Version of Record: https://www.sciencedirect.

com/science/article/pii/S0360128521000526
Manuscript_e187354a8d030abb495d42de2f79eaa5

2 A review of rare earth elements and yttrium in coal ash: content, modes of

3 occurrences, combustion behavior, and extraction methods

4 Biao Fua, James C. Howerb, Wencai Zhang c, Guangqian Luoa, Hongyun Hua, Hong

5 Yaoa*

a
6 State Key Laboratory of Coal Combustion, School of Energy and Power

7 Engineering, Huazhong University of Science and Technology, Wuhan 430074, China

b
8 University of Kentucky, Center for Applied Energy Research, 2540 Research

9 Park Drive, Lexington, KY 40511, United States of America and University of Kentucky,

10 Department of Earth & Environmental Sciences, Lexington, KY 40506, United States of

11 America

c
12 Department of Mining and Minerals Engineering, Virginia Polytechnic Institute

13 and State University, Blacksburg, VA 24061, United States of America

14 Contact information

15 Biao Fu: fubiao1223@gmail.com;James C. Hower: james.hower@uky.edu; Wencai

16 Zhang: wencaizhang@vt.edu; Guangqian Luo: guangqian.luo@mail.hust.edu.cn;

17 Hongyun Hu: hongyunhu@hust.edu.cn

18 Corresponding author: Hong Yao (hyao@mail.hust.edu.cn)

19 Abstract

20 Rare earth elements and yttrium (REY) have attracted considerable attention over the

21 last decade because of their vital roles in clean energy, consumer product, national

22 defense and security applications, among other uses. Due to the retention of REY

23 during coal burning, coal combustion ash is considered as potential alternative sources
1

© 2021 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
24 for REY. Understanding the content, speciation, retention and/or transformation

25 behavior of REY during coal combustion not only expand our knowledge of the

26 combustion behavior of the trace elements in coal, but also provide basis for modeling

27 REY partitioning during coal combustion and for developing economically viable REY

28 recovery technologies. This review makes a critical summary of recent progress in the

29 study of REY in coal ash. The contents and the extraction potentials of REY in coal ash

30 derived from 15 major coal-producing countries worldwide were summarized and

31 evaluated. Various analytical methods for determining REY bulk contents and

32 speciation, together with the solid sample pretreatment, analytical accuracy and

33 precision, advantages and disadvantages were summarized and compared. Modern

34 analytical approaches combined the indirect methods (e.g., sequential extraction) shed

35 light on the physical distribution, mineralogy, and the chemical state of REY in coal ash.

36 Three types of REY occurrences in coal ash, including Si-Al glassy association,

37 discrete minerals or compounds, and organic association (bound with unburned carbon)

38 were defined in the review. The glassy association can be further divided into REY

39 minerals closely bound to glass phases and dispersed throughout the glassy structure.

40 REY partitioning in various emission streams, the size distribution, and their

41 enrichment behavior in coal ash were discussed. Thermal behavior and transformation

42 of various REY forms in coal during combustion process, including organic-associated

43 REY, REY phosphates, REY carbonates, clay-bound REY and among others were

44 summarized. Two possible retention mechanisms of REY by aluminosilicate glass at

45 boiler temperature were proposed: the incorporation of the individual REY phases into

2
46 the glass as inclusions and the diffusion of REY phases throughout Si-Al glass

47 structures in the melting process. Feed coal mineral types, mineral-mineral associations,

48 boiler conditions, and other factors control the retention process. After coal combustion,

49 the speciation of REY in fly ash may be modified by the reactions of REY phases with

50 flue gas components. Further, an overview of REY transformation mechanisms during

51 coal combustion was deeply discussed. Finally, current extraction techniques for REY

52 recovery from coal combustion ash were introduced. Future outlooks and research

53 problems were also identified.

54 Keywords: Rare earth elements and yttrium; Coal combustion ash; Partitioning

55 mechanisms; Extraction technologies

56

57

3
58 Contents
59 1. Introduction ........................................................................................................................... 6
60 1.1 Background of rare earth elements in coal ash study and their significance ................... 6
61 1.2 Scope of the review ......................................................................................................... 7
62 2. Nature of REY-rich coal........................................................................................................ 9
63 2.1. Nature of REY ................................................................................................................ 9
64 2.2. Contents of REY in coal and REY rich coal ................................................................ 13
65 2.2.1. Abundance of REY in coals around the world ...................................................... 13
66 2.2.2. REY rich coal in major coal-producing countries ................................................. 15
67 2.3. REY occurrence in coal ................................................................................................ 19
68 2.3.1 Proportions of organic and inorganic associations of REY in coal ........................ 21
69 2.3.2. Size and spatial distribution of REY-bearing particles in coal .............................. 25
70 3. Determination of REY in coal ash....................................................................................... 29
71 3.1. Bulk analysis of REY concentration ............................................................................ 31
72 3.1.1. Sample preparation for ICP-based techniques....................................................... 33
73 3.1.2. Inductively coupled plasma atomic emission spectrometry (ICP-OES) ............... 43
74 3.1.3. Inductively coupled plasma mass spectrometry (ICP-MS) ................................... 45
75 3.1.4. Instrumental neutron activation analysis (INAA).................................................. 49
76 3.1.5. X-ray fluorescence (XRF) analysis ....................................................................... 51
77 3.2. Speciation analysis ....................................................................................................... 58
78 3.2.1. Indirect methods .................................................................................................... 58
79 3.2.2. Direct microanalysis methods ............................................................................... 64
80 3.2.3 Synchrotron-based techniques ................................................................................ 69
81 4. Concentrations and valuation of REY in coal ash ............................................................... 76
82 4.1. REY concentrations in world coal ash ......................................................................... 76
83 4.2. REY contents in coal ash from major coal-producing countries .................................. 79
84 4.2.1. Contents and spatial distribution ........................................................................... 79
85 4.2.2. Critical REY ratios ................................................................................................ 82
86 4.3. Estimation of REY reserves and economic value ........................................................ 84
87 4.3.1. Reserve of REY in coal ash ................................................................................... 84
88 4.3.2. Valuation of REY in coal ash ................................................................................ 86

4
 89 5. Speciation of REY in coal ash ............................................................................................. 91
90 5.1. Definition and classification of REY forms in coal ash ............................................... 91
91 5.2. Associations with amorphous glassy materials ............................................................ 92
92 5.3. Associations with discrete minerals or compounds ...................................................... 99
93 5.4. Associations with unburned carbon............................................................................ 103
94 6. Behavior of REY in coal-fired power plants ..................................................................... 109
95 6.1. REY partitioning across the coal-fired power plants.................................................. 109
96 6.1.1 Mass balances of REY in power plants. ............................................................... 110
97 6.1.2. Relative distribution of REY among ash streams ................................................ 113
98 6.2. REY partitioning within ash-collection systems ........................................................ 115
99 6.2.1. Variations in REY distribution pattern within ash-collection systems ................ 115
100 6.2.2. Variations of REY in sized fly ash and implications for REY partitioning trend 118
101 6.3. Speciation transformation mechanisms ...................................................................... 121
102 6.3.1. Physical fragmentation ........................................................................................ 121
103 6.3.2. Phase transformation and chemical reactions ...................................................... 123
104 6.4. Further retention mechanisms by fly ash glass ........................................................... 128
105 6.4.1. Origin and formation ........................................................................................... 129
106 6.4.2. Structure and chemistry ....................................................................................... 135
107 6.4.3. Possible incorporation mechanisms of REY into fly ash glass ........................... 138
108 6.5. Overview of REY partitioning mechanisms in coal-fired power plants..................... 146
109 7. Methods for REY recovery from coal ash ......................................................................... 150
110 7.1. Current REY extraction methods ............................................................................... 150
111 7.2. Physical beneficiation................................................................................................. 150
112 7.3. Hydrometallurgical extraction .................................................................................... 153
113 8. Conclusions ....................................................................................................................... 157
114 9. Problems and future outlooks ............................................................................................ 161
115 Acknowledgements ............................................................................................................... 164
116 References ............................................................................................................................. 165

5
117 1. Introduction

118 1.1 Background of rare earth elements in coal ash study and their significance
119 The story of metals recovery from coal ash (i.e. fly ash and bottom ash) can be

120 traced back to the late 19th century, at which vanadium (V) and silver (Ag) were

121 separated from the Poland coal ash [1]. Late in 20th century much attention has been

122 given to coal combustion residues (mainly ash) as potential sources for valuable

123 elements including gold (Au), germanium (Ge), gallium (Ga), and uranium (U) [2]. By

124 the middle of 20th century, U and Ge have been commercially extracted from coal ash

125 residues in the United Kingdom, Soviet Union, and Japan [3–5]. Coal and its ash even

126 became the main U source for the nuclear applications in the post-World War II period

127 [6,7]. Today, the extraction of Ge from the high-Ge coal-derived ash in China and

128 Russian Far East has been the main sources (more than 50%) for world industry [8,9].

129 In case of rare earth elements (herein REYSc represents the sum of the lanthanides, Y,

130 and Sc; REY, REE+Y; REE, if both Y and Sc are not included), several international

131 agencies and national governments have identified REY, especially the middle and

132 heavy REY as the critical metals because of their special applications in the high-tech

133 areas, such as renewable energy technologies, catalysts, permanent magnets,

134 metallurgical additives, and rechargeable batteries [10,11]. The shortage of

135 conventional rare earth ore deposits in recent years, especially the heavy rare earth

136 elements (HREY), stimulated attempts to exploit alternative sources to meet the

137 ever-increasing demand for REY by modern society [11]. Coal and coal ash containing

138 rare metals have again attracted special attention by a number of countries including the

139 United States [12–17], Europe [18–23], China [24–28], Russia [7,29], India [30,31],

140 South Africa [32], Indonesia [33], and among others to alleviate the REY resource

141 depletion crises.

6
142 Coal seams with anomalously high contents of rare and precious metals could be

143 possible sources of REY [7,34]. A number of coals, such as the Kentucky Fire Clay coal

144 in Eastern USA [35,36], southwestern China coal [37–39], and Russian Far East coal

145 [29] have been identified with high REY concentration. For detailed knowledge of

146 REE-rich coal deposits, readers are referred to the review by Seredin and Dai [7], Dai et

147 al. [8], and Zhang et al [40]. To evaluate the coal-related materials as resources for REY,

148 Seredin and Dai [7] proposed the cut-off grade for REY oxides (herein used as REO,

149 sum of La2O3 to Lu2O3 plus Y2O3) in coal ash is 0.1%. Most of REY in coal are

150 inorganic minerals (e.g. REY-phosphates) have high melting, boiling, and thermal

151 decomposition temperature [41], allowing them to be enriched in coal ash after removal

152 of organic matter [42]. The concentration of REO in ash residual after REY-rich coal

153 burning is up to 1-3%, comparable to that of conventional economic ores [43].

154 Moreover, the extraction of REY from coal ash has several advantages over the

155 conventional ores, for example fly ash is readily available waste with typical particle

156 size ranging from 0.1-100 µm, which can avoid mining and milling costs spent in the

157 traditional ores (e.g., carbonatite and ion-adsorbed ores). The recovery of REY from

158 coal ash is also an innovative use of coal fly ash with higher economic value. Besides,

159 coal-derived ash is relatively free of radioactive substances compared to the radioactive

160 elements (Th and U) released with the conventional REE ores utilization [10,44]. In

161 brief, the recovery of REY as well as critical metals from these waste materials, if

162 successful on an industrial scale, can both alleviate the REY depletion crisis worldwide

163 and beneficiate the coal-related industries [7, 8].

164 1.2 Scope of the review

165 An increasing number of literature, scientific reports, and patents have been

166 published to characterize and extract REY from coal ash, since the United States

7
167 Department of Energy (US-DOE) initiated several projects for recovering REY from

168 coal combustion byproducts in 2014 [17]. Before the extraction of REY from coal ash,

169 it is important to make clear that (1) the concentration, physical distribution, and

170 chemical speciation of REY in coal ash and (2) how these characteristics are associated

171 with the REY extractability. In coal-fired power plants, how REY partition from coal to

172 fly ash and to which constituents the REY are bound in coal ash (i.e., fly ash and bottom

173 ash or boiler slag) are dependent upon factors such as the feed coal properties, the

174 thermal stability of REE species, boiler conditions (temperature, O2 contents, and other

175 parameters), and configuration of the flue gas purification devices. A final viable

176 scheme for REY recovery from coal ash requires identification of ash fractions that

177 enrich REY which are easily leached at mild conditions. Understanding the fate of REY

178 during coal burning in coal-fired power plants can provide a knowledge basis for

179 modeling REE partitioning behavior and for designing proper extraction processes to

180 promote REY recovery from coal ash[8,17].

181 The recovery of REY from coal and coal ash is ongoing and will be a hot topic in

182 the foreseeable future. Several review papers are available related to the concentrations,

183 speciation, and genetic origin of REY in coal [6,8,9,17] and the recovery methods of

184 REY from coal and coal byproducts [40,45,46]. However, there is no systematic

185 knowledge on the retention and transformation behavior of REY during coal

186 combustion; and a summary of the contents, distribution, and chemical speciation of

187 REY in coal ash as well as the REY extraction methods is lacking. To better understand

188 the resource base and improve the knowledge about REY in coal ash that can be

189 recovered in the future, this paper serves as a summary review of recent work focusing

190 on the contents, distribution, speciation, and recovery methods of REY in coal ash, with

8
191 special attention to the partitioning behavior of REY during coal combustion. The

192 problems and future outlooks are introduced in each section and the end of the paper.

193 2. Nature of REY-rich coal

194 2.1. Nature of REY

195 As defined by the International Union of Pure and Applied Chemistry [47], rare

196 earth elements consist of 15 lanthanides series (atomic number 57 of lanthanum to 71

197 of lutetium), plus scandium (Sc, atomic number 21) and yttrium (Y, atomic number 21)

198 (Figure 1). The electron configuration for Sc is recognized as [Ar]3d14s2, Y as

199 [Kr]3d14s2, and REE as [Xe]4fn5dm6s2. The similarity in valence electron (outermost

200 shell) is the reason behind the similarities of physicochemical properties among the 17

201 rare earths. Due to differences in electron structures, such as the number of electrons

202 in 4f orbitals, REY shows different properties and each of them has special

203 applications in high-technologies [48–54]. Various classification methods have been

204 developed and used to understand their distinct behavior in different processes, such

205 as in rock geochemical evolution history [43,55–59], mineral processing [60–64], and

206 environmental mobility, etc. [65–71]. For coal-related studies, a classification scheme

207 of rare earth elements has been built by dividing the REY into light REE (La to Sm)

208 and HREE (Eu to Lu), with Y sometimes being included among the HREE [36,72,73].

209 Seredin and Dai [7] proposed a threefold classification that is used to divide the REY

210 into light (LREY; La through Sm), middle (Eu through Dy plus Y), and heavy rare

211 earth elements (Ho through Lu)(Figure 1). The ratio of LREE to HREE (L/H) was

212 used to assess the distribution of REE in coal and coal ash [29]. To further evaluate

9
213 coal as a REY raw materials, a threefold industrial division of REY, i.e., critical (Nd,

214 Eu, Tb, Dy, Y, and Er), uncritical (La, Pr, Sm, and Gd) and excessive (Ce, Ho, Tm, Yb,

215 and Lu) groups is proposed [7,78]. Note that the classification of REY into critical,

216 uncritical, and excessive groups is somewhat subjective and transient, contingent

217 upon the global market demands and the supplies on hand [75]. The price of

218 individual REY fluctuated sharply over the last decade [76–79].

219

220

221 Figure 1. The classification of rare earth elements and yttrium according to their

222 geochemical behavior or market price. (after Seredin and Dai [7] and Hower et al.

223 [17])

224

225 The abundance of REY in natural samples follows the Oddo-Harkins rule, in

226 which elements with even atomic numbers are more abundant than odd-numbered

227 elements [80,81]. This effect can also be reflected in coal and coal fly ash (Fig 2a).

228 Normalization of REY to reference materials is necessary to eliminate the “zigzag”

229 shape caused by the Oddo-Harkins effects and to compare the fractionation of REY in

10
230 different processes. Average concentrations of REY in world coal or coal ash [82]

231 have been normalized to chondrites [83], Upper Continental Crust (UCC) [84], North

232 American Shale Composite [85], and Post-Archaean Australian Shale Composite

233 (PAAS) for geochemical and coal combustion studies. The former (i.e., chondrites

234 normalization) is suggested not to be used in coal-related studies because chondrite is

235 the primitive materials of solar systems which is commonly used in

236 igneous/metamorphic geochemistry. It is known that coal is derived from the upper

237 earth crusts having similar genetic origins to UCC, and therefore, REY in coal or

238 coal-related materials are commonly normalized to the average REY abundance in

239 UCC. In addition to UCC and chondrites, normalization to shale such as the North

240 American Shale Composite (NASC) [85] has been used in some investigations. In

241 presenting REY data, any of the normalization patterns are acceptable.

242

243

244 Figure 2. Plots showing how REE data are presented. (a) Plots of REE

245 abundances in order of atomic number of rare earths, showing “zig-zag” shape

246 constrained by the Oddo-Harkins rule; Promethium (Pm, atomic number (61) is

11
247 omitted because it is not stable in nature; (b) Plots show how normalization of REE

248 concentration data results in smooth patterns for distribution. The average REE

249 concentrations (parts per million) of the materials shown in the legends represents: 1)

250 Chondrites, data from Boynton [83]; 2) the upper continental crust (UCC), data from

251 Taylor and McLennan [84]; 3) the North American Shale Composite (NASC), data

252 from Gromet et al. [85]; and 4) the world hard coal and coal ash, data from Ketris and

253 Yudovich [82].

254

255 When normalizing to UCC, three enrichment types of REY can be identified

256 [7,34]: L-type (LREY, LaN/LuN > 1; N indicates a value corrected against a standard

257 distribution, such as the Upper Continental Crust), M-type (LaN/SmN <1, GdN/LuN >

258 1), and H-type (LaN/LuN <1). Besides the enrichment types, the following ratios can

259 be used to decouple Ce, Eu, and Gd from the other REE in the distribution patterns

260 [7,34]:

261 EuN/EuN* = EuN/(0.67SmN + 0.33TbN) (1)

262 CeN/CeN* = CeN/(0.5LaN+0.5PrN) (2)

263 GdN/GdN* = GdN/(0.33SmN + 0.67TbN)

264 (3).

265 Different distribution patterns and the distinct features of Ce, Eu, and Gd correspond

266 to different processes or origins [34], thus REY are a good tracer which has been

267 widely used in the geochemical and environmental studies [67,86–90]. Other

268 definitions of the REY distribution patterns and REY anomalies (e.g., La and Y

12
269 anomalies) are also proposed [29,36,72,73]; therefore, the reader should be cautious

270 to carefully examine the definitions in any literature.

271 2.2. Contents of REY in coal and REY rich coal

272 2.2.1. Abundance of REY in coals around the world

273 A summary of average REY contents in coal from some coal-producing countries,

274 i.e., China, Columbia, Democratic People’s Republic (DPR) of Korea, North Asia

275 (Siberia, Russia Far East, Mongolia, and Kazakhstan), Turkey, and the United States

276 is shown in Table 1. Among the countries, coals from China contained the most

277 abundant REY, followed by Turkey and other countries. Ketris and Yudovich [82]

278 reported that the average abundance of REY in the world coal was estimated to be

279 68.6 µg g-1, comparable to that of UCC (Fig. 2b). On the basis of data from Ketris and

280 Yudovich [82], Zhang et al. [40] estimated a total of ~ 50 million metric tons of

281 REE reserved in world coal, equating to almost half of the REE reserve of traditional

282 REE-bearing ore deposits. Note that the concentrations of REY significantly vary in

283 different coal deposits and different mines, or even show large variations at a single

284 seam in a single mine [36,91–93]. Rare earth elements in coal are usually at trace

285 level (below 0.1 w.t.%). It is necessary to identify REY-rich coal or enrich REY in

286 coal ash (such to meet the requirement for the industrial extraction [40].

287

13
288 Table 1. A summary of REY concentration in major coal-producing countries or regions around the world.

Columbian coal1 China2 DPR Korea coal3 Turkey coal4 US5 North Asia6

Element Ave Ra N Ave N Ave Ra N Ave N Ave n Ave Ra n

Y 10.8 0.5-36.9 24 18.2 888 7.2 2.9-22.0 50 12.7 13 8.9 7585 - - -

La 9.9 0.2-37.6 24 22.5 392 14.5 3-53.0 50 21.1 13 11.3 6652 14.7 2.7-61.4 4560
Ce 19.9 0.4-73.4 24 46.7 392 27.2 5-102.0 50 39.2 13 21.8 6081 31.4 6.7-141.0 4560
Pr 2.6 0.05-10.1 24 6.4 392 2.9 0.6-10.0 50 4.7 13 2.5 5601 - - -
Nd 9.8 0.2-38.6 24 22.3 392 11.1 2.0-38.0 50 16.8 13 10.4 5946 - - -
Sm 2.2 0.1-9.0 24 4.1 392 2.3 0.6-7.0 50 3.2 13 2.0 5588 3.1 0.46-9.1 4560
Eu 0.5 0.05-2.3 24 0.8 392 0.5 0.1-1.0 50 0.7 13 0.4 5626 0.8 0.09-2.2 4560
Gd 2.2 0.05-8.3 24 0.8 392 1.4 0.4-4.0 50 3.0 13 2.3 5602 - - -
Tb 0.3 0.05-1.2 24 0.6 392 0.3 0.1-0.8 50 0.4 13 0.4 5619 0.5 0.06-2.1 4560
Dy 2.1 0.05-7.7 24 3.7 392 2.0 0.6-5 50 2.4 13 2.3 5607 - - -
Ho 0.4 0.05-1.4 24 0.9 392 0.4 0.1-0.8 50 0.5 13 0.5 5598 - - -
Er 1.1 0.05-4.0 24 1.8 392 1.1 0.3-2.0 50 1.4 13 1.2 5603 - - -
Tm 0.2 0.05-0.6 24 0.6 392 0.3 0.1-0.6 50 0.2 13 0.2 5603 - - -
Yb 1.1 0.1-3.7 24 2.1 392 1.0 0.3-2.0 50 1.3 13 1.0 7269 1.5 0.33-6.4 4560
Lu 0.2 0.05-0.6 24 0.4 392 n.d. n.d. 50 0.2 13 0.2 5587 0.3 0.043-0.8 -
LREY 44.3 - - 102 - - - - 85.1 48.1 - - - -
MREY 15.9 - - 24.2 - - - - 19.4 14.4 - - - -
HREY 3.1 - - 5.9 - - - - 3.6 3.1 - - - -
Total
63.3 - - 132.1 - - - - 108.1 63.31 65.5 - - - -
REY

289 Values in μg g-1. 1 From Huang et al. [94]. 2 From Dai et al. [95]. 3 From Hu et al. [96]. 4 From Karayigit et al. [97]. 5 From Finkelman et

290 al. [98]. 6 From Arbuzov et al. [99], and the countries of North Asia in their study refer to Siberia, Russia Far East, Mongolia, and

291 Kazakhstan. Ave: the average concentration, µg g-1; Ra: the range of element content, µg g-1; N: the number of coal samples.

14
292 2.2.2. REY rich coal in major coal-producing countries

293 Coal is arguably one of the most complicated rocks on earth [100] and is a

294 typical geochemical barrier that can enrich many precious metals, such as REY, U, Ge,

295 Ga, V, platinum group elements (PGEs), and among others [6,9]. The increasing

296 demand for REY in recent years has stimulated the exploration of alternative rare

297 earth sources including REE-rich coal deposits around the world. Seredin and Dai et

298 al. [7] proposed the cut-off grade of REY concentration for potential industrial

299 utilization. The cut-off value of REO content is required to be greater than 0.1% in

300 coal ash, or 800-900 µg g-1 in coal ash for coal seams with thickness above five

301 meters. The US-DOE set a criterion in their work for assessing raw coal as REY

302 resources: the total REY content shall be greater than 300 µg g-1 on a whole dry coal

303 basis. As shown in Table 2, coal geologists have so far identified several typical

304 REE-rich coal beds (> 0.1%, ash-basis), such as the Pavlovka coal from Russian Far

305 East [29], the Fire Clay coal from central Appalachian basin of Eastern United States

306 [35,36], and the Fuisui and Heshan coals from Sichuan basin of southwestern China

307 [38]. These coals and their corresponding combustion ashes have been considered as

308 potential REY resources. In some literature, the REY-rich coals are broadly referred

309 as ‘metalliferous coal deposits’ or ‘coal-hosted ore deposits’, which had been

310 accurately defined by Seredin and Finkelman [6] as coals with trace element contents

311 10-times higher than the world average. This term has been gradually accepted and

312 widely used in many studies thereafter [8,14,17,100–102].

313

15
314 Table 2. Coal samples extremely enriched in REY (µg g-1) from some coal deposits

315 around the world.

Thickness REY
Country Region Deposits Ad % Rank N Reference
-1
(m) µg g

Russia Far East Pavlovka 19.4 Lignite 1.15-1.9 5952 13 [7]

China Guizhou Liuzhi 12.8 Bituminous 10 2491 1 [7]

China Chongqing Songzao 24.8 Bituminous 0.4-2.4 1264 16 [7]

China Guangxi Fusui 29.6 Bituminous 1.3-1.8 1095 2 [103]

USA Kentucky Appalachian 8.7 Bituminous 1.1 1460 6 [36]

Tadjikistan Nazar-Ailok 3.5 Anthracite 1.0-3.0 1836 6 [104]

316 Notes. Ad represents ash yield of the coal on dry-basis,%; N is the number of coal samples.

317

318 To date, the study of REY-rich coals in the United States and China, including

319 the distribution, reserve, geochemical behavior [24,37–39,91,105,106] and related

320 extraction method [27,107,108] have come a long way in the last decade. The United

321 States Geological Survey systematically investigated the spatial distribution of REY

322 in coals from different US coal basins. All the data has been recorded in the USGS

323 Coal Quality database, known as the COALQUAL database with the latest Version

324 3.0 (V3.0) published online (https://ncrdspublic.er.usgs.gov/coalqual/). As shown in

325 Figure 3, samples with REY contents higher than 1000 µg g-1 (ash basis) are densely

326 distributed in the Appalachian and Black Warrior basins. Other regions, including the

327 Illinois, Forest City, and Arkoma basins, etc., are sporadically distributed with

16
328 REY-rich coals. Based on the data of 5378 coal samples from USGS COALQUAL

329 database, Lin et al. [109] estimated that about 9-13% of the samples in the Unites

330 States are promising coals for REY extraction. Bituminous coals from Central

331 Appalachian region, especially from Eastern Kentucky are probably the best option

332 for the beneficial recovery of REY [35,36,109].

333

334 Figure. 3. Coal-hosted REE metallogenic zone across the United States. Black dot

335 represents the sample containing REY greater than 1000 µg g-1 (ash basis). The

336 distribution map of REE-bearing coals was obtained from the web-based

337 COALQUAL V3.0 database at the USGS website

338 (https://ncrdspublic.er.usgs.gov/coalqual/).

339
17
340 Coals from China, especially from southwestern China, contain high

341 concentration of REY [37,38,92,106,110,111]. Based on the location of coal basins

342 and their regional tectonic back-ground, two coal-hosted REE metallogenic zones

343 have been recognized, namely the Erlian-Hailaer REY mineralization belt and the

344 Sichuan-Yunnan-Guangxi REY mineralization belt [106]. As shown in Figure 4, the

345 first region is situated in the Erlian basin of northeastern China, where the REY-rich

346 coals are mainly formed in early-middle Jurassic period. The average REY

347 concentration of coal is 397.3 µg g-1 and the maximum value can be up to 1461.2 µg

348 g-1. The second coal-hosted REY metallogenic belt locates in the southwestern of

349 China, including coal deposits from Sichuan province, Chongqing city, Guangxi

350 Province, and Yunnan Province. Dai et al. [34,37–39], Zhuang et al.[112,113], Zhao

351 [110,114], and among others [115–118] have studied the REY origins and distribution

352 in the coals from the REY mineralization belt. Average abundance of REY in this

353 region is estimated to be in the range of 874 µg g-1 to 2497 µg g-1 (ash basis) and have

354 thus been considered as promising coals for REY recovery [37,38,92]. According to

355 the data from China Geological Survey

356 (https://www.cgs.gov.cn/xwl/ddyw/201802/t20180226_451616.html), the total reserve

357 of REY in these coal-hosted ore approach approximately 11,300,000 Mt.

358 Overall, the REY-enriched coals from the US and China have been used as

359 steam coal for power generation for many years [16,28,119–122]; the combustion

360 process can further enrich REY in the ash which is beneficial to the extraction. While

361 a few coal-hosted REE ore deposits have been identified worldwide, the uneven

18
362 distribution of REY in different coals from different regions still require a full

363 investigation of REY distribution before the industrial extraction and utilization.

364 Taking the distribution of REY in China as an example, while the average REY

365 content in Chinese coal (136 µg g-1) is higher than the world coal [95], many coal

366 deposits in China cannot be used as potential raw materials for REY extraction. The

367 North China coal basin is the most important coal producing area for China industry,

368 however, the average content of REY in the northern Chinese coals is estimated to be

369 111.2 µg g-1 [123], much lower than the extraction criterion [7].

370

371

372 Figure. 4. Coal-hosted REE metallogenic zone across China coal basins, revised

373 based on the data after Ning et al. [106].

374

19
375 2.3. REY occurrence in coal

376 The modes of occurrence of REY in coal are important to anticipate their

377 behavior during coal combustion. REY in coal are closely associated with their

378 geological origins, i.e., detrital inputs from the source rocks by wind and water,

379 mineralized REY-rich fluids, and volcanic ash deposits [8,9,34]. Achieving a detailed

380 knowledge of REY speciation in a coal is difficult because: (1) REY enrichment in

381 coal is usually a function of multiple modes of mineralization resulting in complicate

382 REY associations [35] and (2) the concentration of an individual element is

383 commonly less than 100 µg g-1. Broadly, REY in coal can be classified into inorganic

384 and organic matter associations (Figure 5) [7,132]. The inorganic forms are the

385 common REY associations in coal including individual REY-bearing minerals (such

386 as phosphates, sulfates, carbonates, and fluorocarbonates, etc.), substitutions in the

387 crystalline structure of accessory minerals (zircon, apatite, etc.), adsorbed to clay

388 minerals, and dissolved in pore waters in coal [7,133,134]. For organic associations,

389 REY in low-rank coal might be bound with carbon functional groups (e.g. -COOH

390 and OH) or form organometallics [72,91,124]. To model the partitioning behavior of

391 REY during coal combustion, study of REY forms in coal should answer the

392 following questions: (1) what is the ratio of each REY associations (mass balance of

393 REY forms) in REY-rich coal? (2) what is the size range of REY-bearing particles in

394 coal and how does REY-bearing phase distribute in coal matrix? (3) what are the

395 thermal properties of common REY-bearing phase in coal?

20
396

REY-bearing
INTRINSIC
organometallic compound
ORGANIC
Side chain functional
ADSORBED groups, -COOH, -OH, -
SH, =NH
REY IN COAL

Apatite, zircon, APS

REPLACEMENT
minerals

Adsorbed on existing
ADSORBED
minerals, e.g. clays
INORGANIC
Monazite,xenotime,
DISCRETE MINERALS bastnaesite, kimuraite,
lanthanite, florencite, etc.

DISSOLVED REY ions in pore fluids

397

398 Figure 5. The modes of occurrences of REY in coal.

399

400 2.3.1 Proportions of organic and inorganic associations of REY in coal

401 Besides REY binding forms, the study of different REY associations and their

402 proportions in coal is important to the prediction of REY speciation in coal ash and

403 understanding REY transformation behavior during coal combustion. However,

404 quantitative analysis of each association in coal is hard to be achieved by current

405 techniques. Considering the thermal stability of REY-bearing phase, more attention is

406 given to the distribution ratio of REY in organic matter and inorganic matter in coal.

407 In general, coal minerals are the primary hosts of REY especially in high-rank coals

408 [7,133,135]. A part of REY are considered to be associated with organics in low-rank

409 coals, such as lignite [29,72,127]. The first literature supporting the organic

21
410 occurrences of REY in coal might be from Yershov [128], who favored the organic

411 bonding of Y and Yb in Kizelovsk coal from the former Soviet Union. Seredin and his

412 colleagues [29,129] found that about 50% of REY in the lignite and subbituminous

413 coals in the Russian Far East region can be extracted by 1% NaOH, which are

414 considered to be distributed in the humic matter. The organic association of REY in

415 coal was further reported in various coals from different countries, and confirmed by a

416 series of adsorption experiments (adsorption of REY onto humic and fluvic acids, peat,

417 and coals) [29,72,127].

418 Since early the 1980s, the U.S. Geological Survey (USGS) investigated the

419 modes of occurrence of REY in coals of various ranks (from lignite to bituminous

420 coal) through selective leaching procedure [125,130,131] . Although indirect,

421 sequential extraction experiments can rapidly provide the opportunity to quantify the

422 REE occurrences in coal. The results showed that REY in most coals occur in

423 monazite, xenotime, and other phosphate minerals, and less than 10% of the total

424 REY existing in US lignite was associated with organic matter [132]. On the basis of

425 summarizing their previous studies and using the same sequential leaching protocol

426 on another 20 international coal samples, Finkelman et al. [125] concluded that most

427 of LREY (> 90%) in coal occurred in mineral phase and about 25-30% HREY had an

428 organic association (Table 3). Further selective leaching on low-rank coals found that

429 more REY were associated with organics than that of high-rank coals. Lin et al. [124]

430 estimated that 25% REY were associated with organics in the low-ash central

431 Appalachian coals. Based on the organic solvent extraction on seven coals from

22
432 Shanxi, China, Wang et al. [91] suggested that REE in organic matter were probably

433 adsorbed by hydrogen-containing functional groups and HREE might be directly

434 bound to carbon. Dai et al. [133] also found that aluminosilicates were the primary

435 hosts for REE in bituminous coals from Ningxia Province of China. However, there

436 are still disputes on the distribution of LREE and HREE in different occurrence

437 modes.

23
438

439 Table 3. The quantification of the modes of REY in coal by sequential leaching procedure.

Coal source Coal Rank Elements modes of occurrences References

70% phosphates, 20% clays, 10% carbonates,

LREE
9 from U.S., 1 from Canada, 1 from Brazil, 1 organic association
Bituminous
from the Great Britain, and 2 from Australia 50% phosphates, 20% clays, 30% organic
HREE
association carbonates Finkelman et al.

60% clays, 20% phosphates, 20% carbonates, [125]

LREE
Lignite and organic association
6 from US
subbituminous 50% clays, 25% phosphates, 25% organic
HREE
association, carbonates

U.S. Kentucky coal Bituminous REY 25% of REE bound with organic matter in coal Lin et al. [134]

43-95% aluminosilicates, 0.03-28% carbonates,

LREE
3-7% organic association, 17-23% sulfides

43-95% aluminosilicates, 0.1-33% carbonates,

4 China Ningxia Shitan coal Bituminous HREE Dai et al., [133]
4-23% organic association, 7-33% sulfides

37-93% aluminosilicates, 0.04-29% carbonate,

REE
5-8% organic association, 16-27% sulfides

Lignite and about 50% of the total REY is bound with humic Serdin and
Russia coal
subbituminous matter Shpirt [129]

24
440 In summary, REY in coals have both inorganic and organic associations, and low

441 ash and low rank coals generally have more organically associated REY than high-ash

442 and high-rank coals. However, the proportions of organically- and

443 inorganically-associated REY differ between coal deposits and even with samples

444 from a single coal bed, which are certainly a function of geological settings of coal

445 basins (e.g., source-rock properties, volcanic eruptions, etc.), weathering conditions

446 during peat and coal formation, and depositional environment [7,125]. The share of

447 each REY forms in the balance of the total REY in Table 3 is a general result and can

448 be updated with more studies carrying out in the future. To develop reliable sequential

449 extraction procedure, verification experiments by direct techniques and

450 double-checked by leaching pure substance and certified coal standards are strongly

451 suggested.

452 2.3.2. Size and spatial distribution of REY-bearing particles in coal

453 The size and spatial distribution of REY-bearing particles in coal are essential to

454 understanding their partitioning behavior during coal combustion. According to the

455 chemical components and the modes of occurrences as noted above, REY-bearing

456 mineral grains in coal can be divided into:

457 REY phosphates, including monazite, xenotime, florencite, and water-bearing

458 aluminum phosphates;

459 REY carbonates, including bastnaesite, lanthanite, kimuaite, and others;

460 Isomorphic admixtures in minerals: zircon, apatite, REY-bearing aluminum

461 phosphates, sulfates of the alunite supergroup (APS minerals); and

25
462 Adsorbed onto clays.

463 The size of these REY-bearing particles can vary from several tens of microns

464 (Figure 6a), to several microns (Figure 6b and 6c), to the submicron to nanometer

465 scale (Figure 6d) [37,39,135]. The size and shape of REY particles are associated

466 with their different origins. In addition to the discrete minerals inherited from the

467 source region (i.e., volcanic ash or peralkaline basement rocks of the coal basin)

468 [36,136–138], a number of SEM-EDS studies have shown that the fine authigenic

469 REY-bearing minerals, formed from chemical or biochemical precipitation of

470 REE-bearing mineralized solutions, are typical forms for the majority of

471 REY-enriched coals. These fine-grained clastic and authigenic REY minerals usually

472 have particle size less than 10 μm [14,35,37–39,92,110,111,135,136,139].

473

26
474

475 Figure 6. The REY containing mineral grains in coal. (a) rhabdophane distributed

476 along the bedding planes in the late Permian K1-2b coal from the Huayingshan,

477 Sichuan, southwestern China. Reprinted from [136] with permission from Elsevier (b)

478 REY-bearing carbonate mineral (Ca(Mg)CO3F) in the Late Permian coal from the

479 Heshan Coalfield, southern China. Reprinted from [139] with permission from

480 Elsevier. (c) Ce-, La-, Nd-, Ca-rich phosphate (bright spots) distributed in a kaolinite

481 matrix in Fire Clay coal from Eastern Kentucky. Reprinted from [36] with permission

482 from Elsevier. (d) Mixed monazite and kaolinite grain in the Dean coal, Knox County,

483 Kentucky. Reprinted from [135] with permission from Elsevier.

484

27
485 The fine particle size of REE-bearing mineral grains and the low concentration

486 of individual REY in coal make the quantification of the size distribution of REY

487 mineral grains a challenging task. Based on the computer-controlled scanning electron

488 microscopy techniques (CCSEM), a research team from the University of North

489 Dakota found that 97 REE-bearing particles of 3307 particles in a lignite are all less

490 than 10 μm[140] (Details are available at

491 https://edx.netl.doe.gov/dataset/70fc29f0-91d5-4fed-83bb-93fe0d0c0c66/resource/45a

492 e0e79-9470-4db1-8869-9716afbd7302). Indeed, 75% of the 97 REE-bearing particles

493 are in the size range of 0.4-1.0 μm. The particle size of most of REE-bearing phase in

494 REY-rich coal is considered in the size of < 10 μm and some fractions are even

495 sub-micron particles [140]. This provides important information on predicting size

496 distribution of REY in coal-derived ashes. However, there are several severe

497 limitations to quantify the REY containing minerals in coal via CCSEM technique:

498 Even in the REY rich coal, the lower concentrations of REY, especially the

499 HREY are beyond the detection limits of EDS;

500 The REY minerals are chosen and identified based on their chemical

501 compositions, thus, there is no allowance for identifying an unclassified particle

502 or particles with different chemistry from those in the mineral data files;

503 Interference peaks in some cases appear in the EDS spectra, requiring

504 de-convolution of the overlapping peaks; and

505 Nano-sized REE-bearing minerals might occur in other mineral matrices (such

506 as kaolinite), making the quantification analysis difficult and inaccurate.

28
507 In addition to the particle size distribution, spatial distribution of REY-bearing

508 particles within single coal particle is another important parameter affecting REY

509 combustion behavior in coal-fired power plants. REY bound with included minerals

510 within coal char might be subjected to a more reducing and a higher temperature

511 condition than the excluded minerals [141,142]. Therefore, the decomposition and

512 evaporation of the included REY-bearing minerals might be strengthened. Previous

513 electron microscopy studies revealed that REY-bearing mineral particles seemed to

514 occur in the organic maceral of coal [6,35,100,126,143]. Some of them are dispersed

515 into coal maceral, and a part of REY mineral grains are embedded in other mineral

516 matrix, such as clay minerals and carbonates within a coal particle (Figure 6).

517 Excluded REY-minerals in coal is rarely reported in published literature. The

518 CCSEM study as illustrated in the technical report further demonstrated that the

519 REY-bearing particles mainly occur as included minerals in coal [140].

520

521 3. Determination of REY in coal ash

522 As shown in Figure 7, methods for analysis of REY in coal ash can be divided into

523 the following categories: (1) methods that measure the total REY concentrations and (2)

524 methods that determine speciation of REY in coal ash. Both the bulk elemental analysis

525 and speciation analysis for REY in coal ash are critical to design cost-effective and

526 environmental-friendly extraction methods. The use of various techniques for trace

527 element analysis has been developed for many years; however, a reliable and accurate

528 determination of REY in coal ash still has some problems. This is mainly because (1)

29
529 the complex matrix of the organic/inorganic constituents in coal ash and potential

530 contamination during sampling, preservation, and pretreatment procedure, all of which

531 make REY determination difficult and complicated; and (2) REY associations in coal

532 ash are usually too fine (also few) to be discerned and are heterogeneously distributed

533 among crystalline minerals, amorphous materials, and unburned carbon [144–147],

534 making the accurate speciation analysis very difficult. Therefore, conventional

535 analytical methods used for quantifying REY contents in various materials, including

536 gravimetry, titrimetric, spectrophotometry, and atomic absorption spectrometry (AAS)

537 have limitations, such as high detection limits, low sensitivities, poor accuracy, and

538 time-consuming process. These methods are not suitable for REY determination in

539 complicated coal ash and will not be considered in this review.

540 Previous summary of trace elements (mainly toxic elements) determination in coal

541 and coal ash has been quite extensively made, including reviews and papers by Karr

542 [148], Vouropoulos [149], Huggins [150], Bullock [151], ASTM[152]. Nonetheless,

543 few studies focused on the REY determination in coal-related materials. The following

544 section will review the REY analytical methods in coal ash, based on the available

545 literature published over the last thirty years.

30
 Analysis of
REY in coal
ash

Bulk elemental Speciation

analysis analysis

Instrumental X- Mass
Absorption/Emissio Direct methods Indirect methods
ray/γ-ray n spectroscopic Spectroscopic
methods methods methods

INAA AAS ICP MS Instrumental

microanalysis Electron Synchrotron- Float-sink
microscopy based techniques
techniques
EDXRF ICP OES GD MS Sequential
Bulk leaching
LA ICP MS SEM EDS XANES
WDXRF ETV ICP MS
Statistical
SIMS Bulk EXAFS methods
FIB SEM
Others

TEM/SAED/ µXRF
EDS

µXANES
HRTEM/FF
T/EELS
546
547 Figure 7. Subdivision of REY determination methods for coal ash.

548

549 3.1. Bulk analysis of REY concentration

550 Figure 8 is a statistical illustration of the share of various methods in the

551 analyses of lanthanides. In the last two decades, the most common analysis methods

552 of bulk REY in coal ash are inductively coupled plasma mass spectrometry (ICP-MS)

553 [16,18,19,22,23,27,28,32,33, 93, 97,

554 107,108,119,120,158,165,166,169-171-173,175-182,185-186,190–196, 204, 230,

555 234, 237-238, 246-247, 250-253, 258-259, 262, 264, 270, 272, 364-367], followed

556 by inductively coupled plasma optical emission spectrometry or atomic emission

557 spectrometry (ICP-OES/AES) [15, 22, 30, 31, 33, 122, 140, 159, 161, 163-164, 173,

558 184-186, 231, 236, 363], instrumental neutron activation analysis (INAA) [32, 98,

559 125, 131, 140, 185-186, 211-213, 357], and X-ray fluorescence (XRF) [20-21, 186,

31
560 215-220]. Other common-used techniques for the trace elements analysis in coal ash,

561 including atomic fluorescence spectrometry (AFS) and atomic absorption

562 spectrometry (AAS), are difficult for quantitative determination of REY, because the

563 stable compounds forming at the flame zone leads to low atomization efficiency and

564 potential spectra interferences. Direct analysis techniques, such as laser-induced

565 break-down spectrometry (LIBS) for solid samples, are still under development

566 [153]. Hence, the following text reviewed here will focus on bulk analysis of REY in

567 coal ash by ICP-OES, ICP-MS, XRF, and INAA.

568

569
570

571 Figure 8. The frequency of ICP-MS [16,18,19,22,23,27,28,32,33, 93, 97,

572 107,108,119,120,158,165,166,169-171-173,175-182,185-186,190–196, 204, 230, 234,

573 237-238, 246-247, 250-253, 258-259, 262, 264, 270, 272, 364-367], ICP-OES [15, 22,

574 30, 31, 33, 122, 140, 159, 161, 163-164, 173, 184-186, 231, 236, 363], INAA [32, 98,

575 125, 131, 140, 185-186, 211-213, 357], and XRF [20-21, 186, 215-220] used in REY

576 determination in coal ash, data compiled from the published literature over last thirty

577 years.
32
578

579 3.1.1. Sample preparation for ICP-based techniques

580 For the solid ash particles, sample pretreatment is a key step for preparing a

581 sample for ICP-based analysis. The first principle for sample preparation for trace

582 element analysis requires producing samples that are representative of the original

583 material without alteration, either by losses of volatile fractions or by adventitious

584 additions. Although some researchers attempted to introduce suspended solids/slurries

585 or pulverized coal into the high temperature plasma [154–156], the precision and

586 reproducibility for the elemental determination are poor [150]. REY-containing

587 aqueous solutions obtained from solid ash particles by acid digestion and/or alkaline

588 fusion are required for ICP-based techniques [16,19,22–24,28,42,108,120,152,157–

589 161]. Several sample preparation methods, such as acid dissolution, bomb

590 decomposition, alkaline fusion, and sintering have been widely used in numerous

591 laboratories. Here, a short review of the sample pretreatment methods is made before

592 the review of REY analysis through ICP-based techniques. The pretreatment methods

593 for coal ash includes:

594 Open acid digestion (hot plate digestion)

595 Closed vessel digestion

596 Microwave digestion

597 Alkaline fusion or sintering

598 Open acid digestion is commonly used for sample preparation for REY

599 determination in coal ash by the means of ICP-OES/MS (Table 4). Several national

600 standard methods have set up detailed digestion procedure using open acid digestion

601 method, such as American ASTM D 6357-19 [152], Chinese Standard MT/T

602 1014-2006 [162], and ISO 23380. It generally involves weighing certain amounts of

33
603 coal ash (0.1g or more), transferring the sample to an anti-corrosive container (e.g.,

604 Teflon cups), and adding concentrated acid and/or other oxidation reagents at a fixed

605 mixing ratio. Next the digestion vessel is placed on a hot plate with all the acids,

606 especially HF, being nearly completed evaporated. The obtained wet solid (mainly

607 salts) is re-dissolved with dilute HNO3, filtered, and diluted to a certain volume.

608 Most of ash components, including unburned carbon, aluminosilicates, carbonates,

609 sulfates, and other mineral or mineraloids, can be dissolved with the acid mixtures.

610 However, some REY-bearing particles (such as zircon) may not be dissolved;

611 therefore, higher digestion temperature (130-150 °C) and longer digestion time (at

612 least 5 h) are required for complete dissolution. Obviously, the main advantages of

613 the open hot-plate digestion method over other pretreatment methods are its

614 simplicity and low cost; however, the drawbacks are also evident, including the high

615 amount of acid consumed, acid mist escape, and longer time for the whole digestion

616 process. For example, 40-ml HF plus aqua regia are consumed by the ASTM D 6357

617 method and a longer time, including the ashing, digestion, evaporation, and

618 re-dissolving process, is required compared to other techniques.

619

34
620 Table 4. A summary of the pretreatment methods for REY determinations by the means of hot plate digestion, microwave-assisted digestion,

621 high-pressure closed-vessel digestion, and alkaline fusion-acid digestion methods.

Pretreatment method Procedural description Equipment References Recovery %

0.1g+20 ml HF+20 ml aqua regia, heated at around

Hot plate 130 ℃ for, then evaporated and re-dissolved with ICP-OES [15,159,163,164] 85-115%

dilute 10 ml HNO3 + 30 ml H2O at 110 ℃

0.2-0.5g +20 ml HF+20 ml aqua regia, heated at

Hot plate 130-150 ℃, evaporated; add 1 ml HNO3+20 ml H2O ICP-OES/MS [152] 85-115%

heated at 90-100 ℃

0.1 g + 2 ml HClO4 + 10 ml HF, evaporated and

Hot plate ICP-OES [162] /
added 10 ml HCl +10 ml H2O for re-dissolution

35 mg ash+2HF+2HNO3 at 90-100℃ overnight;

Two-step hot plate re-digestion with 1 HNO3+1 H2O2+5 H2O at ICP-MS [165–167] 89-102%

90-100℃ overnight

0.2 g + 2 HNO3 digested at 90 ℃ for 2h, and the residue was

High pressure closed vessel digested again with 7ml HF at 90 ℃ for 3h, evaporated and ICP-AES/MS [157,168–171] /

digested with 2 ml HClO4+2ml HNO3

35
 50mg+1HF+1HNO3, heated at 190℃ for 24 h, then 6ml HNO3
PTFE-lined stainless-steel bombs ICP-MS [28] 95-105%
(40%) heated at 140℃ for 5h for further digestion

1g + 10 ml HF+10 ml HCl+10ml HNO3, heated at

closed vessel ICP-AES [31] 95-105%
200 ℃ for 6h, evaporated and dissolved in 4M HNO3

0.25g + 2 ml HF + 5 ml aqua regia, heated at 200 ℃, adding 25

Teflon-lined Parr bomb ICP-OES [32] /
ml H3BO3 to react with fluorine

0.05g + 2 ml HNO3+5 ml HF, heated at 240 ℃ for 75 min,

Microwave digestion vaporization at 180 ℃, and re-dissolved with 5 ml HNO3 (50%) at ICP-MS [24,27,158,172] /

180 ℃ for 4h

Microwave digestion 0.1g ash+4 ml aqua regia + 2 ml HF, heated at 180 ℃ for 1 h ICP-MS [161]

Microwave digestion HF+HCl+HNO3, EN 13656 method, EN ISO 17294-2 ICP-MS [166] Close to 100%

250 mg of coal ash + 2 ml H2SO4 digested at 200 ℃ for

Hot plate plus microwave 1 hour on a hotplate, then adding 0.5 ml HF and heated for 15
ICP-OES [173] > 95%
digestion min, and the sample was mixed with 8 ml HCl + 2 ml H2SO4,

microwave digested at 220 ℃

0.2g ash +0.9g lithium borate flux, fused at 1000℃, dissolved in

Alkaline fusion ICP-MS [23]
4% HNO3 + 2%HCl

36
 0.05-0.1 g ash + 0.4 g LiBO2, heated at 1000-1100 ℃ for 20 min,
Alkaline fusion ICP-OES/MS [174] 85-115%
re-dissolved with 30 ml 1 + 9 HNO3

50 mg ash + 400 mg LiBO2 fused at 1100 ℃ for 5 min,

Alkaline fusion ICP-MS [175,176] 85-115%
re-dissolved in 5% HNO3 (USGS method)

0.1g ash + LiBO2 fused at 1025 ℃, dissolved in 4%

Alkaline fusion ICP-MS [33] /
HNO3-2%HCl

Alkaline fusion No details, LiBO2 fusion ICP-AES [22] /

0.1g ash+0.6g Na2O2 fused at 450 ℃（0.5 h), dissolved with 20 ml

Alkaline fusion ICP-MS [16,32] 85.8-101.3%
25% HNO3

5 g +20g NaOH+20 g NaNO3, heated at 600 ℃ for 5h, and

Alkaline fusion-hydrothermal agitated with water at 100-105 ℃ for 6h, the precipitate was
ICP-AES [31] /
treatment mixed with 5% NaOH at 100-105 ℃ for 6h, the final filtered solid

after drying will be dissolved in 4M HNO3

37
622 The closed-vessel digestion method, also known as oxygen-bomb decomposition

623 is widely used in rock dissolution for geochemistry studies. It is based on transferring

624 known amounts of samples and mix acid to a Teflon-lined bomb. Then, the closed

625 vessel is stirred, sealed, and placed in an electric oven at a high temperature (190 °

626 C). After dissolution, it is necessary to drive off the acid (HF) at a certain temperature

627 (e.g., 140 ° C). These procedures can be repeated for achieving a complete

628 dissolution of the solid residual. For REY determination, Querol et al. [170,171]

629 developed a two-step closed-vessel acid-digestion method for trace elements

630 determination in coal and coal fly ash (Table 4). Trace elements are sequentially

631 extracted with HNO3 (volatile elements) and HF-HClO4-HNO3 (non-volatile

632 elements). Taggart et al. [16] proposed a simple closed-vessel method for total REY

633 quantification analysis. Coal ash (34 ± 1 mg) was digested overnight at 90-100 °C

634 with 2-ml each of HNO3 and HF. The acidified sample was then evaporated to dryness

635 and further re-digested overnight at 90-100 °C in a mixture of HNO3-H2O2. The

636 recovery of NIST fly ash 1633c was in the range of 89.2-102.4%. The closed

637 PTFE-lined stainless-steel bombs or Parr bombs used for dissolving rocks at high

638 temperature and high pressure are also used for coal ash digestion [28,171,177].

639 Overall, the recoveries of closed-vessel digestion were satisfied by comparing the test

640 value to the certified values of NIST SRMs (Table 4). It can be used for REY analysis

641 in ash samples, and the amount of acid volume and the operation steps could be

642 reduced compared to the open acid-digestion method. However, the whole time for

643 the digestion is long and, thus, the working efficiency would be affected if many ash

644 samples need to be analyzed. In addition, care must be taken to avoid (1) the loss of

645 volatile REY species in the vapor that might escape when the pressure is released and

646 (2) the loss of elements by adsorption on the walls of the bomb.

38
647 A number of laboratories use microwave digestion systems to dissolve coal ash

648 for REY determination [19,27,119,158,161,169,172,173,178–180]. For microwave

649 digestion, a known amount (~ 0.1g) of coal ashes plus certain volume of mixed acids

650 are digested in a microwave oven. Next, the mixture is heated to remove the HF and

651 re-dissolved with acids. In comparison to open acid digestion and closed vessel

652 digestion, the microwave digestion can greatly reduce the time and acid volume

653 required for the sample dissolution. As shown in Table 4, the longest duration time

654 for microwave digestion reported in the published literature is within two hours and

655 the total volume of mixed acid is generally below 10 ml. Important operation

656 parameters including digestion time, temperature, pressure, acid types, and the ratio

657 of solid-to-acid adopted in the microwave digestion vary significantly among

658 different laboratories. All of these parameters affect the detection of REY by

659 ICP-OES/MS [172,173]. However, detailed pretreatment methods or the data for

660 quality control are not reported in many published studies (see the ‘recovery column’

661 in Table 4). Thus, it is recommended that the accuracy and precision of the analytical

662 procedure should be presented in further research literature. In addition, compared

663 with other pretreatment methods, few international or national standards give

664 certified microwave digestion methods for ash digestion. Considering the great

665 advantages of microwave digestion over other pretreatment methods, future efforts

666 are required to develop standard microwave digestion methods.

667 Finally, alkaline fusion method is also a potential choice for sample preparation

668 for REY determination. It is based on the fusion of ash samples with alkaline

669 materials (lithium meta-/tetraborate, Na2O2, and among others) at high temperature

670 (450 °C or higher). After cooling down, the homogeneous glassy materials are

671 dissolved with HNO3 or HCl. REY in the resulting solutions will be finally analyzed

39
672 by either ICP-OES or ICP-MS. Among the alkaline fusion methods, lithium

673 meta/tetraborate alkaline sintering has been the most popular methods for preparing

674 a liquid sample for REY quantification. Briefly, coal ash is added to lithium

675 metaborate or borate flux, thoroughly mixed, fused in furnace at 1000-1100 °C,

676 and the resulting solids are re-dissolved in diluted HNO3. This method has been

677 written in the ASTM D4503-08 [174] for trace elements determination (REY not

678 included) in solid waste, and has been further used for REY quantification in coal

679 ash in literature [174]. Besides, Na2O2 alkaline sintering was also used in REY

680 determination in ash samples, as proposed and used by several USGS laboratories

681 [174,181,182]. The fusion temperature (450 °C) required in the digestion is

682 significantly lower than that of lithium metaborate or borate fusion method, but acid

683 concentration for the subsequent re-dissolution is higher (25% HNO3) compared to

684 the ASTM alkaline fusion method (5% HNO3). In addition to Li meta/tetraborate

685 alkaline fusion and sodium peroxide sintering, Mondal et al. [31] used alkali fusion

686 (NaOH-NaNO3) method for REY quantification in coal ash.

687 Overall, the due to the encapsulation of REY within aluminonsilicate glass of

688 coal ash (see section 5), the analysis of REY requires reagents that can decompose

689 the glass matrix. In comparison to acid digestion, alkaline fusion is more effective

690 in extracting REY from the matrix. Among the alkaline reagents, Na2O2 and NaOH

691 were the most effective reagents for analysis of REY in coal ashes [167]. Attention

692 should be paid to the high procedural blanks for example the introduction of Ba, Ca,

693 and other interfering elements, since a large volume of flux used in the fusion

694 process could be detrimental to trace analysis of individual rare earth elements. For

695 example, scandium cannot be accurately determined (overestimated) using the

40
696 method described by the U.S. Geological Survey, because the use of Zr crucible can

697 result in Zr interfering with the Sc analysis [16, 32, 167].

698 To further benefit the readers and researchers in measuring REY contents in coal

699 ash, here the DOE-supported round-robin interlaboratory study (RRIS) was served as

700 a typical example to compare the results (also the pros and cons) of different

701 pretreatment methods on REY determination in coal ash. In the round-robin

702 inter-laboratory study [140], thirteen qualified laboratories participated in determining

703 rare earth elements in coal, coal ash, and other related materials for evaluating the

704 accuracy, precision, and intra-/between-laboratory reproducibility of pretreatment

705 methods REY determination in coal ash (Table 4). Eleven materials including coals of

706 different rank, coal fly ashes, coal-bearing rocks, and mine waste as well as four

707 standard reference materials (SRM NIST 1633c, 1632e, 2780a, and USGS SBC-1)

708 were sent to the laboratories and prepared via three procedures:

709 Procedure A: ASTM D6357 method–Mixed acid (i.e., hydrochloric, nitric,

710 and hydrofluoric) digestion with heating followed by ICP-MS/OES analysis;

711 Procedure B: ASTM D4503 – High-temperature lithium borate fusion

712 digestion followed by ICP-MS/OES analysis; and

713 Procedure C: Microwave digestion – Mixed acid (i.e., hydrochloric, nitric,

714 and hydrofluoric) digestion with heating plus the addition of boric acid to

715 neutralize the hydrofluoric acid followed by ICP–MS.

716 The results indicated that Procedure A and Procedure B are effective for total

717 REY extraction from the ash particles (Table 5), showing acceptable recoveries (±

718 15%) for fly ash reference material (NIST 1633c). In terms of the precision, the

719 reproducibility RSD results showed that both open acid digestion (Procedure A) and

720 alkaline fusion (Procedure B) can perform excellently in producing concentrations of

41
721 REY that compared with NAA values with lower RSDs (< 5%); and Procedure A

722 performed slightly better than Procedure B because the introduction of alkali

723 reagents into the solution may result in matrix interferences, which should be

724 reduced by diluting the digestion solution prior to ICP-MS analysis. Procedure C, i.e.,

725 the mixed acid digestion with heat plus the addition of boric acid showed the worst

726 mean recoveries (62-96%) and repeatability (RSD > 5%). It is suggested that both of

727 the acid digestion (without boric acid) and the alkaline fusion methods can be

728 effective in REY determination. In addition to these two procedures, any other

729 pretreatment methods can be used before providing appropriate validation with

730 reference materials.

731 Table 5. The REY recovery rates from NIST 1633 c by using different pretreatment

732 methods.

Procedure A Procedure B Procedure C NAA

Reference
Element Ave R (%) Ave R (%) Ave R (%) Ave R (%)
value
Sc 37.6 ± 0.6 40.7 108% 44.4 118% 32.8 87% 36.5 97%
Y N 89.6 N 102 N 107.8 N N N
La 87.0±2.6 72.5 83% 79.6 91% 71.7 82% 82.6 95%
Ce 180 181 101% 177 98% 172 96% 186 103%
Pr N 19 N 21.9 N 17.9 N N N
Nd 87 81 93% 80 92% 78 90% 88 101%
Sm 19 18 93% 21 109% 15 79% 21 109%
Eu 4.67± 0.07 3.86 83% 4.62 99% 3.2 69% 4.43 95%
Gd NA 17.5 N 21.6 N 15.2 N N N
Tb 3.12 ± 0.06 2.71 87% 3.24 104% 2.26 72% 3.21 103%
18.70 ±
Dy 15.83 85% 19.61 105% 13.26 71% 17.93 96%
0.30
Ho N 2.92 N 3.88 N 2.61 N N N
Er N 8.66 N 10.9 N 7.4 N N N
Tm N 1.26 N 1.48 N 0.9 N N N
Yb 7.7 7.4 97% 9.3 121% 5.5 71% 8.6 111%
Lu 1.32 ± 0.03 1.55 117% 1.42 107% 0.81 62% 1.25 94%
Range 83-117% 91-107% 62-96% 94-111%
Average 95% 104% 78% 100%
733
42
734 Notes. Ave: the average concentration, μg g-1. R: recovery rate, calculated by taking
735 the average mean value reported by the labs for REY and dividing by the reference
736 value reported in the SRM certificates. N: not detected.

737 3.1.2. Inductively coupled plasma atomic emission spectrometry (ICP-OES)

738 ICP-OES, which is also known as inductively coupled plasma atomic emission

739 spectrometry (ICP-AES) has been widely used in analysis of REY in coal and coal

740 ash [15,30,31,122,156,159,163,164,167,173,183–186]. The superior properties of the

741 ICP torch as an excitation source can produce a very high-temperature plasma

742 (7000-10,000 K). Solutions containing REY for analysis are pumped into the

743 nebulizer and sprayed as aerosols into the hot plasma, where it can easily atomize and

744 ionizes the refractory REY. The excited rare earth species naturally emit their unique

745 radiation, which is then detected by the spectrometer. Modern ICP-OES can

746 simultaneously analyze up to 70 elements with polychromatous or array detectors and

747 also enjoys several advantages in comparison to conventional techniques, such as

748 direct current arc spectrometry or alternating current spark spectrometry : (i) the high

749 temperature (above 7000 K) and electron density (1014-1016 cm-3) of the plasma can

750 destroy almost every species in the sample, minimizing the chemical interferences; (ii)

751 low background emission and chemical interferences; (iii) good accuracy and

752 precision; (iv) wide linear response range; and (v) reasonable costs compared to other

753 trace elements analytical techniques, such as INAA and ICP-MS.

754 The accuracy of ICP-OES, like other analytical techniques, decrease

755 significantly when approaching the limits of detection (LOD). To reach an accuracy in

756 the range of 2%, the minimum concentration of the analyte should at least be

757 100-times the LOD. The LOD of the method are sample and matrix dependent, and

758 for lanthanides it is usually in the range of 0.1-100 µg L-1. Usually, the concentration

759 of individual rare earth elements in coal ash is below 100 µg g-1 and a part of REY is
43
760 at sub-µg g-1 level. Thus, quantification of trace amounts of REY in digestion

761 solutions, especially for most HREY (Tb and Ho to Lu) in coal ash is difficult by

762 ICP-OES technique. Readers should be cautioned to the data listed references in

763 Table 4, where most of the ash samples have extremely high total REY contents (>

764 1000 µg g-1).

765 The most important source of instrumental inaccuracy in ICP-OES is the spectral

766 inferences derived from the plasma. The interferences are likely to occur in the

767 analysis of trace REY in coal ash having complex chemical compositions. Especially,

768 the concentrations of Fe, Ca, K, Na, Ti, and Ba in coal ash are several orders of

769 magnitude higher than the REY contents, which may result in the total or partial

770 overlap of analyte (REY) with the interfering species’ lines [187]. Eze. et al. [186]

771 carried out a comparative study of REY determination in coal fly ash using INAA,

772 ICP-OES, XRF, and ICP-MS. Based on analysis of the NIST standard coal fly ash

773 1633b and a South African fly ash, the authors found that REY contents in coal ash

774 determined by ICP-OES are much lower than those by INAA and ICP-MS (Table 6).

775 This is probably caused by the spectra interference or problems occurring in sample

776 preparation process (such as incomplete digestion). The selection of appropriate

777 analytical line and use software for spectra recognition are important to avoid the

778 inferences in REY analysis. In addition, the ion/cation-exchange separation used for

779 matrix elimination in complicated geological samples can be applied to REY

780 determination in coal ash [187–189].

781

782 Table 6. A comparison of REY concentrations in a South African coal fly ash

783 determined by INAA, ICP-OES, LA-ICP-MS, and XRF. Unit: μg g-1. Data from Eze

784 et al. [186].

44
 NAA ICP OES ICP-MS XRF
Elements
Conc. D.L. Conc. Conc. Conc.
Y n.d. / n.d. 52.3 103.71
La 92.23 1.43 0.196 81.66 111.45
Ce 247.33 8.16 20.31 189.78 226.02
Pr n.d. / 9.53 18.35 100.32
Nd 88.4 36.6 n.d. 1.97 n.d.
Sm 18 0.091 3.72 11.93 n.d.
Eu 3.11 0.77 1.24 2.35 n.d.
Gd 27.57 0.72 1.16 10.4 n.d.
Tb 2.26 0.065 n.d. 1.6 n.d.
Dy 35.23 4.85 0.44 9.5 n.d.
Ho n.d. / n.d. 1.97 n.d.
Er n.d. / 4.3 5.38 n.d.
Tm 1.89 0.35 n.d. 0.77 n.d.
Yb 7.64 1.03 n.d. 5.27 n.d.
Lu 1.26 / 0.3 0.72 n.d.
785

786 Note. Conc: the concentration of REY determined by various methods, μg g-1. D.L.:

787 The minimum detection limit, μg g-1. n.d.: not detected. not detected

788

789 3.1.3. Inductively coupled plasma mass spectrometry (ICP-MS)

790 As shown in Figure 8, ICP-MS is one of the most popular techniques in REY

791 determination in coal ash. Just as the ICP source can improve the precision and

792 accuracy of the emission spectroscopy, the same source creates similar improvements

793 in mass spectrometry techniques [197,198]. In comparison to ICP-OES, ICP-MS

794 shows several advantages: (i) lower detection limits (at ppt level); (ii) wider linear

795 response ranges (up to nine orders of magnitude); (iii) much faster in simultaneous

796 determining multi-elements; (iv) simpler spectra that can avoid matrix separation.

797 Overall, ICP-MS is a better choice for REY determination in coal ash than that of

798 ICP-OES, but the high costs would limit its use in some laboratories.

45
799 To obtain accurate and reliable data, several shortcomings of ICP-MS should be

800 considered including the mass spectral interferences and its intolerance to high

801 contents of dissolved solids (matrix effects) in solutions. For REY analysis, the

802 spectral interferences may arise from the formation of oxide and hydroxide ions in the

803 ICP segment. Table 7 summarizes potential spectral inferences species on REY ions

804 in ICP-MS system. As shown in Table 7, the formation of REE oxide ions (REEO+)

805 especially the occurrence of LREEO+ (with higher ionization energy) that maybe

806 generated from the high-temperature plasma can interfere with the determination of

143
807 the HREE, such as NdO+ on 160
Gd+, 144
SmO+ on 160
Gd+, and 143
NdO+ on 159
Tb+.

808 Fortunately, the yield of REO+ to RE+ is only a few thousandths or lower; and

809 therefore, interferences of LREEO+ on HREE can be neglected due to the lower

810 concentrations of individual REY (at ~ µg g-1 level) in coal ash. However, as Si-rich

811 materials, the determination of Sc in coal ash by ICP-MS may be plagued by spectral

812 interferences from the ash matrix constituents such as Si and Ca, Li and B if using

813 lithium metaborate or borate flux as fusion reagents to digest the ash samples [190]. Li

814 et al. [377] reported that the interferences of B and Ca on Sc analysis are negligible

815 when the B and Ca content is below 100 and 200 µg mL-1 in the final solution for

816 ICP-MS analysis. In terms of Si-based interferences, the employment of

817 high-resolution ICP-MS [376], algebraic corrections [377], and collision/reaction cell

818 technique [190] during the determination of Sc in siliceous rocks by ICP-MS can be

819 referred to overcome the polyatomic interferences. In addition, the interferences of

135
820 Ba-bearing polyatomic ions on Eu isotopes, namely the interference of BaO+ on

46
 151
821 Eu+ as well as 137
BaO+ on 153
Eu+ is another problem which must be eliminated if

822 the ash contains high Ba contents [198,199]. Yan et al. [172] used an AG50W-x8

823 cation resin to separate Ba species from the acid digestion solution of coal fly ash

824 samples. Precision and accuracy of their procedure have been demonstrated by repeat

825 analyses of National Institute of Standards Technology (NIST) standard references of

826 coal and fly ash samples. The authors also proposed a ratio of Ba/Eu (1000) as basis

827 for evaluation of the interference of Ba on Eu. If Ba/Eu > 1000, then Ba in coal fly ash

828 can greatly interfere with Eu determination, and measures should be taken to diminish

829 the interferences [172].

830

831 Table 7. The potential spectral inferences in REY determination by ICP-MS. After

832 Ardini et al. [183] and Léveillé [190]

Analyte Interfering Analyte Interfering species Analy Interfering species

species te

73
89
Y+ Ge16O+ 154
Gd+ 154
Sm+, 138Ba16O+ 164
Er+ 164
Dy+, 148
Nd16O+,

148
Sm16O+

123
139
La+ Sb16O+ 155
Gd+ 139
La16O+, 138Ba16O1H+ 166
Er+ 150
Nd16O+, 150Sm16O+

124
140
Ce+ Sn16O+, 156
Gd+ 140
Ce16O+ 167
Er+ 151
Eu16O+

124
Te16O+

142
142
Ce+ Nd+, 126Te16O+ 157
Gd+ 141
Pr16O+ 168
Er+ 152
Sm16O+

125
141
Pr+ Te16O+ 158
Gd+ 142
Ce16O+, 142Nd16O+ 170
Er+ 170
Yb+, 154Gd16O+, 154Sm16O

142
142
Nd+ Ce+, 126Te16O+ 160
Gd+ 160
Dy+,144Nd16O+,144Sm 169
Tm 153
Eu16O+

47
 16
O+ +

144
144
Nd+ Sm+ 159
Tb+ 143
Nd16O+ 170
Yb+ 170
Er+,154Gd16O+, 154Sm16O

130
146
Nd+ Te16O+ 160
Dy+ 160
Gd+,144Nd16O+,144Sm 171
Yb+ 155
Gd16O+

16
O+

148
148
Nd+ Sm+ 161
Dy+ 145
Nd16O+ 172
Yb+ 156
Gd16O+

150
150
Nd+ Sm+, 162
Dy+ 146
Nd16O+ 173
Yb+ 157
Gd16O+

134
Ba16O+

144
144
Sm+ Nd+ 163
Dy+ 147
Sm16O+ 174
Yb+ 158
Gd16O+

133
149
Sm+ Cs16O+ 164
Dy+ 164
Er+, 148
Nd16O+, 176
Yb+ 176
Lu+,176Hf+,160Gd16O+,160D

148
Sm16O+ y16O+

150
150
Sm+ Nd+, 134Ba16O+ 165
Ho+ 149
Sm16O+ 175
Lu+ 159
Tb16O+

136
152
Sm+ Ba16O+ 151
Eu+ 135
Ba16O+ 176
Lu+ 176
Yb+,176Hf+,160Gd16O+,160D

y16O+

154
154
Sm+ Gd+ 153
Eu+ 137
Ba16O+ 45
Sc+ 7
Li38Ar+,10B35Cl+,

28
Si16O1H+,29Si16O+,

44
Ca1H+,13C16O16O+,

12
C16O16O1H+

833

834 The use of sector-field inductively coupled plasma mass spectrometry

835 (SF-ICP-MS) in high resolution mode maybe an effective way to reduce the

836 polyatomic interferences occurring in quadruple inductively coupled plasma mass

837 spectrometry (Q-ICP-MS). Thompson et al. [200] carried out a comparative analysis

48
838 of REE in a series of coal-based wastes by Q-ICP-MS and SF-ICP-MS. As revealed,

839 Q-ICP-MS can handle most of geological materials, including coal ash for the

840 determination of REY; whilst, for ash samples containing large amounts of Ba (Ba/Eu >

841 1000), SF-ICP-MS may be suitable to accurately determine all REY (especially Eu).

842 Other methods that have been used to overcome the spectral interferences for

843 environmental samples and geological matrices, including: (1) measure bivalent ions

844 (Eu2+) [201], use of algebraic corrections [202–204], optimize instrumental

845 parameters (ICP power, carrier gas flow rate, among others) [205–208], and apply

846 de-solvation techniques [204,209]. These methods could be potential solutions to the

847 accurate analysis of REY in coal fly ash. In addition, both matrix effects and spectral

848 interferences for ICP-MS greatly depend on the sample introduction techniques

849 employed. The pneumatic nebulization for sample introduction brings larger amount

850 of liquids (H2O), which is a main source for generating REEO+ in the ICP source.

851 Ultrasonic nebulization and electro-thermal vaporization (ETV) are alternatives to the

852 solution sample introduction system for ICP-MS. Zhang et al. [190] developed an

853 effective electro-thermal vaporization-ICP-MS (ETV-ICP-MS) technique for REE

854 quantification in coal fly ash. With polytetra-fluoroethylene (PTFE) as a chemical

855 modifier to promote REY vaporization, solid coal fly ash sample (< 325 mesh) was

856 directly introduced into the ETV system, vaporized in a graphite furnace, and was

857 carried into ICP torch for ionization. Their results revealed that the spectral

858 interferences of matrix elements including Al, Fe, Si, Na, K, Ca, Mg, and Ba on REE

859 were reduced with sample introduction by ETV.

49
860 3.1.4. Instrumental neutron activation analysis (INAA)

861 Instrumental Neutron activation analysis (INAA) is a robust technique to detect

862 trace elements (including rare earth elements) in coal and coal ash. The working

863 principal of INAA was built upon the nuclear reactions, where the half-life of

864 radioactive isotopes and the γ-ray energy of each elements are unique and can be used

865 as “fingerprint” for elemental identification and quantification [210,211]. It generally

866 involves: (1) activation process, involving irradiation of the stable isotopes of one

867 element in the sample by thermal neutrons, (2) decay process, involving emission of

868 different types of ionizing radiations during the decay of the radioactive isotopes, and

869 (3) detection of the radiation, such as the β-radiation, γ-radiation, and sometimes

870 determine the decay rate of the radioactive nuclei. Accordingly, the overall test time

871 for INAA analysis requires several weeks, much longer than other instruments [198].

872 This is one of main drawbacks of INAA analysis. Besides, the use of INAA have

873 several other disadvantages: (1) it suffers from time-consuming procedures,

874 interferences of major elements in the samples, and high costs of operation [198]; and

875 (2) it cannot analyze the full suite of REY, because the low γ-ray energies for

876 detecting Er and Tm, low sensitivity to neutrons activated from Ho and Y, and no

877 stable Pr isotopes.

878 The major advantages of INAA are also obvious: (1) it is a non-destructive

879 analysis technique, material remains intact; (2) sample preparation is very simple, i.e.,

880 drying and weighing mg of test samples without chemical treatment (e.g., dissolution);

881 (3) it allows simultaneous analysis of ~40 radionuclides; (4) a very sensitive

50
882 technique can determine elements at µg g-1 and sub-µg g-1 concentrations. Therefore,

883 INAA has been used for the development and certification of reference materials at

884 national organizations, such as the National Institute of Standards and Technology

885 (NIST).

886 INAA has been widely used to detect trace elements (including rare earth

887 elements) in coal-related materials [150]. For REY determination in coal ash, Tu et al.

888 [212] determined eight rare earth elements (La, Ce, Nd, Eu, Tb, Dy, Yb, and Sc) in a

889 Chinese coal fly ash by the means of INAA using NIST SRMs 1632a as

890 multi-element standards. In their study [212], the procedure in the INAA analysis

891 involved both short and long irradiations with thermal neutron flux (24 h), the decay

892 of the radionuclides (3 weeks), and the measurement of the gamma-ray by a Ge (Le)

169 182
893 detector (4 h). Besides, the γ-ray lines of Yb overlap with Ta at 179.5 keV and

160
894 Tb with 131Ba at 215.8 keV. Thus, INAA cannot be used in routine analysis for most

895 laboratories with such a long measurement period accompanying high costs and long

896 time. Nevertheless, due to its high sensitivity and ability of multi-elemental and

897 non-destructive analysis, INAA has been used as one of the principal reference

898 methods to determine REY in coal ash [32,186,213]. For example, in the

899 DOE-supported REE characterization projects [140], to evaluate the precision and

900 accuracy of the ICP-based REE determination methods, one laboratory applied INAA

901 as the reference method. As illustrated in Table 6, INAA had the best performance in

902 determining REE in coal ash with average REE recovery approaching 100%. Wagner

51
903 et al. [32] gained comparable values of six REEs (Sc, La, Ce, Eu, Sm, and Tb) in five

904 South African coal fly ash samples analyzed by INAA to that of ICP-MS.

905 3.1.5. X-ray fluorescence (XRF) analysis

906 In addition to INAA, XRF is another nondestructive method for determining

907 REY in the solid ash. It is usually used for analysis of major elements in geological

908 and environmental materials [214]. Some researchers, for examples, Kuhn et al. [215]

909 and Prather et al.[216] tried to extend this technique to trace elements analysis in coal.

910 The main advantages of XRF are: (1) in situ analysis of elements without sample

911 preparation, such ashing, milling, pelletization, or digestion; and thus, the overall

912 measurement time is shorter and less expensive relative to other instrumental

913 techniques; (2) simultaneous analysis of both major and minor components; (3)

914 covering wide ranges from < 1 µg g-1 to 100%; and (4) it is a common laboratory

915 instrument and therefore readily accessible. Therefore, XRF plays an important role in

916 geochemical survey and environmental monitoring because of simple sample

917 preparation process and quick measurements of multi-element within a few minutes

918 [217,218]. Depending upon the detector types, there are general two XRF techniques,

919 i.e., energy-dispersive X-ray fluorescence spectrometry (EDXRF) and

920 wavelength-dispersive X-ray fluorescence spectrometry (WDXRF). For quantitative

921 analysis, WDXRF shows better performance in analysis of trace REY in solid

922 materials, including coal ash [219–221]. However, several problems occur in REY

923 determination by XRF:

924 Sample matrix effects (Absorption/enhancements)

52
925 Spectral interferences

926 Low sensitivity

927 Sample homogeneity and size effects

928 Therefore, the use of XRF techniques for quantifying REY in coal ash is not as

929 popular as other three instrumental methods (Figure 8). For REY analysis in coal ash,

930 the most serious problems of WDXRF are probably matrix effects and overlapping of

931 the analytical lines. For example, the signals of HREE in coal ash may be weakened

932 (absorption effects) because of the absorption of X-rays by light rare earth elements or

933 other major elements in coal ash. In a study by Eze et al. [186], only Sc, La, and Ce

934 were detected by XRF due to no suitable standards for calibration. Table 8 listed the

935 possible interfering lines in the XRF analysis of REY in coal ash. Smoliński et al.[219]

936 developed a method for the quantification analysis of REE based on WDXRF. The

937 matrix effects and spectra interferences during REY determination are corrected by

938 calibration and peak de-convolution of the overlapping spectral lines. Based on the

939 analysis of 16 REY in 12 different coal combustion ash and eight certified reference

940 materials, the relative standard deviations (RSD) and the recoveries of most REY in

941 the standard ash samples are estimated to be 2.1-12.9% and < 10%, respectively.

942 However, the recovery rates of Tb (21.8%), Dy (21.8%), and Er (20.1%) in the

943 standard ash are not satisfied. This is mainly caused by the lower concentrations of

944 HREE in coal ash and thus the Lα lines (2θ angles of Tb and Dy: 56-58°) are easily

945 interfered with Fe Kα (2θ angles: 57.5°) and Fe Kβ lines (2θ angles: 51.7° vs 52.6° of

53
946 Lα line of Er). In general, the study from Smoliński et al. [219] demonstrated that the

947 WDXRF can be used to estimate most REY in coal combustion ash.

948 Portable XRF is generally used as a screening tool in field geological and

949 environmental investigations, for rapidly determining the elemental compositions of

950 ores, rocks, sediments, etc., [222–227]. To quickly identify REY-rich coal or coal ash

951 in the U.S. coalfield and power plants, Bryan et al. [218] studied the use of a portable

952 hand-held XRF to estimate the concentrations of REY. Generally, the results of LREE

953 content obtained from the portable XRF units could be correlated to the laboratory

954 quantitative analysis, with a correlation coefficient of 0.878. Thus, it can be used to

955 screen likely samples for submission to the laboratory. The accuracy, however, is not

956 quantitatively reliable. For example, the detected Y contents in the NIST 1633c via

957 the portable XRF were 5-7 times higher than the certified values.
958

959 Table 8. Lines for REY determination, potential spectral interferences, and analysis

960 uncertainty of REY by XRF technique. Revised after Smoliński et al. [219].

Lines Analytical line 2-theta Background Inferences

Sc Kα 97.74 96.10 /

Y Kα 23.78 24.42 Kβ1Rb

La Lα 82.88 84.36 /

Ce Lα 78.98 79.82 Lβ1Ba; Kβ1Ti

Pr Lα 75.4 75.78 Lβ1La

Nd Lα 72.1 72.56 Lβ1Ce

54
 Sm Lα 66.2 67.08 Lβ2Ce

Eu Lα 63.54 65.72 KαMn

Gd Lα 61.08 61.76 Lβ6Pb

Tb Lα 58.77 60.52 KαFe

Dy Lα 56.57 55.38 Kβ1Mn, KαFe

Ho Lα 54.52 55.44 KαCo

Er Lα 52.58 54.81 Kβ1Fe; KαCo

Tm Lα 50.77 50.3 Kβ1Fe

Yb Lα 49.04 49.7 KαNi

Lu Lα 47.41 46.68 KαNi; Kβ1Co

961

962 Overall, a general conclusion can be reached is that the main analytical

963 techniques have been used for REY determination in last twenty years are ICP-MS,

964 ICP-OES, INAA, and XRF (Figure 8). A comparison of these analytical methods in

965 REY determination in coal ash is summarized in Table 9. Coal ash usually contained

966 low concentration of individual REY (sub-μg g-1 to μg g-1 range) and the major

967 chemical components and inorganic salts could bring matrix effects, spectral

968 interferences, and other instrumental problems for the four analytical techniques.

969 ICP-MS is recommended for determining REY in coal ash because of its high

970 sensitivity, multi-elemental capability, and reasonable costs. The measurement of REY,

971 especially low content HREE by ICP-MS is critical because ICP-MS can directly

972 determine the digestion solutions without separation and pre-concentration procedure.

55
973 As mentioned in 3.1.3, in terms of HREE determination, several methods can

974 overcome the spectral interferences during the ICP-MS analysis of HREY in coal ash

975 digestion solutions: i) monitoring the isobaric REE interferences from oxides of

976 lighter REE (La, Ce, Pr and Nd)) or other elements (BaO and BaOH) on the HREE,

977 for example using the mass ratio of CeO/Ce or ThO/Th, and the value should be kept

978 < 2%, ii) the use of collision cell by introducing pure gas (such as helium) to

979 eliminate the polyatomic and doubly-charged interferences, iii) algebraic correction

980 that can be used to reduce overlapped spectra lines, for example the interference of

981 PrO and NdO on Gd and Tb. In contrast, pre-concentration and matrix separation

982 might be necessary prior to ICP-OES analysis (especially for HREE), for the coal ash

983 containing REY at normal concentration level (contents of total REY below 300 μg

984 g-1). Further work should develop workable pretreatment methods to overcome these

985 problems. INAA could be a reference method for assessing the precision and accuracy

986 in determination of REY by other techniques. XRF, as a rapid and nondestructive

987 method for analysis of solid sample, is suitable used for exploring REE-rich coal ash

988 in field survey.

56
989

990 Table 9. A comparison of the four main instruments (ICP-MS, ICP-OES, INAA, and XRF) for REY determination in the published literature

991 [150,151,186,202,204,209,218,228].

Major interfering Applicability

Sample pretreatment/sample type Detection limit µg g-1 Costs Analysis time for one sample
problems

Most popular method and

Relative
ICP MS Digestion/solution isobaric superpositions 10-6 - 10-4 2-4 min for all element can be used at almost any
expensive
types of ash samples
Second to ICP-MS but
ICP OES Digestion/solution spectral 10 -3 ~ 10 Fair 2 min for 10-15 element have limitations for
low-REE ash samples
60 min or more for all elements; Reference method but
INAA No/solid spectral 10-4 ~1 Expensive the whole period needs several cannot determine all REY
weeks in ash
Screening method for fast
determining the
XRF No/solid matrix 10-1 ~ 100 Fair 10 -120 min for all elements
concentration level of REY
in ash

57
 992 3.2. Speciation analysis

993 3.2.1. Indirect methods

994 Speciation of REY in coal ash assists in understanding their partitioning behavior

995 in coal combustion process and inform REY recoveries strategies [26,229–232].

996 Methods that are used to measure REY occurrences in coal ash can be classified into

997 indirect methods and direct methods (Figure 7). Indirect methods refer to applying

998 the method of statistical calculations, physical separations, and chemical leaching to

999 gaining the affinities of REY to special components of coal ash.

1000 Statistical calculation can provide preliminary but limited REY speciation

1001 information [18,20,22,121,170,175,180]. Physical separation such as float-sink and

1002 magnetic separations are methods for REY separation from bulk fly ash

1003 [12,23,33,108,121,170,175,176,179,185]. These indirect methods are more used for

1004 beneficiation and pre-concentration of REY from fly ash for the further extraction,

1005 having few implications for REY occurrences. The detailed descriptions can be

1006 referred to section 7. Here, we will focus on making a review on the chemical

1007 leaching methods, since it is an effective process to quantify REY speciation and have

1008 implications for REY extraction from coal ash.

1009 Sequential chemical extraction (SCE) has been extensively used for determining

1010 trace elements associations in coal fly ashes. As shown in Table 10, numerous

1011 sequential leaching protocols were designed to study the REY speciation in coal ash

1012 [16,20,27,108,121,145,159,163,171,176,196,233,234]. In these leaching procedurals,

1013 REY associations can generally be classified into three types:

58
1014 Easily-leached fraction: including the water-soluble, ion-exchangeable, and weak

1015 acid-soluble forms, which can be easily released with mild regents such as

1016 deionized water, ammonium acetate, magnesium chloride, dilute HCl, acetic acid,

1017 and sodium acetate, etc.;

1018 Metal-oxide bound fractions and organic/sulfide forms: including Fe/Mn-oxides,

1019 crystalline Fe/Al-oxides forms, amorphous Fe oxides-bound, crystalline Fe

1020 oxides-bound, and organic/sulfide-bound; dissolution of these REE-bearing forms

1021 requires relatively strong acid or chelating agents (e.g., oxalic acid); and

1022 Insoluble forms or residual forms: strong acid/alkaline digestion methods as

1023 indicated in 3.1.1 are necessary for REY leaching.

1024 The major problems of SCE process may arise from the assumption that each

1025 chemical reagent reacts with a special REY-bearing phase. However, this might be

1026 not true in many cases because REY can be leached from multiple phases [145]; and

1027 secondary reactions (e.g. re-adsorption, precipitation, and re-dissolution) may occur in

1028 the leaching steps. Therefore, it is suggested that each step of the SCE process should

1029 be theoretically and/or experimentally verified. As shown in Table 10, Liu et al. [145]

1030 proposed a three-step sequential acid leaching procedure for discerning REY

1031 associations in coal ash. The procedure was validated by the acid leaching

1032 experiments with REY reference compounds and the thermodynamic prediction of

1033 solubility of REE pure compounds at varying pH by geochemical software

1034 PHREEQC. This procedure may be suitable to quantify REY occurrences in fly ashes.

1035 Note that the predicted dissolution curve of some REE-bearing phases by PHREEQC,

59
1036 for example the complete dissolution of monazite at pH = 1.0 seems to be unrealistic

1037 [229]. As indicated by Liu et al. [145], the ash compositions, particle size, and

1038 physical distribution of REY-bearing minerals can affect the REY dissolution kinetics.

1039 Overall, a reliable and convincing quantification of REY species in fly ash by

1040 sequential leaching methods requires cross-validation using direct techniques.

60
1041 Table 10. A summary of various sequential chemical extraction methods used for determining REY speciation in coal fly ash samples.

Leaching procedural Brief description Main conclusions References

4-step sequential ion-exchangeable, acid soluble, metal oxides, organic or sulfide, 75-77% REY bound to silicate and aluminosilicate, with minor [108,121]

extraction and silicate/aluminosilicate form fraction being acid soluble and organic/sulfide form

7-step sequential water soluble, exchangeable, carbonate-bound, Mn 86.1% of REE were associated with the residual phase [176]

extraction oxides-bound, amorphous Fe oxides-bound, crystalline Fe

oxides-bound, organic/sulfide-bound, and residual form

HCl leaching leached by cold dilute HCl (HCl: H2O =2:1), HCl soluble and about 50–55% solubility of Y for the fluidized coal ash, higher than [20]

residual form pulverized-coal combustion ash

4-step sequential leaching water leachable, ion-exchangeable form, acid soluble form, low mobility (< 30% extractable fraction) in fly ash relative to the feed [171]

metal oxides form, organically or sulfide form, and residual coal, REY maybe incorporated in high-temperature aluminosilicate

phases

sequential leaching REE oxides and REE carbonates dissolved at pH 5.5−4; apatite Class F fly ash: 30-70% of REE are associated with the zircon and [145]

61
 dissolved at pH 3.5−2; REE phosphates and hematite dissolved glass phase, 10-40% in apatite, 10% with oxides and carbonates, 10%

at pH < 1.5, and zircon and glass phase with phosphates and hematite; Class C fly ash: 50−60% of REEs are

REE oxides and carbonates, 20-30% in apatite and 20% as phosphates

and hematite

HF leaching 4% HF solution was used to tentatively separate the glassy phase <10% of REY retained in the undissolved crystalline phases, with [27]

and crystalline phase more LREY than HREY; and over 90% of REY are associated with the

amorphous glass in the fly ash

magnetic using magnetic separation - 30% NaOH - 15% HCl to separate REE are enriched in glassy phases and depleted in crystalline minerals [119,234]

separation-alkaline-acid the fly ash into magnetic-rich phase, mullite-corundum, and

leaching glass phases

revised BCR method water soluble (F1), exchangeable/acid soluble (F2), reducible The sum of the F1, F2, F3, and F4 in PRB ash (> 60%) is much higher [233]

(F3), and oxidizable (F4) than in App ash and IL ash (< 15%)

62
 4-step sequential ion exchangeable, carbonates, metal oxides, acid soluble, and the majority of REEs in the calcined samples (600-750℃) occurred as [159,163]

extraction insoluble/silicates metal oxides, HREEs were more likely associated with easily

dissolvable forms than LREEs

1042

63
1043 3.2.2. Direct microanalysis methods

1044 In recent years, to develop effective REY recovery techniques, there are a

1045 number of studies working on in-situ identifying and quantifying REY speciation in

1046 coal fly ash by advanced instrumental analysis [122,144–147,192,230–233,235–238].

1047 As summarized in Figure 7, these direct methods are generally divided into:

1048 Instrumental microanalysis techniques: laser-ablation ICP-MS (LA-ICP-MS),

1049 secondary ion mass spectrometry (SIMS; also called ion microprobe analysis);

1050 Electron microscopy techniques: scanning electron microscopy and

1051 energy-dispersive X-ray spectroscopy (SEM-EDS), electron probe micro analysis

1052 (EPMA), focused ion beam and scanning electron microscopy (FIB-SEM),

1053 transmission electron microscopy (TEM) and related techniques, including

1054 selected area electron diffraction (SAED), fast Fourier transform (FFT), and

1055 electron energy loss spectroscopy (EELS); and

1056 Synchrotron-based techniques: X-ray absorption near edge structure (XANES),

1057 extended X-ray absorption fine structure (EXAFS), microprobe X-ray

1058 fluorescence spectroscopy (µXRF), micro-XANES (µXANES), and

1059 micro-EXAFS (µEXAFS)

1060 SEM-EDS can be used to determine REY occurrences in fly ash powder or

1061 polished pellet. REY-bearing phases usually appear to be very bright in the ash pellets

1062 at backscattered mode, since the brightness of the backscattered imaging depended on

1063 the average atomic number of the material. As a result, SEM-EDS is capable of finding

1064 discrete REY minerals that are dispersed in coal ash [122,144,147,192,236]. EPMA is

64
1065 similar in scope to the SEM methods, and one of the main differences between the two

1066 methods is that element concentrations can be accurately quantified by the wavelength

1067 dispersive spectrometry (WDS). EPMA has significantly lower detection limits

1068 (~0.01wt.%) for REY relative to EDS (~0.1wt.%) and is thus more capable of

1069 determining the dispersed occurrences of REY in trace phases [146,147,192]. However,

1070 the bulk analysis indicated that the concentration of individual REY in coal ash is

1071 generally lower less than 100 μg g-1, especially for some low-content HREE being far

1072 beyond the detectability of EPMA. In addition, REY mineral phases are very

1073 complicated, one single particles (several microns in diameter) may contain nanoscale

1074 multiple REE-bearing phases [144,233]. Detailed and accurate information on trace

1075 REY phase in coal ash can be obtained with TEM and other micro-scale mass

1076 spectrometric techniques.

1077 TEM and related techniques to determine REY occurrences in fly ash have been

1078 extensively used by Hower et al. [122,144,192,231,233,236]. Electron microscopic

1079 examination of coal ash by SEM-EDS and EPMA lay foundation for the use of these

1080 advanced techniques. After the SEM-EDS survey of a small quantity of coal ash,

1081 REE-rich particles are identified and localized, then the individual minerals or mineral

1082 assemblages of interest can be isolated and extracted through in situ Focused Ion

1083 Beam (FIB) technology. Following FIB extraction, the lift-out samples are mounted

1084 on a Cu grid and ion milled to 100 nm or less for the TEM analysis. Selected area

1085 electron diffraction (SAED) is employed for mineralogy identification of the selected

1086 grains; and fast Fourier transform (FFT), computed by image analysis software, is

65
1087 used to determine the lattice spacing in HRTEM micrographs. Chemical identification

1088 in TEM is available in some units via electron energy loss spectroscopy (EELS) and

1089 EDS. A detailed procedural is presented in Figure 9.

1090

1091 Figure 9. A general diagram showing how the transmission electron microscopy

1092 (TEM) and related techniques work on identification of REE-bearing grains in coal

1093 and coal fly ash. After Hower et al. [231,236]. Reproduced with permission from

1094 Elsevier.

1095

1096 The microanalysis based on mass spectrometric techniques have been widely used

1097 in trace element geochemistry study of inorganic minerals and rocks [239–243]. A

1098 major difference in these microanalysis techniques is the rate and depth of penetration

1099 within the sample. In LA-ICP-MS, solid samples are ablated with a laser source to tens

1100 of micrometers in a matter of seconds. Accordingly, LA-ICP-MS is useful in detecting

1101 compositional variation with sample depth. Application of LA-ICP-MS to coal ash

1102 aims principally at quantifying the spatial distribution and association of elements in
66
1103 coal fly ash. Spears [244] Bauer et al. [245], and Kostova et al. [246] used

1104 LA-ICP-MS [244] to determine the major element contents in epoxy-mounted coal fly

1105 ash samples. Piispanen et al.[246] compared the LA-ICP-MS results of major

1106 elements in fly ash samples to the results by SEM-EDS, finding a consistent result

1107 between the two methods. Trace elements of environmental concern, such as As, Se,

1108 and Pb, have been successfully determined in fly ash pellets by LA-ICP-MS, as

1109 reported by Spears [244]. For REY determination, Hood et al. [192] applied

1110 LA-ICP-MS to determine the concentration of Sc, Y, La, Ce, Pr, Nd, and Sm in a set

1111 of fly ash samples derived from Kentucky Fire Clay coal using NIST SRMs 610, 612,

1112 and 614 as calibration standards. The matrix and the homogeneity of NIST reference

1113 materials differ from that of fly ash so that the results are semi-quantitative.

1114 Thompson et al. [237,247] explored the use of varying laser spot size (32-60 μm) to

1115 determine the REY concentration in coal fly ash. Instead of selecting a specific

1116 element as an internal standard, the quantification of REY in randomly selected laser

1117 spots were achieved by normalizing all elements to 100% against NIST SRM 610

1118 glass reference. The overall average REY concentration of the randomly ablated spots

1119 in NIST SRM 1633b is comparable to the certified values.

1120 Recent studies have tried to use SIMS to discern REY and other rare metal

1121 occurrences in coal ash. For example, like REY, lithium (Li) is also a critical metal

1122 which might be recovered from some coal ash. To gain insights on the speciation of Li

1123 in coal ash, Hu et al. [248] used SIMS to analyze the distribution of Li in high-Al coal

1124 fly ash, using Bi1+ or Bi3++ as the primary ion beam with a spot size of 0.2 μm. In SIMS,

67
1125 a primary beam of incident ions sputters secondary ions from the sample at a much

1126 slower rate, typically leading to penetration depths of only about 1-5 µm. Compared to

1127 smaller ion microprobes, the SHRIMP-RG (Sensitive High Resolution Ion Microprobe

1128 – Reverse Geometry) ion microprobe and other SHRIMP configurations employ a large

1129 format magnetic sector mass spectrometer with high mass resolution, thereby

1130 minimizing the effect of isobaric interferences in SIMS analysis. Kolker et al. [146]

1131 analyzed REE distribution and occurrences within ash particles using Stanford-USGS

1132 SHRIMP-RG ion microprobe. An oxygen negative-ion beam with spot size of 15 μm

1133 was used to quantify eleven REE in 19 US and international coal fly ashes. NIST

1134 glass standards (SRM 611 and 613) were used as internal standard for calibration and

1135 the overall analytical uncertainty for REE was reported to be within 10–20%.

1136 Overall, microanalysis techniques can provide important information on REY

1137 species and interparticle distribution differences in complex ash matrices. SEM-EDX

1138 and EPMA are the most common techniques used for geochemistry study in many

1139 laboratories but have limited detection limits and lower resolution (compared to TEM)

1140 for analyzing trace metals in coal ash. For REY determination, these two methods can

1141 be used as preliminary steps for other advanced microscale techniques, such as TEM,

1142 LA-ICP-MS, and SIMS. In both LA-ICP-MS and SIMS, they have appropriate

1143 detection limits at the µg g-1 to sub-µg g-1 level, which is important to characterize

1144 grain-scale REY distribution in coal ash. However, a major problem of these two

1145 methods is obtaining a calibration standard which should be: (1) homogenous on a

1146 grain scale, (2) a good matrix match to the material being analyzed, and (3) avoid

68
1147 spectral interferences of other isotopes [239,244,249]. In addition, the working spot

1148 size of LA-ICP-MS or SIMS might be too large to distinguish REY associations in the

1149 fly ash particles. These large-scale techniques gloss over the actual mineral and

1150 (potentially) non-mineral associations at nanoscale. TEM and related techniques, such

1151 as SAED can provide more details of REY occurrences in the nanoscale examination,

1152 which is an important supplement to the association of REY in coal fly ash.

1153 3.2.3 Synchrotron-based techniques

1154 Synchrotron X-ray spectroscopy-based analysis is another powerful technique to

1155 discern the speciation of REY in coal ash. For those elements of relatively high

1156 abundance in fly ash samples (such as La, Ce, Nd, and Y), X-ray absorption

1157 spectroscopy of a fly ash sample can be used to assess bulk-level element speciation.

1158 For example, bulk XANES has been used to identify major phases of Ce and Y,

1159 primarily through linear combination fitting (LCF) models of the sample spectrum with

1160 spectra of reference compounds [145,232,233]. The analyses generate relative

1161 percentages of the element as different reference species. This approach generally

1162 requires careful selection of known reference materials that could be reasonable

1163 approximations of the elemental forms in fly ash. Thus, the selection of reference

1164 materials must consider the conditions of fly ash formation and collection. The

1165 selection process can be especially challenging if the individual elements are

1166 distributed diffusely within aluminosilicate glasses. In this case, the localized

1167 coordination state of REY is not readily represented by a commercially available

1168 material (see section 6.4), and is not easy to mimic in laboratory-synthesized materials.

69
1169 The utility of XANES analyses is often limited by the relatively high

1170 concentration requirements of the method, particularly as coal fly ash is also enriched

1171 in other elements that might be interfering (Table 11). Quantitative identification of

1172 REY species also requires sufficient differences in the spectral features of candidate

1173 reference compounds, which are difficult to achieve. In a previous application of

1174 yttrium K-edge XANES for coal fly ash, the spectral features of an yttrium carbonate

1175 mineral (Y2(CO3)3·3H2O) and monazite standard were remarkably similar to Y-doped

1176 glass and monazite reference materials (Figure 10). Thus, in the fitting of XANES

1177 spectra for coal fly ash samples, the replacement of one of these references with the

1178 other did not substantially alter the quality of the model fits. These subtleties emphasize

1179 the need for careful and cautious interpretation of XANES results. In other words, often

1180 the results are limited to qualitative descriptions rather than quantitative interpretations.

1181 In addition to XANES technique, the use of EXAFS spectroscopy can determine the

1182 local bonding environment of the absorber atom (~5 Å) and provide structural

1183 information such as coordination number and bond distance that can differentiate

1184 compounds (e.g., between Y-O and Y-P structures) and complement results of XANES

1185 analysis. However, due to the low REE concentrations and line inferences, there have

1186 been no study for REE in coal fly ash using EXAFS. Therefore, more work with

1187 EXAFS or other similar bulk elemental speciation methods is required to improve the

1188 determination REE speciation in coal ash.

1189

70
1190 Table 11. Rare earth elements in coal fly ash that could be analyzed for bulk

1191 speciation by X-ray absorption spectroscopy and potential interfering elements to

1192 consider in coal fly ash. Data from Taggart et al. [233].

Potential
Concentration in fly Absorption Emission
Element interfering
ash energy energy
elements

5732 eV 4840.2 eV
-1
Cerium (Ce) 100 – 200 µg g V, Ba
(L3-edge) (Lα1)

5483 eV 4651.0 eV
-1
Lanthanum (La) 70 – 110 µg g Ti
(L3-edge) (Lα1)

6208 eV 5230.4 eV
-1
Neodymium (Nd) 60 – 90 µg g Possibly Ce
(L3-edge) (Lα1)

17038 eV 14958 eV
-1
Yttrium (Y) 80 – 100 µg g None likely
(K-edge) (Lα1)

71
1193

1194 Figure 10. Bulk Y XANES spectra for yttrium reference compounds. Note t

1195 hat the spectral features for the Y-doped glass, monazite, and Y2(CO3)3∙3H2O wer

1196 e similar to each other. After Taggart et al. [233]. Reproduced with permission fr

1197 om the Royal Society of Chemistry. Further permissions related to the material ex

1198 cerpted should be directed to the Royal Society of Chemistry at https://pubs.rsc.or

1199 g/en/content/articlelanding/2018/EM/C8EM00264A.

1200 A summary and comparison on the advantages and limitations of various methods

1201 for determining REY speciation is given in Table 12. As shown in Table 12, the

1202 identification and quantification of REE in coal ash is still challenging due to low REY

1203 contents and complex ash matrices. One should keep in mind that due to the limitations
72
1204 of each method; no single scale/ technique can provide all of the answers. Depending

1205 on the ash properties (particle size, crystallinity, REE contents, and among others),

1206 readers should be cautious to the speciation results derived from different analytical

1207 techniques. Comprehensive results from both indirect and direct methods are

1208 recommended for studies on the speciation of REY in coal ash.

73
1209 Table 12. A comparison of direct and indirect methods on REY speciation analysis.

Methods Scale Advantages and information provided Detection limits

SEM-EDS Submicron Chemical analysis of areas of interest; Particle ~ 0.1wt.% Low spatial re

morphology and distribution limitation REE

EPMA Submicron Chemical analysis of areas of interest; lower ~ 0.01wt.% Low spatia

detection limit than EDS

SEM-FIB Nanometer Precision chemical analysis; milling allows /

measurements in three dimensions

TEM Nanometer Precision chemical analysis; milling allows /

measurements in three dimensions; SAED/FFT

SIMS Submicron Low detection limit at μg g-1 or sub-μg g-1 level; sub-μg g-1 Working sp

distribution of REE within ash at trace level

LA-ICP-MS Submicron Low detection limit at μg g-1 or sub-μg g-1 level; sub-μg g-1 Working sp

74
 distribution of REY within ash at trace level

μXRF and Micro Identification of REE-bearing phases; ~ 0.01wt.%

μXANES Distribution of REY at trace level

Bulk XANES Bulk Oxidation states of REY; Quantification of ~ 0.01wt.% High detectio

REY-bearing phases by linear combination on REE dis

fitting problem

Sequential Bulk Quantification of REE associations; low cost Depending on the analytical instruments Quantifica

extraction and REE extractability empirical int

Physical Bulk Preliminary information of REE affinity Depending on the analytical instruments

separation

1210

75
1211 4. Concentrations and valuation of REY in coal ash

1212 4.1. REY concentrations in world coal ash

1213 In the review by Ketris and Yudovich [82], the average REY content in the world

1214 coal ashes is estimated to be 404.5 µg g-1, approximately three-times greater than that

1215 in the earth’s crust (180.9 µg g-1). However, the results from Ketris and Yudovich [82]

1216 were the equivalent value of REY in coal on an ash basis, rather than the determined

1217 values in real coal combustion ash produced at power plants. In order to better

1218 evaluate the reserve of REY from the existing coal combustion ash, current study

1219 compiled data of 581 ash samples derived from different coal-fired power plants in 15

1220 countries (Table 13). Considering the number of samples available from each country,

1221 six countries including Spain, England, Poland, Bulgaria, Romania, and Finland are in

1222 the group of Europe; and four countries including South Africa, India, South Korea,

1223 and Indonesia are classified into the group of “other countries”. As a result, the

1224 average concentration of REY in world coal ash is estimated to be 435.45 µg g-1,

1225 slightly higher than the reported value by Ketris and Yudovich [82]. Coal ash of China

1226 (473 µg g-1) and the U.S. (459.61 µg g-1) contains higher contents of REY than that of

1227 Europe (278.65 µg g-1) and other countries (298 µg g-1). This might be caused by the

1228 limited number of samples from these two regions, where no data is available on the

1229 coal ash generated from the combustion of coals containing high REY contents. The

1230 rare earth oxides (REO) have been used as a proxy for estimating the abundance of

1231 metals in ores. Some coal ash, for example, fly ash from Eastern Kentucky

1232 pulverized-coal fired power plants [250] has been reported to have extremely high

76
1233 REE concentrations (1202-1667 µg g-1), comparable to the REO content of some

1234 conventional REE ores, such as the ion-adsorbing type rare earth ores and low-grade

1235 hydrothermal ores associated with alkaline igneous activities.

77
1236 Table 13. A summary of the abundances of REY in coal ash from different countries around the world. Unit: µg g-1
a b c d e
China US Europe Others World
average range n average range n average range n average range n average range n
Y 57.4 1.5-233.0 217 68.2 6.2-214.0 257 33.1 5.1-60.7 72 36.4 2.2-77.0 20 58.5 1.5-233.0 566
La 84.4 5.8-441.0 224 73.5 21.9-265.0 257 45.0 2.6-103.0 80 54.5 9.6-112.0 20 73.1 2.6-441.0 581
Ce 176.8 21.9-955.0 224 161.6 44.0-534.0 257 98.7 5.2-221.0 80 103.6 16.2-222.0 20 156.8 5.2-955.0 581
Pr 19.1 2.7-117.0 224 17.2 0.10-63.0 257 11.2 0.7-26.0 80 14.5 2.3-48.0 20 17.0 0.1-117.0 581
Nd 75.3 9.7-436.0 224 74.3 21.4-234.0 257 56.2 3.1-264.0 80 44.3 8.8-86.5 20 71.2 3.1-436.0 581
Sm 13.7 1.7-79.0 224 14.3 4.4-50.5 257 8.6 0.5-16.0 80 11.1 1.7-53.0 20 13.2 0.5-79.0 581
Eu 2.5 0.3-8.0 224 3.0 bdl-7.0 257 1.9 0.2-4.0 80 2.1 0.5-3.5 20 2.6 bdl-8.0 581
Gd 14.1 1.7-73.0 224 14.8 4.6-40.4 257 9.3 0.5-20 80 8.0 1.8-16.8 20 13.5 0.5-73.0 581
Tb 2.0 0.17-11.0 224 2.3 0.7-6.6 257 1.6 0.1-7.0 80 1.4 0.3-2.51 20 2.0 0.1-11.0 581
Dy 11.3 0.9-55.0 224 12.3 3.8-44.2 257 7.0 0.5-15.0 80 7.4 1.5-14.2 20 11.0 0.5-55.0 581
Ho 2.2 0.2-11.0 224 2.2 0.1-8.8 257 1.2 bdl-2.9 80 1.7 0.3-2.9 20 2.1 bdl-8.8 581
Er 6.5 0.4-31.0 224 7.7 2.2-24.5 257 4.4 0.3-15.0 80 11.0 0.9-120.0 20 6.9 0.3-120.0 581
Tm 0.9 bdl-5.0 224 1.0 bdl-6.0 257 0.5 bdl-1.1 80 0.8 0.1-1.6 20 0.9 bdl-6.0 581
Yb 5.9 0.3-30.0 224 6.5 2.1-21.8 257 3.9 0.4-9.0 80 4.3 0.8-7.8 20 5.8 0.3-30.0 581
Lu 0.9 bdl-5.0 224 0.9 bdl-3.0 257 0.6 bdl-2.0 80 0.9 0.1-2.0 20 0.8 bdl-5.0 581
LREY 371.5 62.9-2028.0 340.8 101.6-1140.7 218.8 12.1-630.0 226.7 42.6-463.1 331.3
MREY 86.5 7.4-380.0 100.5 30.94-299.4 49.4 1.3-91.0 54.3 13.3-114.0 29.2
HREY 16.5 1.4-82.0 18.3 5.7-61.0 10.4 0.9-25 17.1 2.4-122.0 16.5
TREY 473.0 74.1-2490 459.6 117.3-1295.3 278.7 14.3-4884.0 298.0 58.2-597.9 435.5
Critical
0.39 0.30-0.76 0.37 0.28-0.5 0.34 0.24-0.48 0.35 0.24-0.51 0.36 0.24-0.76
REY ratio

1237 Notes. n: the number of ash samples.

a
1238 China [24-28, 107, 108, 119, 121, 160, 161, 169, 178-180, 196, 234, 252]

b
1239 US [12, 15-17, 42, 120, 122, 140, 144, 145, 146, 147, 158, 167, 191, 231-233, 238, 250-251, 258-259, 262, 264, 271 ]

c
1240 Europe [18-23, 97, 166, 168, 170, 171, 186, 191, 213, 246]

d
1241 Others [31-33, 185]

e
1242 World [82]

1243

78
1244 4.2. REY contents in coal ash from major coal-producing countries

1245 4.2.1. Contents and spatial distribution

1246 The concentration of REY in coal ash is largely dependent on the geological

1247 origin of the parent coal. The burning of coal from some coal-hosted ore deposits can

1248 further enrich REY in coal ash and may produce the potential “artificial REY ores”

1249 (REY-rich coal ash). Thus, prioritization of ash sources for REY recovery requires a

1250 detailed knowledge on REY concentrations of different types of coal ash from

1251 different origins or regions. The information is also vital in order to explore possible

1252 strategies to leach REY from the ash during recovery operations and understand the

1253 potential valuation of the coal ash with respect to the REY reserves. In view of this,

1254 previous findings in the published literature are summarized and discussed in the

1255 review, with emphasis on ash samples from U.S. and Chinese power plants as two

1256 examples.

1257 Extensive analysis of REY contents has been performed on ash samples from

1258 U.S. power plants (Table 13). Taggart et al. [16] made a broad survey study of REY

1259 contents in over 100 U.S. coal fly ash samples produced from coals from three major

1260 coal basins, namely the Appalachian basin, Illinois basin, and Powder River basin

1261 (Figure 11A). The total REY content and trends in the U.S. fly ashes are:

1262 REY contents differed significantly by coal basins, showing a decreasing order of

1263 Appalachian basin coal ash (591 µg g−1) > Illinois Basin coal ash (403 µg g−1) >

1264 Powder River coal ash (337 µg g−1);

79
1265 Relative to the UCC, REY distribution pattern of coal fly ash displayed different

1266 enrichment modes among the three basins, suggesting the characterization of

1267 REY in coal ash mainly inherited from the feed coals.

1268 Afterwards, fly ash samples from burning Illinois basin and Powder River basin have

1269 been detailed investigated by Hower et al. [251] and Huang et al. [120], respectively.

1270 The results confirmed the REY contents for Illinois Basin coal-sourced and Powder

1271 River basin coal-sourced fly ashes were less than those of Central Appalachian

1272 coal-sourced fly ashes. Meanwhile, a scattered distribution of the REY in coals of

1273 each region is also observed due to the mine-to-mine variations of REY contents in

1274 parent coals and the varying combustion conditions.

1275

1276 Figure 11. Comparison of total REY contents (A) and the critical REY ratios (B) in

1277 coal combustion ash at power plants from three major U.S. coal basins: Appalachian

1278 basin (App, triangles in blue), Illinois basin (IL, circles in red), and Powder River

1279 basin (PRB, diamonds in black). After Taggart et al. [16]. Reproduced with

1280 permission from the American Chemical Society. Further permissions related to the

80
1281 material excerpted should be directed to the American Chemical Society at

1282 https://pubs.acs.org/doi/abs/10.1021/acs.est.6b00085.

1283

1284 For Chinese coal ashes, compiled data from published literature on REY

1285 concentrations from different provinces and/or regions from China is discussed. A

1286 total of 227 sets of data related to REY analysis (including 11 samples from the first

1287 author’s laboratory) were collected from 14 provinces or cities in China

1288 [24,25,27,28,93,117,119,161,169,178–180,252,253]. The spatial distribution of REY

1289 contents in coal ashes among different provinces or cities of China is compared

1290 (Figure 12).

1291 Based on the available data and the geographic locations, provinces with

1292 different REY concentrations in coal ashes can be divided into three zones (Figure

1293 12). The total REY differed among these zones, following the decreasing order of

1294 Zone Ⅰ (623.5 μg g-1) > Zone Ⅱ (430.9 μg g-1) > Zone Ⅲ (318.6 μg g-1). Zone Ⅰ

1295 consists of four provinces or cities located in southwestern China (Sichuan,

1296 Chongqing, Yunnan, and Guizhou). This is consistent with the distribution of REY

1297 contents in Chinese coal basins, where feed coals mined from the southwestern China

1298 are rich in REY. Zone Ⅱ consists of five provinces from central China (Shanxi,

1299 Shandong, Henan, Hubei, and Anhui); and one province from southern China

1300 (Guangdong province) is also included, because the feed coal for power generation is

1301 mainly derived from the central regions (e.g., Shanxi province). REY concentrations

1302 in coal ashes from Zone Ⅱ are much lower than that of Zone Ⅰ. Inner Mongolia and

81
1303 Xinjiang autonomous regions, as two important coal producing area located in the

1304 Northeastern and Northwestern China are grouped into Zone Ⅲ. Although some coals

1305 from Inner Mongolia, such as REY-rich Pennsylvanian coals from Jungar and

1306 Daqingshan coalfields are reported to have high concentration of REY [254], the

1307 compiled results at present study suggest that REY are not significantly enriched in

1308 the coal ashes from Zone Ⅲ. This is probably caused by (i) the “mineralized coal”

1309 from Zone Ⅲ fed into the utility boilers depleted REY and (ii) insufficient published

1310 data related to the abundances of REY in the coal ashes from this region.

1311

1312

1313 Figure 12. Distribution of rare earth elements in coal ash across China. Data was

1314 compiled from literature [24,25,27,28,93,117,119,161,169,178–180,252,253]

1315

1316 4.2.2. Critical REY ratios

1317 In addition to the total REY concentration, the ratio of critical rare earth element

1318 (selected MREY and HREY) is another important parameter influencing the potential
82
1319 values of rare earth ores. As shown in Table 13, the average ratio of critical REY (Nd,

1320 Eu, Tb, Dy, Y, and Er) in the world coal ash is 36%, which is higher than some

1321 conventional REY ores. Accordingly, coal ashes that are enriched in critical REY

1322 would provide alternative sources to the conventional rare earth ores.

1323 Similar to the REY contents, the critical REY fraction also heavily depends on

1324 the origins of feed coal. For the U.S. coal ashes, the critical REY fraction was greatest

1325 for the Appalachian basin and Illinois basin-sourced coal ashes (Figure 11B) than

1326 ashes from Powder River basin. There are no statistically differences in the critical

1327 REY ratios between Appalachian basin and Illinois basin coal ashes [16].

1328 The ratio of critical REY in different zones in China is displayed in Figure 13.

1329 The data demonstrated that the critical REY content differed significantly among the

1330 three regions (Kruskal-Wallis test, p = 0.005). Ashes from Zone Ⅱ had higher critical

1331 REY ratios than those from Zone Ⅰ (median difference of 3.4%) and Zone Ⅲ (median

1332 difference of 7.3%). Regardless of the differences, the ratio of critical-to-excessive

1333 REY of Chinese coal ashes is higher than that of selected conventional REE ores

1334 currently being mined, such as the super large Bayan Obo Fe-REE-Nb ore deposit in

1335 Inner Mongolia of China (< 10% critical REY). As shown in the UCC normalization

1336 curve of REY in coal ash, REY in coal combustion ash show a medium- and heavy-

1337 REY distribution pattern, which is different from the typical REE ores (LREE rich

1338 type).

1339 Overall, both of the total REY and critical REY contents are mainly determined

1340 by the origin of feed coals. Coal ash samples produced in U.S. and China would

83
1341 provide three times the critical REY of total REY extracted per kg than that of

1342 conventional ores. The high fraction of in coal ash may represent a major advantage

1343 over the conventional ores if the extraction costs per kg of can match that of REE

1344 mining and extraction.

1345

1346

1347 Figure 13. The ratio of critical-to-excessive REY in China coal ashes from different

1348 zones (A) and the normalization curve relative to UCC (B). Data was compiled from

1349 literature [24,25,27,28,93,117,119,161,169,178–180,252,253].

1350

1351 4.3. Estimation of REY reserves and economic value

1352 4.3.1. Reserve of REY in coal ash

1353 The “workable reserve” of REY in coal combustion ash depends not only on

1354 their concentration but also on the quantity of available ash. In the United States,

1355 around 59.6% of the 45.7 Mt of coal combustion ash produced annually in the US is

1356 beneficially reused, whereas the remainder was generally discarded [255]. Taggart et

1357 al. [16] estimated that the annual production of REO in unused portion of coal fly ash

1358 was 4000 t for the Appalachian Basin, 1280 t for the Illinois basin, and 3630 t for the
84
1359 Powder River basin. The unused fly ash comprises a REY reserve that is similar in

1360 scale to the annual production of REY at the Mountain Pass mine (4769 tons in 2014).

1361 China is the world’s largest consumer of coal, with annual usage surpassing that

1362 of the US, Japan, and European Union combine [256]. Annual amount of generated

1363 combustion residual from thermal power plant is huge. In 2017, about 70% of 686 Mt

1364 was reused for beneficial purposes [257]. The remainder is disposed in landfills and

1365 thus the annual ash production plus the fly ash already in storage constitute a huge

1366 potential REY resource. Based on Taggart et al. [16], this study made an similar

1367 assessment of REY reserves in the unused coal ash annually produced by Chinese

1368 power plants. Assuming that coal-fired power generation of each provinces mirrors

1369 the relative coal ash production by region, then the annual total amount of coal ash

1370 produced from the Zone Ⅰ, Zone Ⅱ, and Zone Ⅲ was estimated to be around 32.8,

1371 260.5, 90.0 million tons; and the amount of unused ash accounted for 30% of the total

1372 ash, that is, 9.8 million tons for Zone Ⅰ, 78.2 million tons for Zone Ⅱ, and 27.0 million

1373 tons for Zone Ⅲ (Table 13).

1374

1375 Table 14. A summary of annual coal ash production, unused ash amount, annual

1376 power generation by region, and the calculated ash amount.

Annual coal ash production 686 million tons

Annual unused ash 206 million tons

Annual power generation 4663 billion kW•h

85
 Power

generatio Ratio Total Ash (metric Subtotal (metric

Region
n 10^8 (%) tons) tons)

kW•h

Chongqing 469 1.0 6,898,483

Sichuan 384 0.8 5,656,225

32,770,686
Guizhou 1134 2.4 16,690,071

Yunnan 240 0.5 3,525,907

Shanxi 2589 5.6 38,092,179

Anhui 2318 5.0 34,108,058

Jiangxi 939 2.0 13,821,034

Shandong 4914 10.5 72,294,376 260,542,273

Henan 2576 5.5 37,898,149

Hubei 1043 2.2 15,346,416

Guangdong 3329 7.1 48,982,061

Xinjiang 2369 5.1 34,851,927

Inner 90,028,186
3750 8.0 55,176,259
Mongolia

1377

1378 4.3.2. Valuation of REY in coal ash

1379 Based on the data in Table 14, the annual total tonnage of REO in the discarded

1380 coal ash is estimated to be 8029 t for the southwestern provinces of Zone Ⅰ, 42,114 t

86
1381 for the central provinces of Zone Ⅱ, and 9769 t for Zone Ⅲ, respectively. If all of

1382 REO in the unused coal ash is extracted, the total amount could be around 59,909 t

1383 (Table 14). This coal ash reserve of REO annually (the year of 2019) produced from

1384 China power plants is nearly half of the annual China REY mine production (132

1385 thousand Mt of REO in 2019; Data sources:

1386 https://www.statista.com/statistics/268011/top-countries-in-rare-earth-mine-productio

1387 n/). According to the 2020 prices for REO (http://www.reht.com/h-nd-426.html), the

1388 total value of the REO in the discarded ash reserve from China is estimated at 14.8

1389 billion yuan (Table 15), which is equivalent to U.S. 2.2 billion dollars based on the

1390 annual average exchange rate of Chinese Yuan to US dollar (0.15 USD). The REY

1391 value in Chinese coal ash is lower than that of US coal ash as estimated by Taggart et

1392 al. (4.3 billion dollars) [16]. Main reasons are the large variations of REY prices with

1393 time, different REY contents (especially critical REY) and Sc is excluded from the

1394 valuation due to insufficient Sc data in coal ash. Note that only a part of coal ash

1395 could be extracted for rare earths products, as indicated by the large variations of REY

1396 concentration in different coal ashes [258]. Accordingly, the valuation of REY

1397 reserved in coal ashes is likely smaller since not all of the unused ash can meet the

1398 requirements of being used as extraction sources. Besides the total value, the value of

1399 REO per ton of coal ash was also calculated to provide a basis for assessing the use of

1400 coal ash as an alternative REY source. As shown in Table 15, it is estimated that the

1401 per-ton REO values is 180.1 yuan for Zone Ⅰ, 111.9 yuan for Zone Ⅱ, and 66.3 yuan

1402 for Zone Ⅲ. The main contributors to the value of REO are some of the critical rare

87
1403 earths. The high prices of Nd (374 yuan/kg), Tb (4950 yuan/kg), and Dy (1950

1404 yuan/kg) contribute to near 70% of the total value of the whole REY value in coal ash.

1405 The result highlights the importance of the high ratio of critical REY in ash, which is

1406 one of the main advantages over other conventional rare earth ores.

1407 Overall, the value of REY in coal ash primarily depends on several factors, i.e.,

1408 the total REY concentration in coal ash, the ratio of critical REY, the separation

1409 technology, and the global market price of REO [18,281,295]. The valuations will be

1410 adjusted, depending upon these factors especially the extraction techniques that are

1411 currently developed. The costs for REY recovery from coal ash, such as transportation

1412 and extraction processes, should be scaled regarding the valuation. To add values and

1413 reduce the overall costs, the selection and pre-concentration of REY-rich coal ash as

1414 well as co-extraction of other metals or reuse the wastes after extraction should be

1415 considered and incorporated in the extraction procedure (see the discussion in section

1416 7). There are huge amounts of “old ash” preserved in landfill sites (e.g., ash ponds),

1417 that have been accumulating for many years could be a source but is not considered in

1418 the valuation. Considering the state of the coal-fired generating industry (continues to

1419 fall) in the developed countries, such as in the U.S., this preserved ash would be a

1420 more viable resource. Waste management regulations would additionally drive the

1421 scientific community and relevant enterprises to explore the REY recovery from coal

1422 ash, particularly for the REY-rich ash disposed in potential agriculture lands [16,122].

88
1423 Table. 15. Total mass and estimated value of REO as well as the REO value per metric ton of fly ash produced annually in China.

a
Total REO (t/yr) Total mass 2020 prices Total value Chinese yuan REO/t ash

Elements Zone I Zone Ⅱ Zone 3 Ⅲ (t/yr) yuan per kg million yuan/yr Zone 1 Zone 2 Zone 3

Y 3646 18378 4578 26602 22.0 585.2 2.96 1.64 1.15

La 4556 24764 5336 34656 10.8 374.3 1.81 1.08 0.68

Ce 10019 50564 12099 72683 10.8 785.0 3.99 2.21 1.51

Pr 1064 5559 1293 7915 354.6 2806.7 13.91 7.99 5.32

Nd 4091 23942 5043 33076 374.0 12370.4 56.42 36.29 21.74

Sm 794 3567 990 5351 12.0 64.2 0.35 0.17 0.14

Eu 188 1144 235 1567 300.0 470.0 2.08 1.39 0.81

Gd 769 3305 979 5053 187.0 944.9 5.30 2.51 2.10

Tb 114 706 127 948 4950.0 4690.1 20.81 14.17 7.29

Dy 634 3017 774 4425 1950.0 8628.0 45.59 23.84 17.32

89
 Ho 126 639 140 905 435.0 393.6 2.02 1.13 0.70

Er 349 2391 421 3161 170.0 537.4 2.19 1.65 0.82

Tm 50 378 38 466 7150.0 3331.7 13.29 10.94 3.21

Yb 318 1703 447 2469 100.0 246.9 1.17 0.69 0.51

Lu 47 322 54 423 4750.0 2010.3 8.20 6.20 2.99

Total 26764 140380 32554 199698 20776.2 38238.6 180.10 111.91 66.30

Unused ash 8029 42114 9766 59909

a
1424 2020 price: the price of light rare earths and heavy rare earths derived from China Northern Rare Earth (Group) High-Tech Co., Ltd

1425 (http://www.reht.com/h-nd-426.html) and China Southern Rare Earth Group Co., Ltd.

1426 (http://www.zgnfxt.com/n91/n277/n279/c10814/content.html).

1427

90
1428 5. Speciation of REY in coal ash

1429 5.1. Definition and classification of REY forms in coal ash

1430 The combustion of coal in a utility boiler can significantly alter the speciation of

1431 REY in coal. As reviewed in section 3.2, the speciation of REY in coal ash has been

1432 extensively studied by techniques at the bulk [16,108,230] and micro scales

1433 [122,144,146,192,230,233,236]. Glassy materials, crystalline minerals, and unburned

1434 carbon are the main components comprising coal ash. REY have been found to be

1435 associated with these three main phases. Based on the REY chemistry, mineralogy,

1436 and physical distribution within fly ash, the speciation of REY in coal ash is classified

1437 as: amorphous glassy associations; discrete minerals, both the major REE-bearing

1438 minerals and trace associations in other minerals, or compounds; and organic

1439 associations (bound to unburned carbon). The third type seems to be elusive with only

1440 a few TEM cases study finding it as inclusions dispersed into the amorphous and

1441 graphitic carbons, so that more studies and evidences are needed to make it clear and

1442 credible (Figure. 14).

1443

91
1444
1445 Figure 14. The classification of REY speciation in coal ash.

1446

1447 5.2. Associations with amorphous glassy materials

1448 Aluminosilicate glasses are the dominant phases in coal-derived Class F fly

1449 ashes. A number of studies have indicated that REY associated with glassy materials

1450 in coal ash might be the dominant form [20,27,121,122,145,146,196,231–233]. As

1451 shown in Figure 15, correlation analysis of REY contents versus major elements

1452 contents in fly ashes found that REY showed a positive correlation to SiO2 and Al2O3.

1453 Sequential chemical extraction studies on coal fly ash samples from Southwestern

1454 China [27], Eastern United States [260], and northeast Spain [171] indicated that more

92
1455 than 70% of REY was hard to leach in various acid solutions and was considered to

1456 be bound with silicates and aluminosilicates.

1457

1458

1459 Figure 15. The relationship between the concentrations of REY and major

1460 components of coal ash. Data was compiled from literature

1461 [18,25,32,120,169,178,186,191,252,259,261].

1462

1463 Amorphous Si-Al glassy associations can be further divided into those having

1464 REY dispersed throughout the whole glass grains and individual REE-bearing phases

1465 that are closely bound with Si-Al glasses (Figure 14). Scale is a key parameter to

1466 discern these associations and various microscale techniques, including SEM-EDS,

1467 EPMA, TEM and the related techniques, SIMS, synchrotron-based micro XRF

1468 imaging and micro XANES, and among others are necessary for identifying these

1469 REY forms. REY dispersed throughout the aluminosilicate glass have been reported

1470 in several instrumental studies. Using Ce as a proxy for all of the REY, an EPMA

1471 study of REY-enriched Kentucky fly ashes by Hower et al. [238] was conducted, and

93
1472 the WDS mapping observed that REY was distributed throughout the glassy ash

1473 particles, such as Si-Al glassy cenospheres and frothy glassy particles. Note that at

1474 that scale, EPMA may not tell if the Ce was truly dispersed or in fine minerals. The

1475 distribution of REY in individual fly ash Si-Al glass grains were determined using the

1476 Stanford/USGS SHRIMP-RG ion microprobe [146]. The uniform distribution of REY

1477 in single aluminosilicate particles to that of bulk fly ash is indicative of REY

1478 partitioning into the aluminosilicate glasses. In a study by Stuckman et al. [232], a

1479 suite of four coal fly ash and three bottom ash samples derived from the Appalachian

1480 basin was studied by synchrotron micro-X-ray fluorescence and near-edge absorption

1481 spectroscopy (μ-XRF and μ-XANES). The μ-XRF mapping showed that REY were

1482 dispersed throughout the aluminosilicate glass in irregular shapes with large grain size

1483 (> 40 μm). In another XANES study [233], linear combination fits of bulk Y-XANES

1484 spectra of nine fly ashes from the central Appalachian basin, Illinois basin, and

1485 Powder River basin found that 22-76% of REE occur as Y-doped glass. After

1486 leaching these ash samples with 1M oxalic acid, the bulk ash fits were dominated

1487 by Y-doped glass.

1488 The close association of REY mineral particles with glassy matrices is another

1489 common REY occurrence in fly ash. In the study of Liu et al. [230], REY-rich

1490 phosphate particle (xenotime in ~10 μm size) was observed by SEM-EDS in a US

1491 coal-derived fly ash, and the REE minerals were found to be totally encapsulated or at

1492 the edge of Si-Al glasses. In combination of SEM-EDS and LA-ICP-MS, Thompson

1493 et al. [237] observed that small grains (~10 μm) of monazite, zircon, and apatite were

94
1494 commonly embedded within Al-Si glassy cenospheres, Al-Si spheres, or other

1495 aluminosilicate glass forms and the particle size of these REY-bearing glassy phases

1496 was mostly in the range of 50-80 μm (Figure 16). Synchrotron micro-X-ray

1497 fluorescence and near-edge absorption spectroscopy (μ-XRF and μ-XANES) also

1498 provide direct evidences [218]. In their study, Ce hotspots, an area (4–30 μm) denoted

1499 by the yellow to red color on the maps representing high XRF signal intensity, were

1500 mostly found within glass phase (13 out of 19) in μ-XRF maps. In six of 13 Ce

1501 hotspots, the Ce hotspots were encapsulated in irregular-shaped aluminosilicate grains.

1502 The co-localization of Ce with Si, Al, Ca, P, Ti, and Nd suggested that monazite is

1503 embedded in glass phase. Similarly, as a proxy element, Y was found to be

1504 co-localized with a range of elements, including Si, P, Ca, Fe, and other REE by

1505 μ-XRF mapping techniques [145,233]. Further analysis by μ-XANES reveals

1506 REY phosphates, REY oxides, or REY carbonate making up these Y hotspots.

1507

95
1508

1509 Figure 16. SEM-BSE images of REY-bearing minerals encapsulated within glassy

1510 particles in a fly ash from a pulverized coal power plant in Ohio, USA. After

1511 Thompson et al. [237]. Reproduced with permission from Elsevier.

1512

1513 The use of TEM and related techniques, such as SAED and EDS can provide

1514 insights into the glass-REY mineral associations. TEM studies of REY-rich Kentucky

1515 fly ash samples have revealed a variety of REY-bearing minerals and amorphous

1516 phases to be present within the ash [122,144,192,231,236]. For example, Hower et al.

1517 [122] found ~10 nm xenotime (YPO4) particles included within monazite in the

1518 ponded ash from the burning of eastern Kentucky coal; the composite mineral

1519 assembles are within glass. The nanosized REY phosphates within glass phase were

1520 further confirmed by X-ray pattern of the separated REE grain after FIB extraction.

96
1521 Figure 17 illustrates a STEM image of a REY rich region, EDS mapping showing the

1522 La, Ce, and Nd region surrounded by complex mineral assemblages.

1523
1524 Figure 17. Images showing how a scanning transmission electron microscope (STEM)

1525 worked on identifying the fine REY-bearing mineral grains with the aid of SEM-BSE

1526 techniques. (a) SEM-BSE identified the REY-rich area for preparing the TEM lamella

1527 (yellow box); (b) STEM image of the selected area extracted by FIB technique high

1528 angle annular dark field (HAADF) mode. After Hower et al. [231]. Reproduced with

1529 permission from Elsevier.

1530 A mineralogic examination of Figure 17 particle is seen on Figure 18, where the

1531 lattice interlayer spacing and the diffraction pattern, as derived from FFT and SAED
97
1532 indicates a Ce orthophosphate monazite, with a monoclinic unit cell. In addition to

1533 monazite, xenotime, zircon, and among other, crichtonite-group minerals, possibly

1534 davidite ((Ce,La)(Y,U)(Ti,Fe)20O38) or (Ce,La)(Y,U, Fe)(Ti,Fe)20(OH,O)38)) were

1535 found within Si-Al glass in a stoker ash from the combustion of REY-rich coal

1536 (Figure 19) [25]. However, some REY-rich areas have no discernable minerals, as

1537 indicated by SAED or FFT imaging for the crystallinity [144]. This may indicate that

1538 REY disperse into the glassy structures. The restrictions of the current technique

1539 determine that it cannot detect exceptionally fine minerals, overlapped, and/or

1540 disparate mineral grain orientations if REY present as the complicated forms.

1541

98
1542

1543 Figure 18. (a) Low-magnification bright field TEM image from the region shown in

1544 Fig. 18; (b) REY-containing area indicated in a yellow box; (c) Lattice fringe

1545 interlayer spacings determined by FFT; (d) SAED quantification obtained from the

1546 same region. After Hower et al. [231]. Reproduced with permission from Elsevier

99
1547

1548 Figure 19. (A) TEM bright field image of a REE-bearing mineral, possibly davidite

1549 ((Ce,La)(Y,U)(Ti,Fe)20O38) or (Ce,La)(Y,U, Fe)(Ti,Fe)20(OH,O)38)) based on their

1550 diffraction pattern from SAED. (B) Element maps of Al, Ti, and Fe. (C) Element maps

1551 of Al, La, and Ce. After Hower et al. [236]. Reproduced with permission from

1552 Elsevier

1553

1554 5.3. Associations with discrete minerals or compounds

1555 Discrete REY-bearing minerals/compounds are another common REY forms in

1556 coal ash. While aluminosilicate glass is the main ash components, it can observe that

1557 some discrete REY-bearing minerals/compounds have no contact with the glass

1558 phases. In comparison to the amorphous glassy associations as noted above, discrete

1559 REY minerals or/and compounds are not exclusively enclosed within the glassy
100
1560 matrix of coal ash, showing certain independence from the glass phases (Figure

1561 20-21). Typical discrete REY phosphates, REY carbonates, REY oxides, REY

1562 silicates, REY sulfates, REY-bearing iron oxides, etc., have been observed in coal fly

1563 ashes. By using polarized reflected-light and oil-immersion optics, Figure 20 (A)

1564 shows a monazite grain in a Kentucky Fire Clay coal-derived fly ash, and Figure 20

1565 (B) illustrates a zircon (a potential Y source) grain in fly ash from Powder River Basin

1566 subbituminous coal-derived fly ash. These two large mineral particles might be

1567 extreme examples of the discrete mineral/compound associations, since most REY

1568 particles in ash are too small to be observed using fly ash petrographic methods.

1569 Figure 21 illustrates a EDS mapping result showing that discrete REY-bearing

1570 phosphate minerals rhabdophane (Ce,La)(PO4·H2O) and monazite

1571 (Ce,La,Nd,Th)(PO4,SiO4), as well as REY-enriched calcium oxide were identified in

1572 a coal fly ash from combustion of coal from a Ohio pulverized coal-fueled power

1573 plant. Discrete pure REY phosphates (∼30-μm size) and minor REY contained in

1574 lime have been observed in two US coal fly ashes (Liu et al., 2019).

1575

1576

101
1577 Figure 20. (A) Monazite (m) with glassy particles in an Eastern US bituminous

1578 coal-derived fly ash. (B) Zircon (z) in fly ash from Powder River Basin

1579 subbituminous coal-derived fly ash. Images from Hood et al. [192].

1580 https://doi.org/10.4177/CCGP-D-17-00002.1.Copyright [2017], University of

1581 Kentucky Center for Applied Energy Research and the American Coal Ash

1582 Association. Reproduced with permission from Coal Combustion and Gasification

1583 Products.

1584

1585

1586 Figure 21. (A) SEM-BSE image of a fly ash derived from an Ohio pulverized coal

1587 power plant. (B) Mineral phases of fly ash (segmented image) are Al/Si-rich particles

1588 (blue), Fe-oxide (red), rhabdophane/monazite (yellow)), and CaO-rich (pink). After

1589 Montross et al. [147]. Reproduced with permission from Elsevier.

1590 Quantification of discrete REY-bearing minerals or compounds in fly ash is

1591 important for evaluation of REE extraction efficiency and some synchrotron

1592 microscopy and spectroscopy may provide some important information in this aspect

1593 [145,232,233]. The LCF of bulk Y XANES spectroscopy of fly ash from three major
102
1594 coal basins of the U.S. suggest that bulk ash fits were comprised primarily of

1595 Y2 O3 (18% to 51%) and Y-doped glass (22% to 76%) [233]. While in another Y

1596 XANES study [145], no Y-doped glass was detected in the LCF results; instead

1597 REY oxides (35-55%), REY-bearing hematite (20−30%), and REY phosphates

1598 (20−25%) or apatite (20−35%) consisted of the REY species in fly ash. Both of

1599 the bulk Y studies only used Y-doped glass as reference compounds, meaning

1600 that REY species closely bound with glassy materials (e.g., encapsulation or

1601 embedding) may not be detected by the method. This may cause the different

1602 fitting results of micro-scale synchrotron spectroscopy (μ-XANES) of REE hotspots

1603 in the same fly ash from that of bulk XANES. In Liu et al.’s study [145], twelve Y

1604 hotspots with particle size from < 10 μm to 50 μm in fly ashes was fitted and the LCF

1605 results of Y μXANES showed that five of the hotspots were identified as REY oxides,

1606 three as REY phosphates (churchite/xneotime), one as REE in Ca-rich Si-Al glass,

1607 and the rest had no results because of indistinguishable LCF data profile. Taggart et al.

1608 [233] observed great variations in μXANES fits among Appalachian basin, Illinois

1609 basin, and Powder River basin ashes. Results from XANES at both bulk and

1610 micro scale can be informative for knowing the speciation of REY in fly ash

1611 and contribute to the confirmation of certain amounts of discrete REY minerals

1612 or compounds in fly ash but is still challenging to provide a complete and

1613 quantitative information.

103
1614 5.4. Associations with unburned carbon

1615 Unburned carbon in fly ash has been proved to be important trapper for many

1616 trace metals, such as Hg, As, Se, and among others [262,263]. In a recent study by

1617 Hower et al. [264], the froth flotation of fly ashes collected from one power plant

1618 burning Pennsylvanian-age Kentucky coal generates low density, C-rich particles. The

1619 concentration of REY in the carbon froth concentrate (512 ppm) is comparable to the

1620 feed ash (464 ppm), indicating that the fly ash carbon may host certain amounts of

1621 REY. TEM studies found that graphitic and fullerene-like carbons, often containing

1622 included/entrained nanoscale minerals and/or metals, occur in the fly ash resulting

1623 from the burning of bituminous coal [122,192,235,265]. In accordance with the

1624 established occurrences of trace metals with fly ash carbon deposited on

1625 aluminosilicate glass particles, Hower et al. [122] and Hood et al. [192] observed

1626 association of REY with the nanosized carbons by HR-TEM technique. In Hood et

1627 al.’s study [192], a thin-layer carbon shell containing Ce and Nd is noted (Figure 22).

1628 The nano-sized carbon deposits onto a Si-Al-O glassy sphere. Similarly, in another

1629 study by Hower et al. [122], an complex ash constituent of Al-Si spheres, Fe-bearing

1630 minerals, and unburned carbon can be seen; the Al-Si glass sphere with an Fe-O core,

1631 possibly spinel, is surrounded by Ce-Nd-Sm-Y-bearing graphitic carbon (Figure 23).

1632 The original construction implied that the REY are part of the graphite structure.

1633 However, the TEM resolution is still insufficient to distinguish if REY are in the same

1634 particles or occur individually or in varying concentrations. Figure 24 shows a

1635 detailed view of the graphitic carbon at the scale of 10-nm including few-nano-size

104
1636 Ce and Fe particles [235]. The sample is a C-rich concentrate product produced by the

1637 froth flotation of a fly ash. Note that the locations of the Ce Lβ peaks are close to that

1638 of the La, Pr, Sm, and Gd peaks, which cannot be discerned by the software. It is

1639 suggested that these REY might also be present in the particles [235].

1640

1641

1642 Figure 22. HR-TEM image (upper left) of Al-Si glass sphere surrounded by graphitic

1643 carbon deposits, as indicated by EDS mapping of C, Nd, and Ce in the upper right,

1644 lower left, and lower right images, respectively. The fly ash is from the combustion of

1645 an Eastern US bituminous coal. Images from Hood et al. [192].

1646 https://doi.org/10.4177/CCGP-D-17-00002.1. Copyright [2017], University of

1647 Kentucky Center for Applied Energy Research and the American Coal Ash

1648 Association. Reproduced with permission from Coal Combustion and Gasification

1649 Products.

1650
105
1651

1652 Figure 23. HR-TEM image (right) of Al-Si glass sphere surrounded by graphitic

1653 carbon deposits. The composite of 50 EDS scans (left) finds that the carbon contains

1654 Ce, Nd, Sm, and Y.

1655 Images from Hower et al. [122].

1656 https://doi.org/10.4177/CCGP-D-17-00002.1.Copyright [2017], University of

1657 Kentucky Center for Applied Energy Research and the American Coal Ash

1658 Association. Reproduced with permission from Coal Combustion and Gasification

1659 Products.

1660

1661
106
1662 Figure 24. (A) HR-TEM image of a graphitic carbon contains Ce and Fe (yellow box),

1663 as indicated by EDS spectra (B). Images from Hower et al. [122].

1664 https://doi.org/10.4177/CCGP-D-17-00002.1. Copyright [2017], University of

1665 Kentucky Center for Applied Energy Research and the American Coal Ash

1666 Association. Reproduced with permission from Coal Combustion and Gasification

1667 Products.

1668

1669 In addition to the association of metals with carbon surrounding Si-Al glassy ash

1670 particles, fly ash carbon also encases the Fe-spinels along with rare earth elements.

1671 Figure 25 illustrates the carbon deposit on the surface of the several spinel grains

1672 entraining Nd, Ce, Sm, and Y. The ash sample is a magnetic fraction of a beneficiated

1673 fly ash [192]. TEM imaging and the EDS mapping of the Fe-spinel minerals at a

1674 larger magnification further indicated that the spinel minerals occur in association

1675 with REY (Figure 26). The enrichment of some REY, for example Gd and Pr has

1676 been noted in the magnetic fractions of fly ashes from Eastern Kentucky power plants

1677 [12,122]. In their study [12], despite that the total REY contents of the magnetic

1678 concentrate (327 μg g-1) being lower than the clean product fraction (< 200 mesh, 485

1679 μg g-1), the Gd content shows a enrichment trend in the magnetic fraction (55 μg g-1

1680 for the magnetic fraction versus 18 μg g-1 REY for the <200-mesh fraction).

1681 Praseodymium is another element with a notably increased content in the magnetic

1682 fraction (35 μg g-1 versus 19 μg g-1).

1683

107
1684

1685 Figure 25. HR-TEM image of spinel (sp) surrounded by an amorphous carbon with

1686 Ce, Nd, Sm, and Y-bearing inclusions. The area scanned by EDS is shown by the

1687 rectangle on the lower right. The fly ash is a magnetic fraction of a fly ash from the

1688 combustion of an Eastern US bituminous coal. Images from Hower et al. [122].

1689 https://doi.org/10.4177/CCGP-D-17-00002.1.Copyright [2017], University of

1690 Kentucky Center for Applied Energy Research and the American Coal Ash

1691 Association. Reproduced with permission from Coal Combustion and Gasification

1692 Products.

1693

1694 Overall, results of the TEM combined with the EDS mapping indicated that

1695 nanosized particles containing a broad suite of REY were found in amorphous and

1696 graphitic carbons surrounding and binding either aluminosilicate glass or iron spinel

1697 grains in coal ash. It seems that REY occurs as sub-micron, nano-scale mineral grains

1698 associated with fly ash carbons. In another sample from the study of Hower et al [122],

1699 some Ce oxide was observed as 1-2-nm spheres encased with carbon, suggesting that

1700 the particles may be formed in the ash by nucleation and growth, consistent with the
108
1701 observation of REY inclusions in graphitic carbons, which might form during

1702 incomplete combustion processes, not as the fractured remnants of minerals from the

1703 coal. However, unlike the REY minerals seen in the fly ashes, the association of REY

1704 with unburned carbon is very difficult to resolve and the formation mechanisms seem

1705 to be elusive. In contrast the visible individual mineral inclusions in the carbons, some

1706 REY either dispersed throughout (Figure 22) or incorporated into the fly ash carbons

1707 (Figure 24) are too small to be detected as independent mineral grains or really bound

1708 to the unburned carbon (like Hg-C structure). More research and evidence are needed

1709 for confirming those associations and the formation mechanisms in the future.

1710

1711

1712 Figure 26. (A) TEM imaging of Fe-spinel minerals (sp) in fly ash from the

1713 combustion of an Eastern US bituminous coal blend. (B) Composite EDS image

109
1714 based on the elements mapped individually in (C). Unpublished images from Hower

1715 et al. [122]. https://doi.org/10.4177/CCGP-D-17-00002.1. Copyright [2017],

1716 University of Kentucky Center for Applied Energy Research and the American Coal

1717 Ash Association. Reproduced with permission from Coal Combustion and

1718 Gasification Products.

1719

1720 6. Behavior of REY in coal-fired power plants

1721 6.1. REY partitioning across the coal-fired power plants

1722 When considering the fate of REY in coal-fired power plants, it is important to

1723 estimate the distribution of REY in various product streams. This determines which

1724 types of coal combustion byproducts are suitable for REY recovery upon the coal

1725 combustion. Generally, inputs of REY into a utility combustion system include coal

1726 and any additives, such as limestone for desulphurization, and urea for denitrification;

1727 the outputs from a coal-fired power plant include bottom ash, fly ash, gypsum, filter

1728 sludge, flue gas, and escaped fine particulate matter [266].

1729 6.1.1 Mass balances of REY in power plants.

1730 Table 16 gives a summary of the distribution of REY among different emission

1731 streams of industrial coal-fired power plants. Several filed measurements and

1732 laboratory combustion tests estimated the overall mass balance of REY in various

1733 streams, finding the mass balance ratios greatly varied in different plants. In a work

1734 by Wu et al. [180], the distribution of REY in four China coal-fired power plants was

1735 investigated. The mass balance obtained for a 300-MW circulated fluidized bed boiler

110
1736 (CFB) and a pulverized coal-fired (PC) power plant were satisfied (79.4-103.4%).

1737 Only 37-79.4% REY recovery were obtained for another two power plants. In another

1738 laboratory-scale case study by Li et al [185], an evaluation of REY partitioning

1739 behavior during coal combustion was performed at three pilot-scale combustion tests

1740 on a 50-kW underfed stoker unit. Good mass balance closures (81-84%) were

1741 achieved for REY (taking Ce as proxy) of the two combustion runs, with another one

1742 having lower ratios (37%).

1743 The mass balance ratio of REY by measuring the REY content in all types of

1744 solid products (e.g., gypsum) and particularly flue gas has rarely been reported. Based

1745 on US Environmental Protection Agency Method 29 (EPA Method 29) for sampling

1746 and determination of trace metals in flue gas , recent field measurements on five

1747 Guizhou coal-fired power plants in Southwest China indicated that the concentration

1748 of REY in gypsums and flue gases was low, with ranges of 5.7-11.7 μg g-1 and

1749 5.9-21.4 μg m-3, respectively [28]. The balance ratio for REY in the five power plants

1750 was acceptable (90.3-113.7%). Overall, a closed balance for REY in coal power plants

1751 is difficult to obtain and the evaluated ratios are subject to large errors. The lower

1752 mass balance ratios for some measurements may arise because (1) the collected

1753 combustion byproducts do not correspond the coal burned in the furnace, this usually

1754 occurs in utility boiler experiencing large deviations, such as the boiler load,

1755 combustion temperature, residence time, varying feed coal sources, and among others;

1756 (2) REY emitted as gaseous species are not determined; and (3) contaminants in

1757 sample preparation and analytical errors in REY determination.

111
1758 Table 16. The distribution of REY across the whole power plants.

Boiler type Solid fuel Adsorbents Bottom ash fly ash Gypsum Stack gas References
Fuel property % Con Type Conc Conc f Conc f Conc f Conc μ f Balance
μg g-1 μg g-1 μg g % μg g-1 % μg g-1 % g m-3 % %
-1

a 50-kW stok bituminous coal; 6 - - 282 97 69 3 - - - - 81.0 [185]

er-fired labor Ar: 3.1; Vr: 39.
atory combus 26
tion unit bituminous coal; 16 - - 102 96 70 4 - - - - 35.0 [185]
Ar: 12.5; Vr: 3
6.44
bituminous coal; 16 - - 354 98 99 2 - - - - 84.0 [185]
Ar: 3.1; Vr: 37.
18
a 300-MW C coal washing wa 132.7 - - 251.1 58 331.04 42 - - - - 103.4 [180]
FB boiler stes; Aad: 49.6
5%. Vad: 18.19
a 1000-MW bituminous coal; 65.4 - - 319.9 6 545.1 94 - - - - 37.0 [180]
PC boiler Aad:25.05; Vad:
28.21
a 600-MW P bituminous coal; 63.6 - - 357.2 7 556.9 93 - - - - 79.4 [180]
C boiler Aad: 15.54; Vad:
29.54
a 606-MW P lignite; Aad:13.48 40.9 - - 368.8 13 432.9 87 - - - 50.0 [180]
C boiler ; Vad: 38.7
2*300 MW C coal washing wa 260.3 limesto 8.8 514.5 31.1 573.5 68.9 8.1 0.05 5.9 0.01 105.1 [28]
FB boilers stes; Aad: 45.15, ne
Vad: 18.77
4*600 MW P bituminous coal; 147.2 limesto 5.3 413.1 13 486.1 86.5 7.9 0.42 6.3 0.04 90.3 [28]
C boilers Aad:30.68; Vad: ne
18.07
4*600 MW P bituminous coal; 189.0 limesto 11.2 451.4 5.7 497.7 93.7 7.2 0.54 6.9 0.03 113.7 [28]
C boilers Aad:39.56; Vad: ne
15.05
4*300 MW P anthracite; Aad: 3 468.6 limesto 10.8 1257 15.22 1225 84.02 17.1 0.73 18.1 0.02 104.9 [28]
C boilers 8.08; Vad: 8.77 ne
2*300 MW P anthracite; Aad: 3 205.4 limesto 4.7 576.9 14.4 588.3 84.7 5.7 0.74 21.4 0.08 110.3 [28]
C boilers 1.72; Vad: 9.1 ne

1759 Conc: REY concentrations; f: mass fractions of REY in the combustion byproducts; r: as-received basis; ad: dry-basis.

112
1760

1761 6.1.2. Relative distribution of REY among ash streams

1762 Following the summary of Smith [267], Meij [266], and Clarke [268], and

1763 modified from Hower et al. [262], three groups of elements are broadly classified

1764 according to their volatility and partitioning behavior during coal combustion (Figure

1765 27). Due to the high boiling temperature, REY in coal are hardly volatilized and can

1766 be categorized as non-volatile elements. As shown in Table 16, almost all of REY

1767 partitioned into the solid ash after coal combustion. The mass fractions of REY in

1768 gypsum plus flue gas were calculated to be less than 1% [28]. Therefore, coal ash has

1769 been regarded as a primary byproduct for REY recovery. Moreover, it seems that fly

1770 ash may have more advantages for REY recovery than bottom ash in terms of the

1771 REY concentration and “reserve”. Mardon and Hower [250] arranged a set of

1772 REE-rich Fire Clay coals from a single seam of a single mine burned in a 220-MW

1773 wall-fired boiler for tracking the fate of trace elements in coal combustion. Fly ashes

1774 collected from the economizer, mechanical hopper, and electrostatic precipitator

1775 hopper have higher REY contents (average value of 1422 μg g-1, ash-basis) than

1776 bottom ash (1203 μg g-1, ash-basis). The slight higher REY contents in fly ash than

1777 bottom ash were also reported in many other power plants [24,27,119,161,178] . The

1778 mass ratio of fly ash to bottom ash for utility PC boilers varied significantly, but

1779 generally around 4:1. As a result, during coal combustion, most of REY in coal could

1780 transfer into the fly ash. In the studies of Wu et al. [180] and Li et al. [28], REY in

1781 coal fly ash derived from nine Chinese power plants show higher mass ratio of 42-94%

113
1782 than that of bottom ash (6-58%) [28,180]. These results could be useful for selecting

1783 REY sources from coal combustion byproducts. Note that an uneven distribution

1784 between fly ash and bottom ash does not follow the classification of trace element in

1785 Figure 27 [266]. The variation of REY contents in fly ash as a function of particle

1786 sizes and collection arrays in power plants is not only controlled by the REY volatility

1787 but other factors as will be discussed below.

1788

1789

1790 Figure 27. Distribution of elements among bottom ash, fly ash, and flue gas. After

1791 Hower et al [262]. Reproduced with permission from Elsevier. Boiling points after

1792 Wang et al.[269]. This is a generalized concept and actual performance in pollution

1793 control systems may vary. The elements emphasized in this paper are shown in bold

1794 italics.

1795

114
1796 6.2. REY partitioning within ash-collection systems

1797 6.2.1. Variations in REY distribution pattern within ash-collection systems

1798 The distribution of REY content within fly ash collection system, i.e.,

1799 electrostatic precipitator (ESP), baghouse or fabric filter (FF) is firstly dependent on

1800 their concentrations in feed coal. After coal combustion, the partitioning of several

1801 volatile elements (Group 2&3 in Figure 27) within ESP or FF is a function of flue gas

1802 temperature and particle size at the collection point [272] . As with the decreased flue

1803 gas temperature and ash particle size, the concentration of volatile trace elements

1804 trend towards the back row of the ash collection system. Unlike the volatile trace

1805 elements (e.g., Hg, As, and Se), rare earth elements have very low volatility; therefore,

1806 significant variations in REY content may not be expected among the fly ashes

1807 collected from different zones of individual fly ash collection systems. The study of

1808 an Appalachian-high-REE-bituminous-coal fired power plants with a two-row

1809 mechanical (cyclone)/three-row ESP ash collection system is a good case showing the

1810 REY distribution in fly ash capture system [250]. The fractionation of REY in the

1811 mechanical and ESP fly ashes, while variable, shows less temperature-related

1812 distribution than observed in the volatile elements. Rather, the total REY fall within a

1813 narrow range of 1214-1668 μg g-1 (Figure 28 (a)). A decrease in the ratio of LREE

1814 (La through Eu) to HREE (HREE: Gd through Lu) can be observed from mechanical

1815 rows (> 8) through the last row of ESP (< 6) (Figure 28 (b)). Hower et al. [167] also

1816 noted the fractionation between LREE and HREE after examining the chemistry of

1817 paired feed coal and corresponding ash from five different anthracite-burning stoker

115
1818 boilers. Light lanthanides decreased in concentration relative to heavy REE, as seen in

1819 the average coal LREE/HREE (9.8) against the average ash LREE/HREE (9.4). The

1820 decrease in the average Ce /Ce∗ (3.7 to 3.4) while increase in Gd /Gd∗ (1.9 to 2.1)

1821 from coal to ash also indicates a shift in LREE and HREE during coal combustion.

1822

1823

1824 Figure 28. (a) Variation of REY in fly ashes at different collection points of the ash

1825 collection system. (b) Variation of light REE to heavy REE ratio in fly ashes at

1826 different collection points of the ash collection system. Mech-1: 1st -row mechanical

1827 hopper ash; Mech-2: 2nd -row mechanical hopper ash; ESP-1: 1st row ESP ash; ESP-2:

1828 2nd -row ESP ash; ESP-3: 3rd -row ESP ash. Data was compiled from Hower et al

1829 [250,262].

1830

1831 Similar to the REY variation in the study of Hower [250,262], the enrichment of

1832 heavy lanthanides towards the back rows of the ash collection systems is notable in

1833 another three Kentucky power plants plus power plants from Bulgaria [259]. There

1834 are three- to five-row mechanical/ESP-ash collection system installed within these

1835 power plants, burning coals of varying ranks (lignite, subbituminous coal, and
116
1836 bituminous coal), with Plant P exclusively burning a small amount of waste tires with

1837 high-S coal. When the last-row ESP lanthanides are normalized to the first-row REY

1838 concentrations, all cases exhibit an enrichment in the HREE (Figure 29, after Hower

1839 et al. [259]). The ratio of critical to total REY also increases from the front ESP rows

1840 to the rear rows.

1841 Overall, there is no significant partitioning of REY within ash collection systems.

1842 Nonetheless, with respect to the enrichment of HREE in coal fly ash especially the

1843 back rows of ash collection system, the detailed mechanisms are still not well

1844 understood. It is known that the flue gas temperature and particle size decrease with

1845 increased distance from the boiler. In the study of Liu et al. [158], the distribution of

1846 REY in a series of fly ash derived from the combustion of coal at a 100-MW

1847 pulverized-coal-fired power plant in southeastern Kentucky was measured. The ratio

1848 of LREE/HREE decreases in economizer (7.15) through mechanical hoppers (6.45) to

1849 the last row of ESP (4.74), suggesting the enrichment of HREE would be associated

1850 with either a decrease in flue gas temperature or in the fly ash particle size. Fly ash

1851 petrology may provide some evidences explaining the increased HREE contents in

1852 fine fly ash, which might be due to [262] (1) the result of HREE having an affiliation

1853 to the organics in feed coal; and (2) the capture of the fine REY minerals into the

1854 glassy particles varying in size, chemistry, and structure, demonstrated by fly ash

1855 petrology study of sized fly ash, which will be discussed below.

1856

117
1857

1858 Figure 29. The normalization of REY contents in the last row ash to the first-row

1859 ash. Data was compiled from Hower et al [259].

1860

1861 6.2.2. Variations of REY in sized fly ash and implications for REY partitioning

1862 trend

1863 The particle size distribution of REY in fly ash is important to REY beneficiation

1864 from bulk ash into sized ash containing high contents of REY [21,166,259]. To

1865 understand the variation of REY contents versus particle size, the size dependence of

1866 REY has been investigated in several studies

1867 [20,21,23,24,33,119,158,161,170,178,192]. Apart from the finest fly ashes (minus

1868 500-mesh), the REY concentrations in sized fly ash (100-500 mesh) of individual

1869 hoppers, as examined by Liu et al. [158] noted above, have no defined trend (Figure

1870 30a). For the fly ash with particle size lower than 200-mesh size, the ratio of

118
1871 LREE/HREE decreases with the decrease in ash size for all the ash hoppers (Figure

1872 30a). Similar to Liu et al. (2014), the enrichment of REY and decrease in

1873 LREE/HREE in the finest fly ash by sieving is also reported by several investigations

1874 [21,23,167,192]. However, one exception study of the sized fly ash from Jungar

1875 power plant found that the concentration of REY in the finer fly ash were not

1876 significantly enriched and only increased by a factor of less than 2.5. For fly ash with

1877 much smaller particle size (< 20 μm), the enrichment of REY in the finer sized ash

1878 was noted in two cases study [166,170]. With an air-based classifier, fly ash was

1879 separated into size down to 2.2 μm (the whole size ranging from 2.2 to 43.2 μm), and

1880 REY exhibited an increase in the finest size fraction but depleted in the second-finest

1881 sized ash (2.2-9.7μm) relative to the bulk fly ash [166]. In another study [170], fly ash

1882 was separated into 0.3-2.5 and 2.5-10 μm size range using a cascade impactor. The

1883 contents of REY increased from 352.4 µg g-1 in the fine fly ash (2.5 -10 μm) to 384 µg

1884 g-1 in the coarse ash (2.5-10 μm).

1885

1886

119
1887 Figure 30. (a) Variations of REY contents and LREE/HREE in size fractions of

1888 each of the ash collection systems in a U.S. Kentucky power plant [158,259].

1889 (b)Variations of REY contents and LREE/HREE in size fractions of economizer

1890 hopper ash from Jungar power plant, Inner Mongolia, China [24].

1891

1892 Overall, as non-volatile elements, REY show an enrichment in finer fly ash but a

1893 relatively random distribution in the whole size range, and the enrichment of HREY

1894 (decreasing LREE/HREE) towards the finer fly ash is notable. The normalized

1895 distribution pattern of REY versus the particle size provides clues for REY partition

1896 among various sized ash (Figure 31). Here, the sized ash from the same hoppers from

1897 a southern Kentucky power plant perhaps gives the best examples, since coal feed to

1898 the boiler was specially arranged from a single coal seam from a single mine, thus

1899 eliminating the variations from multiple coal sources on REY fractionation during

1900 coal combustion [158,250]. As shown in Figure 31, all sized fractions show similar

1901 distribution pattern, characterized by medium REY enrichment mode and positive

1902 anomalies of Eu and Gd. This suggested that there is no notable fractionation among

1903 different size fractions. Difference in fly ash petrology may account for the

1904 LREE/HREE distributions. For the two most abundant fractions of mechanical ash,

1905 LREE/HREE significantly differs between the 100-325 mesh ash (7.30-7.37) and the

1906 fraction < 325 mesh (5.35-5.94). The 100-325 mesh ash has 70.1% glass compared to

1907 85.8% glass in the <325-mesh fraction. As summarized in section 5, REY bound with

1908 glassy materials is the dominant association in coal fly ash. Variations in the amount

120
1909 and size of REY minerals between and in the microscopically identifiable fly ash

1910 constituents can contribute to the differences in REY contents across the

1911 ash-collection array and the sized ash fractions.

1912

1913

1914 Figure 31. The normalized REY distribution pattern (relative to REY in UCC) in

1915 fly ash from the same hopper in a southern Kentucky power plant: (a) different size

1916 fractions of fly ash from the economizer hopper and (b) different size fractions of fly

1917 ash from the mechanical hopper. Data was compiled from Liu et al. [158] and Hower

1918 et al. [250].

1919

1920 6.3. Speciation transformation mechanisms

1921 6.3.1. Physical fragmentation

1922 Phosphate minerals such as monazite and xenotime are the common

1923 REY-bearing minerals found in coal (see section 2.3). These phosphate minerals have

1924 high melting temperature point, depending upon their elemental components:

1925 monazite (RPO4) melts at 1916°C (R = Sm) and up to at 2072°C (R = La);

1926 Er-xenotime melts at 1896°C and Y-xenotime at 1995°C, respectively [273].

121
1927 Accordingly, monazite and xenotime particles with high thermal stability would

1928 remain intact during coal combustion and, after removal of organic matter, should be

1929 more commonly seen in coal fly ash than that in the feed coal. However, the

1930 mineralogy study of fly ash at micro-scale does not support this deduction. The

1931 microanalysis of REY-rich coal, for example, the well-known REY-bearing Fire Clay

1932 coal [36,46,135], can readily find discrete monazite particles and other REY-bearing

1933 minerals (such as zircon) at several microns disseminated through the coal matrix. In

1934 contrast, the fly ash has few REY mineral grains that can be seen by optical/electron

1935 microscopy [144,147,192,231,236]. If monazite and other REY-bearing phases are

1936 scarcely to be observed at the scale of optical/electron microscopy, where are the rare

1937 earth elements in fly ashes? Hood et al. [192] demonstrated that pure monazite grains

1938 (hundreds of micrometers) will shatter into micron and sub-micron particle at 1400 ºC

1939 (Figure 32). A possible mechanism of the heat-induced fragmentation is the

238 232
1940 expansion of He produced from the decay of U and Th [274–276]. Among the

1941 other important REY minerals in coal, REE-bearing carbonates may also experience

1942 fragmentation at boiler temperature because the release of carbon dioxide [230].

1943 Zircon will decompose to zirconia and cristobalite in the temperature range of

1944 1285-1720 ° C [277]. Apatite, fluorapatite, and zircon may also experience

1945 fragmentation at boiler temperatures. Very fine REE phosphate minerals,

1946 REY-bearing CaO, and other unknown phases at a few μm to nm are observed in fly

1947 ash, suggesting the REY mineral fragmentation contributes to the final distribution

1948 and occurrences in coal ash.

122
1949

1950
1951 Figure. 32 (A) Original pure monazites observed under optical microscopy. (B)

1952 Monazite after heating to 1400 ℃ observed under optical microscopy. Images from

1953 Hood et al. [192]. https://doi.org/10.4177/CCGP-D-17-00002.1.Copyright [2017],

1954 University of Kentucky Center for Applied Energy Research and the American Coal

1955 Ash Association. Reproduced with permission from Coal Combustion and

1956 Gasification Products.

1957

1958 6.3.2. Phase transformation and chemical reactions

1959 While the REY minerals are thermally stable, REE-bearing phases in coal may

1960 experience complicated chemical transformation during coal combustion. For

1961 example, the XANES results of Ce in fly ash samples have identified that the

1962 redox-sensitive trivalent REE (Ce) could be changed into high valence state (RE4+)

1963 [145,232]. Thermal decomposition and conversion of individual REE-bearing phases

1964 in coal can provide insights into the transformation behavior of REE during coal

1965 combustion. Liu et al. [230] used an in-situ XRD combined with a thermal analyzer to

123
1966 monitor the phase transformation of typical pure REY phases at the temperature range

1967 of 25-1000°C (Figure 33) and found that (1) REE-organic compounds and REE

1968 carbonates will decompose (T > 300 ℃) into oxides; (2) water-bearing REE

1969 phosphates will lose water after heating with no Ce oxidation; (3) REE phosphates

1970 such as monazite and xenotime have no structure change; but (4) REE-bearing

1971 phosphates (e.g. apatite) and zircon show characteristics of melting on mineral surface.

1972 A summary of the thermal conversion pathways of typical REE forms in coal is

1973 illustrated in Table 17.

1974

1975

1976 Figure 33. Thermal decomposition curve of individual pure REE-bearing phases at

1977 the temperature range of 25-1500 ℃ via thermogravimetric analysis under air

1978 atmosphere. After Liu et al. [230]. Reproduced with permission from Elsevier.

1979
124
1980

1981 Note that the combustion temperature (< 1000 ℃) of the experimental devices

1982 limits Liu et al’s [230] observations and conclusions. First of all, the flame

1983 temperature for the most-commonly used pulverized coal-fired boiler varies from

1984 1200-1800 ℃. When the combustion temperature is above 1000 ℃, the stable REE

1985 mineral phases may shatter into small grains (such as monazite) or chemically

1986 decompose into other phases (such as zircon), as mentioned above. In addition, as an

1987 important carrier for REY in coal, apatite is not stable and can transform into other

1988 phases with temperature being greater than 1000 ℃; and the decomposition

1989 temperature is dependent on the anion position in the hexagonal axis showing

1990 decrease order of F(fluorapatite) > Cl (chlorapatite) > OH (hydroxyapatite) [278]

1991 (Table 17). Secondly, there are still many of other REE minerals or non-mineral

1992 phases, such as REY carbonates (kimuratie and lanthanite) and REE-bearing clay

1993 phases requiring an in-depth study for their combustion behavior. As noted in 2.3,

1994 most of REE mineral grains are included within coal macerals, thus the combustion

1995 atmosphere (reducing or oxidizing) should be considered in the future work for a

1996 better understanding of the REY speciation transformation behavior in coal

1997 combustion process.

1998

1999 Table 17. Overview of reported thermal decomposition and transformation of

2000 common REE-phases during coal combustion. Compiled from Liu et al. [230], Hood

2001 et al. [192], Tõnsuaadu et al. [278].

125
REY speciation Thermal conversion pathways

in coal Reference Phase transformation Oxidation

compound state

Organic REE-ligin REE-ligin → REE-oxides Ce (Ⅲ)→ Ce

associations (Ⅳ)

Carbonates Y2(CO3)3 Y2(CO3)3 → Y2O3 Ce (Ⅲ)→ Ce

(Ⅳ)

Ce2(CO3)3 Ce2(CO3)3 → CeO2 Ce (Ⅲ)→ Ce

(Ⅳ)

(Ce,La)CO3(F,OH) (Ce, La) CO3(F, OH) → Ce (Ⅲ)→ Ce

(Ce, La)O2 (Ⅳ)

REE-doped calcite CaCO3 → CaO Ce (Ⅲ)→ Ce

(Ⅳ)

Phosphates Hydrated YPO4 YPO4 ·2 H2O → YPO4 No change

Hydrated CePO4 CePO4 ·H2O → CePO4 No change

Monazite size reduction via Ce(Ⅲ)→

fragmentation (> 1400 ºC) Ce(Ⅳ)

Xenotime No change No change

Calcium apatite Fluorapatite melts at 1644 Partial

ºC; oxidation

Chlorapatite structure

change; begin 200ºC and melts

at 1530 ºC;

Hydroxyapatite

dehydroxylate at 900 ºC and

decompose above 1200 ºC;

Melting > 1000 ℃

Silicates Zircon No change below 1000 ºC; No change

ZrSiO₄ → ZrO2-t + SiO2 (>

126
 1285 ºC)

2002

2003 After coal burning, fly ash will be exposed to flue gas with varying temperature

2004 along the process path from the flame zone (1300 -1700 ºC) to the inlet of ESP or FF

2005 (around 150 ºC). Flue gas is a complicated oxidizing system, in which the gas

2006 compositions mainly consist of N2, CO2, H2O, and O2; minor amounts of NOX, SOX,

2007 CO, HCl, and Cl2; and trace amounts of gaseous heavy metals such as Hg [279–281].

2008 In the complex environment of flue gas, the REY associations contained within fly

2009 ash could be impacted due to the reactions occurring between fly ash and flue gas.

2010 There are about 30s are available for the occurrences of the gas-solid reactions [282].

2011 A good example of the reactions between flue gas components and REY speciation

2012 can be the presence of Ce (IV) in coal ash, despite that Ce (III) might be the dominant

2013 valence state (> 85%) in coal-derived fly ash [230,232]. Using the µ-XRF mapping

2014 and μ-XANES techniques, Stuckman et al. [232] observed the co-existence of Ce, S, P,

2015 and Ca in eleven coal ash samples, inferring the formation of REE species such as

2016 REE-bearing sulfates and carbonates in the transport process of coal-derived flue gas

2017 in power plants. Bulk Ce XANES spectra of some fly ash samples also matched the

2018 reference Ce (III) chloride spectra, suggesting a chlorination of REY in the flue gas.

2019 In another study by Taggart et al. [233], bulk Y XANES results indicated that Y

2020 coordination states in class F fly ash samples resemble a combination of Y-bearing

2021 carbonates and other species (Y-doped glass and YPO4); whereas Y spectra of class C

127
2022 fly ash shows characteristics of Y2(SO4)3. Potential reactions of REY species in fly

2023 ash with flue gas components can be described as following [42,253]:

2024 RE2O3 (fly ash) + 3 SO2 + 1.5 O2 → RE2 (SO4)3 (4)

2025 RE2O3 (fly ash) + 3 CO2 + 1.5 O2 → RE2 (CO3)3 (5)

2026 RE2O3 (fly ash) + 3 Cl2 → 2 RECl3 + 1.5 O2

2027 (6)

2028 6.4. Further retention mechanisms by fly ash glass

2029 Apart from organic-bound REY, REY in coal is seldom vaporized into gaseous

2030 form in flue gas during coal combustion [230,251]. Most of REY in coal is

2031 transformed into the solid residual (mainly fly ash) (Table 16). As discussed in

2032 section 5, the instrumental analysis and chemical leaching of REY from coal fly ash

2033 demonstrated that a large fraction of REY is captured and enclosed within the

2034 Si-Al-based glassy materials. This would be understandable since the glass is usually

2035 the most abundant constituent in coal ash [283–285]. However, the detailed retention

2036 mechanisms are unclear. This is largely because (1) the details of the decomposition

2037 and chemical reactions of various coal minerals in combustion process, especially for

2038 a particle containing a mixture of mineral inclusions are poorly understood [160]; and

2039 (2) the morphology, chemistry, structure, and size of glass in coal ash itself are

2040 complex to be characterized [286,287], which will influence the reactivity of fly ash

2041 glass [288–291]. The fly ash glass reactivity is important in many areas, for example

2042 the use of fly ash as a pozzolan in concrete [292,293], the environmental leachability

2043 of hazardous elements released from ash landfilling sites [286,290,294], and the

128
2044 recovery of valuable metals as we discussed in this study. Therefore, the following

2045 section will make a short review on the characteristics of fly ash glass, providing a

2046 basis for understanding the close association of REY with glass and for designing an

2047 efficient REY extraction scheme.

2048 6.4.1. Origin and formation

2049 According to ASTM [295], glass is a type of non-crystalline inorganic material

2050 formed upon the melts cooling. Glass is also regarded as a super-cooled liquid product

2051 obtained from high-temperature melt [296]. From the point of mineralogy and

2052 petrology, fly ash is essentially composed of crystalline minerals, unburnt carbon, and

2053 non-crystalline aluminosilicate glass [297]. For coal ash, quantitative X-ray

2054 diffraction researches [285] and petrographic examinations [297–299] indicate that

2055 the Si-Al glass is the dominant phases in most fly ashes; and it is not a uniform and

2056 homogeneous aluminosilicate material. The petrographic examination of coal ash is

2057 typically performed on an epoxy-bound ash pellet with polarized reflected-light,

2058 oil-immersion optics [300,301], with the unburned carbon usually being the main

2059 target [296,337, 340],[300]. In the case of inorganic compositions of coal ash, Hower

2060 et al. [297,299,301] proposed a classification scheme detailing the inorganic

2061 components as amorphous glass, mullite, crystalline silicates, spinel, lime, sulfates,

2062 and oxidized minerals (Figure 34). Under optical microscopy, typical fly ash glass

2063 includes: amorphous solid and frothy phases (Figure 35A), glassy cenospheres

2064 (Figure 35B), and pleiopheres (Figure 35B). The formation of these different glasses

129
2065 in the context of pulverized coal-fired (PC) combustion involves the mineral reactions

2066 in the melting and cooling history of coal ash, which can be summarized as:

2067 thermal decomposition and transformation of aluminosilicate minerals (mainly

2068 clays) in combustion zone to mullite, spinel, cristobalite and liquid phase, which

2069 occurs exothermically above 950 ℃ [302];

2070 melting of the mineral products to liquid droplets on the surface of the coal

2071 particle and coalescence into large particle as the char burning [287,303–305];

2072 and

2073 quenching of the liquid as it leaves the combustion zone in boiler.

2074

2075

2076 Figure 34. Basic nomenclature of fly ash components based on the optical

2077 petrology method. After Hower et al. [299].

2078

130
2079 Feed coal mineral types and temperature are important factors affecting the

2080 formation of silicate melts in pulverized coal combustion process

2081 [287,297,302,303,305,306]. Clay minerals usually dominate the coal mineralogy and

2082 are the main sources for silicate melts and glass formation. The combustion behavior

2083 of clays and other common minerals including quartz, carbonates, and pyrite in coal

2084 are well-understood [297,302,307–309]. As summarized in Table 18 [297], coal

2085 minerals and their decomposition products having high melting point above the boiler

2086 temperature include: (1) crystalline minerals including quartz, mullite, lime, MgO,

2087 etc., and (2) poorly-crystalline materials such as cristobalite. These minerals and their

2088 intermedia products may survive the whole coal combustion process.

2089

2090

2091 Figure 35. Inorganic components of coal fly ash observed by optical microscopy

2092 under polarized light. A: Solid (s) glass spheres. B: Glass cenospheres (c) and

2093 pleiospheres (p). C: Mullite (m). D: Needle-like spinel — magnetite plus Cr, Ni,

2094 Al-spinels. E: Spinel in a sphere. F: Quartz (q). Basic nomenclature of fly ash

131
2095 components based on the optical petrology method. Field of view about 220 μm on

2096 long axis. After Hower et al. [299]. Reproduced with permission from Elsevier.

2097

2098 Table 18. Generalized scheme of behavior of common coal minerals during

2099 combustion and boiling points of the combustion products. After Hower et al. [297].

Mineral matter Combustion process and combustion products Melting points (℃)

Mullite:1850 ℃
Dehydration and sintering; possible mullite; reactions of
K/Na/Ca Feldspar:
Clays products with Fe2O3 from pyrite; CaO from carbonates, SiO2,
1100-1550 ℃
etc.to form glass or minerals

Quartz: 1750 ℃

Relict; reaction with aluminosilicates from clays, with CaO Na-silicate: 1089 ℃
Quartz
and Fe2O3 to form crystalline and molten silicate phases Ca-silicate: 1540 ℃

Carbonate Dissociation to oxides; reaction with silicates CaO: 2570 ℃

Oxidation to hematite (Fe2O3) or magnetite (Fe3O4); reaction Fe2O3: 1565 ℃

Pyrite
with silicates Fe3O4: 1595 ℃

Ca, Na, etc., in CaSO4: 1450 ℃

Mostly to sulfates; can also react with SiO2, etc. form silicates
carboxylates NaSO4: 884 ℃

2100

2101 However, note that the melting point of a mineral may change, largely depending

2102 upon the mineral compositions, mineral-mineral association and the combustion

132
2103 atmosphere (oxidizing or reducing). For example, the clay mineral decomposition

2104 products typically melt at above 1400 ℃, but the melting can occur at around 980 ℃

2105 for SiO2-Al2O3-K2O eutectics. This is often seen in illite and smecite combustion

2106 process, which have certain amounts of alkaline elements [308]. Besides, the

2107 “included” clay minerals usually co-exist with other minerals within individual coal

2108 particles [310,311]. Table 19 gives the types of mineral-mineral association in a U.K.

2109 bituminous coal as determined by CCSEM technique [310]. As expected, clays are the

2110 dominant associations and almost 40% of the mineral matter are present as multiple

2111 mineral occurrences in a single coal particle. Thus, the mineral reactions in coal

2112 burning are not only self-transformations but thermal reactions of their intermediate

2113 products with other minerals and mineraloids. Melting of the thermal inactive

2114 components (quartz and clay decomposition products) with adding modifiers may

2115 occur as the burnout of coal particles (mineral coalescence) [287,305,310,312]. For

2116 example, the SiO2-Al2O3-CaO system has several eutectics at around 1200-1300 ℃

2117 [313–315]. At 1000-1200 ℃, pyrites co-existing with clay minerals in a single coal

2118 particle (reducing atmosphere) have preference to coalesce with their intermedia

2119 products to form complex eutectics: SiO2-FeO, SiO2-Al2O3-FeO-CaO-MgO,

2120 SiO2-Al2O3-FeO-K2O liquid system [308,313,316,317]. This process may also have

2121 influence on the REY mineral transformation, as will be discussed in next section.

2122

133
2123 Table 19. Quantification of mineral-mineral associations in a pulverized bituminous

2124 coal fed into U.K. power plants by the means of CCSEM. Data from Wigley and

2125 Williamson [310].

Mineral association Percentages %

clay 37.1

Clay + clay +other 14

Clay + clay 12.8

Pyrite 7.8

Quartz 3.8

Fe-clay 3.3

Clay + clay + clay 3

Accessory 2.4

Clay + pyrite 2.2

Calcite 2.1

Pyrite + pyrite 1.2

Clay + accessory 1.0

Clay + Fe-clay 1.0

Other 8.1

2126

2127 To summarize, the melting of inorganic components in boiler is the basis for

2128 glass formation in fly ash. When the coal particles burn in a PC boiler, spherical

2129 droplets and inflated molten particles derived from the combustion of clay and other

134
2130 minerals suspended in the combustion zone are the two notable features. As the

2131 particle leave the boiler, the quenching of the spherical molten particles will form the

2132 spherical glassy particle. If the included minerals within coal particles are fully melted,

2133 the gases released from the decomposition of char, carbonates, sulfide minerals, and

2134 vaporization of pore water may be trapped in the liquid microspheres. Then the gases

2135 can bloat the molten melt up to a 500-μm diameter hollow particle [303,318,319]; and

2136 the size and the thickness of the particle (cenosphere) are a function of the melt

2137 viscosity and gas pressure. The pure mineral decomposition and reactions are

2138 important to understand the formation of silicate liquid but cannot capture the real

2139 interactions among two or more different minerals in coal combustion process. The

2140 details of the mineral-mineral reactions, especially for a mineral mixture containing

2141 one or more types of mineral inclusions are poorly understood [302,306,310,320,321].

2142 Besides, fly ash glass usually co-exist with crystalline minerals, for example

2143 magnetite and mullite crystals presenting as a spherical shell of the fly ash glass

2144 [286,299,302]. These minerals are derived from either recrystallization of the melts or

2145 directly from parent mineral transformations in the thermal conversion process

2146 [302,306,322]. The REY phases that are immediately adjacent to or surrounded by the

2147 mineral phases (such as mullite and quartz) within a dense Al-Si glass would offer

2148 physical resistance to the release of REY from fly ash [122,144,231].

2149 6.4.2. Structure and chemistry

2150 Glasses are usually described as “non-crystalline”, “vitreous”, or “amorphous”

2151 solids [296]. These terms are considered to be synonymous, revealing the state of

135
2152 atoms in the glasses are highly disordered. For fly ash glass, it mainly consists of

2153 silicate and/or aluminosilicate glass, and it can be simply described as“… an

2154 inorganic product of fusion which has been cooled to a rigid condition without

2155 crystallizing.” [323]. Many attempts have been made to understand the structure of

2156 silicate melts and glass materials. It is beyond the scope of this study to review the

2157 extensive papers on glass structure and its formation mechanisms, but generally, the

2158 random network model and crystallite model can be used as two typical models for

2159 the description of silicate glass structure [323]. For a complex system such as coal fly

2160 ash, the random network model proposed by Zachariasen [324] that was established

2161 on the basis of crystal chemistry view of Goldschmidt has been widely used

2162 [287,290,291,298]. The glass structure in coal ash and its relationship with glass

2163 composition are very complex, because the inter- and intra-particle heterogeneity of

2164 fly ash brings difficulties to an absolute understanding of fly ash glass

2165 [287,288,290,305,306]. A basic knowledge is that the structure and chemistry of the

2166 fly ash glass vary with the feed coal properties and the boiler operational conditions.

2167 Structurally, the atoms in silicate glass are disordered arranged in long range but

2168 has short- and medium-range order. This short-range order but long-range disorder

2169 atomic model has been extensively examined by various analytical methods, for

2170 example extended X-ray-absorption fine structure, nuclear magnetic resonance, and

2171 nano-beam electron diffraction [325–331]. The structure disorder in fly ash glass has

2172 been considered as a result of three factors [284]: (1) rapid quenching of silicate melts;

2173 (2) network isomorphic substitution; and (3) cation modification. The tetrahedral

136
2174 [SiO4] is the skeleton of the fly ash glass according to the network theory [324].

2175 Usually, though, fly ash glass has a multi-component chemistry and is more complex

2176 than the vitreous SiO2. In the three-dimensional network structure, the co-melting of

2177 silicon oxides with other network formers such as Al3+, Fe3+, and Ti4+ (with

2178 coordination numbers of 3 or 4) can replace Si (disorder due to substitution). The

2179 introduction of network modifiers, such as Na+, K+, Ca2+, Fe2+, and Mg2+, with

2180 coordination numbers ≥ 6 induces de-polymerization of the network (disorder due to

2181 modification). The intermediates, such as Al, can either reinforce (coordination

2182 number of four) or weakens the network (coordination number of five or six),

2183 depending upon the amount of Al and the modifier elements [284,287,305,307].

2184 Accordingly, both of the substitution and cation modification order would prevail in

2185 the structure of fly ash glass [284,308]. For example, the position of 2θ in XRD

2186 patterns of sodium silicate shows a marked shift in the halo to a higher angle (32-34°)

2187 with increasing Na2O content (0-50%) [323]. In another study, Brindle and McCarthy

2188 [305] extensively examined the relationship between ash glass content and glass

2189 chemistry of 75 European fly ashes. Their result suggested a major influence of alkali

2190 elements on the glass formation during coal combustion [305].

2191 The chemistry of fly ash glass is dominated by SiO2, Al2O3, CaO, and Fe2O3;

2192 with minor MgO, K2O, Na2O, P2O5; and trace elements including rare earth elements.

2193 Based on the structural characteristics of the major components, Hemmings and Berry

2194 [284] summarized that the chemistry of fly ash glass locates between silica fume and

2195 the Ca-rich blast furnace slag:

137
2196 lignite ash bituminous ash has 72% network formers (SiO2 + Al2O3 + Fe2O3) and

2197 28% network modifiers (MgO + K2O + Na2O + CaO);

2198 subbituminous ash has 81-84% network formers (SiO2 + Al2O3 + Fe2O3) and

2199 16-19% network modifiers (MgO + K2O + Na2O + CaO); and

2200 bituminous ash has 84-90% network formers (SiO2 + Al2O3 + Fe2O3) and 10-16%

2201 network modifiers (MgO + K2O + Na2O + CaO).

2202 However, the result is a general summary of chemical compositions of fly ash

2203 glass. Ash particles may greatly vary in their compositions from grain-to-grain

2204 (inter-particle difference) in a fly ash derived from the burning of the same coal at a

2205 power plant. This extreme heterogeneity of fly ash is a result of (1) feed coal

2206 properties, i.e., variations in coal rank and seam to seam changes resulting in the coal

2207 mineralogy that are differentiated largely by the alkaline and alkaline earth element

2208 contents, e.g., lignite usually contains more of K, Na, and Ca than bituminous coal

2209 [283,312]; and (2) variations in the formation conditions and collection points at the

2210 power plants, for example, the boiler condition, load, ESP fields, and among others

2211 [250,262,272,332]. Therefore, it would be more difficult to study the distribution

2212 differences of REY among various glasses in coal ash due to its trace amount in coal

2213 ash.

2214 6.4.3. Possible incorporation mechanisms of REY into fly ash glass

2215 Based on the study of REY speciation in coal ash and the glass formation

2216 mechanisms during coal combustion, two capture mechanisms can be broadly

2217 proposed:

138
2218 the incorporation of individual fine REY-bearing mineral grain into Si-Al melts

2219 (physical encapsulation);

2220 the incorporation of rare earth elements into the aluminosilicate glass structures in

2221 the melting process (physiochemical diffusion).

2222 The first capture mechanism has been demonstrated by numerous electron

2223 microscopy studies [122,144,147,192,231,237,333]. Typical REY-bearing mineral

2224 grains such as monazite, xenotime, zircon, and apatite at varying size (from several

2225 microns to nanometers) are observed to be partly or fully embedded within fly ash

2226 glass (Figure 16 to Figure19). Mineral-mineral associations may be an important

2227 factor influencing the process. As indicated in section 2.3, prior to combustion in

2228 coal-fired boilers, these REE-bearing minerals are almost dispersed within coal

2229 organic maceral as included minerals (Figure 6a and 6b); and usually intermixed

2230 with other major coal minerals (such as clay minerals in Figure 6c and 6d). In this

2231 case, incorporation of these REY-bearing mineral grains into glassy components can

2232 occur when the silicate melts form at PC boiler temperature (1300-1700 ℃). The

2233 stable REY mineral grains or its derivative products will naturally enclosed within the

2234 new-formed glass upon the cooling of the melts, at concentration and particle size

2235 usually below the detection limit of XRD and optical microscopy techniques,

2236 respectively [192,237,238].

2237 Another possible pathway for capturing the REY mineral inclusions may result

2238 from mineral collision and coalescence during the burnout of parent char particles

2239 [310]. Mineral coalescence has been used to describe the ash formation mechanism,

139
2240 and conditions for completing the mineral coalescence generally require [334]: (1) at

2241 least two mineral grains included within one coal particle, (2) the local temperature is

2242 enough for mineral melting, and (3) contact between the molten particles. As

2243 indicated in Table 19, individual coal particles often contain more than one mineral

2244 grains, where REY phases may be present in some coal particles. For PC boilers

2245 operating at normal temperature (> 1300 ℃), the included REY-bearing constituents

2246 and the melted silicates will be exposed as the decomposition and oxidation of coal

2247 char particles. The distance between the ash droplets and REY minerals decreases as

2248 the char combustion progress and finally would contact and merge into one large

2249 particle in case of no fragmentation. Due to thermal stability of REY minerals, the

2250 heterogeneity of the coalescent particles may exist, for example monazite and zircon

2251 still persist in the large cenosphere or solid glassy sphere particles after coalescence

2252 (Figure 16). Note that the coalescence of minerals is limited to the coal char

2253 fragmentation. If there is no char fragmentation, all the included minerals in a single

2254 coal particle can be merged into one large particle (full-coalescence) after the

2255 complete burnout of char [335,336]. But char fragmentation is usually seen in coal

2256 particle combustion which may diminish mineral coalescence [337,338]. A negative

2257 relationship between coal particle fragmentation and mineral coalescence can be

2258 observed [142,339]. Thus, three models including full-coalescence,

2259 partial-coalescence, and no-coalescence have been proposed for different scenarios of

2260 mineral coalescence [322,335,340,341]. The extent of mineral coalescence may

2261 account for the REY associations in coal ash, that is, REY minerals are observed to

140
2262 have fully enclosed, partially enclosed, or have no contact with the glassy particles in

2263 coal ash. Besides char fragmentation, the extent of mineral coalescence is also

2264 influence by the boiler type, mineral types and associations, and mineral particle size

2265 [142].

2266 In the case of the second retention mechanism, although most of the

2267 REY-bearing phases in coal have high melting points, the incorporation of

2268 REY-bearing phases into silicate melts (co-melting) might occur if (1) the local

2269 temperature is high up to the melting interactions; (2) active components (e.g.,

2270 Na2O/CaO) exist in the silicate melts, and interactions between physically- associated

2271 minerals may be likely to bring about thermal chemical reactions at a high

2272 temperature. Evidences from the direct determination of REY in various

2273 aluminosilicate glass particles via SHRIMP-RG analysis may support the REY

2274 incorporation mechanism [146]. The REE distribution patterns in different fly ash

2275 glasses, i.e., Al-rich aluminosilicates, Si-rich aluminosilicates, and Ca- and Fe-rich

2276 aluminosilicates, are similar to that of the bulk ash (Figure 36 a). As shown in Figure

2277 36 b, one zircon particle is found in the ash samples but the chondrite-normalized

2278 REE distribution of zircon exhibits a different enrichment mode (HREE-enrichment

2279 type) from the bulk fly ash (LREE-enrichment type). If the discrete mineral dominates

2280 the REY associations in ash, then the normalized distribution pattern cannot be

2281 consistent among various ash particles (see Figure 31 and 36a). Therefore, the

2282 enrichment of REY in coal ash would be explained by the diffusion of REY into the

2283 fly ash glass at the expenses of REE trace phases in coal [146].

141
2284

2285

2286 Figure 36. (a) Distribution pattern of REE in various aluminosilicate glass particles in

2287 a coal fly ash derived from the combustion of Power River Basin coal. (b)

2288 Distribution pattern of REE in zircon in different coal fly ashes from Jungar power

2289 plant in Inner Mongolia, China and from Central Appalachian Basin, eastern United

2290 States. Data was compiled from Kolker et al. [146] and Dai et al. [24].

2291

2292 The incorporation mechanisms are somewhat analogous to the process in making

2293 rare earth silicate glass and in natural igneous system [342,343]. The incorporation of

2294 rare earth oxides (RE2O3) into aluminosilicate glass usually occurs at around 1500 ℃

2295 (melting temperature) [344–352]. This suggest that similar processes may occur

2296 inside the pulverized coal-fired boilers, where the normal operating temperatures are

2297 maintained in 1300-1700 °C. On the basis of reactions for producing REY-bearing

2298 silicate glass, the REE trace phases, at least for some REY-bearing oxides generated

2299 from decomposition of organic-bound REE or carbonate-bound REE in coal might be

2300 dispersed into the silicate melts. Even for the thermal stable REE phosphates (such as

142
2301 monazite), potential reactions for decomposing monazite would happen when active

2302 components such as CaO exists [353]:

2303 3CaO + 2REPO4 = RE2O3 + Ca3(PO4)3 (7)

2304 The decomposition temperature of monazite by adding CaO is 650-780 ℃,

2305 significantly lower than the coal combustion temperature in power plants. Note that

2306 the reaction rate is limited by the solid diffusion rate; however, the multi-component

2307 coal-derived silicate melts (especially containing high amounts of network modifiers)

2308 would favor the reaction because the mass transfer process could be improved within

2309 a liquid system [353].

2310 Local structure of REY in the silicate glass network can provide more details

2311 about the REY incorporation mechanisms. Various techniques, for example magic

2312 angle spinning nuclear magnetic resonance spectroscopy, Raman spectroscopy, X-ray

2313 and neutron diffraction spectroscopy, X-ray absorption fine structure (XAFS), and

2314 atomistic molecular dynamics simulations have been used to understand the structure

2315 of REY-bearing silicate glass [344,345,347,348,350,350,352,354–357]. As illustrated

2316 in a Y-bearing glass model complexes [348] (Figure 37), the primary units of the

2317 network are basically composed of SiO4 and AlO4 groups connected by bridging

2318 oxygen. Yttrium or Sc have coordination numbers ≥ 5, connecting with their

2319 neighboring cation-centered polyhedral SiO4 and AlOx (x = 4, 5, and 6)

2320 predominantly by corner-sharing, then by edge-sharing and rarely connected by the

2321 polyhedral face [344,345,348,349,356]. The introduction of RE3+ into aluminosilicate

2322 will alter the structure and increase the configurational disorder, which are manifested

143
2323 by: (1) the occurrence of higher-coordination Al species, i.e., AlO5 and AlO6, perhaps

2324 at the expense of AlO4 by the virtual reaction, REO7 + AlO4 → REO6 + AlO5

2325 [352,356]; (2) anomalous features of free O ions and Al-O (NBO) over wider

2326 compositional ranges [344,345,356]; and (3) violations of Loewenstein’s rule by the

2327 presence of Al[4]−O−Al[4] bridges [356,358,359]. These short- and medium-range

2328 disorder will be concurrently increased with the filed strength of RE3+ [349].

2329 Generally, RE3+ ions have been considered to play dual roles in, not only modifying

2330 the network by generating NBO, but also balancing the negatively charged AlO4 unit

2331 [348,349,352,356,360]. For example, Charpentier [349] reported that RE3+ in the

2332 alkaline earth-aluminosilicate glass exhibits dual functionality, i.e., decreasing the

2333 polymerization like a modifier at high temperature but strengthening the glass

2334 network at lower temperature. Therefore, when considering the REY diffusion

2335 mechanisms during coal combustion, the combustion temperature and coal

2336 mineralogy including the mineral types and mineral-mineral associations should be

2337 considered.

2338

144
2339

2340 Figure 37. A structural yttrium (Y)-aluminosilicate glass model, showing how the

2341 [YOx] polyhedral interconnect with their neighboring units through O. Unit 1 is an

2342 [YO6] octahedron sharing corners with three [SiO4] tetrahedral, and one corner each

2343 with [AlO4], [AlO5], [YO5] and [YO6] (unit 2) polyhedra, as well as edges with

2344 [AlO5], [YO5] (unit 3), and [YO7] (unit 4) groups. Totally, [YO6] shares seven corners

2345 and three edges with other cation neighbors through O. After Stevensson et al. [348].

2346 Reproduced with permission from Elsevier.

2347

2348 To summarize, during coal combustion, the enrichment of REY in coal ash is

2349 mainly related to the REY incorporation into silicate melts and, thus, the fly ash

2350 glasses are the main hosts for REY. As shown in Figure 38, two retention

2351 mechanisms are proposed in this review. The capture of REY-bearing trace phases

2352 into glass is manifested by the direct observation of micro- and nano-sized

2353 REE-bearing minerals or compounds enclosed (or partially encapsulated) within

145
2354 aluminosilicate glass [27,121,122,145,147,232,236]. While the coal ash is more

2355 enriched in REY, REY-bearing trace phases are less common in coal ash than in the

2356 parent coal. This might be explained by the size reduction of the REY trace phases

2357 into nano-sized particles due to thermal shock [144,192]. The uniform

2358 chondrite-normalized REE distribution patterns for various fly ash constituents versus

2359 bulk ash [146] and the consistent bulk Ce [232] or Y [145,233] XANES of fly ash

2360 samples resembling Ce-, Y-doped aluminosilicates argue against REY occurrences as

2361 randomly captured and distributed sub-micron domains. Diffusion of REY from trace

2362 phases to the aluminosilicate glass might be another REY capture mechanism. Similar

2363 processes could be observed in REY-glass making industry or in natural magmatic

2364 melt system, but the diffusion mechanism for enriching REY in coal ash seems to be

2365 elusive. In a coal-fired power plant, thermal maxima last on the order of 1-2s, whereas

2366 times in diffusion of RE2O3 into silicate glass is generally longer (15 min or more).

2367 Thus, diffusion of trace REE phases might be difficult in coal combustion systems.

2368 Overall, current research data is still insufficient for a full understanding of REY

2369 enrichment mechanisms in aluminosilicate glass of coal ash. It is necessary to know

2370 whether the physical encapsulation or diffusion mechanism dominate the REY

2371 incorporation process during coal combustion. This is because the location and

2372 structure of REY in glasses greatly influence the extraction efficiency of REY from

2373 coal ash, which is reflected in the chemical durability of the aluminosilicate glass.

2374 Rare earth-bearing glasses, if the diffusion mechanism dominates, the extraction of fly

146
2375 ash glass might be more difficult because rare earth cations in the glass network can

2376 increase the water, acid, and alkaline resistance of glass [361,362].

2377

2378
2379 Figure 38. An overview of REY retention mechanisms by ash glass during coal

2380 combustion.

2381

2382 6.5. Overview of REY partitioning mechanisms in coal-fired power plants

2383 An overview of REY partitioning behavior in coal-fired power plants is

2384 illustrated in Figure 39. The mass balance of REY is an average value based on the

2385 field sampling data from Li et al. [28] as an example explaining the REY fate in

2386 power plant. As shown in Figure 39, due to the high boiling temperature, REY in coal

2387 is hardly volatilized and therefore the stack emission of REY from coal-fired power

2388 plants is negligible. In contrast, most of REY will partition into coal ash. Generally,

2389 unlike volatile trace elements (such as Hg, As, and Se), REY entering the coal-fired
147
2390 boiler seems not be systematically fractionated between coal, fly ash, and bottom ash.

2391 Nonetheless, the relative higher mass percentages of REY in fly ash than bottom ash

2392 can be observed, probably resulting from the (1) the ratio of fly ash to bottom ash

2393 produced in coal combustion process is usually about 4:1; and (2) higher REY

2394 concentrations in fly ash than in bottom ash. Likewise, little partitioning between the

2395 ash-collection rows is observed. However, variation in the light versus heavy REE has

2396 been noted between ESP rows and among sized fractions of fly ashes. Currently, it is

2397 still not clear what is behind the phenomenon, and two reasons may account for it: (1)

2398 HREE have preferences for organic matter in coal, leading to HREE enrichment in

2399 finer fly ash particles via a more intense vaporization-condensation process during

2400 coal combustion; (2) finer fly ash particles from back-rows of ESP system may

2401 contain more glass components than that of the coarse particles from the first ESP

2402 rows.

2403 Detailed partitioning mechanisms of REY from coal to the solid ash involve

2404 thermal decomposition of various REE species in coal, mineral fragmentation, and

2405 incorporation of REY species into silicate melts at furnace temperature (Figure 39).

2406 Most of REY in coal occurs as thermally stable minerals such as REY phosphates and

2407 silicates, indicating that some REY trace phases in coal may survive and remain intact

2408 in the combustion process. However, one cannot readily reach the conclusions that

2409 REY will not be subject to vaporize-condensation process and other thermal reactions

2410 during coal combustion, because 1) some of REY in coal occur as thermal unstable

2411 compounds, such REY-bearing carbonates and organic-associated REY (Table 3); 2)

148
2412 different REY have their special thermal properties and reactivities, as can be

2413 reflected in their wide boiling-point range (1193-3454 °C; Figure 27); and 3) even

2414 for the thermal stable REY-bearing minerals, these minerals, along with other “active”

2415 minerals included within one coal particle can coalesce and react at a

2416 high-temperature condition (Figure 38). It is commonly observed that REY occur in

2417 aluminosilicate glass of coal ash [16,23,121,122,145–147,176,192,231–233,237,238].

2418 The partitioning of REY species into aluminosilicate glass is controlled by (1)

2419 variation in coal origins / sources, both resulting in different coal mineralogical

2420 characteristics and different feed coal chemistry affecting the glass formation; (2)

2421 variations during ash formation and collection at coal-fired power plant, such as the

2422 combustion conditions of the boiler, load, ash collection device configurations, etc.

2423 After coal combustion, the interactions between REY in fly ash and flue gas

2424 compositions may further influence the final speciation of REY in coal ash.

2425

149
2426

2427 Figure 39. An overview of REY partitioning behavior during coal combustion in power plants. The mass fractions of REY (%) in coal,

2428 adsorbents (limestone), bottom ash, fly ash, FGD gypsum, and in stack gas are compiled from Li et al. [28].

2429

150
2430 7. Methods for REY recovery from coal ash

2431 7.1. Current REY extraction methods

2432 Many laboratory experimental studies have been conducted in recent years to

2433 recover REY from coal combustion ash, and the investigated methods can be

2434 classified as: (1) physical beneficiation [23,33,108,122,166,175,363], and (2)

2435 hydrometallurgical extraction [30,31,122,364,365]. Moreover, pretreatment by

2436 roasting with the addition of appropriate additives has also been employed in

2437 combination with hydrometallurgical extraction to achieve higher recovery values of

2438 REE from coal combustion ash [107,165,366]. As reported in the literature, the

2439 highest content of REY in coal combustion ash, especially coal fly ash, is usually

2440 associated with fine-particle size, medium-density, and non-magnetic fractions

2441 [23,33,108,175]. Therefore, preconcentrates containing more REY relative to feed ash

2442 materials can be obtained using physical beneficiation methods, such as sieving,

2443 gravity separation, magnetic separation, and froth flotation.

2444 7.2. Physical beneficiation

2445 A summary of physical beneficiation performances is presented in Table 20. As

2446 shown in Table 20, preconcentrates of higher REY content have been generated from

2447 several coal fly ashes through physical beneficiation; however, the enrichment ratios

2448 of REY content in the preconcentrates to that in the feeds are exceedingly low. This

2449 fact suggested the low efficiency of using physical beneficiation to preconcentrate

2450 REY from coal combustion ash. In order to achieve a higher enrichment ratio and

2451 produce a product of higher grade, different physical beneficiation methods can be

151
2452 collectively applied. For example, Blissett et al. [23] obtained a product containing

2453 637 ppm of REY from a fly ash with 505 ppm of REY, using flotation, magnetic

2454 separation, and classifying cyclone to remove unburned carbon, magnetic particles,

2455 and coarse particles, respectively. Pan et al. [108] used a process consisting of size

2456 fractionation at 38 µm and magnetic separation of the undersize material.

2457 Non-magnetic material containing 1025 ppm of REY was obtained from a fly ash

2458 with 782 ppm of REY. Using preconcentrated coal combustion ash as a higher-grade

2459 feedstock reduces the cost for REY recovery. Besides, physical beneficiation can (1)

2460 produce a more uniform feed to the chemical processing, (2) eliminate oversize

2461 material that would be inefficient to process, (3) eliminate carbons that can adsorb

2462 chemicals, and (4) eliminate magnetics which usually have lower REY.

2463

2464 Table 20. A summary of the physical beneficiation of REE from coal fly ash

2465 (Conc., ER, and Re represent the content, enrichment ratio, and recovery of REE in

2466 beneficiation concentrates).

Conc. Re
Sample Location Recovery Method ER Reference
-1
(µg g ) (%)

Fly ash China Sieving 550 1.86 18.6 [24]

Sieving 570 1.19 NA

Fly ash UK Flotation – magnetic [23]

637 1.26 NA
separation – size classification

Fly ash USA Sieving 593 1.16 30 [363]

152
 Density fractionation 522 1.02 95

Magnetic separation 666 1.06 0.7

Froth flotation 920 1.46 2

Sieving 670 1.08 38

Fly ash USA Magnetic separation 630 1.01 96 [175]

Density fractionation 1100 1.77 4

Fly ash USA Magnetic separation - sieving 366 1.13 82 [176]

Fly ash Poland Sieving 271 1.17 50 [166]

Fly ash China Sieving 603 1.13 NA [26]

Fly ash China Sieving 557 1.07 32

[121]
Fly ash China Sieving 499 1.04 57

Sieving 896 1.15 35

Magnetic separation 879 1.12 65

Fly ash China [366]
Density fractionation 855 1.09 30

Sieving – magnetic separation 1025 1.31 31

Fly ash Indonesia Sieving 249 1.10 58

[33]
Fly ash Indonesia Magnetic separation 278 1.12 92

2467 Note: Con, ER, and Re represent content, enrichment ratio, and recovery, respectively.

2468

2469 7.3. Hydrometallurgical extraction

2470 While REY can be enriched from coal combustion ash using physical

2471 beneficiation, in order to generate high-grade REY products hydrometallurgical

153
2472 extraction has to be applied to transfer REY into solution which, in turn, can be

2473 further processed for purification and upgrading. However, due to the inclusion of

2474 REY in the glassy matrix of coal combustion ash, the leaching recovery of REY is

2475 normally low and harsh reaction conditions are required [107,122,365,367]. As

2476 reported in a study performed on a fly ash sample collected from a Chinese power

2477 plant, less than 20% of total rare earth elements were leached using 12M HCl or 16M

2478 HNO3 [367]. The low leaching recovery of REE agrees with findings from sequential

2479 chemical extraction characterization studies reported in literature [26,121,368]. The

2480 majority of REE in coal combustion ash is associated with aluminum-silicate phases

2481 which can only be dissolved by digestion. However, there are some exceptions.

2482 Nearly 100% of rare earth recovery has been achieved for high calcium-containing fly

2483 ashes such as samples derived from Power River Basin coals in the U.S. [364]. Coal

2484 combustion ashes generated from fluidized bed combustors showed relatively high

2485 leachabilities, likely due to the lower combustion temperatures than pulverized or

2486 stoker coal-fired boilers [15].

2487 Given the low leachability of REY in most fly ashes, pretreatment by roasting

2488 with additives (e.g., NaOH, Na2CO3, and CaO) or reacting with strong alkaline

2489 solutions have been applied to dissolve REY from ash into solution

2490 [27,108,165,176,364,366,369]. In both cases, the amorphous glassy structure of coal

2491 combustion ash is destroyed, and difficult-to-leach rare earth minerals such as

2492 monazite are decomposed, collectively contributing towards high REY recoveries. For

2493 example, Taggart et al. [165] compared different roasting additives and found that

154
2494 NaOH and Na2O2 performed better than other chemicals, and >90% of total REY can

2495 be recovered followed by leaching under mild conditions such as 1-2M HNO3. Strong

2496 alkaline solutions, such as 6.25 M NaOH [364], 3 M NaOH [369], and 40% NaOH

2497 [27], are able to decompose aluminum-silicate glassy matrix into soluble species, such

2498 as H2SiO42-, H2Si2O62-, H2Si3O82-, and Al(OH)4- [370,371]. Therefore, REY

2499 encapsulated in the matrix can be released and mostly converted into rare earth

2500 hydroxides under such strong alkaline conditions [372], which can be easily leached

2501 under weakly acidic conditions. Based on the above discussions, the most effective

2502 way to recover REY from coal combustion ash is combining physical and chemical

2503 methods. By taking advantage of the fractionation characteristics of REY in the

2504 particles of different physical natures, a high-grade feed material can be obtained from

2505 coal combustion ash for downstream pretreatment and acid leaching processes. Using

2506 this concept, Pan et al. [108] developed a process, consisting of size classification,

2507 magnetic separation, roasting, and acid leaching units, which enables 80% of REY in

2508 the physical separation product to be recovered.

2509 Nearly all existing studies reported in the literature failed to incorporate

2510 economic analysis. This analysis is essential for the commercial deployment of

2511 technologies and processing flowsheets developed based on laboratory experimental

2512 results. As shown in Table 21, nearly all the hydrometallurgical and pyrometallurgical

2513 methods reported in the literature consume an exceedingly high dosage of chemicals,

2514 suggesting that operating costs for the recovery of REY from coal combustion ash

2515 will be high, leading to low economic viability. The situation is worsened by the fact

155
2516 that as an alternate resource, coal combustion ash has a much lower REY content

2517 relative to conventional rare earth ore deposits. As presented in a report by [373], the

2518 per tonne cost to process fly ash in a commercial plant using alkaline followed by acid

2519 leaching is approximately $140. This cost is too high to make the recovery

2520 economically viable if only REE is produced as the saleable product. In order to

2521 improve the economic viability for REE recovery from coal combustion ash, it needs

2522 to be combined with other beneficial uses, such as aluminum recovery and zeolite

2523 production [374,375]. A technical roadmap for recovering REY from coal combustion

2524 ash is presented in Figure 40. Different options are available, such as

2525 preconcentration – acid leaching – purification, preconcentration – alkaline leaching –

2526 acid leaching – purification, and preconcentration – roasting – leaching – purification.

2527 For a specific ash material, a comprehensive experimental program is required to find

2528 the most efficient and cost-effective flowsheet.

2529

2530 Table 21. Studies reported in the literature with higher recoveries of REY from

2531 coal fly ash.

Sample Location Method Performance References

Fly ash USA Reacting with 6.25 mol/L REY recoveries of up [364]

NaOH at 85℃, followed by to 85% were obtained

acid leaching with 20% HCl

Fly ash USA Reacting with 5 mol/L NaOH REY was enriched [176]

at 100℃ from 325 to 877 ppm

Fly ash USA Roasting with NaOH (1:1 by REY recoveries [233]

weight) at 350℃, followed by of >70% were

156
 acid leaching with 1-2 mol/L obtained for all ash

HNO3 samples

Fly ash China Roasting with Na2CO3 (1:1 by A REY recovery of [107]

weight) at 850℃, followed by 72.78% was obtained

acid leaching with 3 mol/L

HCl

Fly ash China Reacting with a 40% NaOH A REY recovery of [27]

solution at 150℃, followed by 88.15% was obtained

acid leaching with 8 mol/L

HCl

Fly ash USA Roasting with Na2CO3 (1:1 by >90% of REY was [366]

weight) at 850℃, followed by recovered

sequential leaching with water

and 6 mol/L HCl

Fly ash China Reacting with 3 mol/L NaOH A REY recovery of [369]

at 80℃, followed by acid 95.5% was obtained

leaching with 3 mol/L HCl

2532

157
2533

2534 Figure 40. Technical roadmap for REY recovery from coal combustion ash

2535 formulated based on findings from existing studies reported in the literature.

2536

2537 8. Conclusions

2538 Coal ash is one of the most promising sources for rare earth elements and yttrium

2539 (REY). The present work aims for (1) making a comprehensive summary of REY

2540 contents, distribution, speciation, and analytical methods for REY in coal ash; (2) a

2541 deeper understanding of REY partitioning behavior in particular the speciation

2542 transformation and retention mechanisms during coal combustion; and (3) the recovery

2543 methods developed in recent years. The distribution of REY among various streams

2544 across the whole power plants and the retention mechanisms of REY by fly ash glasses

2545 are deeply discussed.

2546 The most fundamental control on REY in coal ash is their amount in the parent

158
2547 coal. REY in coal is most commonly in phosphate minerals but can occur in a number

2548 of other minerals (carbonates, silicates, etc.) as “included” minerals and in organic

2549 association in low-rank coals. The concentration of REY greatly varies in different coal

2550 deposits or even coal seams, and the combustion of “metalliferous coal” can produce

2551 REY-rich ash meeting the cut-off grade for industrial extraction. A detailed knowledge

2552 of REY in feed coals is essential for providing both a background and basis for

2553 recovering REY from the resulting ash as alternative sources.

2554 Methods to characterize REY in coal ash can inform REY extraction strategies.

2555 A summary and comparison on the advantages and limitations of various methods for

2556 determining REY contents and speciation is made in this study. Methods that are used

2557 for total REY content determinations mainly include ICP-MS, followed by ICP-OES,

2558 INAA, and XRF. ICP-MS is recommended for determining REY in coal ash because

2559 of its high sensitivity and multi-elemental capability. INAA could be a reference

2560 method for estimating the precision and accuracy in determination of REY by other

2561 techniques. XRF, as a rapid and nondestructive method for analysis of solid sample, is

2562 suitable for field survey screening coal ash for laboratory examination and REY

2563 recovery. Sequential leaching can quantify REY associations but often have difficulties

2564 in accurately distinguished REY speciation in coal ash. Speciation of relatively

2565 abundant REY such as Ce and Y can be completed using synchrotron-based XANES on

2566 either the bulk- or micro-scale. However, the similar spectral features of reference

2567 materials restrict the results to qualitive descriptions. TEM is useful in analyzing

2568 nanometer-scale REY-bearing domains, which may differ considerably from macro-

159
2569 and micro-scale REY hosts.

2570 The contents and reserve of REY in coal ash are the primary factor

2571 determining the extraction potentials of REY from fly ash. This study updates the

2572 reported value from Ketris and Yudovich [82], which was obtained based on the REY

2573 contents in coal. Rather, current work compiled data of 581 coal ashes produced at

2574 coal-fired power plants from 15 countries around the world. The estimated average

2575 REY contents in world coal-derived fly ash are 435.4 µg g-1. The concentration of REY

2576 in coal ash greatly varies, which heavily depends on the geological origin of feed coal.

2577 REY contents in Chinese coal ash (473 µg g-1) and U.S. coal ash (459.61 µg g-1) are

2578 higher than those in European (278.65 µg g-1) and other countries (298 µg g-1).

2579 Coal-derived fly ashes from China and U.S. are two typical examples showing the

2580 variations of REY with feed coal origins. In comparison to coal ash from other regions,

2581 the central Appalachian basin coal-derived ashes and southwestern China coal ashes

2582 are significantly enriched REY, providing three times the critical REY of total REY

2583 extracted per kg than that of conventional ores.

2584 The speciation of REY in coal ash is another key factor affecting the extraction

2585 efficiencies of REY from coal ash. This study classifies the speciation of REY in coal

2586 ash as amorphous glassy associations, discrete REY minerals or compounds, and as

2587 REY bound with unburned carbon. Rare earth elements are mostly found to be

2588 associated with aluminosilicate glass in coal ash. Two possible forms account for the

2589 Si-Al glassy associations, i.e., individual REY-bearing phases that are closely bound

2590 with aluminosilicate glasses and REY dispersed throughout the fly ash glass. The

160
2591 former association has been extensively demonstrated via electron microscopy

2592 observations or synchrotron-based XANES analysis. Direct determinations by the

2593 means of ion microprobe indicate that REY in fly ash samples is dispersed into Si-Al

2594 glasses formed during melting at boiler temperatures. Discrete REY minerals are

2595 usually more difficult to be detected in coal ash than that in coal, with particle size

2596 ranging from several microns to ~nm, or even to sub-nanometers (as shown by TEM).

2597 The association of REY with unburned carbon is difficult to be distinguished. Only

2598 nanosized particles containing REY are found in amorphous and graphitic carbons

2599 surrounding Al-Si glass and Fe-rich particles in fly ashes.

2600 Rare earth elements and yttrium are not volatile elements; and their mass

2601 distributions among different emission streams in power plants show an enrichment in

2602 the combustion ash, especially in fly ash. There is no significant REY fractionation

2603 among different ESP fly ashes or in the sized ash based on their normalized distribution

2604 pattern relative to the upper continental crust (UCC). However, a significant variation

2605 in the light REE versus heavy REE is noted between ESP rows and among sized

2606 fractions of fly ashes. Variations in the amount and size of REY minerals between and

2607 in the microscopically-identifiable fly ash constituents can contribute to the

2608 differences in REY contents across the ash-collection array.

2609 The speciation of REY in coal can be greatly altered during and after coal

2610 combustion. The thermally stable REY-bearing minerals, such as phosphates, silicates,

2611 etc. in coal may not melt at boiler but, instead, shatter into very fine grains. Thermal

2612 decomposition and oxidation of REY-organic compounds, REY carbonates,

161
2613 water-bearing REY phosphates and sulfates, and other minerals have been observed in

2614 both pure mineral combustion experiments and direct XANES determinations. Most of

2615 REY is retained in aluminosilicate glass with very few REY phases emitted as gaseous

2616 forms. The heat-induced fragmented REY minerals, neo-formed minerals, or the

2617 original REY minerals in coal are partitioned into the Al-Si glass by two mechanisms:

2618 the incorporation of the individual REY phases into the glass as inclusions and the

2619 diffusion of REE phases throughout Si-Al glass structures in the melting process. The

2620 retention of REY by aluminosilicate glass is controlled by the feed-coal mineral types,

2621 mineral-mineral associations, boiler conditions, etc. After coal combustion at boiler

2622 temperatures, the speciation of REY in may be further influenced and modified by the

2623 reactions of REY phases with flue gas components.

2624 REY recovery from coal combustion ash normally requires harsh conditions

2625 because of the encapsulation of REY-rich particles in the glassy matrix. Various

2626 methods have been tested in existing studies, such as physical beneficiation, roasting,

2627 alkaline leaching, and acid leaching. Based on the review, using physical beneficiation

2628 first to generate concentrates with higher REY content followed by hydrometallurgical

2629 processing can improve the economic viability of the recovery.

2630 9. Problems and future outlooks

2631 A number of works have been devoted to the REY contents in coal, but the

2632 contents in coal ash, especially the REY contents in the ponded ash are not known.

2633 Portable XRF has been developed for screening coal ashes for REY recovery in field

2634 survey, but its precision and accuracy should be further improved.

162
2635 While a number of techniques have been used to determine the REY

2636 associations in coal or coal ash, methods that can be used for quantifying various REY

2637 associations in coal or coal ash should be developed. The quantification of REY

2638 minerals in coal by CCSEM is a good start, but the size and the trace amounts of REY

2639 phases (at ~ µg g-1 or sub-µg g-1 level) in heterogenous coal ash make the

2640 quantification results unsatisfactory.

2641 The current TEM studies observed the nanosized REY inclusions occurring

2642 in the bituminous coal-derived fly ash. More research and studies are needed for

2643 confirming those associations by extending the work to higher and lower coal ranks.

2644 Current studies have investigated the high-temperature reactions of pure REY

2645 species are limited to below 1000 ℃ and in an air atmosphere, future work on

2646 behavior of the main REY species in coal at a wider temperature range, a more

2647 oxidizing and reducing atmosphere, and in the presence of acid gas components

2648 should be expanded to modeling the REY transformation behavior.

2649 One of the main reasons for a lower extraction efficiency of REY from coal

2650 ash in particular Class F fly ash is that most of REY partitioned into aluminosilicate

2651 glasses in the combustion process. The ash glass reactivities including their

2652 dissolution rates and chemical durability are determined by the fly ash glass structures

2653 and compositions. Initial fundamental work on the diffusion of REY into glasses in

2654 coal ash should determine what types of glasses host REY in coal ash and the

2655 formation conditions, with consideration of the effects of mineral-coal and

2656 mineral-mineral associations.

163
2657 Recovery of REY from coal combustion ash is viable from a technical point

2658 of view per existing studies; however, the economic viability of REY recovery from

2659 coal combustion ash has not been sufficiently assessed. More efforts are required to

2660 evaluate capital and operating expenses to generate a certain quantity of REY.

2661 Moreover, REY recovery needs to be combined with other beneficial uses of coal

2662 combustion ash, such as aluminum recovery, zeolite preparation, and supplementary

2663 cementitious materials.

164
2664 Acknowledgements

2665 This work was supported by the National Natural Science Foundation of China

2666 (52006082) and Chinese Postdoctoral Science Foundation (2019M662586). This study

2667 was completed as part of U.S. Department of Energy contracts DE-FE0027167 and

2668 DE-FE0026952; U.S. Department of Energy contract DE-FE0029007 to the

2669 University of North Dakota Energy & Environmental Research Center with a

2670 subcontract to the University of Kentucky; and U.S. National Science Foundation

2671 grants CBET-1510965 and CBET-1510861 to Duke University and the University of

2672 Kentucky, respectively. This work used shared facilities at the Virginia Tech National

2673 Center for Earth and Environmental Nanotechnology Infrastructure (NanoEarth), a

2674 member of the National Nanotechnology Coordinated Infrastructure (NNCI),

2675 supported by NSF (ECCS 1542100). Additional microbeam analyses were conducted

2676 at the University of Kentucky Electron Microscopy Center.

2677

2678

2679

2680

2681

2682

2683

2684

165
2685 References

2686 [1] Jensch E. On the metal content of Upper Silesian coals. Chem Ind Berl 1887;10.

2687 [2] Goldschmidt VM. Rare elements in coal ashes. Ind Eng Chem 1935;27:1100–2.

2688 [3] Swaine DJ. Trace elements in coal. Butterworth-Heinemann Ltd, Oxford, UK.;

2689 1991.

2690 [4] Goldschmidt VM. Geochemistry. Clarendon Press, Oxford; 1954.

2691 [5] Thompson AP, Musgrave JR. Germanium, produced as a byproduct, has become of

2692 primary importance. J Miner Met Mater Soc 1952;4:1132–7.

2693 [6] Seredin V V., Finkelman RB. Metalliferous coals: A review of the main genetic and

2694 geochemical types. Int J Coal Geol 2008;76:253–89.

2695 https://doi.org/10.1016/j.coal.2008.07.016.

2696 [7] Seredin V V., Dai S, Sun Y, Chekryzhov IY. Coal deposits as potential alternative

2697 sources for lanthanides and yttrium. Int J Coal Geol 2012;94:67–93.

2698 https://doi.org/10.1016/j.coal.2011.11.001.

2699 [8] Dai S, Yan X, Ward CR, Hower JC, Zhao L, Wang X, et al. Valuable elements in

2700 Chinese coals: a review. Int Geol Rev 2018;60:590–620.

2701 https://doi.org/10.1080/00206814.2016.1197802.

2702 [9] Seredin V V., Dai S, Sun Y, Chekryzhov IY. Coal deposits as promising sources of

2703 rare metals for alternative power and energy-efficient technologies. Appl Geochemistry

2704 2013;31:1–11. https://doi.org/10.1016/j.apgeochem.2013.01.009.

2705 [10] Massari S, Ruberti M. Rare earth elements as critical raw materials: Focus on

2706 international markets and future strategies. Resour Policy 2013;38:36–43.

2707 [11] Chegwidden J, Kingsnorth DJ. Rare Earths—An Evaluation of Current and Furture

2708 Supply. Int Assoc Genocide Sch New Salem, MA, Accessed Apr 2011;3:2018.

2709 [12] Hower JC, Groppo JG, Joshi P, Preda D V., Gamliel DP, Mohler DT, et al.

2710 Distribution of lanthanides, yttrium, and scandium in the pilot-scale beneficiation of fly

2711 ashes derived from eastern kentucky coals. Minerals 2020;10:1–17.

2712 https://doi.org/10.3390/min10020105.

166
2713 [13] Zhang W, Noble A, Yang X, Honaker R. Lithium leaching recovery and

2714 mechanisms from density fractions of an Illinois Basin bituminous coal. Fuel

2715 2020;268:117319. https://doi.org/10.1016/j.fuel.2020.117319.

2716 [14] Hower JC, Qian D, Briot NJ, Hood MM, Eble CF. Mineralogy of a rare earth

2717 element-rich Manchester coal lithotype, Clay County, Kentucky. Int J Coal Geol

2718 2020;220:103413. https://doi.org/10.1016/j.coal.2020.103413.

2719 [15] Honaker RQ, Zhang W, Werner J. Acid Leaching of Rare Earth Elements from Coal

2720 and Coal Ash: Implications for Using Fluidized Bed Combustion to Assist in the Recovery

2721 of Critical Materials. Energy and Fuels 2019;33:5971–80.

2722 https://doi.org/10.1021/acs.energyfuels.9b00295.

2723 [16] Taggart RK, Hower JC, Dwyer GS, Hsu-Kim H. Trends in the Rare Earth Element

2724 Content of U.S.-Based Coal Combustion Fly Ashes. Environ Sci Technol 2016;50:5919–26.

2725 https://doi.org/10.1021/acs.est.6b00085.

2726 [17] Hower JC, Granite EJ, Mayfield DB, Lewis AS, Finkelman RB. Notes on

2727 contributions to the science of rare earth element enrichment in coal and coal combustion

2728 byproducts. Minerals 2016;6. https://doi.org/10.3390/min6020032.

2729 [18] Adamczyk Z, Komorek J, Białecka B, Nowak J, Klupa A. Assessment of the

2730 potential of polish fly ashes as a source of rare earth elements. Ore Geol Rev

2731 2020;124:103638.

2732 [19] Valentim B, Abagiu AT, Anghelescu L, Flores D, French D, Gonçalves P, et al.

2733 Assessment of bottom ash landfilled at Ceplea Valley (Romania) as a source of rare earth

2734 elements. Int J Coal Geol 2019;201:109–26. https://doi.org/10.1016/j.coal.2018.11.019.

2735 [20] Bartoňová L, Serenčíšová J, Čech B. Yttrium partitioning and associations in

2736 coal-combustion ashes prior to and after their leaching in HCl. Fuel Process Technol

2737 2018;173:205–15. https://doi.org/10.1016/j.fuproc.2018.01.011.

2738 [21] Lanzerstorfer C. Fly ash from coal combustion: Dependence of the concentration of

2739 various elements on the particle size. Fuel 2018;228:263–71.

2740 https://doi.org/10.1016/j.fuel.2018.04.136.

2741 [22] Folgueras MB, Alonso M, Fernández FJ. Coal and sewage sludge ashes as sources of

2742 rare earth elements. Fuel 2017;192:128–39. https://doi.org/10.1016/j.fuel.2016.12.019.

167
2743 [23] Blissett RS, Smalley N, Rowson NA. An investigation into six coal fly ashes from

2744 the United Kingdom and Poland to evaluate rare earth element content. Fuel 2014;119:236–

2745 9. https://doi.org/10.1016/j.fuel.2013.11.053.

2746 [24] Dai S, Zhao L, Hower JC, Johnston MN, Song W, Wang P, et al. Petrology,

2747 mineralogy, and chemistry of size-fractioned fly ash from the Jungar power plant, Inner

2748 Mongolia, China, with emphasis on the distribution of rare earth elements. Energy and Fuels

2749 2014;28:1502–14. https://doi.org/10.1021/ef402184t.

2750 [25] Liu H. Compostion of the coal combustion and utilization of the fly ash generated

2751 from two major coal-fired power plants in Chongqing city, Southwest China. Ph.D Thesis.

2752 China University of Minining & Technology, Bejing, 2015.

2753 [26] Pan J, Zhou C, Liu C, Tang M, Cao S, Hu T, et al. Modes of Occurrence of Rare

2754 Earth Elements in Coal Fly Ash: A Case Study. Energy and Fuels 2018;32:9738–43.

2755 https://doi.org/10.1021/acs.energyfuels.8b02052.

2756 [27] Wang Z, Dai S, Zou J, French D, Graham IT. Rare earth elements and yttrium in coal

2757 ash from the Luzhou power plant in Sichuan, Southwest China: Concentration,

2758 characterization and optimized extraction. Int J Coal Geol 2019;203:1–14.

2759 https://doi.org/10.1016/j.coal.2019.01.001.

2760 [28] Li Z, Li X, Zhang L, Li S, Chen J, Feng X, et al. Partitioning of rare earth elements

2761 and yttrium (REY) in five coal-fired power plants in Guizhou, Southwest China. J Rare

2762 Earths 2019;38:1257–64. https://doi.org/10.1016/j.jre.2019.12.013.

2763 [29] Seredin V V. Rare earth element-bearing coals from the Russian Far East deposits.

2764 Int J Coal Geol 1996;30:101–29. https://doi.org/10.1016/0166-5162(95)00039-9.

2765 [30] Kumari A, Parween R, Chakravarty S, Parmar K, Pathak DD, Lee J chun, et al.

2766 Novel approach to recover rare earth metals (REMs) from Indian coal bottom ash.

2767 Hydrometallurgy 2019;187:1–7. https://doi.org/10.1016/j.hydromet.2019.04.024.

2768 [31] Mondal S, Ghar A, Satpati AK, Sinharoy P, Singh DK, Sharma JN, et al. Recovery

2769 of rare earth elements from coal fly ash using TEHDGA impregnated resin.

2770 Hydrometallurgy 2019;185:93–101. https://doi.org/10.1016/j.hydromet.2019.02.005.

168
2771 [32] Wagner NJ, Matiane A. Rare earth elements in select Main Karoo Basin (South

2772 Africa) coal and coal ash samples. Int J Coal Geol 2018;196:82–92.

2773 https://doi.org/10.1016/j.coal.2018.06.020.

2774 [33] Rosita W, Bendiyasa IM, Perdana I, Anggara F. Sequential particle-size and

2775 magnetic separation for enrichment of rare-earth elements and yttrium in Indonesia coal fly

2776 ash. J Environ Chem Eng 2020;8:103575. https://doi.org/10.1016/j.jece.2019.103575.

2777 [34] Dai S, Graham IT, Ward CR. A review of anomalous rare earth elements and yttrium

2778 in coal. Int J Coal Geol 2016;159:82–95. https://doi.org/10.1016/j.coal.2016.04.005.

2779 [35] Hower JC, Eble CF, Dai S, Belkin HE. Distribution of rare earth elements in eastern

2780 Kentucky coals: Indicators of multiple modes of enrichment? Int J Coal Geol 2016;160–

2781 161:73–81. https://doi.org/10.1016/j.coal.2016.04.009.

2782 [36] Hower JC, Ruppert LF, Eble CF. Lanthanide, yttrium, and zirconium anomalies in

2783 the Fire Clay coal bed, Eastern Kentucky. Int J Coal Geol 1999;39:141–53.

2784 https://doi.org/10.1016/S0166-5162(98)00043-3.

2785 [37] Dai S, Xie P, Jia S, Ward CR, Hower JC, Yan X, et al. Enrichment of U-Re-V-Cr-Se

2786 and rare earth elements in the Late Permian coals of the Moxinpo Coalfield, Chongqing,

2787 China: Genetic implications from geochemical and mineralogical data. Ore Geol Rev

2788 2017;80:1–17. https://doi.org/10.1016/j.oregeorev.2016.06.015.

2789 [38] Dai S, Liu J, Ward CR, Hower JC, French D, Jia S, et al. Mineralogical and

2790 geochemical compositions of Late Permian coals and host rocks from the Guxu Coalfield,

2791 Sichuan Province, China, with emphasis on enrichment of rare metals. Int J Coal Geol

2792 2016;166:71–95. https://doi.org/10.1016/j.coal.2015.12.004.

2793 [39] Dai S, Xie P, Ward CR, Yan X, Guo W, French D, et al. Anomalies of rare metals in

2794 Lopingian super-high-organic-sulfur coals from the Yishan Coalfield, Guangxi, China. Ore

2795 Geol Rev 2017;88:235–50. https://doi.org/10.1016/j.oregeorev.2017.05.007.

2796 [40] Zhang W, Rezaee M, Bhagavatula A, Li Y, Honaker R. A Review of the Occurrence

2797 and Promising Recovery Methods of Rare Earth Elements from A Review of the Occurrence

2798 and Promising from Coal and Coal By-Products. Int J Coal Prep Util ISSN 2015;35:295–

2799 330. https://doi.org/10.1080/19392699.2015.1033097.

169
2800 [41] Borzone G, Raggio R, Ferro R. Thermochemistry and reactivity of rare earth metals.

2801 Phys Chem Chem Phys 1999;1:1487–500. https://doi.org/10.1039/a900312f.

2802 [42] Hower JC, Groppo JG, Henke KR, Hood MM, Eble CF, Honaker RQ, et al. Notes on

2803 the potential for the concentration of rare earth elements and yttrium in coal combustion fly

2804 ash. Minerals 2015;5:356–66. https://doi.org/10.3390/min5020356.

2805 [43] Bao Z, Zhao Z. Geochemistry of mineralization with exchangeable REY in the

2806 weathering crusts of granitic rocks in South China. Ore Geol Rev 2008;33:519–35.

2807 [44] McLellan BC, Corder GD, Ali SH. Sustainability of rare earths—An overview of the

2808 state of knowledge. Minerals 2013;3:304–17.

2809 [45] Zhao Z, Cui L, Guo Y, Li H, Cheng F. Recovery of gallium from sulfuric acid leach

2810 liquor of coal fly ash by stepwise separation using P507 and Cyanex 272. Chem Eng J

2811 2020;381:122699. https://doi.org/10.1016/j.cej.2019.122699.

2812 [46] Zhang W, Yang X, Honaker RQ. Association characteristic study and preliminary

2813 recovery investigation of rare earth elements from Fire Clay seam coal middlings. Fuel

2814 2018;215:551–60. https://doi.org/10.1016/j.fuel.2017.11.075.

2815 [47] Damhus T, Hartshorn RM, Hutton AT. Nomenclature of inorganic chemistry:

2816 IUPAC recommendations 2005. Chem Int 2005;27:25–6.

2817 [48] Li C, Lin J. Rare earth fluoride nano-/microcrystals: synthesis, surface modification

2818 and application. J Mater Chem 2010;20:6831–47.

2819 [49] Tan TTY. Rare earth nanotechnology. 1st Editio. Jenny Stanford Publishing; 2012.

2820 https://doi.org/10.1201/b12310.

2821 [50] Lian H, Hou Z, Shang M, Geng D, Zhang Y, Lin J. Rare earth ions doped phosphors

2822 for improving efficiencies of solar cells. Energy 2013;57:270–83.

2823 https://doi.org/10.1016/j.energy.2013.05.019.

2824 [51] Pan H, Ren Y, Fu H, Zhao H, Wang L, Meng X, et al. Recent developments in

2825 rare-earth free wrought magnesium alloys having high strength: A review. J Alloys Compd

2826 2016;663:321–31. https://doi.org/10.1016/j.jallcom.2015.12.057.

2827 [52] Akah A. Application of rare earths in fluid catalytic cracking: A review. J Rare

2828 Earths 2017;35:941–56.

170
2829 [53] Cheisson T, Schelter EJ. Rare earth elements: Mendeleev’s bane, modern marvels.

2830 Science (80- ) 2019;363:489–93. https://doi.org/10.1126/science.aau7628.

2831 [54] Zhao H, Xia J, Yin D, Luo M, Yan C, Du Y. Rare earth incorporated electrode

2832 materials for advanced energy storage. Coord Chem Rev 2019;390:32–49.

2833 https://doi.org/10.1016/j.ccr.2019.03.011.

2834 [55] Gammons CH, Wood SA, Jonas JP, Madison JP. Geochemistry of the rare-earth

2835 elements and uranium in the acidic Berkeley Pit lake, Butte, Montana. Chem Geol

2836 2003;198:269–88. https://doi.org/10.1016/S0009-2541(03)00034-2.

2837 [56] Wood SA, Williams-Jones AE. The aqueous geochemistry of the rare-earth elements

2838 and yttrium 4. Monazite solubility and REE mobility in exhalative massive

2839 sulfide-depositing environments. Chem Geol 1994;115:47–60.

2840 https://doi.org/10.1016/0009-2541(94)90144-9.

2841 [57] Wood SA. The aqueous geochemistry of the rare-earth elements and yttrium. 1.

2842 Review of available low-temperature data for inorganic complexes and the inorganic REE

2843 speciation of natural waters. Chem Geol 1990;88:99–125.

2844 https://doi.org/10.1016/0009-2541(90)90080-Q.

2845 [58] Peccerillo A, Taylor SR. Geochemistry of eocene calc-alkaline volcanic rocks from

2846 the Kastamonu area, Northern Turkey. Contrib to Mineral Petrol 1976;58:63–81.

2847 https://doi.org/10.1007/BF00384745.

2848 [59] Peng S, Kusky TM, Jiang XF, Wang L, Wang JP, Deng H. Geology, geochemistry,

2849 and geochronology of the Miaowan ophiolite, Yangtze craton: Implications for South

2850 China’s amalgamation history with the Rodinian supercontinent. Gondwana Res

2851 2012;21:577–94. https://doi.org/10.1016/j.gr.2011.07.010.

2852 [60] Hidayah NN, Abidin SZ. The evolution of mineral processing in extraction of rare

2853 earth elements using liquid-liquid extraction: A review. Miner Eng 2018;121:146–57.

2854 https://doi.org/10.1016/j.mineng.2018.03.018.

2855 [61] Jordens A, Cheng YP, Waters KE. A review of the beneficiation of rare earth

2856 element bearing minerals. Miner Eng 2013;41:97–114.

2857 https://doi.org/10.1016/j.mineng.2012.10.017.

171
2858 [62] Neumann R, Medeiros EB. Comprehensive mineralogical and technological

2859 characterisation of the Araxá (SE Brazil) complex REE (Nb-P) ore, and the fate of its

2860 processing. Int J Miner Process 2015;144:1–10.

2861 https://doi.org/10.1016/j.minpro.2015.08.009.

2862 [63] Borai EH, El-Ghany MSA, Ahmed IM, M.Hamed M, El-Din AMS, Aly HF.

2863 Modified acidic leaching for selective separation of thorium, phosphate and rare earth

2864 concentrates from Egyptian crude monazite. Int J Miner Process 2016;149:34–41.

2865 https://doi.org/10.1016/j.minpro.2016.02.003.

2866 [64] Schulz B, Merker G, Gutzmer J. Automated SEM mineral liberation analysis (MLA)

2867 with generically labelled EDX spectra in the mineral processing of rare earth element ores.

2868 Minerals 2019;9:527. https://doi.org/10.3390/min9090527.

2869 [65] Migaszewski ZM, Gałuszka A. The characteristics, occurrence, and geochemical

2870 behavior of rare earth elements in the environment: A review. Crit Rev Environ Sci Technol

2871 2015;45:429–71. https://doi.org/10.1080/10643389.2013.866622.

2872 [66] Herrmann H, Nolde J, Berger S, Heise S. Aquatic ecotoxicity of lanthanum - A

2873 review and an attempt to derive water and sediment quality criteria. Ecotoxicol Environ Saf

2874 2016;124:213–38. https://doi.org/10.1016/j.ecoenv.2015.09.033.

2875 [67] Ramos SJ, Dinali GS, Oliveira C, Martins GC, Moreira CG, Siqueira JO, et al. Rare

2876 Earth Elements in the Soil Environment. Curr Pollut Reports 2016;2:28–50.

2877 https://doi.org/10.1007/s40726-016-0026-4.

2878 [68] United States Environmental Protection Agency. Rare Earth Elements : A Review of

2879 Production , Processing , Recycling, and Associated Environmental Issues. EPA

2880 600/R-12/572. 2012.

2881 [69] Rim KT. Effects of rare earth elements on the environment and human health: A

2882 literature review. Toxicol Environ Health Sci 2016;8:189–200.

2883 https://doi.org/10.1007/s13530-016-0276-y.

2884 [70] Zhang FS, Yamasaki S, Kimura K. Rare earth element content in various waste ashes

2885 and the potential risk to Japanese soils. Environ Int 2001;27:393–8.

2886 https://doi.org/10.1016/S0160-4120(01)00097-6.

172
2887 [71] Wei B, Li Y, Li H, Yu J, Ye B, Liang T. Rare earth elements in human hair from a

2888 mining area of China. Ecotoxicol Environ Saf 2013;96:118–23.

2889 https://doi.org/10.1016/j.ecoenv.2013.05.031.

2890 [72] Eskenazy GM. Aspects of the geochemistry of rare earth elements in coal: An

2891 experimental approach. Int J Coal Geol 1999;38:285–95.

2892 https://doi.org/10.1016/S0166-5162(98)00027-5.

2893 [73] Schofield A, Haskin L. Rare-earth distribution patterns in eight terrestrial materials.

2894 Geochim Cosmochim Acta 1964;28:437–46.

2895 https://doi.org/10.1016/0016-7037(64)90117-6.

2896 [74] Seredin V V., Chekryzhov IY. Ore potentiality of the Vanchin Graben, Primorye,

2897 Russia. Geol Ore Depos 2011;53:202–20. https://doi.org/10.1134/S107570151103007X.

2898 [75] Hatch GP. Dynamics in the global market for rare earths. Elements 2012;16:295–6.

2899 https://doi.org/10.2113/gselements.8.5.341.

2900 [76] Paulick H, Machacek E. The global rare earth element exploration boom: An

2901 analysis of resources outside of China and discussion of development perspectives. Resour

2902 Policy 2017;52:134–53. https://doi.org/10.1016/j.resourpol.2017.02.002.

2903 [77] Ding K. The rare earth magnet industry and rare earth price in China. EPJ Web Conf.,

2904 2014. https://doi.org/10.1051/epjconf/20147504005.

2905 [78] Baldi L, Peri M, Vandone D. Clean energy industries and rare earth materials:

2906 Economic and financial issues. Energy Policy 2014;66:53–61.

2907 https://doi.org/10.1016/j.enpol.2013.10.067.

2908 [79] Lucas J, Lucas P, Le Mercier T, Rollat A, Davenport W. Rare Earth Production, Use

2909 and Price. Rare Earths 2015. https://doi.org/10.1016/b978-0-444-62735-3.00002-4.

2910 [80] Oddo G. Die molekularstruktur der Radioaktiven atome. Zeitschrift Für Anorg

2911 Chemie 1914;87:253–68.

2912 [81] Harkins WD. The Evolution of the Elements and the Stability of Complex Atoms. J

2913 Am Chem Soc 1917;39:856–79.

2914 [82] Ketris MP, Yudovich YE. Estimations of Clarkes for Carbonaceous biolithes: World

2915 averages for trace element contents in black shales and coals. Int J Coal Geol 2009;78:135–

2916 48. https://doi.org/10.1016/j.coal.2009.01.002.

173
2917 [83] Boynton W V. Cosmochemistry of the rare earth elements: meteorite studies. In:

2918 Henderson P, editor. Dev. geochemistry, vol. 2, Elsevier; 1984, p. 63–114.

2919 [84] Taylor SR, McLennan SM. The Continental Crust: its Composition and Evolution.

2920 An Examination of the Geochemical Record Preserved in Sedimentary Rocks. 1st ed.

2921 Oxford : Blackwell Scientific; 1985.

2922 [85] Gromet LP, Haskin LA, Korotev RL, Dymek RF. The “North American shale

2923 composite”: Its compilation, major and trace element characteristics. Geochim Cosmochim

2924 Acta 1984. https://doi.org/10.1016/0016-7037(84)90298-9.

2925 [86] Haley BA, Klinkhammer GP, McManus J. Rare earth elements in pore waters of

2926 marine sediments. Geochim Cosmochim Acta 2004;68:1265–79.

2927 https://doi.org/10.1016/j.gca.2003.09.012.

2928 [87] Hanchar JM, van Westrenen W. Rare earth element behavior in zircon-melt systems.

2929 Elements 2007;3:37–42. https://doi.org/10.2113/gselements.3.1.37.

2930 [88] David A. The Rare Earth Elements: Fundamentals and Applications. 1st ed. Wiley

2931 Press; 2013.

2932 [89] Chakhmouradian AR, Wall F. Rare earth elements: Minerals, mines, magnets (and

2933 more). Elements 2012. https://doi.org/10.2113/gselements.8.5.333.

2934 [90] Tyler G. Rare earth elements in soil and plant systems - A review. Plant Soil

2935 2004;267:191–206. https://doi.org/10.1007/s11104-005-4888-2.

2936 [91] Wang W, Qin Y, Sang S, Zhu Y, Wang C, Weiss DJ. Geochemistry of rare earth

2937 elements in a marine influenced coal and its organic solvent extracts from the Antaibao

2938 mining district, Shanxi, China. Int J Coal Geol 2008;76:309–17.

2939 https://doi.org/10.1016/j.coal.2008.08.012.

2940 [92] Dai S, Zhou Y, Zhang M, Wang X, Wang J, Song X, et al. A new type of Nb (Ta)–

2941 Zr (Hf)–REE–Ga polymetallic deposit in the late Permian coal-bearing strata, eastern

2942 Yunnan, southwestern China: Possible economic significance and genetic implications. Int J

2943 Coal Geol 2010;83:55–63.

2944 [93] Li B, Zhuang X, Li J, Querol X, Font O, Moreno N. Geological controls on

2945 mineralogy and geochemistry of the Late Permian coals in the Liulong Mine of the Liuzhi

174
2946 Coalfield, Guizhou Province, Southwest China. Int J Coal Geol 2016;154–155:1–15.

2947 https://doi.org/10.1016/j.coal.2015.12.003.

2948 [94] Huang Q, Talan D, Restrepo JH, Baena OJR, Kecojevic V, Noble A.

2949 Characterization study of rare earths, yttrium, and scandium from various Colombian coal

2950 samples and non-coal lithologies. Int J Coal Geol 2019;209:14–26.

2951 https://doi.org/10.1016/j.coal.2019.04.008.

2952 [95] Dai S, Ren D, Chou CL, Finkelman RB, Seredin V V., Zhou Y. Geochemistry of

2953 trace elements in Chinese coals: A review of abundances, genetic types, impacts on human

2954 health, and industrial utilization. Int J Coal Geol 2012.

2955 https://doi.org/10.1016/j.coal.2011.02.003.

2956 [96] Hu J, Zheng B, Finkelman RB, Wang B, Wang M, Li S, et al. Concentration and

2957 distribution of sixty-one elements in coals from DPR Korea. Fuel 2006;85:679–88.

2958 https://doi.org/10.1016/j.fuel.2005.08.037.

2959 [97] Karayigit AI, Gayer RA, Querol X, Onacak T. Contents of major and trace elements

2960 in feed coals from Turkish coal-fired power plants. Int J Coal Geol 2000;44:169–84.

2961 https://doi.org/10.1016/S0166-5162(00)00009-4.

2962 [98] Finkelman RB. Trace and minor elements in coal. Org. Geochem., Springer; 1993, p.

2963 593–607.

2964 [99] Arbuzov SI, Chekryzhov IY, Finkelman RB, Sun YZ, Zhao CL, Il’enok SS, et al.

2965 Comments on the geochemistry of rare-earth elements (La, Ce, Sm, Eu, Tb, Yb, Lu) with

2966 examples from coals of north Asia (Siberia, Russian far East, North China, Mongolia, and

2967 Kazakhstan). Int J Coal Geol 2019;206:106–20. https://doi.org/10.1016/j.coal.2018.10.013.

2968 [100] Dai S, Hower JC, Finkelman RB, Graham IT, French D, Ward CR, et al. Organic

2969 associations of non-mineral elements in coal : A review. Int J Coal Geol 2020;218:103347.

2970 https://doi.org/10.1016/j.coal.2019.103347.

2971 [101] Dai S, Finkelman RB. Coal as a promising source of critical elements: Progress and

2972 future prospects. Int J Coal Geol 2018;186:155–64.

2973 https://doi.org/10.1016/j.coal.2017.06.005.

175
2974 [102] Chehreh Chelgani S, Hower JC. Relationships between noble metals as potential

2975 coal combustion products and conventional coal properties. Fuel 2018;226:345–9.

2976 https://doi.org/10.1016/j.fuel.2018.04.041.

2977 [103] Dai S, Zhang W, Ward CR, Seredin V V., Hower JC, Li X, et al. Mineralogical and

2978 geochemical anomalies of late Permian coals from the Fusui Coalfield, Guangxi Province,

2979 southern China: Influences of terrigenous materials and hydrothermal fluids. Int J Coal Geol

2980 2013;105:60–84. https://doi.org/10.1016/j.coal.2012.12.003.

2981 [104] Valiev YY, Gofen GI, Pachatzanov DN. Trace elements in the Jurassic anthracites of

2982 the Nazar-Aylok deposit (Central Tadzhikistan). Geochemistry Int 1993;49:243–50.

2983 [105] Wang X, Wang X, Pan Z, Yin X, Chai P, Pan S, et al. Abundance and distribution

2984 pattern of rare earth elements and yttrium in vitrain band of high-rank coal from the Qinshui

2985 basin, northern China. Fuel 2019;248:93–103. https://doi.org/10.1016/j.fuel.2019.03.054.

2986 [106] Ning SZ, Huang S, Zhu S. Mineralization zoning of coal-metal deposits in China (in

2987 Chinese). Chin Sci Bull 2019;64:2501–13. https://doi.org/10.1360/N972019-00377.

2988 [107] Tang M, Zhou C, Pan J, Zhang N, Liu C. Study on extraction of rare earth elements

2989 from coal fly ash through alkali fusion – Acid leaching. Miner Eng 2019;136:36–42.

2990 https://doi.org/10.1016/j.mineng.2019.01.027.

2991 [108] Pan J, Nie T, Vaziri Hassas B, Rezaee M, Wen Z, Zhou C. Recovery of rare earth

2992 elements from coal fly ash by integrated physical separation and acid leaching.

2993 Chemosphere 2020;248:126112. https://doi.org/10.1016/j.chemosphere.2020.126112.

2994 [109] Lin R, Soong Y, Granite EJ. Evaluation of trace elements in U.S. coals using the

2995 USGS COALQUAL database version 3.0. Part I: Rare earth elements and yttrium (REY).

2996 Int J Coal Geol 2018;192:1–13. https://doi.org/10.1016/j.coal.2018.04.004.

2997 [110] Zhao L, Dai S, Nechaev VP, Nechaeva E V, Graham IT, French D, et al. Enrichment

2998 of critical elements (Nb-Ta-Zr-Hf-REE) within coal and host rocks from the Datanhao mine,

2999 Daqingshan Coalfield, northern China. Ore Geol Rev 2019;111:102951.

3000 [111] Dai S, Chekryzhov IY, Seredin V V., Nechaev VP, Graham IT, Hower JC, et al.

3001 Metalliferous coal deposits in East Asia (Primorye of Russia and South China): A review of

3002 geodynamic controls and styles of mineralization. Gondwana Res 2016;29:60–82.

3003 https://doi.org/10.1016/j.gr.2015.07.001.
176
3004 [112] Zhuang X, Su S, Xiao M, Li J, Alastuey A, Querol X. Mineralogy and geochemistry

3005 of the Late Permian coals in the Huayingshan coal-bearing area, Sichuan Province, China.

3006 Int J Coal Geol 2012;94:271–82.

3007 [113] Li B, Zhuang X, Querol X, Moreno N, Córdoba P, Shangguan Y, et al. Geological

3008 controls on the distribution of REY-Zr (Hf)-Nb (Ta) enrichment horizons in late Permian

3009 coals from the Qiandongbei Coalfield, Guizhou Province, SW China. Int J Coal Geol

3010 2020:103604.

3011 [114] Zhao L, Dai S, Graham IT, Wang P. Clay mineralogy of coal-hosted Nb-Zr-REE-Ga

3012 mineralized beds from Late Permian Strata, Eastern Yunnan, SW China: Implications for

3013 paleotemperature and origin of the micro-quartz. Minerals 2016;6:45.

3014 [115] Liu J, Song H, Dai S, Nechaev VP, Graham IT, French D, et al. Mineralization of

3015 REE-Y-Nb-Ta-Zr-Hf in Wuchiapingian coals from the Liupanshui Coalfield, Guizhou,

3016 southwestern China: Geochemical evidence for terrigenous input. Ore Geol Rev

3017 2019;115:103190.

3018 [116] Duan P, Wang W, Liu X, Sang S, Ma M, Zhang W. Differentiation of rare earth

3019 elements and yttrium in different size and density fractions of the Reshuihe coal, Yunnan

3020 Province, China. Int J Coal Geol 2019;207:1–11.

3021 [117] Wang Q, Yang RD, Bao M. Rare earth elements stratigraphic significance in late

3022 Permian coal measure from Bijie City, Guizhou Province, China. J Rare Earths

3023 2008;26:760–4.

3024 [118] Li D, Tang Y, Deng T, Chen K, Liu D. Geochemistry of rare earth elements in

3025 coal—a case study from Chongqing, southwestern China. Energy Explor Exploit

3026 2008;26:355–62.

3027 [119] Dai S, Zhao L, Peng S, Chou CL, Wang X, Zhang Y, et al. Abundances and

3028 distribution of minerals and elements in high-alumina coal fly ash from the Jungar Power

3029 Plant, Inner Mongolia, China. Int J Coal Geol 2010;81:320–32.

3030 https://doi.org/10.1016/j.coal.2009.03.005.

3031 [120] Huang Z, Fan M, Tian H. Rare earth elements of fly ash from Wyoming’s Powder

3032 River Basin coal. J Rare Earths 2020;38:219–26. https://doi.org/10.1016/j.jre.2019.05.004.

177
3033 [121] Pan J, Zhou C, Tang M, Cao S, Liu C, Zhang N, et al. Study on the modes of

3034 occurrence of rare earth elements in coal fly ash by statistics and a sequential chemical

3035 extraction procedure. Fuel 2019;237:555–65. https://doi.org/10.1016/j.fuel.2018.09.139.

3036 [122] Hower JC, Groppo JG, Henke KR, Graham UM, Hood MM, Joshi P, et al. Ponded

3037 and Landfilled Fly Ash as a Source of Rare Earth Elements from a Kentucky Power Plant.

3038 Coal Combust Gasif Prod 2017;9:1–21. https://doi.org/10.4177/ccgp-d-17-00003.1.

3039 [123] Ren D, Zhao F, Dai S, Zhang J, Luo K. Geochemsitry of Trace Elements in coal.

3040 Science Press; 2006.

3041 [124] Lin R, Bank T, Roth E, Granite E, Soong Y. Organic and inorganic association of

3042 rare earth elements in coal. Int J Coal Geol 2017;179:295–301.

3043 [125] Finkelman RB, Palmer CA, Wang P. Quantification of the modes of occurrence of

3044 42 elements in coal. Int J Coal Geol 2018;185:138–60.

3045 [126] Finkelman RB, Dai S, French D. The importance of minerals in coal as the hosts of

3046 chemical elements : A review 2019;212:103251.

3047 [127] Greta E. Sorption of trace elements on xylain: An experimental study. Int J Coal

3048 Geol 2015;150–151:166–9. https://doi.org/10.1016/j.coal.2015.08.013.

3049 [128] Yershov V. Rare earth elements in the coals of the Kizelovskii Coal Basin.

3050 Geochemistry 1961;3:306–8.

3051 [129] Seredin V V, Shpirt M. Rare earth elements in the humic substance of metalliferous

3052 coals. Lithol Miner Resour 1999;34:244–8.

3053 [130] Finkelman R. Modes of occurrence of trace elements in coal.U.S. Geological Survey,

3054 Open-file Reoprot. U.S. Geological Survey, Open-file Reoprot; 1981.

3055 [131] Palmer CA, Crowley SS, Mroczkowski SJ, Finkelman RB, Bullock JH. Quantifying

3056 the modes of occurence of trace elements in coal using sequential laboratory leaching

3057 techniques. US Geol Surv Circ 2004.

3058 [132] Finkelman RB. The origin, occurrence, and distribution of the inorganic constituents

3059 in low-rank coals. Proc. Basic Coal Sci. Work. US Dep. Energy, Houston, TX, 1982, p. 69–

3060 90.

3061 [133] Dai S, Ren D. Occurrence and sequential chemical extraction of rare earth element in

3062 coals and seam roofs. J China Univ Min Technol 2002;31:349–53.
178
3063 [134] Lin R, Bank T, Roth E, Granite E, Soong Y. Organic and inorganic association of

3064 rare earth elements in coal. 34th Annu Int Pittsburgh Coal Conf Coal - Energy, Environ

3065 Sustain Dev PCC 2017 2017;2017-Septe:295–301.

3066 [135] Hower JC, Berti D, Hochella MF, Mardon SM. Rare earth minerals in a “no tonstein”

3067 section of the Dean (Fire Clay) coal, Knox County, Kentucky. Int J Coal Geol 2018;193:73–

3068 86. https://doi.org/10.1016/j.coal.2018.05.001.

3069 [136] Dai S, Luo Y, Seredin V V., Ward CR, Hower JC, Zhao L, et al. Revisiting the late

3070 Permian coal from the Huayingshan, Sichuan, southwestern China: Enrichment and

3071 occurrence modes of minerals and trace elements. Int J Coal Geol 2014;122:110–28.

3072 https://doi.org/10.1016/j.coal.2013.12.016.

3073 [137] Zhao ZG, Tang XY, Li BF. Geochemistry of rare earth elements of coal in Huaibei

3074 coalfield. Geochimica 2000;29:578–83.

3075 [138] Zheng L, Liu G, Zhang H, Gao L, Xue J, Chou CL. Study on geochemistry of rare

3076 earth elements (REEs) in Permian coal from Huaibei coalfield. Geol J China Univ

3077 2006;12:41–52.

3078 [139] Dai S, Zhang W, Seredin V V., Ward CR, Hower JC, Song W, et al. Factors

3079 controlling geochemical and mineralogical compositions of coals preserved within marine

3080 carbonate successions: A case study from the Heshan Coalfield, southern China. Int J Coal

3081 Geol 2013;109–110:77–100. https://doi.org/10.1016/j.coal.2013.02.003.

3082 [140] Zygarlicke CJ, Folkedahl BC, Nyberg CM, Feole IK, Kurz BA, Theakar NL, et al.

3083 Rare Eare elements (REEs) in U.S. coal-based resources: sampling, characterization, and

3084 round-robin interlaboratory

3085 study.https://edx.netl.doe.gov/dataset/rare-earth-elements-in-u-s-coal-based-resources,

3086 rfp-10982-fe0029007-final-report-RRIS-UND.pdf (acce. 2019.)

3087 [141] Wall TF. Mineral matter transformations and ash deposition in pulverised coal

3088 combustion. Symp Combust 1992;24:1119–26.

3089 https://doi.org/10.1016/S0082-0784(06)80132-3.

3090 [142] Xu M, Yu D, Yao H, Liu X, Qiao Y. Coal combustion-generated aerosols:

3091 Formation and properties. Proc Combust Inst 2011;33:1681–97.

3092 https://doi.org/10.1016/j.proci.2010.09.014.
179
3093 [143] Hower JC, Berti D, Hochella MF, Rimmer SM, Taulbee DN. Submicron-scale

3094 mineralogy of lithotypes and the implications for trace element associations: Blue Gem coal,

3095 Knox County, Kentucky. Int J Coal Geol 2018;192:73–82.

3096 https://doi.org/10.1016/j.coal.2018.04.006.

3097 [144] Hower JC, Qian D, Briot NJ, Santillan-Jimenez E, Hood MM, Taggart RK, et al.

3098 Nano-scale rare earth distribution in fly ash derived from the combustion of the fire clay coal,

3099 Kentucky. Minerals 2019;9:206.

3100 [145] Liu P, Huang R, Tang Y. Comprehensive Understandings of Rare Earth Element

3101 (REE) Speciation in Coal Fly Ashes and Implication for REE Extractability. Environ Sci

3102 Technol 2019;53:5369–77. https://doi.org/10.1021/acs.est.9b00005.

3103 [146] Kolker A, Scott C, Hower JC, Vazquez JA, Lopano CL, Dai S. Distribution of rare

3104 earth elements in coal combustion fly ash, determined by SHRIMP-RG ion microprobe. Int

3105 J Coal Geol 2017;184:1–10. https://doi.org/10.1016/j.coal.2017.10.002.

3106 [147] Montross SN, Verba CA, Chan HL, Lopano C. Advanced characterization of rare

3107 earth element minerals in coal utilization byproducts using multimodal image analysis. Int J

3108 Coal Geol 2018;195:362–72. https://doi.org/10.1016/j.coal.2018.06.018.

3109 [148] Karr Jr C. Analytical methods for coal and coal products. vol. III. 1st ed. Academic

3110 Press, Inc., New York, NY; 1979.

3111 [149] Vourvopoulos G. Elemental Analysis of Coal and Its By-Products. 1st ed. World

3112 Scientific; 1992.

3113 [150] Huggins FE. Overview of analytical methods for inorganic constituents in coal. Int J

3114 Coal Geol 2002;50:169–214. https://doi.org/10.1016/S0166-5162(02)00118-0.

3115 [151] Bullock JH, Cathcart JD, Betterton WJ. Analytical methods utilized by the United

3116 States Geological Survey for the analysis of coal and coal combustion by-products. U.S.

3117 Geological Survey; 2002.

3118 [152] American Society of Testing Materials. ASTM D6357-19, Standard Test Methods

3119 for Determination of Trace Elements in Coal , Coke , and Combustion Residues from Coal

3120 Utilization Processes by Inductively Coupled Plasma Atomic Emission Spectrometry ,

3121 Inductively Coupled Plasma Mass Spectrometry. 100 Barr Harbor Drive, PO Box C700,

180
3122 West Conshohocken, PA 19428-2959. United States: 2020.

3123 https://doi.org/10.1520/D6357-11.2.

3124 [153] Phuoc TX, Wang P, McIntyre D. Detection of rare earth elements in Powder River

3125 Basin sub-bituminous coal ash using laser-induced breakdown spectroscopy (LIBS). Fuel

3126 2016;163:129–32. https://doi.org/10.1016/j.fuel.2015.09.034.

3127 [154] Riley JT, Werth JL, Lewis LM, Mertens MJ. ICP analysis of coal slurries. Elem Anal

3128 Coal Its By-Products 1992:124–44.

3129 [155] McCurdy DL, Wichman MD, Fry RC. Rapid coal analysis. Part II: Slurry

3130 atomization DCP emission analysis of NBS coal. Appl Spectrosc 1985;39:984–8.

3131 [156] Ng KC, Zerezghi M, Caruso JA. Direct Powder Injection of NBS Coal Fly Ash in

3132 Inductively Coupled Plasma Atomic Emission Spectrometry with Rapid Scanning

3133 Spectrometric Detection. Anal Chem 1984;56:417–21.

3134 https://doi.org/10.1021/ac00267a026.

3135 [157] Querol X, Whateley MKG, Fernández-Turiel JL, Tuncali E. Geological controls on

3136 the mineralogy and geochemistry of the Beypazari lignite, central Anatolia, Turkey. Int J

3137 Coal Geol 1997;33:255–71. https://doi.org/10.1016/S0166-5162(96)00044-4.

3138 [158] Liu J, Dai S, He X, Hower JC, Sakulpitakphon T. Size-dependent variations in fly

3139 ash trace element chemistry: Examples from a Kentucky power plant and with emphasis on

3140 rare earth elements. Energy and Fuels 2017;31:438–47.

3141 https://doi.org/10.1021/acs.energyfuels.6b02644.

3142 [159] Zhang W, Honaker R. Enhanced leachability of rare earth elements from calcined

3143 products of bituminous coals. Miner Eng 2019;142:105935.

3144 https://doi.org/10.1016/j.mineng.2019.105935.

3145 [160] Zhao L, Dai S, Finkelman RB, French D, Graham IT, Yang Y, et al. Leaching

3146 behavior of trace elements from fly ashes of five Chinese coal power plants. Int J Coal Geol

3147 2020;219:103381. https://doi.org/10.1016/j.coal.2019.103381.

3148 [161] Ma Z, Shan X, Cheng F. Distribution Characteristics of Valuable Elements, Al, Li,

3149 and Ga, and Rare Earth Elements in Feed Coal, Fly Ash, and Bottom Ash from a 300 MW

3150 Circulating Fluidized Bed Boiler. ACS Omega 2019;4:6854–63.

3151 https://doi.org/10.1021/acsomega.9b00280.
181
3152 [162] Zhang K, Pi Z. MT/T 1014-2006. Determination of major and miner elements in coal

3153 ash by Inductively Coupled Plasma-Atomic emission spectrometry. China National Coal

3154 Quality Supervision Testing Center; Beijing: 2007.

3155 [163] Zhang W, Honaker R. Characterization and recovery of rare earth elements and other

3156 critical metals (Co, Cr, Li, Mn, Sr, and V) from the calcination products of a coal refuse

3157 sample. Fuel 2020;267:117236. https://doi.org/10.1016/j.fuel.2020.117236.

3158 [164] Zhang W, Honaker R. Calcination pretreatment effects on acid leaching

3159 characteristics of rare earth elements from middlings and coarse refuse material associated

3160 with a bituminous coal source. Fuel 2019;249:130–45.

3161 https://doi.org/10.1016/j.fuel.2019.03.063.

3162 [165] Taggart RK, Hower JC, Hsu-Kim H. Effects of roasting additives and leaching

3163 parameters on the extraction of rare earth elements from coal fly ash. Int J Coal Geol

3164 2018;196:106–14. https://doi.org/10.1016/j.coal.2018.06.021.

3165 [166] Lanzerstorfer C. Pre-processing of coal combustion fly ash by classification for

3166 enrichment of rare earth elements. Energy Reports 2018;4:660–3.

3167 https://doi.org/10.1016/j.egyr.2018.10.010.

3168 [167] Hower JC, Hood MM, Taggart RK, Hsu-Kim H. Chemistry and petrology of paired

3169 feed coal and combustion ash from anthracite-burning stoker boilers. Fuel 2017;199:438–46.

3170 https://doi.org/10.1016/j.fuel.2017.03.007.

3171 [168] Karayigit AI, Ozlem Y, Iserli S, Querol X, Mastalerz M, Oska R, et al. Mineralogy

3172 and geochemistry of feed coals and combustion residues from tunçbilek and eyitömer

3173 coal-fired power plants in western Turkey. Coal Combust Gasif Prod 2019:18–31.

3174 https://doi.org/10.4177/CCGP-D-18-00011.1.

3175 [169] Li J, Zhuang X, Querol X, Font O, Moreno N, Zhou J. Environmental geochemistry

3176 of the feed coals and their combustion by-products from two coal-fired power plants in

3177 Xinjiang Province, Northwest China. Fuel 2012;95:446–56.

3178 https://doi.org/10.1016/j.fuel.2011.10.025.

3179 [170] Querol X, Fernández J, Soler A. Trace elements in coal and their behaviour during

3180 combustion in a large power station. Fuel 1995;74:331–43.

182
3181 [171] Querol X, Juan R, Lopez-soler A, Fernandez-turiel JL, Ruiz CR. Mobility of trace

3182 elements from coal and combustion wastes. Fuel 1996;75:821–38.

3183 [172] Yan X, Dai S, Graham IT, He X, Shan K, Liu X. Determination of Eu concentrations

3184 in coal, fly ash and sedimentary rocks using a cation exchange resin and inductively coupled

3185 plasma mass spectrometry (ICP-MS). Int J Coal Geol 2018;191:152–6.

3186 https://doi.org/10.1016/j.coal.2018.03.009.

3187 [173] Krishna MVB, Venkateswarlu G, Karunasagar D. Development of a simple and

3188 robust microwave-assisted decomposition method for the determination of rare earth

3189 elements in coal fly ash by ICP-OES. Anal Methods 2017;9:2031–40.

3190 https://doi.org/10.1039/c7ay00286f.

3191 [174] American Society of Testing Materials. D4503-08,Standard Practice for Dissolution

3192 of Solid Waste by Lithium Metaborate Fusion. 100 Barr Harbor Drive, PO Box C700, West

3193 Conshohocken, PA 19428-2959. United States: ASTM International; 2008.

3194 [175] Lin R, Howard BH, Roth EA, Bank TL, Granite EJ, Soong Y. Enrichment of rare

3195 earth elements from coal and coal by-products by physical separations. Fuel 2017;200:506–

3196 20. https://doi.org/10.1016/j.fuel.2017.03.096.

3197 [176] Lin R, Stuckman M, Howard BH, Bank TL, Roth EA, Macala MK, et al. Application

3198 of sequential extraction and hydrothermal treatment for characterization and enrichment of

3199 rare earth elements from coal fly ash. Fuel 2018;232:124–33.

3200 https://doi.org/10.1016/j.fuel.2018.05.141.

3201 [177] Zhang Y, Jiang Z, He M, Hu B, Thompson RL, Bank T, et al. Determination of Trace

3202 Rare Earth Elements in Coal Fly Ash and Atmospheric Particulates by Electrothermal

3203 Vaporization Inductively Coupled Plasma Mass Spectrometry with Slurry Sampling.

3204 Environ Pollut 2007;148:459–67. https://doi.org/10.1039/c1ja10140d.

3205 [178] Wei Q, Song W. Mineralogical and Chemical Characteristics of Coal Ashes from

3206 Two High-Sulfur Coal-Fired Power Plants in Wuhai, Inner Mongolia, China. Minerals

3207 2020;10. https://doi.org/10.3390/min10040323.

3208 [179] Yang J, Zhao Y, Zyryanov V, Zhang J, Zheng C. Physical-chemical characteristics

3209 and elements enrichment of magnetospheres from coal fly ashes. Fuel 2014;135:15–26.

3210 https://doi.org/10.1016/j.fuel.2014.06.033.
183
3211 [180] Wu G, Wang T, Zhang Y, Jiawe W, Weiping P. Study on the Enrichment of Rare

3212 Earth Elements Between Coals and Their By-products at Coal-fired Power Plants. Proc

3213 CSEE 2020;40:1963–71.

3214 [181] Meier AL, Lichte FE, Briggs PH, Bullock Jr JH. Coal ash by inductively coupled

3215 plasma-atomic emission spectrometry and inductively coupled plasma-mass spectrometry.

3216 Anal Methods Man Miner Resour Surv Progr US Geol Surv 1996:109.

3217 [182] Lichte FE, Meier AL, Crock JG. Determination of the Rare-Earth Elements in

3218 Geological Materials by Inductively Coupled Plasma Mass Spectrometry. Anal Chem

3219 1987;59:1150–7. https://doi.org/10.1021/ac00135a018.

3220 [183] Ardini F, Soggia F, Rugi F, Udisti R, Grotti M. Comparison of inductively coupled

3221 plasma spectrometry techniques for the direct determination of rare earth elements in digests

3222 from geological samples. Anal Chim Acta 2010;678:18–25.

3223 https://doi.org/10.1016/j.aca.2010.07.036.

3224 [184] Zhang W, Noble A. Mineralogy characterization and recovery of rare earth elements

3225 from the roof and floor materials of the Guxu coalfield. Fuel 2020;270.

3226 https://doi.org/10.1016/j.fuel.2020.117533.

3227 [185] Li Z, Clemens AH, Moore TA, Gong D, Weaver SD, Eby N. Partitioning behaviour

3228 of trace elements in a stoker-fired combustion unit: An example using bituminous coals

3229 from the Greymouth coalfield (Cretaceous), New Zealand. Int J Coal Geol 2005;63:98–116.

3230 https://doi.org/10.1016/j.coal.2005.02.007.

3231 [186] Eze CP, Fatoba O, Madzivire G, Ostrovnaya TM, Petrik LF, Frontasyeva M V., et al.

3232 Elemental composition of fly ash: a comparative study using nuclear and related analytical

3233 techniques. Chemistry-Didactics-Ecology-Metrology 2014;18:19–29.

3234 https://doi.org/10.2478/cdem-2013-0014.

3235 [187] Djingova R, Ivanova J. Determination of rare earth elements in soils and sediments

3236 by inductively coupled plasma atomic emission spectrometry after cation-exchange

3237 separation. Talanta 2002;57:821–9.

3238 [188] Jarvis I, Jarvis KE. Rare-earth element geochemistry of standard sediments: A study

3239 using inductively coupled plasma spectrometry. Chem Geol 1985.

3240 https://doi.org/10.1016/0009-2541(85)90078-6.
184
3241 [189] Roelandts I, Deblond A. Rare-earth element composition of Devonian sediments

3242 from southern Belgium: application of an inductively coupled plasma-atomic emission

3243 spectrometry method. Chem Geol 1992;95:167–76.

3244 https://doi.org/10.1016/0009-2541(92)90053-8.

3245 [190] Léveillé L, Drouin L, Constantin M, Bazin C, Larivière D. Scandium analysis in

3246 silicon-containing minerals by inductively coupled plasma tandem mass spectrometry.

3247 Spectrochim Acta Part B 2016; 118: 112-18. http://dx.doi.org/10.1016/j.sab.2016.02.014

3248 [191] Adamczyk Z, Komorek J, Lewandowska M, Nowak J, Białecka B, Całusz-Moszko J,

3249 et al. Ashes from bituminous coal burning in fluidized bed boilers as a potential source of

3250 rare earth elements. Gospod Surowcami Miner / Miner Resour Manag 2018;34:21–36.

3251 https://doi.org/10.24425/118652.

3252 [192] Madison M, Ross K, Ryan C, Kevin R, John G, Jason M, et al. Rare Earth Element

3253 Distribution in Fly Ash Derived from the Fire Clay Coal , Kentucky. Coal Combust Gasif

3254 Prod 2017;9:22–33. https://doi.org/10.4177/CCGP-D-1.

3255 [193] Vassilev S V., Vassileva CG. Contents and associations of rare earth elements and

3256 yttrium in biomass ashes. Fuel 2020;262:116525.

3257 https://doi.org/10.1016/j.fuel.2019.116525.

3258 [194] Shao L, Chang L, Finkelman RB, Wang W, Liu J, Li J, et al. Distribution of rare

3259 earth elements in PM10 emitted from burning coals and soil-mixed coal briquettes. J

3260 Environ Sci (China) 2020;97:96–101. https://doi.org/10.1016/j.jes.2020.05.010.

3261 [195] Middleton A, Park DM, Jiao Y, Hsu-kim H. Major element composition controls

3262 rare earth element solubility during leaching of coal fly ash and coal by-products. Int J Coal

3263 Geol 2020:103532. https://doi.org/10.1016/j.coal.2020.103532.

3264 [196] Tang M, Zhou C, Zhang N, Pan J, Cao S, Hu T, et al. Extraction of rare earth

3265 elements from coal fly ash by alkali fusion–acid leaching: Mechanism analysis. Int J Coal

3266 Prep Util 2019;1:1–20. https://doi.org/10.1080/19392699.2019.1623206.

3267 [197] Ganjali MR, Gupta VK, Faridbod F, Norouzi P. Overview of Lanthanides Series

3268 Determination. Lanthanides Ser Determ by Var Anal Methods 2016:83–9.

3269 https://doi.org/10.1016/b978-0-12-804704-0.00004-9.

185
3270 [198] Zawisza B, Pytlakowska K, Feist B, Polowniak M, Kita A, Sitko R. Determination of

3271 rare earth elements by spectroscopic techniques: A review. J Anal At Spectrom

3272 2011;26:2373–90. https://doi.org/10.1039/c1ja10140d.

3273 [199] Sandoval Rojano WJ, dos Anjos T, Duyck CB, Saint’Pierre TD. Determination of

3274 rare earth elements in environmental samples with high concentrations of barium by

3275 quadrupole inductively coupled plasma mass spectrometry. Microchem J 2019;149:1–6.

3276 https://doi.org/10.1016/j.microc.2019.104026.

3277 [200] Thompson RL, Bank T, Roth E, Granite E. Resolution of rare earth element

3278 interferences in fossil energy by-product samples using sector-field ICP-MS. Fuel

3279 2016;185:94–101. https://doi.org/10.1016/j.fuel.2016.07.093.

3280 [201] Jarvis KE, Gray AL, McCurdy E. Avoidance of spectral interference on europium in

3281 inductively coupled plasma mass spectrometry by sensitive measurement of the doubly

3282 charged ion. J Anal At Spectrom 1989;4:743–7. https://doi.org/10.1039/JA9890400743.

3283 [202] Raut NM, Huang LS, Lin KC, Aggarwal SK. Uncertainty propagation through

3284 correction methodology for the determination of rare earth elements by quadrupole based

3285 inductively coupled plasma mass spectrometry. Anal Chim Acta 2005;530:91–103.

3286 https://doi.org/10.1016/j.aca.2004.08.067.

3287 [203] Vaughan MA, Horlick G. Correction procedures for rare earth element analyses in

3288 inductively coupled plasma-mass spectrometry. Appl Spectrosc 1990;44:587–93.

3289 [204] Riondato J, Vanhaecke F, Moens L, Dams R. Determination of rare earth elements in

3290 environmental matrices by sector-field inductively coupled plasma mass spectrometry.

3291 Fresenius J Anal Chem 2001;370:544–52.

3292 [205] Vaughan MA, Horlick G. Effect of sampler and skimmer orifice size on analyte and

3293 analyte oxide signals in inductively coupled plasma-mass spectrometry. Spectrochim Acta

3294 Part B At Spectrosc 1990;45:1289–99.

3295 [206] Louie H, Soo SY-P. Use of nitrogen and hydrogen in inductively coupled plasma

3296 mass spectrometry. J Anal At Spectrom 1992;7:557–64.

3297 [207] Tao H, Miyazaki A. Decrease of solvent water loading in inductively coupled

3298 plasma mass spectrometry by using a membrane separator. J Anal At Spectrom 1995;10:1–

3299 5.
186
3300 [208] Günther D, Longerich HP, Jackson SE, Forsythe L. Effect of sampler orifice

3301 diameter on dry plasma inductively coupled plasma mass spectrometry (ICP-MS)

3302 backgrounds, sensitivities, and limits of detection using laser ablation sample introduction.

3303 Fresenius J Anal Chem 1996;355:771–3.

3304 [209] Hu Y, Vanhaecke F, Moens L, Dams R, Del Castilho P, Japenga J. Determination of

3305 the aqua regia soluble content of rare earth elements in fertilizer, animal fodder phosphate

3306 and manure samples using inductively coupled plasma mass spectrometry. Anal Chim Acta

3307 1998;373:95–105.

3308 [210] Hevesy G, Levi H. Action of Slow Neutrons on Rare Earth Elements. Nature

3309 1936;137:185.

3310 [211] Greenberg RR, Mackey EA, Becker DA. Radioanalytical Methods : Data Quality ,

3311 Method Validation and Use of Standard Reference Materials ( continued )The application of

3312 instrumental neutron activation analysis for the certification of the new NIST fly ash SRM. J

3313 Radioanal Nucl Chem 1995;193:7–14.

3314 [212] Tu S, Hanf W, Lieser KH. Multielement analysis of Chinese environmental

3315 reference materials by instrumental neutron activation analysis. J Radioanal Nucl Chem

3316 Artic 1984;83:407–12.

3317 [213] Smolka-Danielowska D. Rare earth elements in fly ashes created during the coal

3318 burning process in certain coal-fired power plants operating in Poland - Upper Silesian

3319 Industrial Region. J Environ Radioact 2010;101:965–8.

3320 https://doi.org/10.1016/j.jenvrad.2010.07.001.

3321 [214] Gluskoter HJ, Ruch RR, Miller WG, Cahill RA, Dreher GB. Trace elements in coal:

3322 occurrence and distribution. Final report,June 1974--June 1976. Urbana, IL: Illinois State

3323 Geological Survey; 1977.

3324 [215] Kuhn JK, Harfst WF, Shimp NF. X-ray fluorescence analysis of whole coal. Trace

3325 Elem. Fuel, ACS Publications; 1975, p. 66–73.

3326 [216] Prather JW, Guin JA, Tarrer AR. X-ray fluorescence analysis of trace elements in

3327 coal and solvent refined coal. Energy & Fuels 1990;4:440–2.

3328 [217] Bettinelli M, Taina P. Rapid analysis of coal fly ash by x‐ray fluorescence

3329 spectrometry. X‐Ray Spectrom 1990;19:227–32. https://doi.org/10.1002/xrs.1300190505.

187
3330 [218] Bryan RC, Richers D, Ackman T. Study on the Utilization of Portable Hand-Held

3331 XRF Spectroscopy as a Screening Tool for Rare Earth Elements in Coal and Coal Waste

3332 Products. Document No: 114-910178X-100-REP-R001-01. Tetra Tech, Inc., Pittsburgh, PA

3333 661: 2015.

3334 [219] Smoliński A, Stempin M, Howaniec N. Determination of rare earth elements in

3335 combustion ashes from selected Polish coal mines by wavelength dispersive X-ray

3336 fluorescence spectrometry. Spectrochim Acta - Part B At Spectrosc 2016;116:63–74.

3337 https://doi.org/10.1016/j.sab.2015.12.005.

3338 [220] Chuparina E V., Chubarov VM, Paradina LP. A comparative determination of major

3339 components in coal power plant wastes by wavelength dispersive X-ray fluorescence using

3340 pellet and fused bead specimens. Appl Radiat Isot 2019;152:162.

3341 https://doi.org/10.1016/j.apradiso.2019.06.040.

3342 [221] Yang X, Kozar D, Gorski D, Marchese A, Pagnotti J, Sutterlin R, et al. Using yttrium

3343 as an indicator to estimate total rare earth element concentration: a case study of

3344 anthracite-associated clays from northeastern Pennsylvania. Int J Coal Sci Technol 2020.

3345 https://doi.org/10.1007/s40789-020-00316-1.

3346 [222] Hughes R, Barker SLL. Using portable XRF to infer adularia halos within the Waihi

3347 Au-Ag system, New Zealand. Geochemistry Explor Environ Anal 2018;18:97–108.

3348 [223] Richards MJ. Realising the potential of portable XRF for the geochemical

3349 classification of volcanic rock types. J Archaeol Sci 2019;105:31–45.

3350 [224] Higueras P, Oyarzun R, Iraizoz JM, Lorenzo S, Esbrí JM, Martínez-Coronado A.

3351 Low-cost geochemical surveys for environmental studies in developing countries: Testing a

3352 field portable XRF instrument under quasi-realistic conditions. J Geochemical Explor

3353 2012;113:3–12.

3354 [225] Carr R, Zhang C, Moles N, Harder M. Identification and mapping of heavy metal

3355 pollution in soils of a sports ground in Galway City, Ireland, using a portable XRF analyser

3356 and GIS. Environ Geochem Health 2008;30:45–52.

3357 [226] Melquiades FL, Appoloni C. Application of XRF and field portable XRF for

3358 environmental analysis. J Radioanal Nucl Chem 2004;262:533–41.

188
3359 [227] Kalnicky DJ, Singhvi R. Field portable XRF analysis of environmental samples. J

3360 Hazard Mater 2001;83:93–122.

3361 [228] Kiss LT. X-Ray Fluorescence Determination of Brown Coal Inorganics. Anal Chem

3362 1966;38:1731–5. https://doi.org/10.1021/ac60244a024.

3363 [229] Zhang W, Noble A, Yang X, Honaker R. A Comprehensive Review of Rare Earth

3364 Elements Recovery from Coal-Related Materials. Minerals 2020;10.

3365 [230] Liu P, Yang L, Wang Q, Wan B, Ma Q, Chen H, et al. Speciation transformation of

3366 rare earth elements (REEs) during heating and implications for REE behaviors during coal

3367 combustion. Int J Coal Geol 2020;219. https://doi.org/10.1016/j.coal.2019.103371.

3368 [231] Hower JC, Qian D, Briot NJ, Henke KR, Hood MM, Taggart RK, et al. Rare earth

3369 element associations in the Kentucky State University stoker ash. Int J Coal Geol

3370 2018;189:75–82. https://doi.org/10.1016/j.coal.2018.02.022.

3371 [232] Stuckman MY, Lopano CL, Granite EJ. Distribution and speciation of rare earth

3372 elements in coal combustion by-products via synchrotron microscopy and spectroscopy. Int

3373 J Coal Geol 2018;195:125–38. https://doi.org/10.1016/j.coal.2018.06.001.

3374 [233] Taggart RK, Rivera NA, Levard C, Ambrosi JP, Borschneck D, Hower JC, et al.

3375 Differences in bulk and microscale yttrium speciation in coal combustion fly ash. Environ

3376 Sci Process Impacts 2018;20:1390–403. https://doi.org/10.1039/c8em00264a.

3377 [234] Zhang Z, Junming S, He Y, Liu G, Wang Y, Cao H. Distribution of some major and

3378 trace elements in high alumimum fly ash. GEOCHMICA 2006;35:660–6.

3379 https://doi.org/10.1017/CBO9781107415324.004.

3380 [235] Hower JC, Groppo JG. Rare Earth-bearing particles in fly ash carbons: Examples

3381 from the combustion of eastern Kentucky coals. Energy Geosci 2020.

3382 https://doi.org/10.1016/j.engeos.2020.09.003.

3383 [236] Hower JC, Cantando E, Eble CF, Copley GC. Characterization of stoker ash from the

3384 combustion of high-lanthanide coal at a Kentucky bourbon distillery. Int J Coal Geol

3385 2019;213:103260. https://doi.org/10.1016/j.coal.2019.103260.

3386 [237] Thompson RL, Bank T, Montross S, Roth E, Howard B, Verba C, et al. Analysis of

3387 rare earth elements in coal fly ash using laser ablation inductively coupled plasma mass

189
3388 spectrometry and scanning electron microscopy. Spectrochim Acta - Part B At Spectrosc

3389 2018;143:1–11. https://doi.org/10.1016/j.sab.2018.02.009.

3390 [238] Hower J, Groppo J, Joshi P, Dai S, Moecher D, Johnston M. Location of Cerium in

3391 Coal-Combustion Fly Ashes: Implications for Recovery of Lanthanides. Coal Combust

3392 Gasif Prod 2013;5:73–8. https://doi.org/10.4177/ccgp-d13-00007.1.

3393 [239] Wirtz T, De Castro O, Audinot J-N, Philipp P. Imaging and analytics on the helium

3394 ion microscope. Annu Rev Anal Chem 2019;12:523–43.

3395 [240] Valley JW, Reinhard DA, Cavosie AJ, Ushikubo T, Lawrence DF, Larson DJ, et al.

3396 Nano-and micro-geochronology in Hadean and Archean zircons by atom-probe tomography

3397 and SIMS: New tools for old minerals. Am Mineral 2015;100:1355–77.

3398 [241] Schmitt AK, Zack T. High-sensitivity U–Pb rutile dating by secondary ion mass

3399 spectrometry (SIMS) with an O2+ primary beam. Chem Geol 2012;332:65–73.

3400 [242] Horn I, Hinton RW, Jackson SE, Longerich HP. Ultra‐trace element analysis of

3401 NIST SRM 616 and 614 using laser ablation microprobe‐inductively coupled plasma‐mass

3402 spectrometry (LAM‐ICP‐MS): A comparison with secondary ion mass spectrometry

3403 (SIMS). Geostand Newsl 1997;21:191–203.

3404 [243] Wiese Jr RG, Muir IJ, Fyfe WS. Trace-element siting in iron sulfides in Ohio coals

3405 determined by secondary ion mass spectrometry (SIMS). Int J Coal Geol 1990;14:155–74.

3406 [244] Spears DA. The use of laser ablation inductively coupled plasma-mass spectrometry

3407 (LA ICP-MS) for the analysis of fly ash. Fuel 2004;83:1765–70.

3408 https://doi.org/10.1016/j.fuel.2004.02.018.

3409 [245] Bauer G, Achleitner B, Bonta M, Friedbacher G, Limbeck A. Analysis of single fly

3410 ash particles using laser ablation ICP-MS–an approach achieving lateral elemental

3411 distribution information via imaging. RSC Adv 2017;7:20510–9.

3412 [246] Kostova I, Vassileva C, Dai S, Hower JC. Mineralogy, geochemistry and mercury

3413 content characterization of fly ashes from the Maritza 3 and Varna thermoelectric power

3414 plants, Bulgaria. Fuel 2016;186:674–84. https://doi.org/10.1016/j.fuel.2016.09.015.

3415 [247] Roth EA, Thompson RL, Lin R, Bank TL, Granite E. Distributions and Associations

3416 of Rare Earth Elements in Fly Ash Using Laser Ablation Sector Field ICP-MS. AICHE Natl.

3417 Meet. San Fr., United States; 2016.

190
3418 [248] Hu P, Hou X, Zhang J, Li S, Wu H, Damø AJ, et al. Distribution and occurrence of

3419 lithium in high-alumina-coal fly ash. Int J Coal Geol 2018;189:27–34.

3420 https://doi.org/10.1016/j.coal.2018.02.011.

3421 [249] Limbeck A, Galler P, Bonta M, Bauer G, Nischkauer W, Vanhaecke F. Recent

3422 advances in quantitative LA-ICP-MS analysis: challenges and solutions in the life sciences

3423 and environmental chemistry. Anal Bioanal Chem 2015;407:6593–617.

3424 [250] Mardon SM, Hower JC. Impact of coal properties on coal combustion by-product

3425 quality : examples from a Kentucky power plant. Int J Coal Geol 2004;59:153–69.

3426 https://doi.org/10.1016/j.coal.2004.01.004.

3427 [251] Hower JC, Groppo JG, Hsu-kim H, Taggart RK. Distribution of rare earth elements

3428 in fly ash derived from the combustion of Illinois Basin coals. Fuel 2021;289:119990.

3429 https://doi.org/10.1016/j.fuel.2020.119990.

3430 [252] Shao P. Paragenetic Association and Synergistic Separation of Li-Ga-REE

3431 Multielements in High-Alumina Coal and Coal Ash: A Case Study of Datong Coalfield.

3432 China University of Mining and Technology Press, 2019.

3433 [253] Zhuang X, Querol X, Zeng R, Xu W, Alastuey A, Lopez-Soler A, et al. Mineralogy

3434 and geochemistry of coal from the Liupanshui mining district, Guizhou, south China. Int J

3435 Coal Geol 2000;45:21–37.

3436 [254] Dai S, Zou J, Jiang Y, Ward CR, Wang X, Li T, et al. Mineralogical and geochemical

3437 compositions of the Pennsylvanian coal in the Adaohai Mine, Daqingshan Coalfield, Inner

3438 Mongolia, China: Modes of occurrence and origin of diaspore, gorceixite, and ammonian

3439 illite. Int J Coal Geol 2012;94:250–70. https://doi.org/10.1016/j.coal.2011.06.010.

3440 [255] American Coal Ash Association. Coal combustion products production and use

3441 statistics, ACAA 2019-Survey-Results. Lexington, Kentucky: 2019.

3442 [256] The 69th annual edition of the BP Statistical Review of World Energy.

3443 https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy-econo

3444 mics/statistical-review/bp-stats-review-2020-full-report.pdf. vol. 69. 1 St James’s Square,

3445 London SW1Y 4PD: 2020.

191
3446 [257] Ding J, Ma S, Shen S, Xie Z, Zheng S, Zhang Y. Research and industrialization

3447 progress of recovering alumina from fly ash: A concise review. Waste Manag 2017;60:375–

3448 87. https://doi.org/10.1016/j.wasman.2016.06.009.

3449 [258] Hower J, Groppo J, Jewell RF, Wiseman JD, Duvallet T, Oberlink A, et al.

3450 Distribution of Rare earth elements in the pilot-scale processing of fly ashes derived from

3451 eastern Kentucky coals Comparisons of the feed and processed ashes. Fuel 2021;in

3452 press:xxx–xxx.

3453 [259] Hower JC, Dai S, Seredin V V, Zhao L, Kostova IJ, Silva LFO, et al. A Note on the

3454 Occurrence of Yttrium and Rare Earth Elements in Coal Combustion Products. Coal

3455 Combust Gasif Prod 2013;5:39–47. https://doi.org/10.4177/CCGP-D-13-00001.1.

3456 [260] Piispanen MH, Arvilommi SA, Broeck B Van den, Nuutinen LH, Tiainen MS,

3457 Perä mä ki PJ, et al. A comparative study of fly ash characterization by LA-ICP-MS and

3458 SEM-EDS. Energy & Fuels 2009;23:3451–6.

3459 [261] Chehreh Chelgani S, Hower JC. Estimating REY content of eastern Kentucky coal

3460 samples based on their associated ash elements. J Rare Earths 2018;36:1234–8.

3461 https://doi.org/10.1016/j.jre.2018.02.015.

3462 [262] Hower JC, Fu B, Dai S. Geochemical partitioning from pulverized coal to fly ash and

3463 bottom ash. Fuel 2020;279:118542. https://doi.org/10.1016/j.fuel.2020.118542.

3464 [263] Hower JC, Groppo JG, Graham UM, Ward CR, Kostova IJ, Maroto-Valer MM, et al.

3465 Coal-derived unburned carbons in fly ash: A review. Int J Coal Geol 2017;179:11–27.

3466 https://doi.org/10.1016/j.coal.2017.05.007.

3467 [264] Hower JC, Groppo JG, Joshi P, Preda D V, Gamliel DP, Mohler DT, et al.

3468 Distribution of Lanthanides, Yttrium, and Scandium in the Pilot-Scale Beneficiation of Fly

3469 Ashes Derived from Eastern Kentucky Coals. Minerals 2020;10:105.

3470 [265] Hower JC, Graham UM, Dozier A, Tseng MT, Khatri RA. Association of the sites of

3471 heavy metals with nanoscale carbon in a Kentucky electrostatic precipitator fly ash. Environ

3472 Sci Technol 2008;42:8471–7. https://doi.org/10.1021/es801193y.

3473 [266] Meij R. Trace element behavior in coal-fired power plants. Fuel Process Technol

3474 1994;39:199–217. https://doi.org/https://doi.org/10.1016/0378-3820(94)90180-5.

192
3475 [267] Smith RD, Campbell JA, Felix WD. Atmospheric trace element pollutants from coal

3476 combustion. Battelle Pacific Northwest Labs., Richland, WA (USA); 1979.

3477 [268] Clarke LB. The fate of trace elements during coal combustion and gasification: an

3478 overview. Fuel 1993;72:731–6.

3479 [269] Wang Z. Geochemistry of rare earth elements. Beijing: Science Press (Beijing);

3480 1989.

3481 [270] Yao DX, Zhi XC, Wang X. Geochemical feature and laws of concentrtion and

3482 dispersion of rare earth elements between coals and their fly asn bottom ashes. Geochemica

3483 2003; 32: 491-500.

3484 [271] Fu B, Liu G, Sun M, Hower JC, Mian MM, Wu D, et al. Emission and

3485 transformation behavior of minerals and hazardous trace elements (HTEs) during coal

3486 combustion in a circulating fluidized bed boiler. Environ Pollut 2018;242:1950–60.

3487 https://doi.org/10.1016/j.envpol.2018.07.066.

3488 [272] Mardon SM, Hower JC, O’Keefe JMK, Marks MN, Hedges DH. Coal combustion

3489 by-product quality at two stoker boilers: Coal source vs. fly ash collection system design. Int

3490 J Coal Geol 2008;75:248–54. https://doi.org/10.1016/j.coal.2008.07.004.

3491 [273] Ushakov S V, Helean KB, Navrotsky A. Thermochemistry of rare-earth

3492 orthophosphates. J Am Ceram Soc 2001;16:2623–33.

3493 [274] Cherniak DJ, Watson EB, Thomas JB. Diffusion of helium in zircon and apatite.

3494 Chem Geol 2009;268:155–66.

3495 [275] Farley KA. He diffusion systematics in minerals: Evidence from synthetic monazite

3496 and zircon structure phosphates. Geochim Cosmochim Acta 2007;71:4015–24.

3497 [276] Zeitler PK, Herczeg AL, McDougall I, Honda M. U-Th-He dating of apatite: A

3498 potential thermochronometer. Geochim Cosmochim Acta 1987;51:2865–8.

3499 [277] Quercia G, Perera Y, Tovar H, Rodríguez E. Thermal degradation of zirconium

3500 silicate (ZrSiO4) ferrules. Acta Microsc 2007;16:205–6.

3501 [278] Tõnsuaadu K, Gross KA, Pluduma L, Veiderma M. A review on the thermal stability

3502 of calcium apatites. J Therm Anal Calorim 2012;110:647–59.

3503 https://doi.org/10.1007/s10973-011-1877-y.

193
3504 [279] Senior C. Mercury behavior in coal combustion systems. John Wiley & Sons, Ltd.;

3505 2015.

3506 [280] Williams A, Pourkashanian M, Jones JM. Combustion of pulverised coal and

3507 biomass. Prog Energy Combust Sci 2001;27:587–610.

3508 https://doi.org/10.1016/S0360-1285(01)00004-1.

3509 [281] Xu M, Yan R, Zheng C, Qiao Y, Han J, Sheng C. Status of trace element emission in

3510 a coal combustion process: A review. Fuel Process Technol 2004;85:215–37.

3511 https://doi.org/10.1016/S0378-3820(03)00174-7.

3512 [282] Lyon RK, Freund H. Coal combustion process. US Pat 4,285,283 1981;4.

3513 https://doi.org/10.16533/j.cnki.15-1099/tf.2013.05.012.

3514 [283] Vassilev S V., Vassileva CG. A new approach for the classification of coal fly ashes

3515 based on their origin, composition, properties, and behaviour. Fuel 2007;86:1490–512.

3516 https://doi.org/10.1016/j.fuel.2006.11.020.

3517 [284] Hemmings RT, Berry EE. On the Glass in Coal Fly Ashes: Recent Advances. MRS

3518 Proc 1988;113.

3519 [285] Ward CR, French D. Determination of glass content and estimation of glass

3520 composition in fly ash using quantitative X-ray diffractometry. Fuel 2006;85:2268–77.

3521 https://doi.org/10.1016/j.fuel.2005.12.026.

3522 [286] Henry J, Towler MR, Stanton KT, Querol X, Moreno N. Characterisation of the

3523 glass fraction of a selection of European coal fly ashes. J Chem Technol Biotechnol

3524 2004;79:540–6. https://doi.org/10.1002/jctb.1023.

3525 [287] Vargas S, Frandsen FJ, Dam-Johansen K. Rheological properties of

3526 high-temperature melts of coal ashes and other silicates. Prog Energy Combust Sci

3527 2001;27:237–429. https://doi.org/10.1016/S0360-1285(00)00023-X.

3528 [288] Tian S, Kang Z, Chen L, Fang Y, Zhuo Y, Xu H. Characterization of

3529 Aluminosilicates in Fly Ashes with Different Melting Points Using 27Al Magic-Angel

3530 Spinning Nuclear Magnetic Resonance. Energy and Fuels 2017;31:10068–74.

3531 https://doi.org/10.1021/acs.energyfuels.7b01466.

194
3532 [289] Tian S, Zhuo Y, Chen C. Characterization of the products of the clay mineral thermal

3533 reactions during pulverization coal combustion in order to study the coal slagging

3534 propensity. Energy and Fuels 2011;25:4896–905. https://doi.org/10.1021/ef200502u.

3535 [290] Towler MR, Stanton KT, Mooney P, Hill RG, Moreno N, Querol X. Modelling of

3536 the glass phase in fly ashes using network connectivity theory. J Chem Technol Biotechnol

3537 2002;77:240–5. https://doi.org/10.1002/jctb.581.

3538 [291] San Nicolas RVR, Walkley B, van Deventer JSJ. Fly ash-based geopolymer

3539 chemistry and behavior. Coal Combust Prod Charact Util Benef 2017:185–214.

3540 https://doi.org/10.1016/B978-0-08-100945-1.00007-1.

3541 [292] Schwarz N, Cam H, Neithalath N. Influence of a fine glass powder on the durability

3542 characteristics of concrete and its comparison to fly ash. Cem Concr Compos 2008;30:486–

3543 96. https://doi.org/10.1016/j.cemconcomp.2008.02.001.

3544 [293] Shi C, Wu Y, Riefler C, Wang H. Characteristics and pozzolanic reactivity of glass

3545 powders. vol. 35. 2005. https://doi.org/10.1016/j.cemconres.2004.05.015.

3546 [294] Yan J, Neretnieks I. Is the glass phase dissolution rate always a limiting factor in the

3547 leaching processes of combustion residues? Sci Total Environ 1995;172:95–118.

3548 https://doi.org/10.1016/0048-9697(95)04727-1.

3549 [295] American Society of Testing Materials. C162-04,Standard Terminology of Glass

3550 and Glass Products. Philadelphia, United States: ASTM International; 2004.

3551 [296] Jiang Z-H, Zhang Q-Y. The formation of glass: a quantitative perspective. Sci China

3552 Mater 2015;58:378–425. https://doi.org/10.1007/s40843-015-0048-z.

3553 [297] Hower JC, Henke KR, Dai S, Ward CR, French D, Liu S, et al. Coal Combustion

3554 Products (CCPs): Characteristics, Utilization and Beneficiation. Chapter 2: Generation and

3555 nature of coal fly ash and bottom ash. Elsevier Ltd.; 2017.

3556 https://doi.org/10.1016/B978-0-08-100945-1.00002-2.

3557 [298] Saikia BK, Hower JC, Hood MM, Baruah R, Dekaboruah HP, Boruah R, et al.

3558 Petrological and biological studies on some fly and bottom ashes collected at different times

3559 from an Indian coal-based captive power plant. Fuel 2015;158:572–81.

3560 https://doi.org/10.1016/j.fuel.2015.06.007.

195
3561 [299] Hower JC. Petrographic examination of coal-combustion fly ash. Int J Coal Geol

3562 2012;92:90–7. https://doi.org/10.1016/j.coal.2011.12.012.

3563 [300] Suárez-Ruiz I, Valentim B, Borrego AG, Bouzinos A, Flores D, Kalaitzidis S, et al.

3564 Development of a petrographic classification of fly-ash components from coal combustion

3565 and co-combustion. (An ICCP Classification System, Fly-Ash Working Group –

3566 Commission III.). Int J Coal Geol 2017;183:188–203.

3567 https://doi.org/10.1016/j.coal.2017.06.004.

3568 [301] Hower JC, Suárez-Ruiz I, Mastalerz M. An approach toward a combined scheme for

3569 the petrographic classification of fly ash: Revision and clarification. Energy and Fuels

3570 2005;19:653–5. https://doi.org/10.1021/ef0498061.

3571 [302] Creelman RA, Ward CR, Schumacher G, Juniper L. Relation between coal mineral

3572 matter and deposit mineralogy in pulverized fuel furnaces. Energy and Fuels 2013;27:5714–

3573 24. https://doi.org/10.1021/ef400636q.

3574 [303] Zyrkowski M, Neto RC, Santos LF, Witkowski K. Characterization of fly-ash

3575 cenospheres from coal-fired power plant unit. Fuel 2016;174:49–53.

3576 https://doi.org/10.1016/j.fuel.2016.01.061.

3577 [304] Wall TF, Creelman RA, Gupta RP, Gupta SK, Coin C, Lowe A. Coal ash fusion

3578 temperatures - New characterization techniques, and implications for slagging and fouling.

3579 Prog Energy Combust Sci 1998;24:345–53.

3580 https://doi.org/10.1016/S0360-1285(98)00010-0.

3581 [305] Brindle JH, McCarthy MJ. Chemical constraints on fly ash glass compositions.

3582 Energy and Fuels 2006;20:2580–5. https://doi.org/10.1021/ef0603028.

3583 [306] Durdziński PT, Snellings R, Dunant CF, Haha M Ben, Scrivener KL. Fly ash as an

3584 assemblage of model Ca-Mg-Na-aluminosilicate glasses. Cem Concr Res 2015;78:263–72.

3585 https://doi.org/10.1016/j.cemconres.2015.08.005.

3586 [307] Srinivasachar S, Helble JJ, Boni AA. Mineral behavior during coal combustion1.

3587 Pyrite transformation. Prog Energy Combust Sci 1990;16:281–92.

3588 [308] Srinivasachar S, Helble JJ, Boni AA, Shah N. Mineral behavior during coal

3589 combustion 2. Illite transformations. Prog Energy Combust Sci 1990;16:293–302.

196
3590 [309] Helble JJ, Srinivasachar S, Boni AA. Factors influencing the transformation of

3591 minerals during pulverized coal combustion. Prog Energy Combust Sci 1990;16:267–79.

3592 https://doi.org/10.1016/0360-1285(90)90036-3.

3593 [310] Wigley F, Williamson J. Modelling fly ash generation for pulverised coal

3594 combustion. Prog Energy Combust Sci 1998;24:337–43.

3595 [311] Fu B, Liu G, Sun M, Hower JC, Hu G, Wu D. A comparative study on the

3596 mineralogy, chemical speciation, and combustion behavior of toxic elements of coal

3597 beneficiation products. Fuel 2018;228:297–308. https://doi.org/10.1016/j.fuel.2018.04.085.

3598 [312] Vassilev S V., Vassileva CG. Occurrence, abundance and origin of minerals in coals

3599 and coal ashes. Fuel Process Technol 1996;48:85–106.

3600 https://doi.org/10.1016/S0378-3820(96)01021-1.

3601 [313] Osborn EF, Muan A, Levin EM, Robbins CM, McMurdie HF. Phase equilibrium

3602 diagrams of oxide systems. The system CaO-Al2O3-SiO2. Am Ceram Socicty 1960;15:6–

3603 63.

3604 [314] Alekseev YI, Vereshchagin VI. Formation of crystalline phases in electroceramics of

3605 the system CaO− MgO− Al 2 O 3− SiO 2 (A review). Glas Ceram 1997;54:340–3.

3606 [315] Fudali RF. The Interpretation of Geological Phase Diagrams. J Geol 1974;82:529–

3607 30.

3608 [316] Huffman GP, Huggins FE, Dunmyre GR. Investigation of the high-temperature

3609 behaviour of coal ash in reducing and oxidizing atmospheres. Fuel 1981;60:585–97.

3610 [317] Abbott MF, Austin LG. Studies on slag deposit formation in pulverized-coal

3611 combustors: 6. sticking behaviour of slag drops from three Pennsylvania steam coals. Fuel

3612 1985;64:832–8.

3613 [318] Ranjbar N, Kuenzel C. Cenospheres: A review. Fuel 2017;207:1–12.

3614 https://doi.org/10.1016/j.fuel.2017.06.059.

3615 [319] Li Y, Gao X, Wu H. “Ash Cenosphere from Solid Fuels Combustion. Part 2:

3616 Significant Role of Ash Cenosphere Fragmentation in Ash and Particulate Matter Formation.

3617 Energy and Fuels 2013:822–9. https://doi.org/10.1021/ef302148f.

3618 [320] Kleinhans U, Wieland C, Frandsen FJ, Spliethoff H. Ash formation and deposition in

3619 coal and biomass fired combustion systems: Progress and challenges in the field of ash
197
3620 particle sticking and rebound behavior. Prog Energy Combust Sci 2018;68:65–168.

3621 https://doi.org/10.1016/j.pecs.2018.02.001.

3622 [321] Reinmöller M, Klinger M, Schreiner M, Gutte H. Relationship between ash fusion

3623 temperatures of ashes from hard coal, brown coal, and biomass and mineral phases under

3624 different atmospheres: A combined FactSageTM computational and network theoretical

3625 approach. Fuel 2015;151:118–23. https://doi.org/10.1016/j.fuel.2015.01.036.

3626 [322] Liu Y, Gupta R, Sharma A, Wall T, Butcher A, Miller G, et al. Mineral

3627 matter-organic matter association characterisation by QEMSCAN and applications in coal

3628 utilisation. Fuel 2005;84:1259–67. https://doi.org/10.1016/j.fuel.2004.07.015.

3629 [323] Brown GEJ. X-ray scattering and X-ray spectroscopy studies of silicate melts. Struct

3630 Dyn Prop Silic Melts 1995;32:317–410.

3631 [324] Zachariasen WH. The atomic arrangement in glass. J Am Chem Soc 1932;44.

3632 https://doi.org/10.1021/ja01349a006.

3633 [325] Fujita T, Konno K, Zhang W, Kumar V, Matsuura M, Inoue A, et al. Atomic-scale

3634 heterogeneity of a multicomponent bulk metallic glass with excellent glass forming ability.

3635 Phys Rev Lett 2009. https://doi.org/10.1103/PhysRevLett.103.075502.

3636 [326] Xi XK, Li LL, Zhang B, Wang WH, Wu Y. Correlation of atomic cluster symmetry

3637 and glass-forming ability of metallic glass. Phys Rev Lett 2007;99:95501.

3638 https://doi.org/10.1103/PhysRevLett.99.095501.

3639 [327] Keen DA, McGreevy RL. Structural modelling of glasses using reverse Monte Carlo

3640 simulation. Nature 1990;344:423–5. https://doi.org/10.1038/344423a0.

3641 [328] Hirata A, Guan P, Fujita T, Hirotsu Y, Inoue A, Yavari AR, et al. Direct observation

3642 of local atomic order in a metallic glass. Nat Mater 2011;10:28–33.

3643 https://doi.org/10.1038/nmat2897.

3644 [329] Miracle DB. A structural model for metallic glasses. Nat Mater 2004;3:697–702.

3645 https://doi.org/10.1038/nmat1219.

3646 [330] Bernal JD. Geometry of the Structure of Monatomic Liquids. Nature 1960;185:68–

3647 70. https://doi.org/10.1038/185068a0.

3648 [331] Gaskell PH. A new structural model for transition metal-metalloid glasses. Nature

3649 1978;3:697–702. https://doi.org/10.1038/276484a0.

198
3650 [332] Hower JC, Robl TL, Thomas GA, Hopps SD, Grider M. Chemistry of coal and coal

3651 combustion products from Kentucky power plants: Results from the 2007 sampling, with

3652 emphasis on selenium. Coal Combust Gasif Prod 2009;1:50–62.

3653 https://doi.org/10.4177/ccgp-d-09-00013.1.

3654 [333] Hower JC, Rimmer SM, Mastalerz M, Wagner NJ. Migmatite-like textures in

3655 anthracite : Further evidence for low-grade metamorphic melting and resolidification in

3656 high-rank coals. Geosci Front 2021;12. https://doi.org/10.1016/j.gsf.2020.12.004.

3657 [334] Kang SH. Fundamental Studies of Mineral Matter Transformation During

3658 Pulverized Coal Combustion: Residual Ash Formation. Ph. D. Thesis.Massachusetts

3659 Institute of Technology. Dept. of Chemical Engineering. 1991.

3660 [335] Lee Y. CCSEM analysis of minerals in pulverized coal and ash formation modelling.

3661 Ph. D. Thesis. Massachusetts Institute of Technology. Dept. of Chemical Engineering. 2000.

3662 [336] Padia A. The behavior of ash in pulverized coal under simulated combustion

3663 conditions.Ph. D. Thesis.Massachusetts Institute of Technology. Dept. of Chemical

3664 Engineering. 1976.

3665 [337] Quann RJ, Sarofim AF. A scanning electron microscopy study of the

3666 transformations of organically bound metals during lignite combustion. Fuel 1986;65:40–6.

3667 [338] Ramsden AR. Microscopic investigation into the formation of fly-ash during the

3668 combustion of a pulverized bituminous coal. Fuel 1969;48:121–37.

3669 [339] Vejahati F, Xu Z, Gupta R. Trace elements in coal: Associations with coal and

3670 minerals and their behavior during coal utilization - A review. Fuel 2010;89:904–11.

3671 https://doi.org/10.1016/j.fuel.2009.06.013.

3672 [340] Wang H, Harb JN. Modeling of ash deposition in large-scale combustion facilities

3673 burning pulverized coal. Prog Energy Combust Sci 1997;23:267–82.

3674 https://doi.org/10.1016/s0360-1285(97)00010-5.

3675 [341] Yan L, Gupta RP, Wall TF. The implication of mineral coalescence behaviour on ash

3676 formation and ash deposition during pulverised coal combustion. Fuel 2001;80:1333–40.

3677 https://doi.org/10.1016/S0016-2361(00)00194-0.

3678 [342] Brady JB, Cherniak DJ. Diffusion in minerals: an overview of published

3679 experimental diffusion data. Rev Mineral Geochemistry 2010;72:899–920.

199
3680 [343] Watson EB, Harrison TM. Zircon saturation revisited: temperature and composition

3681 effects in a variety of crustal magma types. Earth Planet Sci Lett 1983;64:295–304.

3682 [344] Jaworski A, Charpentier T, Stevensson B, Edén M. Scandium and Yttrium

3683 Environments in Aluminosilicate Glasses Unveiled by 45Sc/89Y NMR Spectroscopy and

3684 DFT Calculations: What Structural Factors Dictate the Chemical Shifts? J Phys Chem C

3685 2017;121:18815–29. https://doi.org/10.1021/acs.jpcc.7b05471.

3686 [345] Jaworski A, Stevensson B, Edén M. The Bearings from Rare-Earth (RE = La, Lu, Sc,

3687 Y) Cations on the Oxygen Environments in Aluminosilicate Glasses: A Study by

3688 Solid-State 17O NMR, Molecular Dynamics Simulations, and DFT Calculations. J Phys

3689 Chem C 2016;120:13181–98. https://doi.org/10.1021/acs.jpcc.6b02032.

3690 [346] Mao H, Selleby M, Fabrichnaya O. Thermodynamic reassessment of the Y

3691 2O3-Al2O3-SiO2 system and its subsystems. Calphad Comput Coupling Phase Diagrams

3692 Thermochem 2008;32:399–412. https://doi.org/10.1016/j.calphad.2008.03.003.

3693 [347] Wang X, Zhang R, Ren J, Vezin H, Fan S, Yu C, et al. Mechanism of cluster

3694 dissolution of Yb-doped high-silica lanthanum aluminosilicate glass: Investigation by

3695 spectroscopic and structural characterization. J Alloys Compd 2017;695:2339–46.

3696 https://doi.org/10.1016/j.jallcom.2016.11.104.

3697 [348] Stevensson B, Jaworski A, Edén M. The structural roles of Sc and Y in

3698 aluminosilicate glasses probed by molecular dynamics simulations. J Non Cryst Solids

3699 2017;460:36–46. https://doi.org/10.1016/j.jnoncrysol.2016.12.005.

3700 [349] Charpentier T, Ollier N, Li H. RE2O3-alkaline earth-aluminosilicate fiber glasses:

3701 Melt properties, crystallization, and the network structures. J Non Cryst Solids

3702 2018;492:115–25. https://doi.org/10.1016/j.jnoncrysol.2018.04.028.

3703 [350] Florian P, Sadiki N, Massiot D, Coutures JP. 27Al NMR study of the structure of

3704 lanthanum- And yttrium-based aluminosilicate glasses and melts. J Phys Chem B

3705 2007;111:9747–57. https://doi.org/10.1021/jp072061q.

3706 [351] Zheng Q, Smedskjaer MM, Youngman RE, Potuzak M, Mauro JC, Yue Y. Influence

3707 of aluminum speciation on the stability of aluminosilicate glasses against crystallization.

3708 Appl Phys Lett 2012;101. https://doi.org/10.1063/1.4739005.

200
3709 [352] Pahari B, Iftekhar S, Jaworski A, Okhotnikov K, Jansson K, Stevensson B, et al.

3710 Composition-property-structure correlations of scandium aluminosilicate glasses revealed

3711 by multinuclear 45 Sc, 27 Al, and 29 Si solid-state NMR. J Am Ceram Soc 2012;95:2545–

3712 53. https://doi.org/10.1111/j.1551-2916.2012.05288.x.

3713 [353] Wu W, Bian X, Wu Z, Sun S, Tu G. Reaction process of monazite and bastnaesite

3714 mixed rare earth minerals calcined by CaO-NaCl-CaCl2. Trans Nonferrous Met Soc China

3715 2007;17:864–8. https://doi.org/10.1016/S1003-6326(07)60189-7.

3716 [354] Park B, Li H, Corrales LR. Molecular dynamics simulation of La2O3-Na2O-SiO2

3717 glasses. I. The structural role of La3+ cations. J Non Cryst Solids 2002;297:220–38.

3718 https://doi.org/10.1016/S0022-3093(01)00935-8.

3719 [355] Pohlenz J, Rosa AD, Mathon O, Pascarelli S, Belin S, Landrot G, et al. Structural

3720 controls of CO2 on Y, La and Sr incorporation in sodium-rich silicate - carbonate melts by

3721 in-situ high P-T EXAFS. Chem Geol 2018;486:1–15.

3722 https://doi.org/10.1016/j.chemgeo.2017.12.023.

3723 [356] Iftekhar S, Pahari B, Okhotnikov K, Jaworski A, Stevensson B, Grins J, et al.

3724 Properties and structures of RE2O3-Al 2O3-SiO2 (RE = Y, Lu) glasses probed by molecular

3725 dynamics simulations and solid-state NMR: The roles of aluminum and rare-earth ions for

3726 dictating the microhardness. J Phys Chem C 2012;116:18394–406.

3727 https://doi.org/10.1021/jp302672b.

3728 [357] Pozdnyakova I, Sadiki N, Hennet L, Cristiglio V, Bytchkov A, Cuello GJ, et al.

3729 Structures of lanthanum and yttrium aluminosilicate glasses determined by X-ray and

3730 neutron diffraction. J Non Cryst Solids 2008;354:2038–44.

3731 https://doi.org/10.1016/j.jnoncrysol.2007.11.012.

3732 [358] Jaworski A, Stevensson B, Pahari B, Okhotnikov K, Eden M. Local structures and

3733 Al/Si ordering in lanthanum aluminosilicate glasses explored by advanced 27Al NMR

3734 experiments and molecular dynamics simulations. Phys Chem Chem Phys 2012;14:15866–

3735 78.

3736 [359] Okhotnikov K, Stevensson B, Edén M. New interatomic potential parameters for

3737 molecular dynamics simulations of rare-earth (RE= La, Y, Lu, Sc) aluminosilicate glass

201
3738 structures: exploration of RE 3+ field-strength effects. Phys Chem Chem Phys

3739 2013;15:15041–55.

3740 [360] Wang MT, Cheng JS, Li M, He F. Structure and properties of soda lime silicate glass

3741 doped with rare earth. Phys B Condens Matter 2011;406:187–91.

3742 https://doi.org/10.1016/j.physb.2010.10.040.

3743 [361] Wang M, Cheng J, Liu Q, Tian P, Li M. The effect of light rare earths on the

3744 chemical durability and weathering of Na2O-CaO-SiO2 glasses. J Nucl Mater

3745 2010;400:107–11. https://doi.org/10.1016/j.jnucmat.2010.02.018.

3746 [362] Bois L, Barré N, Guillopé S, Guittet MJ, Gautier-Soyer M, Duraud JP, et al.

3747 Dissolution of lanthanide alumino-silicate oxynitride glasses. J Nucl Mater 2000;277:57–66.

3748 https://doi.org/10.1016/S0022-3115(99)00137-3.

3749 [363] Zhang W, Groppo J, Honaker R. Ash beneficiation for REE recovery. 2015 World

3750 Coal Ash Conf. Nashville, TN, 2015.

3751 [364] King JF, Taggart RK, Smith RC, Hower JC, Hsu-Kim H. Aqueous acid and alkaline

3752 extraction of rare earth elements from coal combustion ash. Int J Coal Geol 2018;195:75–83.

3753 https://doi.org/10.1016/j.coal.2018.05.009.

3754 [365] Kashiwakura S, Kumagai Y, Kubo H, Wagatsuma K. Dissolution of Rare Earth

3755 Elements from Coal Fly Ash Particles in a Dilute H2SO4 Solvent. Open J Phys Chem

3756 2013;3:69–75. https://doi.org/10.4236/ojpc.2013.32009.

3757 [366] Pan J, Hassas BV, Rezaee M, Zhou C, Pisupati S. Recovery of rare earth elements

3758 from coal fly ash thorugh sequential chemical roasting, water leaching, and acid leaching

3759 processes. J Clean Prod 2020:124725.

3760 https://doi.org/https://doi.org/10.1016/j.jclepro.2020.124725.

3761 [367] Huang C, Wang Y, Huang B, Dong Y, Sun X. The recovery of rare earth elements

3762 from coal combustion products by ionic liquids. Miner Eng 2019;130:142–7.

3763 https://doi.org/10.1016/j.mineng.2018.10.002.

3764 [368] Nayaka GP, Zhang Y, Dong P, Wang D, Zhou Z, Duan J, et al. An environmental

3765 friendly attempt to recycle the spent Li-ion battery cathode through organic acid leaching. J

3766 Environ Chem Eng 2019;7:102854. https://doi.org/10.1016/j.jece.2018.102854.

202
3767 [369] Wen Z, Zhou C, Pan J, Cao S, Hu T, Ji W, et al. Recovery of rare-earth elements

3768 from coal fly ash via enhanced leaching. Int J Coal Prep Util 2020:1–15.

3769 https://doi.org/10.1080/19392699.2020.1790537.

3770 [370] Panias D, Asimidis P, Paspaliaris I. Solubility of boehmite in concentrated sodium

3771 hydroxide solutions: Model development and assessment. Hydrometallurgy 2001;59:15–29.

3772 https://doi.org/10.1016/S0304-386X(00)00146-8.

3773 [371] Jendoubi F, Mgaidi A, El Maaoui M. Kinetics of the Dissolution of Silica in

3774 Aqueous Sodium Hydroxide Solutions at High Pressure and Temperature. Can J Chem Eng

3775 1997;75:721–7. https://doi.org/10.1002/cjce.5450750409.

3776 [372] Kumari A, Panda R, Jha MK, Kumar JR, Lee JY. Process development to recover

3777 rare earth metals from monazite mineral: A review. Miner Eng 2015;79:102–15.

3778 https://doi.org/10.1016/j.mineng.2015.05.003.

3779 [373] Peterson R, Heinrichs M, Argumedo D, Taha R. Recovery of rare earth elements

3780 from coal and coal byproducts via a closed loop leaching process : Final Report. 2017.

3781 https://doi.org/10.2172/1377818.

3782 [374] Cao DZ, Selic E, Herbell JD. Utilization of fly ash from coal-fired power plants in

3783 China. J Zhejiang Univ Sci A 2008;9:681–7. https://doi.org/10.1631/jzus.A072163.

3784 [375] Querol X, Moreno N, Umaa JC, Alastuey A, Hernández E, López-Soler A, et al.

3785 Synthesis of zeolites from coal fly ash: an overview. Int J Coal Geol 2002;50:413–23.

3786 https://doi.org/10.1016/S0166-5162(02)00124-6.

3787 [376] Robinson P, Townsend AT, Yu Z, Münker C. Determination of scandium, yttrium

3788 and rare earth elements in rocks by high resolution inductively coupled plasma-mass

3789 spectrometry. Geostand Newsl 1999;23:31–46.

3790 https://doi.org/10.1111/j.1751-908X.1999.tb00557.x.

3791 [377] Li J, Zhang Y, Du M, Hao Q, Li J. Study on the intereferences and direct

3792 determination of Sc in metallurgical with inductively coupled plasma mass spectrometry.

3793 Spectrosc Spectr Anal 2017;37:1259–63.

3794
3795

203