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

com/science/article/pii/S0926669018306058
Manuscript_484c7f59094c149a02b06d73b1de8dac

1 Theoretical Estimation of Silo Design Parameters for Fractionated Loblolly

2 Pine Grinds – Moisture Content and Particle Size Effects

3
4
5
6
7
8
9 *Oluwatosin Oginnia and Oladiran Fasinab
10
a
11 School of Natural Resources, West Virginia University, Morgantown WV
b
12 Department of Biosystems Engineering, Auburn University, Auburn AL
13
14
15
16
17
18
19
20
21
22 Corresponding author: Oluwatosin Oginni
23 Email: ojoginni@mix.wvu.edu
24 332 Percival Hall, School of Natural Resources,
25 West Virginia University,
26 Morgantown WV 26506
27 Phone: 334-444-8227
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

© 2018 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
45 Abstract
46
47 Bulk material handling operations are key operations in process industries such as biorefineries,

48 chemicals and pharmaceuticals. To ensure a successful production operation, a consistent and

49 reliable flow of the materials from storage vessels without dust generation and flow obstruction

50 are required. The objective of this study was to investigate the effect of particle size and moisture

51 content on the theoretical estimation of silo design parameters for fractionated loblolly pine grinds.

52 Loblolly pine grinds were fractionated into sizes using screen apertures: 1.40, 1.00, 0.71, 0.50,

53 0.25 mm and pan. The fractionated loblolly pine grinds were adjusted to the moisture level of

54 4.78%, 8.68%, 16.53%, 22.21% and 25.53% (w.b). There was a reduction in the hopper half angle

55 with increase in moisture content and reduction in the fraction size. The hopper half angle also

56 decreased with increase in angle of wall friction. The adjusted hopper outlet sizes varied between

57 1.20 and 28.56 mm. The wall normal and vertical pressure acting on the cylindrical section of the

58 silo increased from 9.35 to 45.42 kPa and 15.34 to 48.91 kPa, respectively, with increase in fraction

59 size and decrease in moisture content. The initial fill and flow induced pressures acting on the

60 hopper section of the silo increased from 15.34 to 48.91 kPa and 24.71 to 78.79 kPa, respectively,

61 with increase in fraction size and decrease in moisture content. Results of this study will be helpful

62 in designing effective storage vessels and understanding the pressures exerted by stored biomass

63 in the storage vessel.

64

65 Keywords: Loblolly pine, critical applied stress, flow function, angle of wall friction

66

67
68
69
70

2
71 1. Introduction

72 Bulk solid handling, transport and storage are key operations in industries such as pharmaceutical,

73 chemical, food processing, and biorefinery. However, poor flow has been identified to be one of

74 the common problems encountered in these key operations. Such a problem contributes to

75 processing plant startup delays, process inefficiencies and equipment downtime (Bradley et al.,

76 2011; Prescott and Barnum, 2000). A consistent and reliable flow of bulk materials from storage

77 vessels is required to ensure a successful production operation.

78 Ratholing and arching are the two major flow problems associated with bulk solid handling. An

79 arch is a stable obstruction that forms within the hopper section (i.e., the converging portion of the

80 bin) usually near the bin outlet. The arch has enough strength to supports the rest of the bin’s

81 contents, hence, preventing discharge of the remaining content (Neikov et al., 2009; Holdich,

82 2002). A rathole is a stable pipe or vertical cavity that empties above the bin outlet. The silo content

83 in the stagnant zones remains stationary until an external force is applied to dislodge it (Neikov et

84 al., 2009; Johanson, 2002). Flow problems such as ratholing and arching can contribute to

85 structural failure and damaging of silos, process inefficiencies, and frequent equipment downtime

86 (Johanson, 2002; Merrow, 1988).

87 According to EISA (2007), over 1 billion dry tons of biomass can be produced annually from the

88 abundant biomass resources in the United States. These vast quantities of biomass can be converted

89 to liquid fuels, biochemicals and bioproducts. However, series of size reduction processes (e.g.

90 chipping, chopping and grinding) are needed to prepare biomass feedstocks for conversion

91 processes. This is because the size of biomass at harvest are several thousand-fold larger than the

92 particle sizes that are optimum for conversion processes. Before biomass grinds are fed into the

93 throat of the conversion plant, the grinds have to be stored in silos and holding containers that can

3
94 create a steady feed of this bulk material. Unless the silos and holding containers are properly

95 designed or sized, biomass grinds will, therefore, exhibit the problems (i.e. ratholing and arching)

96 that typically occur during flow of bulk solid materials out of silos and holding containers. These

97 flow problems will be compounded by the wide distribution in size of biomass grinds (Olatunde

98 et al. 2016) and by the ability of biomass to exchange moisture with the environment. However,

99 the extent of influence of variability in biomass properties on sizing and selection of storage and

100 holding containers have not been documented.

101 Improper design of silos and holding containers can result in biomass degradation, material loss,

102 flow blockage and/or safety problems (Fitzpatrick et al., 2004). Three main parameters that are

103 needed in silo design are the hopper opening size, the minimum hopper angle, and the pressures

104 on the wall of silo and hopper. The hopper opening size is sized to ensure that arching (the flow

105 problem related with mass flow pattern) does not occur and the required flow rate is achieved

106 (Teunou and Fitzpatrick, 1999). The minimum hopper half angle is designed to ensure mass flow,

107 the preferred flow pattern for a consistent and reliable flow. If the hopper half angle is less than

108 the minimum, then it is likely that a funnel flow pattern will exist (Fitzpatrick et al., 2007; Holdich,

109 2002). The structural integrity of the storage and holding containers is dependent on the amount

110 of pressure exerted on the wall of the containers by the stored material. Therefore, understanding

111 and quantifying the pressures are needed in the design, sizing and selection of materials of

112 construction for the storage and holding container (Rotter, 2009). Jenike’s mathematical

113 methodology is the engineering standard practice for estimating these design parameters. A

114 detailed description of the procedure is available in Fitzpatrick et al. (2007).

4
115 Therefore, the objective of this study was to theoretically estimate silo design parameters for

116 loblolly pine grinds and determine the influence of particle size and moisture content on the

117 estimated silo design parameters.

118 2. Material and methods

119 2.1 Sample preparation and flow properties measurement

120 Clean loblolly pine wood chips were received from West Fraser Inc. Sawmill, Opelika, AL. The

121 wood chips were ground using a hammer mill fitted with a 3.18 mm screen (Model 358, New

122 Holland grinder, New Holland, PA). The initial moisture content of the loblolly pine samples after

123 grinding was 8.69% (w.b.) as measured by moisture analyzer (model MB 45, Ohaus Corp., Pine

124 Brook, NJ) that was programmed according to the standard E1756-08 (ASTM, 2015). Ground

125 loblolly pine were theereafter adjusted to five moisture levels – 4.78%, 8.69%, 16.53%, 22.21%

126 and 25.53% (w.b.). To obtain the 4.78% moisture level, the samples were dried with a humidity

127 chamber (ESL-2CA, ESPEC, Hudsonville, MI) set at a temperature of 50 ⁰C and relative humidity

128 of 20%. Moisture contents above 8.69% were obtained by adding calculated quantity of water to

129 the sample. The moisture conditioned samples were stored in an air tight container for 24 h to

130 allow moisture equilibration to take place. The moisture contents of the conditioned samples were

131 thereafter verified with the moisture analyzer.

132 Each moisture adjusted sample was then fractionated into six fractions with a sieve shaker (Model

133 RX-29, WS Tyler, OH) that was fitted with these screens: #12 (1.40 mm aperture), #18 (1.0 mm

134 aperture), #25 (0.71 mm aperture), #35 (0.50 mm aperture), #60 (0.25 mm aperture) and pan.

135 Choice of sieve sizes was based on preliminary study on the particle size distribution of raw ground

136 samples. Detailed information on the characterization of the physico-chemical properties, particle

5
137 size, and flowability of the conditioned and fractionated samples are provided in a previous work

138 by the authors (Oginni et al., 2016).

139 2.2 Hopper design calculation

140 The flow properties of the fractionated loblolly pine grinds were quantified with a flow tester

141 (Brookfield Engineering Laboratories, Inc., Middleboro, MA) using the Jenike Shear Testing

142 method in ASTM Standard D6128 – 16 (ASTM, 2016). The software provided by the equipment

143 was used in calculating the flow properties; unconfined yield strength, major consolidating stress,

144 cohesion, and angle of internal friction. The angle of wall friction test was carried out using three

145 different vane lids namely; mild steel, stainless steel and Tivar 88. Result of the flow properties of

146 the fractionated loblolly pine grinds are discussed in the previous work of the authors (Oginni et

147 al., 2016).

148 Silo hopper design involves the bulk material properties, geometric design (to prevent arching,

149 ratholing and to ensure proper flow pattern) and structural design (distribution of pressure and

150 shear stresses on walls caused by stored materials). Conical hopper geometry that will ensure a

151 mass flow of loblolly pine grinds was designed using the works of Enstad (1981), Janssens (1895)

152 and Jenike (1964). In mass flow, the bulk solid is in motion at every point within the bin whenever

153 material is drawn from the outlet. There is flow of bulk solid along the walls of the cylinder (the

154 upper parallel section of the bin) and the hopper (the lower tapered section of the bin). Mass-flow

155 guarantees complete discharge of the bin contents at predictable flow rates (Roberts, 1994). Details

156 of the symbols used in the equations are presented in Nomenclatures.

157 2.3 Hopper half angle

158 Hopper half angle was determined using the equations 1 & 2 (Enstad, 1981):

π 1 1-sinδ
159 α= – cos -β 1
2 2 2sinδ

6
 -1 sinфw

ℎ 2
фw + sin
sinδ
160 β=
2

161 As a margin of safety in silo design, the hopper half angle is reduced by 3 - 5⁰ to allow for any

162 difference in the actual wall surface or slight variation in the bulk solid (Schulze, 2016).

163 2.4 Minimum hopper outlet size

164 Hopper outlet size was determined by evaluating for cohesive arching tendency. The procedure

165 involves calculation of flow function (FF) and flow factor (ff), which were estimated from

166 equations 3 through 10 (Enstad, 1981). The parameter used to describe the strength of a bulk solid

167 is the flow function. The FF represents the variation of the unconfined yield strength of a powder

168 as a function its major consolidating stress (Roberts, 1994).

Unconfined yield stress

170 Flow Function FF = 3
Major consolidating stress

Y 1+sinδ H(α)
171 Flow factor (ff) = 4
2 X-1 sinα

1 65 m 200 1-m
169 where = 5
H α 130+α 200+α

172 (m =1 for conical hoppers)

2m sinδ sin 2β+α

173 X= 1-sinδ sinα
+1 6

175 A = 2 1- cos β+α m

β+α 1-m
sinα 7

174 B = sinβ sin1+m β+α 8

176 C = 1-sinδ sin2+m β+α 9

A+B
177 Y= 10
C

178 The critical applied stress (σc) represents the critical value of the unconfined yield strength at which

179 a stable arch can be formed. This was obtained by the intersection of the flow function and flow

7
180 factor line as illustrated in Figure 1. The flow factor line is a straight line through the origin with

181 a slope equal to the inverse of ff (Fitzpatrick et al., 2004). The value of the critical applied stress

182 is used in calculating the minimum hopper outlet size (Dmin). However, if there is no intersection

183 of the flow function and flow factor line, the minimum hopper outlet size is estimated by

184 multiplying the geometric mean diameter by 10 (Purutyan et al., 2001). In practice, the actual

185 hopper outlet size is made larger than Dmin in order to achieve the desired flow rate (Roberts, 1994).

186 Therefore, in this study, the hopper outlet size was multiplied by 1.20.

187 2.5 Silo wall loads

188 The silo is divided into two sections: cylindrical and hopper sections (Figure 2). The vertical and

189 wall normal pressure exerted by the ground loblolly pine in 100 feet (30.48 m) high and 50 feet

190 (15.24 m) diameter silo was estimated using Janssen’s equations (Janssen, 1895). Janssen’s

191 equations for estimating the vertical and wall normal pressure in the cylindrical section are given

192 in equations 11 & 12.

ρgA μKC
193 Vertical pressure: Pv = 1- exp - h 11
A

ρgA μKC
194 Wall pressure: Pw = μC
1- exp - A
h 12

195 The vertical and wall normal pressures exerted by loblolly pine grinds during storage and flow on

196 the hopper section of a silo (Figure 3) were estimated using the equations 13 - 18 (Janssen, 1895).

197 The initial fill loads govern the structural design of the hopper in roughly its bottom 2/3rd while

198 flow induced loads govern the upper 1/3rd.

x n γhh x x n
199 Initial fill load: Pvf = Pvft hh
+ n-1 hh
- hh
13

200 n = 2aμh cot (α) 14

x n γhh x x n
201 Flow induced load: Pve = Pvft hh
+ n-1 hh
- hh
15

8
 sin ∅w
202 ε = ∅w + sin-1 sinδ
16

1+ sinδcosε
203 Fe = 17
1 - sinδcos(2α + ε)

204 n = 2(Fe μh cot α +Fe -1) 18

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

9
225 3. Results and discussion

226 3.1 Hopper half angle

227 The hopper half angle was designed for a mass flow pattern in a conical hopper and the values for

228 the angle of wall friction and internal friction (Oginni et al., 2016) were used in the estimation of

229 the hopper half angles following the Jenike’s methodology. The estimated hopper half angle varied

230 between 18.37° and 41.44° for all the moisture conditioned and fractionated loblolly pine grinds.

231 As a margin of safety during hopper design, the hopper half angle was reduced by 3° making the

232 angle varied between 15.37° and 38.44° (Table S1). A 70° hopper angle (equivalent of 35° for

233 hopper half angle) is often used as a rule of thumb for achieving mass flow (Fitzpatrick et al. 2004).

234 Prescott and Barnum (2000) reported a retrofitting of an existing press hopper in a pharmaceutical

235 company to have a minimum hopper half angle of 25°. The minimum hopper half angle gave a

236 sufficiently steep and smooth hopper which resulted in mass flow of the pharmaceutical powders

237 and elimination of ratholing. Hopper half angles obtained in this present study shows that a number

238 of the moisture conditioned and fractionated loblolly pine grinds tested met both criteria. Also, the

239 hopper half angles obtained for the unfractionated loblolly pine tested at 4.78% moisture content

240 met the criteria. This suggests that a mass flow of the loblolly pine grinds at 4.78% moisture

241 content can occur during discharge from a conical hopper.

242 The effects of the screen size and moisture content on the hopper half angle for the three wall

243 surfaces used in this study are shown in Figures 4a- 4c. There was an increase in the hopper half

244 angle with increase in the screen size for the three wall surfaces that were used. Meanwhile, the

245 hopper half angle decreased with an increase in the moisture content. However, for the Tivar 88

246 surface, there was no change in the hopper half angles with change in moisture content until the

247 moisture reached 22.21% before the hopper half angle reduced. This is attributable to the low

10
248 variation in angles of wall friction for Tivar 88 surface across all the moisture levels. Iqbal and

249 Fitzpatrick (2006) studied the effect of temperature on minimum hopper angle for three food

250 powders (flour, tea and whey permeate) and mentioned that all the powders had lower moisture

251 content at higher temperature. The study showed that there was an increase in the hopper angle as

252 the temperature increased except for whey permeate. This implies that the hopper half angle and

253 moisture content are inversely proportional. This deduction supports the results obtained in our

254 present study.

255 With a fundamental understanding that, to ensure mass flow of bulk solid during discharge from a

256 storage silo, a hopper half angle that will give sufficient steep and smooth hopper is required,

257 therefore it can be inferred that a low moisture content should be ensured during storage and

258 discharge from storage vessels. Although, loblolly pine like every other biological material have

259 the tendency to adsorb moisture during storage from the atmosphere, attention will need to be

260 given to the wall surface used in lining the storage vessel. The Tivar 88 wall surface showed a

261 consistent hopper half angle inspite of the varying moisture content (Figure 4c), hence it may serve

262 as a good wall surface material during the design of storage vessels for loblolly pine or other

263 biomass feedstocks.

264 Figure 5 illustrates the plot of angles of wall friction versus hopper half angles obtained for the

265 moisture conditioned and fractionated loblolly pine grinds. The hopper half angle decreased with

266 increase in angle of wall friction. The angles of wall friction recorded for stainless steel and mild

267 steel had a wider spread of hopper half angles compared to Tivar 88 surface which had a small

268 variation in its hopper half angle. The variation in the hopper half angles for the three surfaces is

269 attributable to the screen size and moisture effect on their angles of wall friction. The small

270 variation in Tivar 88 surface across all the moisture levels resulted in the small variation of its

11
271 hopper half angle. Jenike (1960) designed a chart showing the combination of the angle of wall

272 friction and the hopper half angle. The chart indicates allowable hopper half angles for mass flow

273 or funnel flow for given values of wall friction angles. The highest hopper half angle for the

274 fractionated loblolly pine was 38.44° and the highest angle of wall friction recorded was 24.8°.

275 Using the Jenike design chart, the combination of these two parameters showed that mass flow

276 pattern will exist during discharge of ground loblolly pine from a storage silo.

277 3.2 Critical applied stress

278 The intersection of the flow function and the flow factor lines are used in estimating the critical

279 applied stress. Generally, it is assumed that the flow function and flow factor lines will intersect

280 in order to obtain the critical applied stress. However, the flow function lines for the fractions lie

281 below the flow factor lines (Figure 6) resulting in no critical applied stress. This result implies that

282 the applied stress exceeds the unconfined yield strength of the fractions.

283 3.3 Hopper outlet size

284 The non- intersection of the flow function and flow factor lines (leading to design of hopper outlet

285 size to prevent mechanical arching) is in contrast to the results obtained from the flow properties

286 tests, which showed that fractionated loblolly pine grinds are cohesive in nature. However, due to

287 the small sizes obtained for the hopper outlet sizes, there is tendency for cohesive arching to occur

288 at the hopper outlet. Even though the hopper outlet sizes were calculated to prevent mechanical

289 arching, the selection of the hopper outlet sizes must be large enough to prevent formation of

290 cohesive arch as well.

291 The calculation of the minimum hopper outlet size was therefore not based on cohesive arching

292 but rather to prevent formation of mechanical arching. Purutyan et al. (2001) stated that in

293 estimating the minimum hopper outlet size designed to ensure mass flow and prevent mechanical

12
294 arching, the hopper outlet size should be at least six to eight times bigger than the largest particle

295 diameter of the material. The minimum hopper outlet sizes were estimated by multiplying the

296 geometric mean diameter by 10. The actual outlet dimensions were then multiplied by 1.20 (20%

297 greater) to ensure the desired flow rate (Table S1). As expected, the minimum hopper outlet size

298 followed the trend of the geometric mean diameters and it varied between 1.20 and 28.56 mm with

299 1.40 mm fraction having the highest minimum hopper outlet sizes.

300 3.4 Silo wall loads

301 Table S2 presents the estimated vertical and wall pressures exerted on the cylindrical section and

302 the initial fill and flow induced pressures exerted on hopper section of the designed silo with a

303 height of 30.48 m (100 feet) and diameter of 15.24 m (50 feet). The wall pressure acting on the

304 cylindrical section of the silo varied between 9.35 and 45.42 kPa (Figure 8) while the vertical

305 pressure ranged between 15.34 and 48.91 kPa (Figure 9). Also, the initial and flow induced

306 pressures ranged between 15.34 – 48.91 kPa and 24.71 – 78.79 kPa respectively (Figure 10 & 11).

307 The pressures acting on the cylindrical and hopper sections of the silo increased with increase in

308 fraction size and decrease in moisture content. These results imply that during storage of ground

309 loblolly pine in a storage bin, a large quantity of big particles at lower moisture content will exert

310 more pressure on the cylindrical section of the silo wall.

311 Figure 12 illustrates the vertical and wall pressure versus the cylindrical height (from top of the

312 silo to the bottom) for the pan fractions at the 4.78% moisture level. The vertical and the wall

313 pressures exerted on the cylindrical section of the silo by the material increased with height. These

314 pressures increased downward which indicated that there is a tendency for the silo to expand at the

315 lower part of the silo due to the pressures exerted on the silo wall.

13
316 Figure 13 shows the change in the initial fill and flow induced pressures as the hopper height

317 changes. The initial and flow induced pressures represent the pressures exerted on the hopper

318 section of the silo when the silo is filled with bulk solid and during discharge respectively. The

319 estimated initial fill and flow induced pressures for the fractionated loblolly pine grinds are

320 presented in Table S2. The initial fill loads varied between 15.34 and 48.91 kPa while the flow

321 induced pressures varied between 24.71 and 78.79 kPa. The initial fill and flow induced pressures

322 also increased with increase in fraction size and decrease in moisture content. The values obtained

323 for the initial fill and flow induced pressures indicates the estimated peak pressures that can be

324 attained when a silo of height 100 feet (30.48 m) and diameter 50 feet (15.24 m) is filled to

325 capacity.

326 The initial fill and flow induced pressures increased as the hopper height increased. However, the

327 initial fill pressure became constant at a certain hopper height. This shows that the pressure exerted

328 on the hopper wall as the silo is filled with material reaches a maximum and remains constant at a

329 certain hopper height.

330

331

332

333

334

335

336

337 4. Conclusions

14
338 The design of a silo bin to allow easy flow of material represents important aspect of bulk solid

339 handling. The result of this study showed that the hopper half angle (varied between 15.37⁰ and

340 38.44⁰) decreased with an increase in moisture content and decrease in fraction size. The minimum

341 hopper outlet sizes ranged from 1.20 to 28.56 mm. Vertical and wall normal pressures acting on

342 the cylindrical section increased with increase in fraction size and decrease in moisture content.

343 The initial fill and flow induced pressure also increased with increase in fraction size and decrease

344 in moisture content. As a biological material, biomass can adsorb or lose moisture when exposed

345 to the atmosphere. Also, the thermochemical conversion of the biomass requires its size reduction.

346 Results from this study suggest that careful consideration need to be given to the particle size and

347 moisture content of the biomass during the design, sizing and selection of storage, handling and

348 transport vessels for biomass feedstocks to avoid flow problems related to bulk solid handling.

349 Acknowledgments

350 We acknowledge funding from USDA-NIFA project: Southeast Partnership for Integrated

351 Biomass Supply Systems (IBSS) and Alabama Agriculture Experiment Station Hatch Funding

352

353

354

355

356

357

358

15
359 References

360 ASTM, 2015. Standard E1756-08 - Standard Test Method for Determination of Total Solids in

361 Biomass. ASTM International, West Conshohocken, PA.

362 ASTM, 2016. Standard D6128 – 16 - Standard Test Method for Shear Testing of Bulk Solids Using

363 the Jenike Shear Tester. ASTM International, West Conshohocken, PA.

364 Bradley, M.S.A., Berry, R.J., Farnish, R.J., 2011. Methods for Design of Hoppers. Silos, Bins and

365 Bunkers for Reliable Gravity Flow, for Pharmaceutical, Food, Mineral and Other Applications.

366 http://gala.gre.ac.uk/6974/1/WCA091230.pdf (accessed on 28 May 2018).

367 Carson, J.W., Jenkyn, R.T., 1993. Load Development and Structural Considerations in Silo

368 Design, Reliable Flow of Particulate Solids. https://www.inti.gob.ar/cirsoc/pdf/silos/load-dev-

369 silo-design.pdf (accessed 28 May 2018).

370 Enstad, G.G., 1981. A novel theory on the arching and doming in mass flow hoppers. Department

371 of Science and Technology, Christian Michelsen Institute, Bergen, Norway.

372 Energy Independence and Security Act (EISA) of 2007. Public Law 110-140-Dec.19, 2007.

373 https://www.gpo.gov/fdsys/pkg/PLAW-110publ140/pdf/PLAW-110publ140.pdf, (accessed 28

374 May 2018).

375 Fitzpatrick, J.J., Barringer, S.A., Iqbal, T., 2004. Flow property measurement of food powders and

376 sensitivity of Jenike’s hopper design methodology to the measured values. Journal of Food

377 Engineering 61, 399 – 405.

378 Fitzpatrick, J.J., Barry, K., Cerqueira, P.S.M., Iqbal, T., O’Neill, J., Roos, Y.H., 2007. Effect of

379 composition and storage conditions on the flowability of dairy powders. International Dairy

380 Journal. 17, 383 – 392.

16
381 Holdich, R.G., 2002. Fundamentals of Particle Technology. Midland Information Technology and

382 Publishing, Leicestershire, UK, pp 99 – 111.

383 Iqbal, T., Fitzpatrick, J.J., 2006. Effect of storage conditions on the wall friction characteristics of

384 three food powders. Journal of Food Engineering. 72, 273 – 280.

385 Janssen, H.A., 1895. Versuche uber Getreidedruck in Silozellen (Tests on Grain Pressure in Silos).

386 Verein Deutscher Ingenieure, Zeitschr 39, 1045 – 1049.

387 Jenike, A.W., 1964. Storage and Flow of Solids. University of Utah Engineering Experiment

388 Stattion, Bulletin No. 123, November 1964.

389 Johanson, J.R., 2002. Troubleshooting bins, hoppers and feeders. Chemical Engineering Progress,

390 24 – 36.

391 Oginni, O., Fasina, O., Adhikari, S., Fulton, J., 2016. Physical and Flow Properties of Fractionated

392 Loblolly Pine Grinds. Transactions of the ASABE 59 (5), 999 – 1008.

393 Olatunde, G., Fasina, O., Adhikari, S., McDonald, T.P., Duke, S.R., 2016. Size measurement

394 method for Loblolly Pine Grinds and Influence on Predictability of Fluidization. Canadian

395 Biosystems Engineering, 58, 4.1 – 4.10. http://dx.doi.org/10.7451/CBE.2016.58.4.1

396 Neikov, O.D., Lotsko, D.V., Gopienko, V.G., 2009. Powder characterization and testing, in:

397 Neikov, O.D., Naboychenko, S.S., Murashova, I.V., Gopienko, V.G., Frishberg, I.V., Lotsko, D.V.

398 (Eds.), Handbook of Non-Ferrous Metal Powders; Techonologies and Applications. Elsevier Ltd.,

399 Oxford, UK., pp 5 – 43.

400 Prescott, J.K., Barnum, R.A., 2000. On Powder Flowability. Pharmaceutical Technology 24, 60-

401 84.

402 Purutyan, H., Barnum, R., Schimmelpfennig, M., 2001. Fuel-handling consideration when

17
403 switching to PRB coal. Power, 145 (6), 53 – 64.

404 Roberts, A.W., 1994. Developments in silo design for the safe and efficient storage and handling

405 of grain, in: Highley, H., Wright, E.J., Banks, H.J., Champ, B.R. (Eds.), 6th International Working

406 Conferences on Stored-products Protection. CAB International, Wallingford, U.K, pp. 259 - 280.

407 Rotter, J.M., 2009. Silo and Hopper Design for Strength, in McGlinchey (Eds.), Bulk Solids

408 Handling: Equipment Selection and Operation. Blackwell Publishing Ltd., pp. 99-134.

409 Schulze, D., 2016. Storage and Discharge of Bulk Solids, in: Merkus, H.G., Meesters, G.M.H.

410 (Eds), Production, Handling and Characterization of Particulate Materials, Springer International

411 Publishing AG Switzerland, pp 424 – 478

412 Teunou, E., Fitzpatrick, J.J., 1999. Effect of relative humidity and temperature on food powder

413 flowability. Journal of Food Engineering 42, 109 – 116.

414

415

416

417

418

419

420

421

422

423

424

18
425 Figure Captions

426 Figure 1: Evaluation of critical applied stress (CAS) (Fitzpatrick et al., 2004)

427 Figure 2: Silo subdivision into vertical and hopper section

428 Figure 3: Initial fill and flow induced pressure on the hopper section of a silo (Carson and Jenkyn,

429 1993)

430 Figure 4a: Effect of screen size and moisture content on stainless steel hopper half angle

431 Figure 4b: Effect of screen size and moisture content on mild steel hopper half angle

432 Figure 4c: Effect of screen size and moisture content on Tivar 88 hopper half angle

433 Figure 5: Plot of hopper half angles and wall friction angle of fractionated loblolly pine grinds

434 Figure 6: Flow function and flow factor of fraction size 0.25 mm at 16.53% moisture content

435 Figure 7: Effect of screen size and moisture content on hopper opening size

436 Figure 8: Effect of moisture content and screen size on wall pressure acting on the silo cylindrical

437 section at a wall height of 100 feet (30.48 m)

438 Figure 9: Effect of moisture content and screen size on vertical pressure acting on the silo

439 cylindrical section at a wall height of 100 feet (30.48 m)

440 Figure 10: Effect of moisture content and screen size on initial load acting on the silo hopper

441 section at a wall height of 100 feet (30.48 m)

442 Figure 11: Effect of moisture content and screen size on flow induced pressure acting on the silo

443 hopper section at a wall height of 100 feet (30.48 m)

444 Figure 12: Plot of vertical and wall pressure against cylindrical height for the pan fraction at 4.78%

445 moisture level

446 Figure 13: Plot of initial and flow induced loads against hopper height for the pan fraction at 4.78%

447 moisture level

19
448 Nomenclatures

Symbols Nomenclature Unit

g Acceleration due gravity m/s2
δ Angle of internal friction ⁰
φw Angle of wall friction ⁰
ρb Bulk density kg/m3
γ Bulk unit weight of solid = ρg
β Characteristic of the material/wall relationship ⁰
σc Critical applied stress Pa
H(α) Emprical factor depend on shape of hopper and hopper angle -
ff Flow factor -
FF Flow function -
FI Flow index -
h Height of the cylindrical portion of the silo m
α Hopper half angle ⁰
µh Hopper wall friction coefficient -
Pve Mean vertical stress in the silo at the transition after emptying Pa
Pvf Mean vertical stress in the silo at the transition after filling Pa
pvft Mean vertical stress in the solid at the transition after filling
Dmin Minimum hopper outlet size m
Pne Normal pressure on hopper wall while emptying Pa
Pnf Normal pressure on hopper wall while filling Pa
K Pressure coefficient -
x Vertical coordinate from hopper apex
hh Vertical height between the hopper apex and the transition m
Pv Vertical pressure on the cylinderical portion of the silo Pa
µ Wall fricition coefficient for the hopper -
Pw Wall pressure on the cylinderical portion of the silo Pa
449

450

451

452

453

20
454 Supplemental Documents

455 Table S1: Evaluation of hopper half angle, flow factor and minimum hopper opening size
456 (Dmin) for a cylindrical hopper using Jenike’s method
457

Particle Reduced hopper half angle

Screen size Moisture Flow Adjusted
diameter Stainless Mild Tivar
(mm) content (%) factor Dmin (mm)
(mm) steel steel 88
4.78 0.11 32.19 32.15 29.27 1.06 1.32
8.69 0.15 28.82 27.45 28.09 1.04 1.80
Pan
16.53 0.10 26.34 28.55 29.36 1.05 1.20
22.21 0.10 15.37 21.42 28.25 1.09 1.20
4.78 0.85 35.72 34.94 31.17 1.09 10.20
8.69 0.81 32.36 31.51 31.07 1.08 9.72
0.25 16.53 0.70 30.27 29.93 31.28 1.07 8.40
22.21 0.64 23.66 25.28 30.84 1.07 7.68
25.53 0.86 17.39 20.06 28.67 1.09 10.32
4.78 1.16 36.82 35.52 33.07 1.13 13.92
8.69 0.97 34.29 33.26 32.30 1.12 11.64
0.50 16.53 1.01 30.83 29.29 31.97 1.09 12.12
22.21 0.84 26.01 25.33 32.17 1.08 10.08
25.53 1.33 21.23 23.74 31.36 1.09 15.96
4.78 1.38 37.37 35.92 33.95 1.17 16.56
8.69 1.28 32.78 33.46 33.28 1.14 15.36
0.71 16.53 1.33 30.42 29.10 33.08 1.11 15.96
22.21 1.17 27.20 26.36 32.79 1.09 14.04
25.53 1.46 22.08 23.59 33.12 1.11 17.52
4.78 1.72 37.39 35.71 34.72 1.18 20.64
8.69 1.54 34.62 33.38 34.27 1.16 18.48
1.00 16.53 1.69 30.29 29.55 34.12 1.11 20.28
22.21 1.55 26.46 25.69 31.87 1.09 18.60
25.53 1.73 23.18 24.61 33.43 1.10 20.76
4.78 2.16 38.44 36.12 35.97 1.20 25.92
8.69 2.14 35.58 33.60 34.75 1.17 25.68
1.40 16.53 2.17 30.81 30.76 33.97 1.13 26.04
22.21 2.18 27.47 25.70 32.34 1.09 26.16
25.53 2.38 24.93 25.49 34.37 1.12 28.56
4.78 1.31 36.87 35.61 34.11 1.15 15.72
8.69 1.36 34.63 33.13 30.80 1.12 16.32
Raw 16.53 1.33 27.89 29.93 32.76 1.10 15.96
22.21 1.39 25.44 24.36 29.81 1.09 16.68
25.53 2.19 22.59 24.01 32.19 1.10 26.28
458

459

21
460 Table S2: Silo wall loads of fractionated ground loblolly pine

Cylindrical section Hopper section

Screen size Moisture
(mm) content (%) Wall pressure Vertical Initial fill Flow induced
(kPa) pressure (kPa) load(kPa) pressure (kPa)
4.78 21.71 25.07 25.07 41.07
8.69 18.46 22.59 22.59 36.39
Pan
16.53 14.64 18.87 18.87 30.40
22.21 9.35 15.34 15.34 24.71
4.78 28.58 31.33 31.33 50.46
8.69 23.82 27.57 27.57 44.41
0.25 16.53 19.20 23.02 23.02 37.08
22.21 13.54 18.47 18.47 29.76
25.53 10.32 16.16 16.16 26.03
4.78 32.50 35.35 35.35 56.95
8.69 29.89 33.75 33.75 54.37
0.50 16.53 21.63 25.82 25.82 41.59
22.21 16.76 21.84 21.84 35.19
25.53 13.38 19.25 19.25 31.02
4.78 39.33 42.72 42.72 68.82
8.69 32.29 37.64 37.64 60.63
0.71 16.53 24.50 29.52 29.52 47.56
22.21 19.69 25.13 25.13 40.48
25.53 16.05 22.73 22.73 36.61
4.78 41.95 45.67 45.67 73.57
8.69 35.94 40.75 40.75 65.65
1.00 16.53 27.00 32.67 32.67 52.64
22.21 22.28 28.86 28.86 46.49
25.53 19.84 27.45 27.45 44.22
4.78 45.42 48.91 48.91 78.79
8.69 39.68 44.41 44.41 71.54
1.40 16.53 28.69 34.55 34.55 55.65
22.21 24.96 31.70 31.70 51.07
25.53 23.38 31.33 31.33 50.47
4.78 42.09 45.86 45.86 73.87
8.69 39.90 44.75 44.75 72.08
Raw 16.53 25.14 31.72 31.72 51.10
22.21 22.39 29.55 29.55 47.60
Raw 20.58 28.84 28.84 46.46
461

22