Subscriber access provided by University of Newcastle, Australia
Article
Numerical investigation on development of initial ash deposition layer for a high-alkali coal Chao Liu, Zhaohui Liu, Tai Zhang, Xiaohong Huang, Junjun Guo, and Chuguang Zheng Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.6b03043 • Publication Date (Web): 01 Feb 2017 Downloaded from http://pubs.acs.org on February 10, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
1
Numerical investigation on development of initial ash
2
deposition layer for a high-alkali coal
3
Chao Liu, Zhaohui Liu*, Tai Zhang, Xiaohong Huang*, Junjun Guo, Chuguang Zheng
4
State Key Laboratory of Coal Combustion, School of Energy and Power Engineering, Huazhong
5
University of Science and Technology, Wuhan, 430074, China
6
ABSTRACT: Burning high-alkali content coals frequently causes severe fouling and slagging
7
problems, which greatly threat the efficiency and safety of boilers. In this paper, a set of
8
numerical models have been integrated and validated to predict the development of initial
9
deposition on the tube of super-heater for high-alkali yield fly ash particles. Besides the normally
10
used ash particles transport models (inertial force, turbulent eddy effect, thermophoretic force)
11
and adhesion model (alkali coating surface model), a deposit erosion model based on the energy
12
dissipation at impaction has been introduced. The predicted characteristics, such as mass,
13
morphology of deposit, and particle size distribution of deposited ashes, are compared with
14
experimental observations in a 350MWe boiler and a 10kW laboratory furnace respectively. The
15
results show that, for a coal with high-alkali content, the proposed models can well predict the
16
development of initial ash deposit layer, especially the phenomena of windward deposit erosion
17
at high Reynolds number surrounding flow in real facility.
18
KEYWORDS: High-alkali coal; Ash deposition; Slagging and Fouling; Numerical simulation 1
ACS Paragon Plus Environment
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
19 20
Page 2 of 38
1. Introduction Coals with high-alkali content are used as fuel for stationary boiler in many region of the [1,2]
21
world
. In China, Zhundong coal, which has a huge reserve about 390Gt, is a high-alkali
22
content coal with low ignition point, high burning rate and high calorific value
23
high-alkali content in its ash, usually ≥5%, many alkali metal compounds such as sodium sulfate
24
will condense to form a sticky coat on the fly ash particle and heat exchanger surface [4,5,6]. This
25
sticky coat greatly enhances the sticking probability of ash particles impacted on the heating
26
surface. Therefore, severe slagging and fouling on boiler heating surfaces are induced, which
27
leads to more frequent sootblowing or even unplanned boiler shutdown to maintain a safe and
28
stable operation of the boiler. So far, many researches including experimental tests and numerical
29
simulations have been implemented to study the mechanisms of the deposition of ash with
30
high-alkali content [5,6].
[3]
. Due to the
31
By investigating the formation of high-alkali coal ash in different facilities, many
32
researchers concluded that the mineral vapor, such as Na2SO4 (g), will condense on the surface of
33
fly ash particles and the exposed tube wall, when the temperature below about 1250±50K
34
The condensation mixture is molten at the temperature near 1200K, thus forming a sticky coating
35
surface on the fly ash particles and the bare tube surface. As a consequence, the ash deposition
36
was greatly enhanced [5,7,8]. Based on tests on a 1.5 MWth pilot plant, Shimogori et al. found that
37
the Na2O and K2O mass percentages in parent coals also greatly influence the formation of
38
deposit initial layer [9].
39
[6-8]
.
In order to evaluate the potential of ash deposition in boilers, many empirical fouling and 2
ACS Paragon Plus Environment
Page 3 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
40
slagging indices has been developed. Coal ash deformation temperature and softening
41
temperature, the ratio of silica to alumina and the ratio of basic to acidic oxides (B/A) in the ash
42
are some commonly used indices
43
and the fusibility data, these indices can be easily caculated. However, due to the variation of flue
44
gas flow properties, such as velocities and temperature, and different arrangements of heat
45
exchanging surface, the fouling and slagging degree at different positions of boilers can not be
46
well estimated. So recently, more complicated computational fluid dynamics (CFD) based models
47
for characterizing the whole process of coal ash deposition are developed
48
have been successfully used to aid the design and arrangement of heat exchanging devices.
49
Transport of ash particles, adhesion of ash particles and deposit erosion, are three main aspects
50
employed to predict the ash deposition particularly. We will briefly review them below.
[10]
. Based on the fundamental analyze of the ash composition
[11-19]
. Some of them
51
For predicting the ash particles impaction on the tube wall, inertial impaction, turbulent eddy
52
effect, thermophoretic effects, Fick diffusion are regarded as the main mechanisms in previous
53
works
54
employed to get a reasonable simulation on impaction process of ash deposition. For predicting
55
deposition of coal ash with high-silicon content, inertial impaction is always taken as the only
56
way for ash deposition
57
high-alkali content, besides the initial impaction, turbulent eddy effect and thermophoretic effects
58
are always taken into account [23].
[20,21]
. According to the differences in fuel properties, different impaction models are
[16,22]
. For simulating ash deposition in boilers burning biomass fuel with
59
Various models had been developed to predict the coal ash adhesion. Watt et al. [24] proposed
60
the popularly used ash viscosity model, in which ash viscosity was used as the main factor to 3
ACS Paragon Plus Environment
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 38
61
determine the adhesion of ash particles. Based on the correlation between temperature and the
62
viscosity of some individual minerals, the viscosity caculation method had been rectified by
63
Urbain et al. [25] and Senior et al. [26] to obtain predictions with better accuracy. However, in many
64
cases, the sticking probability was also greatly influenced by parameters such as impaction
65
velocity, impaction angle and surface tension of ash particles
66
particles are coated with a sticky surface formed by the molten sodium sulfate, Walsh et al. [4] put
67
forward the alkali coating surface model. In this model, the sticking probablity of different size
68
ash particles with high-alkali content could be reasonably estimated. Based on the Walsh’s model,
69
Lee et al. [30] modified the reflected velocity prediction to improve the prediction accuracy. Walsh
70
and Lee focused mainly on the deposit mass of fly ash particles. However, other characteristics of
71
ash deposition, such as leeward deposition and deposit morphology, are also very important and
72
seldom evaluated.
[27-29]
. By assuming that the fly ash
73
In addition to the sticking process of fly ash particles, ash deposition is also greatly affected
74
by the rebound of ash particles, which always accompanies with the removal of already deposited
75
ash particles [22,31,32]. In previous studies, the critical velocity model derived by the JKR theory [27]
76
and the extra energy model
77
particles, while the deposit erosion had not been taken into consideration. In some investigations
78
[4]
79
erosion of ash deposit. In the work of Strandström et al.
80
particles had been investigated. By comparing with the experimental result, a set of correction
81
factors are applied to modify the erosivity of different size non-sticky particles. The erosion
[33]
have been successfully used to evaluate the rebound of ash
, a constant value of the erosivity (mass removed/mass impacting) has been used to predict the [22]
, the erosion degree of non-sticky
4
ACS Paragon Plus Environment
Page 5 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
82
process related to the energy loss at impaction is also greatly affected by properties of the
83
impaction particles [34]. At the time being, there’s no report on erosion characteristics of sticky
84
particles.
85
In this paper, based on the previous works [4,11,16,19,30], a set of models have been constructed
86
firstly for simulating the transport and adhesion of high-alkali ash particles. A new deposit
87
erosion model, which combines the energy dissipation at impaction and the surface energy of
88
alkali surface coating particles
89
been integrated to predict the ash deposition of a coal with high-alkali content at initial stage.
90
Predicted deposit characteristics, such as the deposit morphology, the deposit mass and the size
91
distribution of deposited ash particles, are examined by the experimental data.
92
2. Numerical models
[4,30]
, has been developed and validated then. These models have
93
The overall computational algorithm is shown in Figure 1. The transport models, the
94
adhesion model for high-alkali ash particles and the deposit erosion model, which describe the
95
whole ash deposition progress, have been integrated in this paper. In order to simplify the models,
96
some key assumptions, such as the ash particles are considerated as inert particles without mass
97
change, and the most of the alkali vapor condensed rather than nucleated, have been adopted in
98
this work. The detail of these models is shown in the following parts.
5
ACS Paragon Plus Environment
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 38
Inertial impaction
Transport of ash particles
Turbulent eddy effect Thermophoresis
Adhesion of ash particles
Alkali coating surface model
Deposit erosion
Walsh’ s model/ New developed erosion model
99 100 101 102
Figure 1. The computational algorithm used to predict the high-alkali ash deposition 2.1 Transport models In order to obtain detail movement characteristics of the ash particles, the ash particles were [19]
103
individually tracked (Single Particle Tracking - SPT
, which makes the trajectories and the
104
deposition charateristics of individual particles can be obtained) in a Lagrangian frame of
105
reference. The force balance equates the particle inertia with the forces acting on the particle, and
106
can be expressed as follows [13,23]: uur du p uur uur uur uur = FD (u f + u ' − u p ) + FT dt uur d x p uur dt = u p
107
where FD =
(1)
18µ CD Re is the drag force per unit particle mass, Re p is the relative Reynolds ρ p d p2 24 uur uur
ρ f dp up − uf
. CD is the spherical drag coefficient, µ is the
108
number, which is defined as Re p =
109
fluid viscosity, ρ p is the particle density, d p is the particle diameter, ρ f is the fluid density,
µ
6
ACS Paragon Plus Environment
Page 7 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
110
uur uur uur u p is the particle velocity(m/s), u f is the fluid phase mean velocity, u ' is the fluid phase
111
fluctuation velocity, x p is the particle position in the fluid field.
uur
112
Large velocity fluctuation exists in the high-temperature flue gas flow in furnace, which is
113
regarded as the turbulent eddy effect. It will lead to random tortuous movement of particularly
114
small fly ash particles. In this study, discrete random walk model with random eddy lifetime is
115
employed to consider the random eddy effect. The integral time for tracked particles is TL = CL ,
116
and the characteristic random lifetime of the eddy is τ = −TL ln(r ) , where r is a uniform random
117
number greater than zero and less than 1, CL is taken as 0.3 for the Reynolds stress model (RSM)
118
used in this paper [35].
k
ε
119
Large temperature gradient exists between the high temperature flue gas and the tube wall at
120
lower temperature, which also influences especially the movement of fine particles (10 µm) with high impaction kinetic energy can hardly be captured and possess
336
high erosivity. So in Case 1, when the gas velocity is high, windward deposit is corroded
337
seriously by large ash particles and becomes very thin. Ash deposit mainly consists of fine ash
338
particles (
