Thus, it shows a small effect on the structure of the slag at ambient pressure.
symmetric vibrations of Si–O–Si. The symmetric stretching of Si–O–Si and Al–O–Si were at 726 cm−1. The shoulder at about 678 cm−1 and 880 cm−1 were the stretching vibrations of Al–O [32,33]. According to the characteristic stretching vibrations of SiO4 tetrahedra, Q-species can be distinguished. Qn represents the degree of polymerization of [SiO4] tetrahedral specie unit, and n is the number of Si–O–Si bridging oxygen, which indicates the degree of polymerization of slag. The convoluted spectral were categorized into five types, namely, monomers with [SiO4]4− framework(Q0), dimers with [Si2O7]6− units(Q1), chains with [SiO3]2− skeleton(Q2), sheets with [Si2O5]2− units(Q3) and three-dimensional network of SiO2(Q4). On the basis of the previous works, the wavenumbers of Q-species were as follows [34]: 880–850 cm−1 for Q0, 920–900 cm−1 for Q1, 1000–950 cm−1 for Q2, and 1100–1050 cm−1 for Q3 and 1200–1060 cm−1 for Q4. The typical deconvoluted peaks along with the maximum randomness in residual distributions are shown in Fig. 7(a), the relative contents of Qn (n = 0–4) are showed in Fig. 7(b). Q2 was the major structural units in the slag under Ar with and without water vapor. The contents of Q0 and Q1 did not change significantly with the increasing water vapor proportion. When the proportion of water vapor increased from 0% to 20%, Q2 increased from 51.40% to 57.30%, and the content of Q3 decreased by 2.56% from 17.44% to 14.88% at the same time. Q4 disappeared when the proportion of water vapor was 20%. The decreasing Q3 and Q4 indicates that the addition of water vapor leads to the decrease of the polymerization degree of silicate network. Q0 + Q1 + Q2/Q3 + Q4 (denoted as R) was applied to describe the
3.4.2. Effect of water vapor on Al–O Fig. 8 shows the 27Al-NMR spectra of the slag under different atmosphere. 27Al-NMR spectra usually have two characteristic peaks: 10–20 ppm for six-coordinate Al([AlO6]9−), which weakens the degree of polymerization with octahedral structure, and 50–100 ppm for fourcoordinated Al([AlO4]5−), which enhances the polymerization degree with tetrahedron structure. Therefore, the higher percentage of Al(IV) in the melt will increase the slag viscosity. The resonance at 64.6 ppm was assigned to Al(IV) and Al(VI) resonance was centered at 14.5 ppm [37,38]. In order to know the effect of water vapor on Al(IV) and Al(VI), a typical deconvoluted peaks of 27Al NMR spectra are shown in Fig. 9(a). Usually, the integrated area was used to present the content of Al(IV)
Fig. 7. (a) Typical fitted FTIR spectra under Ar atmosphere; (b) Fraction of structural units Qn based on FTIR spectra. 24
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 9. (a) A typical deconvolution of
27
Al NMR spectra; (b) [AlO4]5−/([AlO4]5− + [AlO6]9−).
Fig. 10. SEM-EDX results of slag quenched at 1320 °C. (a) Ar; (b) Ar + 10%H2O; (c) Ar + 20%H2O. (□ for Ca2Al2SiO7 and ○for CaAl4O7).
and Al(VI). The fitting results indicate that the signal intensity is not necessarily equivalent to the real amount of Al(IV) and Al(VI) in the structure [39]. Because the slag in this study was a basic slag, Al(IV) was the main resonance in melt (about 90.0%) and Al3+ is mainly [AlO4]5−, which acts as a network former. As shown in Fig. 9(b), the ratio of [AlO4]5−/[AlO4]5−+[AlO6]9− decreased with the increasing water vapor proportion. Under Ar atmosphere, the ratio was 90.37%, and it dropped to 90.27% and 85.63% under Ar + 10% water vapor and Ar + 20% water vapor, respectively. The reducing [AlO4]5− suggests that a part of [AlO4]5− transformed into [AlO6]9− with the addition of water vapor. The decreasing
[AlO4]5− also indicates that the existed of water vapor weakens the polymerization degree of the slag. In the process, a part of Al3+ enters into the network skeleton as [AlO6]9− and acts as a network modifier. Therefore, the slag viscosity decreased with the increasing water vapor content. Furthermore, due to the similarity structures between silicon oxide and aluminum oxide, the interpretations of NMR are difficult [40]. Therefore, most researches about the effect of water vapor on silicate melt are mainly concentrated on the aluminum-free system, and the specific mechanism needs a further study [41].
25
Fuel 237 (2019) 18–27
X. Cao et al.
of water vapor. (4) The average particle size of solid phases in slag decreased with the increasing water vapor proportion. Water vapor impedes the growth of crystals, which is one of reasons for the decrease of the temperature of critical viscosity. Acknowledgements This work was supported by the National Key R&D program of China (Grant Number 2017YFB0602603), Joint Foundation of Natural Science Foundation of China and Shanxi Province (Grant Number U1510201), National Natural Science Foundation of China (Grant Numbers 21476247 and 21761132032), Joint Foundation of Natural Science Foundation of China and Xinjiang (Grant Number U1703252), Shanxi Province Science Foundation (Grant Number 201601D201003), Youth Innovation Promotion Association, CAS, Bureau of International Cooperation, Chinese Academy of Sciences (Grant Number GJHZ1879). Fig. 11. XRD patterns of the slags quenched at 1320 °C. Ar; (b) Ar + 10% water vapor; (C) Ar + 20% water vapor. 1-Ca2Al2SiO7 (gehlenite); 2-CaAl4O7 (calcium dialuminate); 3-FeAl2O4(hercynite).
References [1] Han S, Chen H, Long R, Cui X. Peak coal in China: a literature review. Resour Conserv Recycl 2018;129:293–306. [2] Chen W, Xu R. Clean coal technology development in China. Energy Policy 2010;38(5):2123–30. [3] Pereira EG, da Silva JN, de Oliveira JL, Machado CS. Sustainable energy: a review of gasification technologies. Renewable Sustainable Energy Rev 2012;16(7):4753–62. [4] Minchener AJ. Coal gasification for advanced power generation. Fuel 2005;84(17):2222–35. [5] Lee HH, Lee JC, Joo YJ, Oh M, Lee CH. Dynamic modeling of Shell entrained flow gasifier in an integrated gasification combined cycle process. Appl Energy 2014;131:425–40. [6] Bi D, Guan Q, Xuan W, et al. Combined slag flow model for entrained flow gasification. Fuel 2015;150:565–72. [7] Chen X, Kong L, Bai J, Dai X, Li H, Bai Z, et al. The key for sodium-rich coal utilization in entrained flow gasifier: the role of sodium on slag viscosity-temperature behavior at high temperatures. Appl Energy 2017;206:1241–9. [8] Crnomarkovic N, Repic B, Mladenovic R, Neskovic O, Veljkovic M. Experimental investigation of role of steam in entrained flow coal gasification. Fuel 2007;86(1–2):194–202. [9] Song W. Measurement and simulation of flow properties of coal ash slag in coal gasification. AIChE 2011. [10] He C, Bai J, Kong L, Guhl S, Schwitalla DH, Xu J, et al. The precipitation of metallic iron from coal ash slag in the entrained flow coal gasifier: by thermodynamic calculation. Fuel Process Technol 2017;162:98–104. [11] Hurst HJ, Novak F, Patterson JH. Phase diagram approach to the fluxing effect of additions of CaCO3 on Australian coal ashes. Energy Fuels 1996;10(6):1215–9. [12] Guo Q, Zhou Z, Wang F, Yu G. Slag properties of blending coal in an industrial OMB coal water slurry entrained-flow gasifier. Energy Convers Manage 2014;86:683–8. [13] Tsuruda A, Arma LH, Takebe H. Viscosity measurement and prediction of gasified and synthesized coal slag melts. Fuel 2017;200:521–8. [14] Vargas S, Frandsen FJ, Dam-Johansen K. Rheological properties of high-temperature melts of coal ashes and other silicates. Prog Energy Combust Sci 2001;27(3):237–429. [15] Gao J, Wen G, Huang T, et al. Effects of the composition on the structure and viscosity of the CaO-SiO2-based mold flux. J Non-Cryst Solids 2016;435:33–9. [16] Wu X, Zhang Z, Piao G, He X, Chen Y, et al. Behavior of mineral matters in Chinese coal ash melting during char-CO2/H2O gasification reaction. Energy Fuels 2009;23(5):2420–8. [17] Bell DA, Towler BF, Fan M. Chapter 4 – Gasifiers. Coal Gasification and Its Applications. Boston: William Andrew Publishing; 2011. p. 73–100. [18] Jin W, Wang Y, Zhu L, Wu C, Xu J. Numerical simulation and assessment of a twostage gasifier modified from an opposed multi-burner gasifier. Chem Eng Technol 2014;37(3):483–94. [19] Ahmed MYM. Phase Equilibria Between Iron and Slag in CO/CO2/H2/H2O Atmospheres Relevant to a novel Flash Ironmaking Technology Doctoral dissertation University of Utah; 2013. [20] Kong L, Bai J, Bai Z, Guo Z, Li W. Effects of CaCO3 on slag flow properties at high temperatures. Fuel 2013;109:76–85. [21] Kong L, Bai J, Li W, et al. 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;158:968–75. [22] Ma Z, Bai J, Li W, Bai Z, Kong L. Mineral transformation in char and its effect on coal char gasification reactivity at high temperatures, part 1: mineral transformation in char. Energy Fuels 2013;27(8):4545–54. [23] Grigore M, Sakurovs R, French D, Sahajwalla V. Influence of mineral matter on coke reactivity with carbon dioxide. ISIJ Int 2006;46(4):503–12. [24] Yan T, Bai J, Kong L, Bai Z, Li W, Xu J. Effect of SiO2/Al2O3 on fusion behavior of coal ash at high temperature. Fuel 2017;193:275–83. [25] Kondratiev A, Ilyushechkin A. Flow behaviour of crystallising coal ash slags: shear viscosity, non-Newtonian flow and temperature of critical viscosity. Fuel
3.5. Effect of water vapor on solid phases during cooling The slags quenched at 1320 °C under Ar, Ar + 10% water vapor and Ar + 20% water vapor was analyzed by SEM-EDX, and the results are presented in Fig. 10. Two crystals with different shape (quadrilateral and short column) were observed in the slag. To verify the types of the crystals, EDX combined with XRD analyses (Fig. 11) were conducted. The results showed that the short column crystals were Ca2Al2SiO7 and the cubic crystals could be CaAl4O7. It can be clearly seen that the average size of solid particles was decreased with the increasing of water vapor content, especially for the size of cubic crystals. The average particle size of the cubic crystal was 80–150 um, 50–90 um and 30–70 um when the atmosphere were Ar, Ar + 10% water vapor and Ar + 20% water vapor, respectively. This demonstrates that water vapor inhibits the growth of the crystals in slag during cooling. Previous researchers [42,43] studied the effect of crystal size on the viscosity of several slags with the addition of the solid particles. They found that the slag viscosity increases as the crystal size increases. High content of water vapor leads to smaller-sized crystals, which lowers the slag viscosity. Moreover, the crystal size is closely related to the TCV [44]. A TCV event usually occurs when the size of the crystals significantly increases. The smaller crystals also contributed to the decrease of the TCV. 4. Conclusions In this work, we investigated the influence of water vapor on the slag viscosity behavior at high temperatures. A typical coal used in coal water slurry gasifier was selected. The influence of water vapor on slag structure and solid phase during cooling was discussed in detail. The main conclusions were summarized as following: (1) Due to the low solubility of water vapor at ambient pressure, the presence of water vapor leads to a small decrease in the AFTs, slag viscosity and TCV, while its effect increases with the increasing proportion. (2) Although the effect of water vapor on mineral species is small, the addition of water vapor decreases the content of SiO2, which lowered the AFTs. Meanwhile, the water vapor is beneficial to the formation of amorphous substances, and it does not favor the formation of crystalline minerals. (3) The structural effect of water vapor on Si–O and Al–O is revealed. Water enters into the slag structure as an –OH and destroys the Si–O–Si bonds, which results in a low polymerization degree. Besides, Al3+ in the slag is likely to be [AlO6]9− due to the addition 26
Fuel 237 (2019) 18–27
X. Cao et al.
Solids 2017;473:79–86. [36] Whittington A, Richet P, Holtz F. Water and the viscosity of depolymerized aluminosilicate melts. Geochim Cosmochim Acta 2000;64(21):3725–36. [37] Le Losq C, Neuville DR, Florian P, Henderson GS, Massiot D. The role of Al3+ on rheology and structural changes in sodium silicate and aluminosilicate glasses and melts. Geochim Cosmochim Acta 2014;126:495–517. [38] Lin X, Ideta K, Miyawaki J, Takebe H, Yoon S-H, Mochida I. Correlation between fluidity properties and local structures of three typical asian coal ashes. Energy Fuels 2012;26(4):2136–44. [39] Pardal X, Brunet F, Charpentier T, et al. 27Al and 29Si solid-state NMR characterization of calcium-aluminosilicate-hydrate. Inorg Chem 2012;51(3):1827–36. [40] Xue X, Kanzaki M. Dissolution mechanisms of water in depolymerized silicate melts: constraints from 1H and 29Si NMR spectroscopy and ab initio calculations. Geochim Cosmochim Acta 2004;68(24):5027–57. [41] S.C. Kohn, The dissolution mechanisms of water in silicate melts; a synthesis of recent data, (2000) 64(3):389–408. [42] Wright S, Zhang L, Sun S, et al. Viscosities of calcium ferrite slags and calcium alumino-silicate slags containing spinel particles. J Non-Cryst Solids 2001;282(1):15–23. [43] Wright S, Zhang L, Sun S, et al. Viscosity of a CaO-MgO-Al2O3-SiO2 melt containing spinel particles at 1646K. Metall Mater Trans B 2000;31(1):97–104. [44] Ilyushechkin AY, Hla SS, Roberts DG, Kinaev NN. The effect of solids and phase compositions on viscosity behaviour and TCV of slags from Australian bituminous coals. J Non-Cryst Solids 2011;357(3):893–902.
2018;224:783–800. [26] Wagstaff FE, Richards KJ. Kinetics of crystallization of stoichiometric SiO2 glass in H2O atmospheres. J Am Ceram Soc 1966;49(3):118–21. [27] Gonzalez-Oliver CJR, Johnson PS, James PF. Influence of water content on the rates of crystal nucleation and growth in lithia-silica and soda-lime-silica glasses. J Mater Sci 1979;14(5):1159–69. [28] Fujita S, Sakamoto A, Tomozawa M. Behavior of water in glass during crystallization. J Non-Cryst Solids 2003;320(1–3):56–63. [29] Folkedahl BC, Schobert HH. Effects of atmosphere on viscosity of selected bituminous and low-rank coal ash slags. Energy Fuels 2005;19(1):208–15. [30] Ban-Ya S, Hino M, Nagasaka T. Estimation of water vapor solubility in molten silicates by quadratic formalism based on the regular solution model. ISIJ Int 1993;33(1):12–9. [31] Browning GJ, Bryant GW, Hurst HJ, et al. An empirical method for the prediction of coal ash slag viscosity. Energy Fuels 2003;17(3):731–7. [32] Mohassab Y, Sohn HY. Analysis of slag chemistry by FTIR-RAS and Raman spectroscopy: effect of water vapor content in H2–H2O-CO-CO2 mixtures relevant to a novel green ironmaking technology. Steel Res Int 2015;86(7):740–52. [33] Jiang Y, Lin X, Ideta K, Takebe H, Miyawaki J, Yoon S-H, et al. Microstructural transformations of two representative slags at high temperatures and effects on the viscosity. J Ind Eng Chem 2014;20(4):1338–45. [34] Baasner A, Schmidt BC, Webb SL. The effect of chlorine, fluorine and water on the viscosity of aluminosilicate melts. Chem Geol 2013;357:134–49. [35] Gao E, Wang W, Zhang L. Effect of alkaline earth metal oxides on the viscosity and structure of the CaO-Al2O3 based mold flux for casting high-al steels. J Non-Cryst
27
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Effect of water vapor on coal ash slag viscosity under gasification condition a,b
a
a,⁎
Xi Cao , Lingxue Kong , Jin Bai , Zefeng Ge Wen Lia a b
a,b
a,b
, Chong He
a
T
a
, Huaizhu Li , Zongqing Bai ,
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China University of Chinese Academy of Sciences, Beijing 100049, PR China
A R T I C LE I N FO
A B S T R A C T
Keywords: Entrained flow gasification Slag viscosity Water vapor Slag structure
The entrained flow gasification process employs a high temperature, high pressure slagging gasifier, in which the viscosity of the slag plays a key role in determining operating conditions. Many studies have focused on the viscosity of slag under the reducing atmosphere, especially CO + CO2 and CO + H2. As an important component in the gasification syngas, the effect of water vapor on the slag viscosity temperature properties is not yet clear. In order to keep the stable operation of the gasifier, the effect of water vapor on the viscosity temperature properties of the slag were studied in this work. Ash fusion temperatures (AFTs) decrease with the increasing water vapor proportion. Although mineral species do not change obviously, the content of SiO2 in the ash decreases, and more amorphous substances forms when water vapor is added. The slag viscosity and temperature of critical viscosity (TCV) also decrease with the introduction of water vapor, while the effect is not obvious due to the low solubility at ambient pressure. It is confirmed that water vapor weakens the melts network structure by breaking [Si–O–Si] bonds. Meanwhile, the network modifier, [AlO6]9−, increases with the increasing water vapor proportion. Besides, water vapor inhibits the growth of the crystal. As its proportion increases, the average particle size of crystals decreases, leading to the decrease of the slag viscosity and TCV. The results provide a theoretical basis for the effect of water vapor on the slag viscosity and will be benefit for the operation of entrained flow gasification, especially for the coal water slurry gasification.
1. Introduction Coal, as the main energy material, plays a dominant role in China’s energy consumption and is an important guarantee for the rapid and stable development of the economy [1,2]. Coal gasification offers one of the most versatile and clean ways to convert coal into electricity, hydrogen, and other valuable energy products [3]. It can not only reduce the emission of harmful gases, but also can significantly increase the energy efficiency of coal. According to flow mechanics in the gasifier, coal gasification technologies are divided into fixed-bed (e.g., Lurgi and UGI), fluidized-bed (e.g., U-gas and KBR,) and entrained flow bed gasification (e.g., Shell, GE and OMB) [4,5]. In recent years, entrained flow gasification has become a predominant gasification technology in coal gasification due to its high throughput and feedstock flexibility. The entrained flow gasifiers usually operate under a high temperature (usually higher than 1300 °C) and a high pressure [6]. Entrained flow gasifiers are slagging gasifiers. Under this condition, the evolution behavior of inorganic substances (minerals, ash and slag) in the coal at high temperature, especially the slag tapping, is a key factor
⁎
for long-term running of an entrained flow gasifier. An appropriate slag viscosity property was required for the steady and reliable discharge of slag [7,8]. The high slag viscosity could cause slag blockage, while a low slag viscosity will result in rapid refractory wear [9]. According to the feedstock, entrained flow gasification includes pulverized coal gasification process (dry feed) and coal water slurry gasification process (slurry feed) [10]. For the pulverized coal process which features a membrane wall (such as the Shell gasifier), slag viscosity-temperature behavior is the key parameter to guide the smooth operation of the gasifier. It is generally accepted that slag viscosity should be 2.5–25 Pa·s at the slagging temperature when the temperature is between 1300 and 1500 °C [11]. At the same time, in order to prevent the rapid increase of slag viscosity caused by temperature fluctuations and the corresponding slagging problems in the gasifier, the slagging temperature should be higher than TCV. Besides, the slag should be a glassy slag, of which the viscosity increases continuously as the temperature decreases. For the coal-water slurry gasifier (such as GE and OMB) [12], refractory lined are used for heat insulation, and the thermal resistance is mainly concentrated on the refractory brick layer. Basically, the slag viscosity is
Corresponding author. E-mail address: [email protected] (J. Bai).
https://doi.org/10.1016/j.fuel.2018.09.137 Received 31 July 2018; Received in revised form 12 September 2018; Accepted 25 September 2018 0016-2361/ © 2018 Published by Elsevier Ltd.
Fuel 237 (2019) 18–27
X. Cao et al.
required in the range of 10–25 Pa·s under normal operation to ensure smooth slagging. The slag viscosity-temperature characteristics are affected mainly by the chemical compositions of the coal slag. The other factors affecting slag viscosity-temperature characteristics are temperature, atmosphere and operating parameters of the gasifier [13]. According to the network theory of molten slag at high temperature, the coal ash chemical compositions are divided into three categories: network formers, network modifiers and amphoterics [14]. The network formers (e.g. Si4+ and Ti4+), which become the polymers, increase the slag viscosity; the network modifiers (e.g. Na+, K+, Mg2+, Ca2+ and Fe2+) destroy the polymers, decreasing the slag viscosity; amphoterics (e.g. Al3+ and Fe3+), which can play as network formers or network modifiers in different system [15]. Atmosphere is also an important factor, affecting the slag viscosity. It mainly attributes to reduction of the oxidation state of iron under high temperature. Under reducing atmosphere, a part of the iron that would normally be Fe3+ is reduced to Fe2+. Fe3+ tends to enhance the three-dimensional structure of the melt and increases the slag viscosity [16]. However, Fe2+ is likely to disrupt the connectivity of network by providing non-bridging oxygen (NBO), lowering the slag viscosity. Except for the difference of the feedstock and inner wall for pulverized coal gasifier and coal water slurry gasifier, the compositions of the syngas also varied a lot. For most of the gasifier, the main components of the syngas are CO, CO2, H2 and H2O, while the proportion of each component is different. For example, in Shell gasifier, the ratio of CO, CO2, H2 and H2O in syngas is 62.2%, 2.3%, 31.6% and 4.0%, respectively. In Texaco gasifier, the percentage of CO, CO2, H2 and H2O is 35.4%, 15.5%, 34.4% and 14.7%, respectively [17]. In OMB gasifier, the syngas consists of 37.68% CO, 11.83% CO2, 26.54% H2 and 23.95% H2O [18]. It can be concluded that water vapor is an important component in the gasification syngas. According to statistics of the syngas in the coal-water slurry gasification, the water vapor content in the gasifier and the slagging outlet is about 10–20%. Therefore, it is necessary to investigate the influence of water vapor on the slag viscosity behavior. In this work, a typical coal in the coal-water slurry gasification was selected. The slag viscosity-temperature characteristics under different atmosphere (without and with different proportions of water vapor) were studied. X-Ray diffraction (XRD) was used to analyze the mineralogical compositions of the ash at high temperatures. The slag structures were characterized by Nuclear Magnetic Resonance Spectrometer (NMR) and Fourier Transform Infrared Spectroscopy (FTIR). Besides, solid phases in the slag during cooling were investigated by Scanning Electron Microscope and Energy Dispersive X-ray Spectroscopy (SEMEDX).
Table 2 Chemical compositions of YZ ash (wt%). SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
K2O
Na2O
TiO2
P2O5
15.08
8.55
22.30
33.59
2.37
12.49
0.05
0.79
0.35
0.02
Fig. 1. A schematic diagram of water vapor generator.
according to GB/T212-2008 and GB/T476-2001, and the results are listed in Table 1. The coal ash was prepared at 815 °C in a muffle furnace based on GB/T 1574-2007. X-ray fluorescence (XRF) (Bruker S8 Tiger, Germany) was used to characterize the chemical compositions of the ash, as listed in Table 2. The coal ash was rich in calcium oxide and iron oxide, but the contents of silicon oxide and aluminum oxide were low.
2.2. Water vapor generator A small fixed bed reactor which combined with a micro water pump (AP0010, Sanotac, China) was used to generate water vapor, and a schematic diagram is given in Fig. 1. The reactor was heated by a heating belt and the temperature was kept at 150 °C to ensure that water can be completely vaporized. A K-type thermocouple and a digital temperature controller (XMTD-2001, 0–399 °C) were used to control the temperature. The flow rate of water was accurately controlled by the micro metering pump which provided a flow rate from 0.001 ml/min to 10 ml/min with an accuracy of ± 0.5% [19]. Because the composition of the reducing atmosphere at various temperatures would be affected by possible water gas shift reaction (Eq. (1)), an argon gas (Ar) was selected to carry the water vapor. The total flow rate of the mixture gas was 700 ml/min, and the proportion of water vapor was 10%, 15% and 20%, respectively. In this work, ideal gas equation (pV = nRT) was used to calculate the amount of liquid water.
2. Experimental 2.1. Sample A bituminous coal in coal water slurry gasification, Yanzhou coal (denoted as YZ coal, Shandong province, North China), was selected in this work. The coal sample was crushed and ground to less than 75 μm. The proximate and ultimate analyses of the coal were performed
(1)
CO + H2 O→ CO2 + H2
Table 1 Proximate and ultimate analyses of YZ coal. Proximate analysis (wt%, ad)
Ultimate analysis (wt%, ad)
Moisture
Ash
Volatile
Fixed Carbon
Carbon
Hydrogen
Oxygena
Nitrogen
Sulfur
5.47
5.46
34.63
54.44
74.46
4.45
8.82
0.91
0.43
ad: air dry base. a By difference. 19
Fuel 237 (2019) 18–27
X. Cao et al.
2.3. Ash fusion temperature test
2.6. Characterization of quenched slags
The AFTs of YZ coal ash were measured under different atmosphere (without and with different proportion of water vapor) by the Carbolite CAF 1600 ash fusion point meter (Carbolite Gero Limited, Britain). According to the American standard ASTM D1857-04, the prepared ash cone was heated to 1560 °C at 8 °C/min. The four characteristic temperatures, deformational temperature (DT), softening temperature (ST), hemispherical temperature (HT), and flow temperature (FT), were determined based on the specific shapes of ash cone. The AFT test was repeated several times, and it showed a good reproducibility of results under the same conditions.
2.6.1. XRD analysis In order to identify the mineral compositions of YZ coal ash at high temperatures, an X-ray powder diffract meter (BRUKER D2, Germany) was applied by using Cu Kα radiation (40 kV, 100 mA). The samples were scanned with the step size of 0.02° over a 2θ range of 5–80°. Besides, SIROQUANT software was carried out to quantify the content of minerals in the slag [23]. The software used the full-profile Rietveld method to refine the shape of a calculated XRD pattern against the profile of a measured pattern. A crystalline zinc oxide (ZnO) as the spiking material was added into the samples to determine the content of crystalline minerals and amorphous.
2.4. Slag viscosity measurement 2.6.2. NMR The silicon and aluminum nuclei were obtained using Bruker Avance III 600 MHz Wide Bore spectrometer (14.1T), which was supplied by Kratos Analytical Inc. (Spring Valley, NY). All samples were measured in powdered form. Spinning frequency was 8 and 13 KHz, respectively. The chemical shift of 29Si and 27A1 nuclei were referenced to TMS (tetramethylsilane) and Al(NO3)3 solution as an external reference, respectively. For 29Si NMR, due to the high iron content of slag, no signal was detected in this study. 27A1 spectra were taken at 77.48 MHz for single pulse experiments.
The viscosities of YZ slag were measured by a Theta high-temperature rotating viscometer (Theta Industries, Inc., USA) under different atmosphere (without and with different proportions of water vapor). The ash samples were firstly melted in an electric furnace at a temperature 100 °C higher than the liquidus temperature (Tliq) which was calculated by FactSage to get a homogeneous slag. Tliq of coal ash is above the temperature at which the last solid phase disappears. Then the pre-melted slag was crushed to less than 3 mm for the viscosity measurement. The start temperature for test was higher than Tliq to make sure the sample totally melted. The maximum temperature of the furnace viscometer can be up to 1680 °C. The alumina rotors and cylinder crucibles were used during the process. A standard reference material 717A glass was used to calibrate the parameters of the rotor and crucible [20]. The procedure was as following: the furnace was heated to 1400 °C at 10 °C/min, and kept for 10 min, then heated to 1600 °C at 5 °C/min and hold for 10 min to be thermally in equilibrium. Then the spindle was lowered into the melt. After reaching equilibrium, the temperature was lowered linearly at 3 °C/min. The viscosity and temperature were recorded continuously at an interval of 0.1 °C, and the viscosity temperature curve was obtained. The spindle was raised out from the slag immediately when the viscosity was over 300 Pa·s or a significant solidification of the melt occurred [21]. Then the slag was cooled to room temperature for further characterization. In order to verify the results of the experiment, the slag viscosity measurement was repeated at least three times. The repeatability of the slag viscosity results is higher than 90%.
2.6.3. FTIR A Bruker Vertex 70 Fourier transform infrared spectrometer (Bruker, Germany) was used to investigate the structure of silicate melts under different atmosphere. The KBr drifts technique was used in the experiment. The powder sample was mixed with KBr with the ratio of 1:200 (weigh ratio). The finely grinded mixture was pressed into a thin tablet under 12 MPa. The spectra were recorded over the wave number range of 400–4000 cm−1 with a resolution of 0.4 cm−1. 2.6.4. SEM-EDX A JSM-7001F SEM was employed to characterize the solid phase in slag during cooling under different atmosphere. The morphology of the slag was observed in back-scattering mode (BSE), and EDX was used to analyze the chemical compositions of the solid phases. 3. Results and discussion 3.1. AFTS of YZ coal ash
2.5. Preparation of quenched slags
The AFTs of coal ash were measured under Ar, Ar + 10% water vapor, Ar + 15% water vapor, Ar + 20% water vapor, as shown in Fig. 2. The AFTs decreased slightly with increasing the proportion of
A series of quenched slags were prepared at 900–1400 °C under different atmosphere (Ar, Ar + 10% water vapor, Ar + 20% water vapor) in a horizontal tube furnace to analyze the mineral transformation during the heating. The maximum temperature of the furnace can reach 1650 °C, and the precision of temperature was ± 1 °C. When the temperature reached 800 °C and maintained for 10 min, the water vapor was input. At low temperature, water will condense, which will result in damage of corundum tube. When the furnace was heated to target temperature, the sample was pushed into the furnace and hold for 20 min. To maintain the slag composition and slag structure at high temperatures, the sample was pulled out and immersed into the ice water immediately [22]. Then the experimental gases were switched to Ar by stopping the water pump and the furnace was cooled to room temperature. Quenched slags at high temperatures were also prepared in the horizontal furnace to explore the slag structure and solid phases during cooling. The heating operation was the same as mentioned above. To ensure the totally melt of the sample, the slag was heated to 1550 °C and hold for 10 min. Then the temperature lowered at 3 °C/min to the target temperature. Afterwards, the sample was taken out and quenched in the ice water. All quenched slags were treated for further test.
Fig. 2. AFTs measured under Ar with water vapor. 20
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 3. Mineral compositions of the ashes under Ar with water vapor. 1-CaCO3 (calcite); 2-Fe2O3 (hematite); 3-SiO2 (quartz); 4-CaSO4 (anhydrite); 5-Fe3O4 (magnetite); 6-Ca2Al2SiO7 (gehlenite).
3.2. Effect of the water vapor on mineral evolution
water vapor, while its effect on DT was more obvious than that on ST, HT, and FT. For example, ST, HT and FT were 1280 °C, 1284 °C and 1290 °C under Ar, and they dropped to 1273 °C, 1277 °C and 1279 °C under Ar + 20% water vapor, respectively. When the proportion of water vapor was 10%, DT was 1270 °C, and it was 1272 °C under Ar. However, it decreased by 25 °C from 1272 °C to 1247 °C when the proportion of water vapor was 20%. The results demonstrate that water vapor is favor of decreasing the AFTs of the coal ash. Furthermore, DT was closely related to the formation of the initial liquid phase in the coal ash [24]. The decrease in DT implies that the increasing water vapor lowers the formation temperature of initial liquid phase.
Fig. 3 shows the mineral transformation of the coal ash under different atmosphere in the range of 900–1400 °C. Although the ash samples were prepared under different atmosphere, mineral species were similar with each other at the same temperature. The major minerals at 900 °C and 1000 °C were calcite (CaCO3), quartz (SiO2), anhydrite (CaSO4), and hematite (Fe2O3). According to the coal ash compositions, the content of CaO is 33.59%, most of the calcium in the ash existed in the form of CaCO3 and CaSO4. As the temperature increased to 1100 °C, the minerals in coal ash were mainly SiO2, CaSO4, Fe3O4 (magnetite) and Ca2Al2SiO7. Ca2Al2SiO7 was formed owing to 21
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 4. Minerals contents in the ashes under Ar with water vapor at different temperatures.
reaction between SiO2, Al2O3 and CaO. CaCO3 disappeared because it decomposed to CaO and CO2. At 1200 °C, the main minerals were Fe3O4 and Ca2Al2SiO7. SiO2 and CaSO4 disappeared gradually due to the formation of Ca2Al2SiO7. When the temperature reached to 1300 °C and 1400 °C, gehlenite and magnetite were the main minerals in the coal ash. Therefore, the main reactions can be listed as follows:
The contents of minerals in the coal ash under Ar with water vapor were further analyzed by Siroquant software. From Fig. 4, the content of SiO2 decreases slightly with the increasing water vapor proportion. At 900 °C, the SiO2 content decreased from 4.3% to 3.8% and 3.1% when the proportion of water vapor increases from 0% to 10% and 20%. The melting temperature of SiO2 was 1610 °C, and the high content of SiO2 can lead to a higher AFTs. This is a reason for the decrease of the AFTs with the addition of water vapor. The CaCO3 content dropped by 8.9% from 37.6% to 28.7%, while the content of CaSO4 raised by 2.1% from 6.5% to 8.6% under Ar and Ar + 20% water vapor, respectively. The opposite tendency was presented between CaCO3 and
CaCO3 → CaO + CO2
CaSO4 → CaO + SO3 Al2O3 + SiO2 + 2CaO → Ca2Al2SiO7 22
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 5. (a) Slag viscosity-temperature curves under Ar with water vapor. (b) slag viscosity at above 1460 °C; (c) TCV of the slag under Ar with water vapor.
It has been widely accepted that the coal ash slag was silicate melt as glass and magmas. Water molecule can enter into the slag structure as an –OH and destroy the Si–O–Si bonds, leading to depolymerization of the melt [26]. The depolymerization is attributed to the decrease of the slag viscosity. Therefore, the effect of water vapor on slag viscosity is similar with NBO which breaks the network structure, reducing the slag viscosity [27,28].
CaSO4, because CaO which is from the decomposition of CaCO3 reacted with SO3 to form CaSO4 which had a high decomposition temperature. Subsequently, CaSO4 will decompose to form Ca2Al2SiO7. Meanwhile, the content of Ca2Al2SiO7 decreased and amorphous substance increased with the increasing water vapor proportion. For example, at 1200 °C, the content of Ca2Al2SiO7 were 11.9%, 10.3%, and 8.0% under Ar, Ar + 10% water vapor, Ar + 20% water vapor. The content of amorphous substance was 62.7%, 65.7%, and 70.3%. The similar results were observed at other temperatures except for the ashes at 1400 °C, at which temperature the content of Ca2Al2SiO7 and amorphous substance remained a constant. The decrease of crystalline substances content and the increasing amorphous substance improve the proportion of liquid phase in the slag [25]. Therefore, a larger fraction of liquid phase in the slag accounts for the decrease of AFTs.
H2 O + Si−O−Si → (−Si−OH HO−Si−) The small effect of water vapor on slag viscosity and TCV was mainly due to the low solubility of water at ambient pressure. From previous studies, the solubility of water at ambient pressure is under 1 wt% [29]. According to the relative research in magmas systems, the influence of water was significant compared with that on coal ash slag. The fact is that the effect observed in magmas was at a high pressure, even over 1GPa. Under a high pressure, the water solubility can be several hundred orders of magnitude [30] and the influence would be much more pronounced. In fact, the real pressure in an entrained flow gasifier (e.g., 8.5 MPa, for a slurry-fed gasifier) is lower than that in the magmas system which was happened in the deep earth in nature [31]. Therefore, it can be speculated that the effect of water vapor on the viscosity of molten slag should be greater under gasification conditions than that at ambient pressure.
3.3. Viscosity temperature behavior of slags The viscosity results of the YZ slag under Ar, Ar + 10% water vapor and Ar + 20% water vapor were measured, as shown in Fig. 5(a–c). From the viscosity temperature curves (Fig. 5(a)), the slag exhibited the behavior of the crystalline slag under Ar with different proportion of water vapor. The viscosity rises rapidly when temperature is lower than TCV. Meanwhile, the slag viscosity decreased with the increasing proportion of water vapor, as shown in Fig. 5(b). At 1465 °C, the slag viscosity was 2.47 Pa·s under Ar. However, it dropped to 1.62 Pa·s and 0.69 Pa·s when the proportion of water vapor was 10% and 20%. Besides, the TCV decreased from 1430 °C to 1370 °C and 1350 °C when the proportion of water vapor increased from 0% to 10% and 20%, as shown in Fig. 5(c). These demonstrate that water vapor lowers slag viscosity and TCV of the slag, but the influence was slight.
3.4. Effect of water vapor on slag structure at high temperature 3.4.1. Effect of water vapor on Si–O As shown in Fig. 6, the characteristic bands of Si–O symmetric stretching are between the wavenumbers 2000–500 cm−1. The bands at about 800 cm−1 and 960 cm−1 were assigned to the asymmetric and 23
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 8. 27Al NMR of spectra under Ar with water vapor. (a) Ar; (b) Ar + 10% water vapor; (c) Ar + 20% water vapor. Fig. 6. FTIR spectra under different atmosphere. (a) Ar; (b) Ar + 10% water vapor; (c) Ar + 20% water vapor.
degree of polymerization. As shown in Fig. 7(b), it is noted that R increased slightly from 3.83% to 3.90% (0.13%) when the proportion of water vapor increased from 0% to 10%, but it increased drastically to 5.72% when the proportion of water vapor was 20%. It demonstrates that water vapor breaks the silicate network structure by decreasing the polymerization degree of melt. However, the effect of water vapor on the slag structure was small in this study. Previous studies [35,36] have found that compared with a high polymerized melt (like acidic melts), water causes a smaller change for a low depolymerized melt (usually basic melts) even at high temperature and high pressure. Because the ash used in this work is a basic slag for the high content of calcium and iron, it has a low polymerization degree. Thus, it shows a small effect on the structure of the slag at ambient pressure.
symmetric vibrations of Si–O–Si. The symmetric stretching of Si–O–Si and Al–O–Si were at 726 cm−1. The shoulder at about 678 cm−1 and 880 cm−1 were the stretching vibrations of Al–O [32,33]. According to the characteristic stretching vibrations of SiO4 tetrahedra, Q-species can be distinguished. Qn represents the degree of polymerization of [SiO4] tetrahedral specie unit, and n is the number of Si–O–Si bridging oxygen, which indicates the degree of polymerization of slag. The convoluted spectral were categorized into five types, namely, monomers with [SiO4]4− framework(Q0), dimers with [Si2O7]6− units(Q1), chains with [SiO3]2− skeleton(Q2), sheets with [Si2O5]2− units(Q3) and three-dimensional network of SiO2(Q4). On the basis of the previous works, the wavenumbers of Q-species were as follows [34]: 880–850 cm−1 for Q0, 920–900 cm−1 for Q1, 1000–950 cm−1 for Q2, and 1100–1050 cm−1 for Q3 and 1200–1060 cm−1 for Q4. The typical deconvoluted peaks along with the maximum randomness in residual distributions are shown in Fig. 7(a), the relative contents of Qn (n = 0–4) are showed in Fig. 7(b). Q2 was the major structural units in the slag under Ar with and without water vapor. The contents of Q0 and Q1 did not change significantly with the increasing water vapor proportion. When the proportion of water vapor increased from 0% to 20%, Q2 increased from 51.40% to 57.30%, and the content of Q3 decreased by 2.56% from 17.44% to 14.88% at the same time. Q4 disappeared when the proportion of water vapor was 20%. The decreasing Q3 and Q4 indicates that the addition of water vapor leads to the decrease of the polymerization degree of silicate network. Q0 + Q1 + Q2/Q3 + Q4 (denoted as R) was applied to describe the
3.4.2. Effect of water vapor on Al–O Fig. 8 shows the 27Al-NMR spectra of the slag under different atmosphere. 27Al-NMR spectra usually have two characteristic peaks: 10–20 ppm for six-coordinate Al([AlO6]9−), which weakens the degree of polymerization with octahedral structure, and 50–100 ppm for fourcoordinated Al([AlO4]5−), which enhances the polymerization degree with tetrahedron structure. Therefore, the higher percentage of Al(IV) in the melt will increase the slag viscosity. The resonance at 64.6 ppm was assigned to Al(IV) and Al(VI) resonance was centered at 14.5 ppm [37,38]. In order to know the effect of water vapor on Al(IV) and Al(VI), a typical deconvoluted peaks of 27Al NMR spectra are shown in Fig. 9(a). Usually, the integrated area was used to present the content of Al(IV)
Fig. 7. (a) Typical fitted FTIR spectra under Ar atmosphere; (b) Fraction of structural units Qn based on FTIR spectra. 24
Fuel 237 (2019) 18–27
X. Cao et al.
Fig. 9. (a) A typical deconvolution of
27
Al NMR spectra; (b) [AlO4]5−/([AlO4]5− + [AlO6]9−).
Fig. 10. SEM-EDX results of slag quenched at 1320 °C. (a) Ar; (b) Ar + 10%H2O; (c) Ar + 20%H2O. (□ for Ca2Al2SiO7 and ○for CaAl4O7).
and Al(VI). The fitting results indicate that the signal intensity is not necessarily equivalent to the real amount of Al(IV) and Al(VI) in the structure [39]. Because the slag in this study was a basic slag, Al(IV) was the main resonance in melt (about 90.0%) and Al3+ is mainly [AlO4]5−, which acts as a network former. As shown in Fig. 9(b), the ratio of [AlO4]5−/[AlO4]5−+[AlO6]9− decreased with the increasing water vapor proportion. Under Ar atmosphere, the ratio was 90.37%, and it dropped to 90.27% and 85.63% under Ar + 10% water vapor and Ar + 20% water vapor, respectively. The reducing [AlO4]5− suggests that a part of [AlO4]5− transformed into [AlO6]9− with the addition of water vapor. The decreasing
[AlO4]5− also indicates that the existed of water vapor weakens the polymerization degree of the slag. In the process, a part of Al3+ enters into the network skeleton as [AlO6]9− and acts as a network modifier. Therefore, the slag viscosity decreased with the increasing water vapor content. Furthermore, due to the similarity structures between silicon oxide and aluminum oxide, the interpretations of NMR are difficult [40]. Therefore, most researches about the effect of water vapor on silicate melt are mainly concentrated on the aluminum-free system, and the specific mechanism needs a further study [41].
25
Fuel 237 (2019) 18–27
X. Cao et al.
of water vapor. (4) The average particle size of solid phases in slag decreased with the increasing water vapor proportion. Water vapor impedes the growth of crystals, which is one of reasons for the decrease of the temperature of critical viscosity. Acknowledgements This work was supported by the National Key R&D program of China (Grant Number 2017YFB0602603), Joint Foundation of Natural Science Foundation of China and Shanxi Province (Grant Number U1510201), National Natural Science Foundation of China (Grant Numbers 21476247 and 21761132032), Joint Foundation of Natural Science Foundation of China and Xinjiang (Grant Number U1703252), Shanxi Province Science Foundation (Grant Number 201601D201003), Youth Innovation Promotion Association, CAS, Bureau of International Cooperation, Chinese Academy of Sciences (Grant Number GJHZ1879). Fig. 11. XRD patterns of the slags quenched at 1320 °C. Ar; (b) Ar + 10% water vapor; (C) Ar + 20% water vapor. 1-Ca2Al2SiO7 (gehlenite); 2-CaAl4O7 (calcium dialuminate); 3-FeAl2O4(hercynite).
References [1] Han S, Chen H, Long R, Cui X. Peak coal in China: a literature review. Resour Conserv Recycl 2018;129:293–306. [2] Chen W, Xu R. Clean coal technology development in China. Energy Policy 2010;38(5):2123–30. [3] Pereira EG, da Silva JN, de Oliveira JL, Machado CS. Sustainable energy: a review of gasification technologies. Renewable Sustainable Energy Rev 2012;16(7):4753–62. [4] Minchener AJ. Coal gasification for advanced power generation. Fuel 2005;84(17):2222–35. [5] Lee HH, Lee JC, Joo YJ, Oh M, Lee CH. Dynamic modeling of Shell entrained flow gasifier in an integrated gasification combined cycle process. Appl Energy 2014;131:425–40. [6] Bi D, Guan Q, Xuan W, et al. Combined slag flow model for entrained flow gasification. Fuel 2015;150:565–72. [7] Chen X, Kong L, Bai J, Dai X, Li H, Bai Z, et al. The key for sodium-rich coal utilization in entrained flow gasifier: the role of sodium on slag viscosity-temperature behavior at high temperatures. Appl Energy 2017;206:1241–9. [8] Crnomarkovic N, Repic B, Mladenovic R, Neskovic O, Veljkovic M. Experimental investigation of role of steam in entrained flow coal gasification. Fuel 2007;86(1–2):194–202. [9] Song W. Measurement and simulation of flow properties of coal ash slag in coal gasification. AIChE 2011. [10] He C, Bai J, Kong L, Guhl S, Schwitalla DH, Xu J, et al. The precipitation of metallic iron from coal ash slag in the entrained flow coal gasifier: by thermodynamic calculation. Fuel Process Technol 2017;162:98–104. [11] Hurst HJ, Novak F, Patterson JH. Phase diagram approach to the fluxing effect of additions of CaCO3 on Australian coal ashes. Energy Fuels 1996;10(6):1215–9. [12] Guo Q, Zhou Z, Wang F, Yu G. Slag properties of blending coal in an industrial OMB coal water slurry entrained-flow gasifier. Energy Convers Manage 2014;86:683–8. [13] Tsuruda A, Arma LH, Takebe H. Viscosity measurement and prediction of gasified and synthesized coal slag melts. Fuel 2017;200:521–8. [14] Vargas S, Frandsen FJ, Dam-Johansen K. Rheological properties of high-temperature melts of coal ashes and other silicates. Prog Energy Combust Sci 2001;27(3):237–429. [15] Gao J, Wen G, Huang T, et al. Effects of the composition on the structure and viscosity of the CaO-SiO2-based mold flux. J Non-Cryst Solids 2016;435:33–9. [16] Wu X, Zhang Z, Piao G, He X, Chen Y, et al. Behavior of mineral matters in Chinese coal ash melting during char-CO2/H2O gasification reaction. Energy Fuels 2009;23(5):2420–8. [17] Bell DA, Towler BF, Fan M. Chapter 4 – Gasifiers. Coal Gasification and Its Applications. Boston: William Andrew Publishing; 2011. p. 73–100. [18] Jin W, Wang Y, Zhu L, Wu C, Xu J. Numerical simulation and assessment of a twostage gasifier modified from an opposed multi-burner gasifier. Chem Eng Technol 2014;37(3):483–94. [19] Ahmed MYM. Phase Equilibria Between Iron and Slag in CO/CO2/H2/H2O Atmospheres Relevant to a novel Flash Ironmaking Technology Doctoral dissertation University of Utah; 2013. [20] Kong L, Bai J, Bai Z, Guo Z, Li W. Effects of CaCO3 on slag flow properties at high temperatures. Fuel 2013;109:76–85. [21] Kong L, Bai J, Li W, et al. 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;158:968–75. [22] Ma Z, Bai J, Li W, Bai Z, Kong L. Mineral transformation in char and its effect on coal char gasification reactivity at high temperatures, part 1: mineral transformation in char. Energy Fuels 2013;27(8):4545–54. [23] Grigore M, Sakurovs R, French D, Sahajwalla V. Influence of mineral matter on coke reactivity with carbon dioxide. ISIJ Int 2006;46(4):503–12. [24] Yan T, Bai J, Kong L, Bai Z, Li W, Xu J. Effect of SiO2/Al2O3 on fusion behavior of coal ash at high temperature. Fuel 2017;193:275–83. [25] Kondratiev A, Ilyushechkin A. Flow behaviour of crystallising coal ash slags: shear viscosity, non-Newtonian flow and temperature of critical viscosity. Fuel
3.5. Effect of water vapor on solid phases during cooling The slags quenched at 1320 °C under Ar, Ar + 10% water vapor and Ar + 20% water vapor was analyzed by SEM-EDX, and the results are presented in Fig. 10. Two crystals with different shape (quadrilateral and short column) were observed in the slag. To verify the types of the crystals, EDX combined with XRD analyses (Fig. 11) were conducted. The results showed that the short column crystals were Ca2Al2SiO7 and the cubic crystals could be CaAl4O7. It can be clearly seen that the average size of solid particles was decreased with the increasing of water vapor content, especially for the size of cubic crystals. The average particle size of the cubic crystal was 80–150 um, 50–90 um and 30–70 um when the atmosphere were Ar, Ar + 10% water vapor and Ar + 20% water vapor, respectively. This demonstrates that water vapor inhibits the growth of the crystals in slag during cooling. Previous researchers [42,43] studied the effect of crystal size on the viscosity of several slags with the addition of the solid particles. They found that the slag viscosity increases as the crystal size increases. High content of water vapor leads to smaller-sized crystals, which lowers the slag viscosity. Moreover, the crystal size is closely related to the TCV [44]. A TCV event usually occurs when the size of the crystals significantly increases. The smaller crystals also contributed to the decrease of the TCV. 4. Conclusions In this work, we investigated the influence of water vapor on the slag viscosity behavior at high temperatures. A typical coal used in coal water slurry gasifier was selected. The influence of water vapor on slag structure and solid phase during cooling was discussed in detail. The main conclusions were summarized as following: (1) Due to the low solubility of water vapor at ambient pressure, the presence of water vapor leads to a small decrease in the AFTs, slag viscosity and TCV, while its effect increases with the increasing proportion. (2) Although the effect of water vapor on mineral species is small, the addition of water vapor decreases the content of SiO2, which lowered the AFTs. Meanwhile, the water vapor is beneficial to the formation of amorphous substances, and it does not favor the formation of crystalline minerals. (3) The structural effect of water vapor on Si–O and Al–O is revealed. Water enters into the slag structure as an –OH and destroys the Si–O–Si bonds, which results in a low polymerization degree. Besides, Al3+ in the slag is likely to be [AlO6]9− due to the addition 26
Fuel 237 (2019) 18–27
X. Cao et al.
Solids 2017;473:79–86. [36] Whittington A, Richet P, Holtz F. Water and the viscosity of depolymerized aluminosilicate melts. Geochim Cosmochim Acta 2000;64(21):3725–36. [37] Le Losq C, Neuville DR, Florian P, Henderson GS, Massiot D. The role of Al3+ on rheology and structural changes in sodium silicate and aluminosilicate glasses and melts. Geochim Cosmochim Acta 2014;126:495–517. [38] Lin X, Ideta K, Miyawaki J, Takebe H, Yoon S-H, Mochida I. Correlation between fluidity properties and local structures of three typical asian coal ashes. Energy Fuels 2012;26(4):2136–44. [39] Pardal X, Brunet F, Charpentier T, et al. 27Al and 29Si solid-state NMR characterization of calcium-aluminosilicate-hydrate. Inorg Chem 2012;51(3):1827–36. [40] Xue X, Kanzaki M. Dissolution mechanisms of water in depolymerized silicate melts: constraints from 1H and 29Si NMR spectroscopy and ab initio calculations. Geochim Cosmochim Acta 2004;68(24):5027–57. [41] S.C. Kohn, The dissolution mechanisms of water in silicate melts; a synthesis of recent data, (2000) 64(3):389–408. [42] Wright S, Zhang L, Sun S, et al. Viscosities of calcium ferrite slags and calcium alumino-silicate slags containing spinel particles. J Non-Cryst Solids 2001;282(1):15–23. [43] Wright S, Zhang L, Sun S, et al. Viscosity of a CaO-MgO-Al2O3-SiO2 melt containing spinel particles at 1646K. Metall Mater Trans B 2000;31(1):97–104. [44] Ilyushechkin AY, Hla SS, Roberts DG, Kinaev NN. The effect of solids and phase compositions on viscosity behaviour and TCV of slags from Australian bituminous coals. J Non-Cryst Solids 2011;357(3):893–902.
2018;224:783–800. [26] Wagstaff FE, Richards KJ. Kinetics of crystallization of stoichiometric SiO2 glass in H2O atmospheres. J Am Ceram Soc 1966;49(3):118–21. [27] Gonzalez-Oliver CJR, Johnson PS, James PF. Influence of water content on the rates of crystal nucleation and growth in lithia-silica and soda-lime-silica glasses. J Mater Sci 1979;14(5):1159–69. [28] Fujita S, Sakamoto A, Tomozawa M. Behavior of water in glass during crystallization. J Non-Cryst Solids 2003;320(1–3):56–63. [29] Folkedahl BC, Schobert HH. Effects of atmosphere on viscosity of selected bituminous and low-rank coal ash slags. Energy Fuels 2005;19(1):208–15. [30] Ban-Ya S, Hino M, Nagasaka T. Estimation of water vapor solubility in molten silicates by quadratic formalism based on the regular solution model. ISIJ Int 1993;33(1):12–9. [31] Browning GJ, Bryant GW, Hurst HJ, et al. An empirical method for the prediction of coal ash slag viscosity. Energy Fuels 2003;17(3):731–7. [32] Mohassab Y, Sohn HY. Analysis of slag chemistry by FTIR-RAS and Raman spectroscopy: effect of water vapor content in H2–H2O-CO-CO2 mixtures relevant to a novel green ironmaking technology. Steel Res Int 2015;86(7):740–52. [33] Jiang Y, Lin X, Ideta K, Takebe H, Miyawaki J, Yoon S-H, et al. Microstructural transformations of two representative slags at high temperatures and effects on the viscosity. J Ind Eng Chem 2014;20(4):1338–45. [34] Baasner A, Schmidt BC, Webb SL. The effect of chlorine, fluorine and water on the viscosity of aluminosilicate melts. Chem Geol 2013;357:134–49. [35] Gao E, Wang W, Zhang L. Effect of alkaline earth metal oxides on the viscosity and structure of the CaO-Al2O3 based mold flux for casting high-al steels. J Non-Cryst
27